Hawafiuw *7: .t... . a 3...... .2. ‘f 5 1.... c1335.. 1:. J .1 .In . Vat... :3. '1. ~ I: if? "4:15:25 .. 3. » y. 3”,”; ”min "2 . ant»... «.- - 23"."? . 5 :u. hi... 9.4.. :32 a): 3 11071 . 4 £25....- » Liuelfl. . 3.19.. It)! :51, s , :I‘IHIA .2... :3...“ xi“ «fin?! . :52. x! x: .1}... .5. x x x! z 5. Z. 4’91!.>J Itvztis I N)! t .V $139,314.? .r. . | is. ,4“, , s 3.15.}. it......! 1. . $11. 8. .v....;Z..q: yAvtfiv , nutv.‘ x. 9.}... J. .. 1. reto’sp :0 itinvovl: Hm l « a $7. 1;: .vx! ‘1‘; t a; $an ‘ v.¢.__.3...¢ g THESIS : .1 Ill/ll“ /II///I.i/ Milli/’17 llll/llIl/ll " This is to certify that the dissertation entitled THE ECONOMIC EFFECTS OF ASIA-PACIFIC ECONOMIC COOPERATION (APEC) AND ASIA-BASED FREE TRADE AREA (AF-11): A COMPUTATIONAL GENERAL EQUILIBRIUM APPROACH presented by Inkyo Cheong has been accepted towards fulfillment ' of the requirements for Ph . D degree in Economics C/{Mfla {grégwfl Major professor DateZé oQgWV/v [9?F MSU is an Affirmative Action/Equal Opportunity Institution 0-12771 LIBRARY MIchIgan State Unlverslty PLACE ll RETURN BOX to roman this checkout from your record. TO AVOID FINES mum on or More dd. duo. DATE DUE DATE DUE DATE DUE JUN t 5 WW THE E( THE ECONOMIC EFFECTS OF ASIA-PACIFIC ECONOMIC COOPERATION (APEC) AND ASIA-BASED FREE TRADE AREA (AF -1 l) : A COMPUTATIONAL GENERAL EQUILIBRIUM APPROACH By Inkyo Cheong A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR of PHILOSOPHY Department of Economics 1995 DEDICATED TO MY PARENTS, BROTHERS, SISTER, \NIFE, MISUK, AND TWO SONS, ARION AND JUNO THE Tl several in' and the A: to strength EV: area \muld adjustment Thus, I intc free trade a Equilibriu” cOmbinario . CGE mode Willi lncrea countries in WSSlbIe Cat The Seem ‘0 be i ABSTRACT THE ECONOMIC EFFECTS OF ASIA-PACIFIC ECONOMIC COOPERATION (APEC) AND ASIA-BASED FREE TRADE AREA (AF-1 1) : A COMPUTATIONAL GENERAL EQUILIBRIUM APPROACH By Inkyo Cheong The possibilty of a free trade area in the Asia-Pacific region has been discussed at several international conferences, such as the Pacific Economic COOperation Conference and the Asia Pacific Economic Conference. The trends of the world economy are likely to strengthen economic cooperation in the area. Even though many authors suggest that trade liberalization in the Asia-Pacific area would accelerate development, there is very little empirical evidence about the adjustment process which would follow the formation of such a free trade area (PTA). Thus, I intend to study the possibilities of Asia-Pacific free trade (APEC) and Asia-based free trade area (AF-11 FTA). I will perform simulations with a computational general equilibrium (CGE) model, in order to look at welfare changes under different combinations of member countries with APEC and AF-l I. Since perfectly-competitive CGE models tend to underestimate the welfare effects of a FTA, I build a CGE model with increasing returns to scale and firm-level product differentiation. If a group of countries in the Pacific Rim can improve welfare by forming a FTA, this group will be a possible candidate for a new FTA in the Pacific Rim. The main results can be summarized as follows: (1) The groupings of regions seem to be important for a formation of a free-trade area in the Pacific-Rim region. From our simul Australia the ASE.J the model competiti‘ llre result our simulations, the highest probable regional cooperation scenario will be a FTA of Australia/New Zealand, China/Hong Kong, the Asian newly-industrialized countries, and the ASEAN nations except Thailand. (2) The introduction of imperfect competition into the model projects large discrepancies between the simulations from the perfectly- competitive CGE model and the model with an imperfectly-competitive component. (3) The results of the simulations are very robust with respect to the choices of parameters. Copyright by Inkyo Cheong 1995 chairpers at Michig and sugge working \ l a and Steve departmer Kwon, for administrn hailed kin An BUSiness (_ gram. My im'Etluable ; S”Pports. u dlssmation ACKNOWLEDGMENTS I wish to express my sincere gratitude to Charles Ballard, my dissertation chairperson, for his dedicated guidance and kindness throughout my study of economics at Michigan State University. I am especially grateful for his support in several ways, and suggestions of many ideas developed in this dissertation. I have really enjoyed working with him. I also want to thank the other dissertation committee members, Lawrence Martin and Steven Matusz for their thoughtful comments on my dissertation. Friends in the department deserve my special recognition, especially HyungSeung Lee and YoungMin Kwon, for their friendship and discussions on economic issues. I appreciate the administrative staff in the department of economics, especially Ann Feldman, who has helped kindly and generously all students in the department. Another special thanks go to Dr. Tamer Cavusgil and the staff of the International Business Center, MSU, for their generous financial support with a Ph.D. Dissertation grant. My best thanks must go to my parents, parents in law, brothers and sister, for their invaluable supports to me. I wish to express my another greatest thanks to my wife, Misuk, and my two beloved sons, Arion and Juno, for their understanding and invaluable supports. Without their encouragement and patience, I could nOt have finished my dissertation. vi TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES CHAPTER I - INTRODUCTION CHAPTER II - THE DESCRIPTION OF THE MODEL Computational General equilibrium (CGE) Modeling The Overall Description of the Model Consumer Preferences Production Sector Total Perceived Demand Elasticities Market-Clearing Conditions Price Linkages CHAPTER III - DATA, PARAMETERS AND SIMULATIONS The Modification of the GTAP Data and Parameters The Simulation Method CHAPTER IV - THE INTERPRETATIONS OF RESULTS The Simulation of GTAP Model Imperfectly-Competitive Model CHAPTER V - CONCLUSION APPENDIX LIST OF REFERENCES vii viii xi 13 SO 59 77 119 134 Table I Table 2 Table 3 Table 4 Table 5 Table 6 Table 7 Table 8 Table 9 Table 10 Table 1] Table 12 Table 13 Table 14 Table 15 Table 1 Table 2 Table 3 Table 4 Table 5 Table 6 Table 7 Table 8 Table 9 Table 10 Table 11 Table 12 Table 13 Table 14 Table 15 LIST OF TABLES Members of APEC and Aggregation Mappings of Regions Lists of Industries/Commodities and Mappings in our Study The Matrix of Regional Exports Parameters for the Elasticity of Substitution Welfare Changes [GTAP Model, GTAP Parameter] Equivalent Variation [GTAP Model, GTAP Parameter] Changes of Income [GTAP Model, GTAP Parameter] Changes of Prices [GTAP Model, GTAP Parameter] Changes of Welfare [IMC Model, GTAP Parameter, Cournot, 100] Changes of Welfare [IMC Model, GTAP Parameter, Bertrand, 100] Changes of Income [IMC Model, GTAP Parameter, Cournot, 100] Changes of Income [IMC Model, GTAP Parameter, Cournot, 100] Changes of Prices [IMC Model, GTAP Parameter, Cournot, 100] Changes of Prices [IMC Model, GTAP Parameter, Cournot, 100] Changes of Numbers of RPR Firms [IMC Model, GTAP Parameter, Cournot, 100] viii 80 81 82 84 85 86 87 88 89 90 91 92 93 94 95 Table 1 Table 1 Table 1‘. Table 1‘. Table 20 Table 31 Table 22 Table 23 Table 24 Table 25 Table 26 Table 27 Table 28 Table 29 Table 30 Table 16 Table 17 Table 18 Table 19 Table 20 Table 21 Table 22 Table 23 Table 24 Table 25 Table 26 Table 27 Table 28 Table 29 Table 30 Changes of Numbers of TME Firms [IMC Model, GTAP Parameter, Cournot, 100] Changes of Numbers of RPR Firms [IMC Model, GTAP Parameter, Bertrand, 100] Changes of Numbers of TME Firms [IMC Model, GTAP Parameter, Bertrand, 100] Changes of Welfare [IMC Model, 59% Parameter, Cournot, 100] Changes of Welfare [IMC Model, 59% Parameter, Bertrand, 100] Changes of Welfare [IMC Model, 99% Parameter, Cournot, 100] Changes of Welfare [IMC Model, 99% Parameter, Bertrand, 100] Changes of Welfare [IMC Model, GTAP Parameter, Cournot, 25] Changes of Welfare [IMC Model, GTAP Parameter, Bertrand, 25] Changes of Welfare [IMC Model, GTAP Parameter, Cournot, 50] Changes of Welfare [IMC Model, GTAP Parameter, Bertrand, 50] Changes of Welfare [IMC Model, GTAP Parameter, Cournot, 200] Changes of Welfare [IMC Model, GTAP Parameter, Bertrand, 200] Changes of Output for SVC [GTAP Model, GTAP Parameter] Changes of Output for AGR [GTAP Model, GTAP Parameter] ix 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 Table 3 Table 3 Table 3; Table 3-‘ Table 35 Table 36 Table 37 Table 38 Table 31 Table 32 Table 33 Table 34 Table 35 Table 36 Table 37 Table 38 Changes of Output for LMN [GTAP Model, GTAP Parameter] Changes of Output for RPR [GTAP Model, GTAP Parameter] Changes of Output for TME [GTAP Model, GTAP Parameter] Changes of Output for SVC [IMC Model, GTAP Parameter, Bertrand, 100] Changes of Output for AGR [IMC Model, GTAP Parameter, Bertrand, 100] Changes of Output for LMN [IMC Model, GTAP Parameter, Bertrand, 100] Changes of Output for RPR [IMC Model, GTAP Parameter, Bertrand, 100] Changes of Output for TME [IMC Model, GTAP Parameter, Bertrand, 100] 111 112 113 114 115 116 117 118 Figure 1 Figure 3 Figure 3 Figure 4 Figure 1 Figure 2 Figure 3 Figure 4 LIST OF FIGURES "Armington" Assumption and Firm-Level Product Differentiation 25 The Structure of Household Demand 28 The Production Side for the Imperfectly-Competitive Sectors 34 Illustration of Euler Method 56 xi attracted economy completit the estabi issues raj; Economie May... 199 Al COUnlrIES. and SUStair Table 1. I Canbem . meeting! at SingapOre gTOLtp To a 1 CHAPTER I INTRODUCTION In recent years, economic integration and cooperation in the Pacific Rim have attracted widespread attention. Drobnick (1992) says that the trends of the world economy are likely to strengthen economic cooperation in the Asia-Pacific region. The completion of a single market in Europe and the open door policy of China may hasten the establishment of greater Pacific-rim economic integration and cooperation. The issues raised have been explored at a number of conferences. These include the Pacific Economic Cooperation Conference (26th International General Meeting, Seoul, Korea, May, 1993), and the Asia-Pacific Economic Cooperation Conference (APEC). APEC was established in 1989 as an informal grouping of 12 Asia-Pacific countries, to better manage the effects of growing interdependence in the Pacific region and sustain economic growth. Currently, APEC has 18 member countries, as shown in Table 1. Foreign and economic ministers met for the first time to discuss APEC, in Canberra, Australia, in November, 1989. Remarkable progress was made at the third meeting, at Seoul, Korea, in 1991, and a permanent secretariat was established, at Singapore in September, 1992. After that, APEC has grown from an informal discussion group to a formalized organization, providing an institution for discussion on a broad range 01 Seattle. APEC n It was in an econt addition ‘t’ision s 2lSt eenu liberaliza educatior 15. 1994. BOgor me leaders at:- leaders' dt intesrmen for develo an imPOITa The 2 range of economic issues. The United States chaired the 5th APEC ministerial meeting in Seattle, Washington, November 17-19, 1993, and President Clinton hosted a historic APEC national leaders’ meeting at Blake Island, near Seattle, on November 19-20, 1993. It was important, in the sense that APEC’s ministerial meeting was upgraded to include an economic and political leaders’ meeting for Asia-Pacific economic cooperation, in addition to the ministerial meeting. The attendees at the Blake Island meeting issued a ‘vision statement’ on their “common goals for the Asia-Pacific region leading up to the 21St century: expand their economic dialogue; advance global and regional trade liberalization; deepen business sector participation in APEC ; establish cooperation in ”1 education and on development of small and medium size enterprises, ..... The 2nd economic leaders’ meeting was held in Bogor, Indonesia, on November 15, 1994, following the APEC ministerial meeting, at Jakarta on November 11-12. The Bogor meeting produced a blueprint for APEC’s trade liberalization agenda. APEC leaders agreed to remove trade and investment barriers in the next quarter century. The leaders’ declaration highlights the fact that APEC nations support free and open trade and investment in the Asia-Pacific region by 2010 for industrialized economies, and by 2020 for developing economies. Even though the accord is not a legal commitment, it can be an important milestone for the region, as it pursues the free-trade goal. The literature contains a great deal of discussion of economic integration in the Pacific region. For example, English (1989) predicted that the possible form of economic ' Quoted from “Focus on Asia-Pacific Economic Cooperation: APEC Economic Leader’ Meeting lnitiatives” October 28, 1994, Bureau of Public Affairs, US. Department of State. coopefi five me the Uni associat Islands' of Asia- and the r T trade “it ($228 bil Pacific re the U.S.. APEC. Tl Free Trad tolOin Na pOSlllOn. a Ham'SIICI percent an ‘ Canada F] Padfic det 3 cooperation in the Pacific basin “would be a Pacific Free Trade Association involving the five most developed countries of the region (Australia, Canada, Japan, New Zealand and the United States) accompanied by varieties of unilateral or collective agreement of association with Korea, ASEAN, China, and perhaps Taiwan and the South Pacific Islands.” K00 (1990) writes that “Korea has been and will continue to be very supportive of Asia-Pacific cooperation,” explaining why APEC brings potential benefits to Korea and the regional economy. The Pacific-rim region is an important trade partner for the United States. US. trade with the Pacific region ($344 billion) was greater than that with Western Europe ($228 billion) in 1992. Three million jobs were created from US. exports to the Asia- Pacific region during the period of 1989 to 1993.2 For the sustained creation of jobs in the US, it is suggested that the US. should hurry to open the fast-growing economies of APEC. The establishment of the European Community (BC) and the North American Free Trade Area (NAFTA) will inevitably direct some trade inward. Mexico’s decision to join NAFTA will push Asian developing countries into an unfavorable competitive position, and these countries will reduce their market share in North America. Cox and Harris (1992) showed that the rest of world was expected to lose market share by 3 percent and 1.26 percent in the US. and Canada, respectively, as a result of the US. - Canada FTA (CAF TA) agreement. There has been an increase in the market shares of Pacific developing countries in North America in the 1970’s and 1980’s. Thus, Pacific 2 Refer to Business Week, “It’s Time To Open All Asia’s Markets,” November 14, 1994. Rim Ct accelei NAPL- is a dev countric improve success I Japan's r the positi the memt policies h thOUgh Ja P€netratin (NTBs),3 NTBS We] intangible CXpeqs la and Taivv a galHS. and plmuted bv . OCUIEmem 4 Rim countries are likely to suffer from trade diversion as a result of NAFTA. This may accelerate the desire of Pacific-rim countries to form a free trade area (FTA). Under NAFTA, these market losses are expected to be bigger than under CAF TA, since Mexico is a developing country, taking the markets of North America, to which Asian developing countries have exported labor-intensive products before NAFTA. Even though all countries in the region have common goals, such as welfare improvements and the creation of jobs by forming a new free trade area, the possibility of success for a FTA can be questioned. There exists a disagreement about the effects of Japan’s participation in the Pacific-rim FTA. Lee and Roland-Holst (1994) emphasize the positive role of Japan, since Japan’s purchasing power would increase employment in the member countries. Moreover, since the Korean War, Japan’s industry and trade policies have been manipulated through a mixture of tariff and non-tariff barriers. Even though Japanese tariffs can be explicitly reduced, American firms face problems in penetrating the Japan market, due to Japan’s subtle application of non-tariff barriers (NTBs),3 which avoid violating GATT regulations. Even though reductions of some NTBs were negotiated in the Tokyo Round of multilateral trade agreements, Japan’s intangible trade barriers have provoked the most foreign complaints. Choo (1992) expects Japan to increase its trade surpluses as a result of the market opening of Korea and Taiwan, while his calculations indicate that the United States would not collect big gains, and U.S.-Japan trade deficits would not be reduced. 3 For Japan’s NTBs to trade in manufactures, see Christelow (1990). Among the NTBs that have been practiced by the Japanese are product standards, testing procedures, distribution systems, and government procurement procedures. charact agree It (LSITC there is anti-Arr with KO such an : ASEAN hand. the and thus. some Olht compam] Korea Si, Pe labOf-interl PTOducts, prOdUCIs li bi’l‘I’t‘en A 4\ jouoted ITOr ASEAN “a Singapore. an WWW. It table 2. 5 Asia-Pacific countries have highly diverse regional, cultural, and political characteristics. Therefore, it is expected to be difficult that all Asia-Pacific countries agree to form a single free trade area. The United States International Trade Commission (U SITC, 1989) reports that the US. has no formal diplomatic relations with Taiwan, and there is no official mechanism for the US. to enter into negotiations with Taiwan, while anti-American atmosphere may be a serious problem, in U.S.’s negotiating F TA issues with Korea. USITC writes that“. . .the majority of US. and foreign . . . did not think such an agreement (U .S.-ASEAN PTA) was workable,” and “ . . . an (U .S. -) F TA ASEAN is not feasible because it would be too difficult to administer.”4 On the other hand, the Association of South-East Asian Nations (ASEAN)5 was already established, and thus, ASEAN can be a basis for the formation of a new free trade area, accepting some other countries in the region, and adding them to ASEAN. ASEAN may be compared with the Asian Newly Industrializing Economies (NIEs, which are Hong Kong, Korea, Singapore, and Taiwan) in several points. Pearson (1994) reports that the ASEAN countries have recently started to produce labor-intensive manufactured products and natural-resource-based manufactured products. The Asian NIEs have exported more capital-intensive and technology-intensive products than ASEAN, with the exception of Singapore.6 China was found to settle in between ASEAN and the NIEs. In addition, ASEAN is endowed with rich natural ’ Quoted from USITC (1989), pp. 3-3 and 3-7, respectively. 5 ASEAN was formed in 1967. The original members were Indonesia, Malaysia, the Philippines, Singapore, and Thailand. Brunei joined ASEAN in 1984. ASEAN established the ASEAN Free Trade Area in January, 1993, and they decided to eliminate trade barriers in 10 years in September, 1994. 6 See table 2.3 in p. 43, Pearson (1994). TCSOUTC accum four N1} states th can be a on the m intraregi I complem CCOIIOITIlt to the NH goods to [Echnolog adt'anced indUStrial New Zeal nations, '1 nations in PTCdICI [ha With their EM would acce 6 resources, while the NIEs have relatively highly-educated labor forces, and greater capital accumulation. Yang (1994) suggests the regional cooperation of the nine countries in East Asia: four NIEs, four ASEAN (Indonesia, Malaysia, Philippines, Thailand), and China. He states that “greater cooperation among the industrializing countries/areas of the region can be an addition to, rather than a substitute for, ASEAN.” Yang’s reasoning is based on the mutual complementarity among nine countries, as well as the high growth of intraregional trade among these countries , “as a condition of success.” Another complementarity can be found between Australia and New Zealand and the NIE economies. Australia and New Zealand have been major suppliers of intermediate goods to the NIEs (for example, ore, wool, and coal). The NIEs have exported manufactured goods to Australia and New Zealand, and Australia and New Zealand have exported high— technology products to Asian nations. Australia and New Zealand are economically- advanced countries with high-technology industry and high per capita income. Their industrial structure does not seem to compete with ASEAN nations, since Australia and New Zealand produce highly-capital-intensive manufactured goods, compared to ASEAN nations. Thus, one F TA in the Pacific-rim region would be the Asia-based F TA of 11 nations including Australia and New Zealand (AF -1 1). Kreinin and Plummer (1994) predict that Australia and New Zealand should endeavor to form closer economic links with their Asia-Pacific neighbors, as a result of the formation of NAFTA. Even though many authors suggest that trade liberalization in Asia-Pacific area would accelerate development, there is very little empirical evidence about the adjustm stud)" th simulati welfare 11. If a this grot effects 0 competit to lower the possi' t1’lifltgula: allows fa, Primary 1. is, the ma Prices in ( COHStam t the Chang; that anal}? (Dauphin I 7 adjustment process which would follow the formation of such a F TA. Thus, I intend to study the possibilities of Asia-Pacific free trade and AF -1 l FTA. I will perform simulations with a computational general equilibrium (CGE) model, in order to look at welfare changes under different combinations of member countries with APEC and AF- 11. If a group of countries in the Pacific Rim can improve welfare by forming a FTA, this group will be a possible candidate for a new FTA in the Pacific Rim. The general equilibrium framework is most appropriate for analyzing the welfare effects of the formation of a free trade area. Firstly, a new FTA will imply more competition between industries for demand. More competitiveness may induce producers to lower the prices of their products, and general equilibrium models allow us to measure the possible welfare change, while providing more accurate welfare evaluations than the triangular calculations of partial equilibrium. Secondly, the general equilibrium approach allows factor prices to vary and thus, relative price changes in intermediate inputs and primary inputs will presumably affect the firrn’s ratio of average to variable costs. That is, the material components of variable costs will be optimized, based on new factor prices in each equilibrium. On the other hand, partial equilibrium analyses assume constant factor prices. However, it is generally believed that prices will be changed with the changes of economic environment. To illustrate this point, we may note two papers that analyze the effects of the FTA of Canada and US. using partial equilibrium methods (Dauphin (1978), and Magun et al.,(1987)). The approach these papers employed is as follows : Firstly, they simulate the macroeconomic impacts of the unilateral and bilateral removal of tariffs and some NTBs, and then determine the amounts of factor price changes exogenr interest. models 1 models ; sectors. equilibri address I estimatet attention eCOttomit Source 0t" imperfecr that com: been a gm mdUStriaI Ha trade “bee, mOde]. Tl] i \ 8 changes and the import and export prices. Next, these price changes are entered as exogenous changes in the model, and a solution is obtained for changes in the variables of interest. Thus, their solutions do not reflect the full effects of the new FTA. International trade modelers have widely used computational general equilibrium models for analyzing such issues as trade liberalization and fiscal reform, since CGE models allow us to track the resulting resource allocation movements between economic sectors. In particular, trade liberalization has increasingly been analyzed in a general equilibrium context. However, in the early computational general equilibrium models that were used to address the issues of trade liberalization and economic integration, welfare effects were estimated to be very small.7 These results pushed economic modelers to pay more attention to possible model misspecification. Their concern was centered on scale economies, since constant-returns-to-scale technology does not capture an important source of welfare gains from trade arising from the presence of economies of scale and imperfect competition. This concern is reinforced by the increasing empirical evidence that countries with similar factor endowments have large volumes of trade. And there has been a growing literature which has explored the issues of international trade and industrial organization. Harris (1984) showed the possibility of increasing the estimated welfare effects of trade liberalization, by including scale economies and imperfect competition in the model. The assumptions of perfect competition and constant-retums-to-scale technology 7 For a review of these early studies of trade liberalization, see Harris (1984). were TC the earl entry ar. fiInctior I area UOL trade lib foreign t deterrnin decline a productit scale incI Dixit and Hunter. x and Sch”. Tl exIllored ; Razinug We (19 9 were regarded as the main sources for the modest welfare effects of trade liberalization in the earlier CGE modeling about trade liberalization. Under perfect competition with free entry and exit, individual firms were operating at the minimum of their average cost functions before trade barriers were reduced, and, thus, the formation of a new free-trade area does not bring a large welfare improvement. Welfare gains will be underestimated if trade liberalization enlarges the size of the market and lets domestic firms compete with foreign competitors. The adoption of scale economies will play an important role in the determination of the trade patterns and welfare effects of a FTA as long as average costs decline as their outputs increases, since fewer resources will be needed per unit of production of goods. The literature on international trade under increasing returns to scale includes Cox and Harris (1985, 1986, and 1992), de Melo and Robinson (1989), Dixit and Norman (1988), Harris (1984), Helpman (1981), Helpman and Razin (1983), Hunter, Markusen, and Rutherford (1992), Markusen and Wigle (1989), and Mercenier and Schmitt (1992). Theoretical models with Chamberlinian monopolistic competition have been explored in Brown (1991), Dixit and Norman (1988), Helpman (1981), Helpman and Razin (1983), Krugman (1981, 1991), Markusen and Svensson (1986), Markusen and Wigle (1989), and Nguyen and Wigle (1992). Initially, firms are assumed to operate at the long-run equilibrium, earning zero profits. Now, trade liberalization would allow foreign competitors to sell their products, which would force domestic prices to decrease. Lower prices would increase the quantity of goods demanded. But there would exist barriers for new firms to enter the industry, because of fixed costs. With higher demand and th' is not 2 firms F the “CI of pol ic a static. and pric simulatit Lineariz. results. v Euler’s n reElons. 0f trade I i 1993). vte monom“ order to 1' r I 9 See Herte JohallSen , l5 linear SI dogenOUS 10 and the same number of firms in the industry, the firms might make positive profits. This is not an equilibrium. In the new long-run equilibrium allowing for free entry and exit, firms produce more and take advantage of scale economies. The basic model I am using for the perfectlyzcompetitjle CGE part of this paper is the “Global Trade Analysis Project (GTAP),”8 which is designed to simulate the effects of policy changes in a computational general equilibrium international trade model. It is a static, Walrasian general equilibrium model that endogenously determines quantities and prices, solved using the Johansen (1960) simulation approach.9 A Johansen simulation will be carried out by solving the linearized equations of the model. Linearization of a non-linear model may give good approximations to the true simulation results, which can be obtained from a multi-step simulation (for example, by using Euler’s method or Gragg’s method). Unfortunately, GTAP assumes that all sectors are perfectly competitive in all regions. As noted above, perfect competition tends to underestimate the welfare effects of trade liberalization. Following Harris (1984), and Cox and Harris ( 1985, 1986, and 1992), we will replace perfect competition with imperfect competition, and incorporate monopolistic competition. The GTAP model will be modified into a simple version, in order to introduce monopolistic competition. The details about the modifications of GTAP will be given in chapter 2. Then, the demand structure will be modified, so that product differentiation at the firm level is used to replace GTAP’s product differentiation at the 8 See Hertel, et a1. (1993). 9 Johansen (1960) approximated his model by a system of linear equations in changes of the variables. This linear system was then solved by matrix manipulation, giving the approximate effects on the k endogenous variables of changes in the (I - k) exogenous variables. national incorpot technolc Merceni competi increasii transport bases. cc input-cu regions. statistics. Submissi. mapping: economic aggregate OI and AR] the GTAP git-en. We SlmUlation GTAP mo 11 national level. Section 2.3 describes the demand structure for consumers. We also incorporate firm-level product differentiation. A detailed description of the production technology in our model is given in section 2.4. Following Cox and Harris (1992) and Mercenier and Schmitt (1992), we will divide production sectors between perfectly- competitive sectors and monopolistically-competitive ones. The latter sectors will have increasing-retums-to-scale technology, with fixed costs. We use the GTAP data base, which includes matrices describing bilateral trade, transport, and protection. These matrices link the 24 country / regional economic data bases, covering the whole world. Each regional data base is derived from each country’s input-output tables. The disaggregated GTAP data base consists of 37 sectors and 24 regions. International trade data in GTAP is based on United Nations D series trade statistics. Export subsidy and protection data are obtained from the original country submissions to the GATT for the Uruguay Round. Lists of disaggregated sectors and mappings of commodities for our study are given in Table 2. Since we study the economic effects of F TA under APEC and AF-l l, the GTAP data base will be aggregated into 13 regions, as shown in Table 1. Our study will be centered on comparing the welfare effects of FTA under APEC and AF -1 1. Table 1 contains the member countries of APEC and the countries/regions in the GTAP data base. In the third column, the country (region) mappings for our study are given. We will aggregate the GTAP data base into a l3-region data base, for the simulations of APEC and AF -1 1 FTA. Each FTA will be simulated with the original GTAP model and with our modified GTAP model, in order to study the sensitivities of the has GT1 TM regi SCCl betv total .- regit llllm servi NIE: el I 0 IV 3T5 ES. AS 12 the welfare changes to different model specifications. Industry sectors are aggregated, based on similarities of the sizes of scale economies and sectoral characteristics. All GTAP sectors in each region will be aggregated into five sectors. Two sectors (RPR and TME)10 will be assumed to have scale economies.ll The five production sectors in each region will consist of one service sector, one agricultural sector, and three manufacturing sectors, as in Table 2. In Table 3a, we list the benchmark values of exports of goods and services between 13 regions in our model in millions of 1992 US. dollars. The numbers are the total values of exported goods and services from source regions (row) to destination regions (column). If the source region is equal to the destination region, then the numbers are the total values of domestic uses of domestically produced goods and services. It is shown that there are small volumes of trade between Canada/Mexico and NIEs/ASEAN nations. Relatively high volumes of goods are traded between NIE nations and ASEAN countries. In Table 3b, we list the benchmark values of exports (excluding domestic uses of domestically-produced goods and services) of goods and services for 13 regions in our model in millions of 1992 US. dollars. Regional shares of world trade is also provided in the bottom in Table 3. '0 RPR is a production sector for resources, plastic and refinery, and TME stands for transportation vehicles, machinery and equipment. See Table 2 for details. it See Prattem (1988) for the magnitudes of scale economies for sectors. 2.1. I economi more €Ct model pr the basic be Studie cOllCCt in 13 CHAPTER II THE DESCRIPTION OF THE MODEL 2.1. Computational General Equilibrium (CGE) Modeling CGE models have been used extensively to capture the essential features of an economic situation. A CGE model is a simplified computer representation of one or more economies. Each economy has consumers, producers, and governments. The CGE model provides a framework in which widely different policies can be examined. Once the basic model has been specified and implemented with actual data, various policies can be studied with minor modifications. The consumers in the model supply factors of production and, in return, they collect income from production sectors. They purchase goods from producers. They pay taxes to government and save the rest of income after the expenditure for final consumption. The consumer solves a utility-maximization problem, with the budget constraint : . . _ N Maxrmrze u (di'dg’ ’dr 'Sr) subject to (1) z . Yr=2i=1pr*di' + 5r where l d2 dr' r r. T€Sp€t COIISUIll. savings maximi/ producer The next and impt CXplainin emmple. were mOC “limiter 0 data Th “Vantage CGE mod of lnputs i can aceOUI N ’3‘ A"mné’lot e the Se l4 didg, ,drll/ stand for consumption demand for aggregated good 1, 2, N in region r, respectively, and p; is the composite price index of good i in region r, inclusive of consumer taxes, which will be described in section 2.7. Yr and S, are income and savings in region r, respectively. The consumer in each region will solve the maximization problem. Region r’s aggregate demand for good i, d; , is an aggregation of domestically- produced good i and an aggregation of imported good i from other regions in the model. The next problem is to divide the aggregated consumption of good i into domestic goods and imports. At this stage, the “Armington” assumption12 has been a basic tool for explaining product differentiation to match CGE models to data on trade flows, for example, GTAP, and Cox and Harris (1985, 1986). If goods of the same good category were modeled as homogeneous, countries would specialize in the production of a small number of goods, and cross-hauling of the same good will not be observed in real trade data. The approach of product differentiation by country of origin has several advantages over alternatives. First, the Armington specification helps multi-regional CGE models to converge into equilibritun fast and easily, since firms determine sourcing of inputs independently of the prices of domestic goods, due to separability.13 Second, it can account for cross-hauling, where each region imports and exports the goods of the '2 Armington (1969) suggested that products are differentiated by the country of origin. '3 See the section for “Behavioral Equations” in Hertel and Tsigas (1994). same p: CV60 at view th. places e | conside' Aming modeler produce assumpt Where 0 imports, i mcr Star lndlcates "DPORS I research: 15 same product category. International trade data show a large amount of cross-hauling even at a very high level of disaggregation. This can be explained with Armington’s view that the goods of the same product category are regarded as different goods, if the places of production of the goods are different. One example is that a car made in US. is considered to be a different good from a German-produced automobile. Third, the Armington specification needs one more set of elasticities of substitution. Then, modelers can assign different values to the elasticities of substitution for domestically- produced goods and imports, depending on the researcher’s purpose. Armington’s assumption will be represented as follows. 0 i , 0' d7 _ 0' f/ % fl (If h; m + mg, m , (2) where q, is the elasticity of substitution between domestically-produced goods and imports, hi. is region r’s consumption demand for domestic (home country) good i, and mgr stands for demand for imported good i for consumption in region r. The subscript c indicates final consumption. In eq. (2), the consumer’s preference for domestic goods or imports will depend on the elasticity, cm, which will be assigned exogenously by researchers. One way of aggregating imports is to use a CBS function, such as "'0': = lame /l 1’ "I 16 where REG is the set of all regions in the model. 6... is the elasticity of substitution between imported goods. m clsr is region r’s consumption of good i from region s. With this equation, the sources of aggregate consumption of good i will be identified, and will be matched to the data. Similarly, a CBS. function will be defined for the composite price index in the following equation: -l—o -l—0‘ %—o ‘1’ pg? ] , (4) P:,=[pcr + whose notations are similar to the aggregated consumption demand, except for replacing C with P to denote a composite price index. The superscripts d and m for prices are used to represent the prices of domestically-produced and imported goods. The same elasticity as in eq. (2) should be used for eq. (4), but subscripts for the elasticity were omitted for simplicity. Then, the conditional demand equation for the domestic goods will be i O' h‘, = d',’ —f,j.- . (5) Cf The demand equation for aggregated imports is i 6 Per "11 pCP’ (6) i i mr 2 dr. mi The composite price index for aggregated imported commodity i in region r, p cr , will be calculated with the CBS. equation : . ”1.1—0 1-0' 173;, = [ Z pcslr :| a (7) mi . . . . . . . where pcsr IS the consumer’s prrce for imported good I from region s In region r. Subscripts for the elasticity were restrained. The equation for imported goods by source will be mi 6 i __, l a pcr (8) mcsr mcr mi ° csr But the Armington assumption bears criticism : (1) The Armington assumption implies that products are differentiated by country of origin, and this differentiation is exogenous to the model. Today, the world economy tends to be unified, and thus, foreign/domestic distinctions have been blurred. (2) The Armington approach was found to underestimate the effects of trade liberalization in Norman (1990). Norman concludes that the “Armington” approach is “a poor substitute for explicit incorporation of oligopolistic interaction and product differentiation at the firm level.” Another approach is based on theoretical work by Dixit and Stiglitz (1977). Their idea is to assume that products are differentiated not by the origin of country but by the producing firm. Consumers purchase goods, considering the brand names of products. For example, a BMW is regarded as a different car than a Mercedes-Benz. Firm-level product differentiation is necessarily linked to imperfect competition, while the Armington assumption does not necessarily require imperfect competition. In fact. will div Howevt (capital optimal product: Where _- good i ir‘ l'Egion r. intermed intermcd “here 0' ,1' _.I andall at! 18 In a dynamic model, consumers save so that they can enjoy future consumption. In fact, saving elasticity could be positive, negative, or zero. Thus, the economic agent will divide his life-time income between current consumption and future consumption. However, in static CGE models, savings will be represented as purchase of investment (capital) goods. Producers will minimize the total cost of production, and this will result in the optimal combination of intermediate goods and value added. The general form of the production function is q:= MIN ( ZliszzirHoZZ-nisVA: )9 (9) where z]; is the demand for aggregate intermediate good j used in the production of good i in region r, and VA: is the value added employed for the production sector i in region r. Like the aggregated consumption demand, the aggregation formula for the intermediate goods and value added will be necessary, and the conditional demand for intermediate goods will be an aggregation of domestically-produced goods and imports. Researchers have used a CBS. aggregation for intermediate goods, as follows : .. ”oz—y ”oz—y “A4 z’,’= d]; “' + m]; , (10) where d 54 and mg are production sector i’s demand for domestically-produced good j and an aggregation of imported good j for intermediate inputs. The subscript f is used for impon T domestit fOT the p “lib (he inlefined as fOIIOul n our Q 19 producer’s intermediate demand, and 0', is the elasticity of substitution for the producer between the imported intermediate good and the domestic intermediate good. Aggregated intermediate demand for imported good j will be defined similarly to aggregated consumption for the imported good, as 0 m "1 0%m ‘4 ji A. m}: = 2 m]... , (11) S GREG where mg is production sector i’s demand for imported good j from region s in region r. 0' ,,, is defined at eq. (3).'4 As in the consumer’s case, C.E.S. equations will be specified for aggregating the import prices from all sources, pg; , and for the producer’s price, p; , aggregated over domestically-produced goods, p31,]: , and aggregated imports, p'j’gi , where the notations for the prices for firms are the same as those for intermediate demand for producers. With the composite price index, producers will choose the optimal amounts of aggregated intermediate demand for domestically-produced goods and imported goods, respectively, as follows : ji at]; =2)!” 51:. . (12) pfi '4 In our CGE modeling with the Armington assumption, the same values of the elasticity of substitution for imports will be used for final consumption and intermediate goods, due to the lack of data. Then. and (l 3 identiii problem (6) deter Pruduce the elast cOSts, vvl lmperfec Value 8d. their mar Firms Wi: Pl’lmaxy T ts\ DElailed 20 co 6 I P} mji Pfi m}; = Zli * (13) Then, the amounts of domestic goods and imports will be chosen, according to eqs. (12) and (13), respectively. The sources of imports for imported intermediates will be identified with the following equation: I)? 6 me = met" —— - no Pfir The same procedure applies for final consumption: From the utility maximization problem, the Optimal amount of each good consumed will be decided. Equations (5) and (6) determine the division of aggregated consumption of each good into domestically- produced goods and imports, with substitution between sources of goods depending on the elasticity of substitution. The sources of imports will be traced with eq. (7). For the perfectly-competitive sectors, price will be simply equal to average total costs, which implies that there will be zero pure profits, while if the model is modeled as imperfectly competitive, then an extra equation will be necessary to assign some part of value added to be fixed costs.'5 And with fixed costs, firms will mark up their prices over their marginal costs. Value added will be a CBS. aggregation of labor and capital. Firms will employ labor and capital, according to the elasticity of substitution between primary production factors. '5 Detailed descriptions for imperfectly-competitive models will be given, in sections 2.3, 2.4, and 2.5. 21 To close the model, we need market-clearing conditions : Primary production factors should be fully employed. The output of each production sector in each region should be equal to the sum of exports and domestic use for final consumption and intermediate use, and imports in each region should be equal to the sum of final consumption and intermediate use. Numerical expressions for market-clearing conditions will be provided in section 2.6. If we solve all equations for consumers and producers simultaneously, satisfying the market-clearing conditions, we have an equilibrium which replicates observed data. Then, the policy changes can be simulated by changing the relevant policy parameters and recalculating a new equilibrium. With this procedure, we can predict the effects of policy changes, such as the effects of a bilateral reduction of tariffs on regional income.‘6 CGE models use the elasticities of substitution, given by macroeconomic and econometric studies. CGE analysts calibrate the parameters of the CGE model, so that the benchmark equilibrium reproduces the transactions observed in the data. And they will do the sensitivity test with a different set of parameters. Therefore, the numerical results of the models should be interpreted in the light of their chosen parameters and data. For the base-line case, we take the elasticity of substitution from the GTAP data base, but we calibrate the elasticity for different sets of parameters, as explained in section 3.1. '6 Taxes and tariffs will be discussed in section 2.7. 2.2. increasi econoni model I model a scale. \‘ reduces part of p lmponar 13th rev 1 in eaCh Tl have an\ prOdUCIIL 22 2.2. The Overall Description of the Model This paper presents a CGE world trade model with imperfect competition and increasing-returns-to-scale technology, in order to study the welfare effects of the economic integration proposals for the Pacific-rim region. The basic framework of the model originates with the 1994 version of GTAP. The standard version of the GTAP model assumes perfect competition in all regions and all sectors, with constant returns to scale. We simplify the GTAP model in the following ways: (1) Removing the consumption structure of the government in each region. This reduces the number of variables. Instead, government consumption will be regarded as a part of private consumption of final goods. Reducing the numbers of variables is important, since our model will be solved, by inverting matrix of variables. All tax and tariff revenues are assumed to be rebated to the household. This means that taxes will not have any income effects, as shown in Ballard (1990). Total demand for each commodity in each region will be the sum of private consumption and the intermediate demand for production sectors in each region. (2) Eliminating re-exports via Hong Kong,17 in order to reduce computational and analytical difficulties. If re-exports are included in the model, then we will need one more dimension of the variables of domestic uses of domestically-produced goods, which increases the number of columns of matrix. The GTAP data base contains the trade data ’7 An example of re-export will be Hong Kong’s re-exports of agricultural products to USA, who exported the same products to Hong Kong. The GTAP data base will be modified to add the trades of re-exports to domestic consumption of domestically-produced goods. for re-e COTlSUlT model. product labor an with agt the agrit the desc model. 1 endOgcn JOhanser allows {C and Clem; PIOdqu- GCOnOmk in all COU TI | MOM“ 23 for re-exports. But in the model, re-exports will be counted as a part of domestic consumption of domestically-produced goods. (3) The transportation sector is defined to be a world service sector in the GTAP model. But in our model, we eliminate it to reduce the complexity of the model. (4) The GTAP model assumes that the agricultural sector has 3 primary production factors (land, labor, and capital), while other manufacturing sectors have only labor and capital. Our paper is much more concerned with manufacturing sectors than with agricultural sectors. Thus, we remove land from the primary production factors for the agricultural sector. The cost of land used will be added into capital. Section 3.2 has the description of the relevant modification of the GTAP data base, suitable for the model. The model used here is a static, Walrasian general equilibrium model that endogenously determines quantities and prices, solved by using a descendant of the Johansen (1960) simulation approach. It is a multi-sector and multi-region model, which allows for the analysis of the effects of policy changes on regional welfare, production and demand per agent and per region, equilibrium prices, rates of return to factors of production, etc. Two initial assumptions are: (1) there are no pure profits in any economic activity (producing, importing, exporting, transporting, etc.), and (2) all sectors in all countries will be assumed to be in equilibrium. Three sectors are assumed to be perfectly competitive, and the rest of the sectors are to be imperfectly competitive. This is a general case adopted by CGE analysts, such as Cox and Harris (1985), Mercenier and Schmitt (1992), and Brown, Deardorff, and Stern ( GTAP GTAP SfClOTS dnvent econom average ( adopted COtttpeti that don will ass‘ CXpons and Sch 23. . 24 Stern (1992). We will add the following components into the simplified version of the GTAP model : (1) Imperfectggmpefition. Some of the perfectly-competitive sectors of the GTAP model will be replaced with imperfect competition. Monopolistically-competitive sectors will be characterized by free entry and exit. As a result, their net profits will be driven to zero. (2) W. Since we want to study the welfare effects of scale economies, the imperfectly-competitive sectors are assumed to have fixed costs, such that average total costs decline as output per firm increases. (3) W. The firm level product differentiation adopted here will be similar to that of Mercenier and Schmitt. In their model, perfectly- competitive sectors are modeled as having the “Armington” specification, by assuming that domestically-produced goods and imports are imperfect substitutes. But our model will assume no “Armington” assumption even for perfectly-competitive sectors, since re- exports were removed in our model. This distinguishes our model from that of Mercenier and Schmitt. 2.3. Consumer Preferences The major difference between GTAP and our model is that we replace the “Armington” assumption with firm-level product differentiation. Figure 1 describes the 25 “Armington” Specification Firm-Level Product Differentiation l i i i // \ "1 n2 - - - ”r i i i mclr mc2r ' ' ' chr differe: houschlI consun the hi gl the bottl this sec model. : Without goods a: aSSUlll€L side of l- tts COUnI Source TH Othenvig S€Cl10n 2 Section 2 formu cOnsumer househok 26 different specifications of product differentiation in the demand structure for the household in each region. In the GTAP model, economic agents divide their consumption of composite commodities into domestically-produced goods and imports at the highest nest of the utility function. Then, the sources of imports will be identified by the bottom nest of the utility function. This is shown at the left-hand side of Figure 1. In this section, we describe the imperfectly-competitive model, since a perfectly-competitive model, such as the GTAP model, was described in section 2.1. With firm-level product differentiation, consumers select commodities directly, without a middle procedure of dividing the composite commodity between domestic goods and imports, as with the “Armington” assumption. That is, economic agents are assumed to differentiate commodities at the firm level, which is shown at the right-hand side of Figure 1. Thus, consumers look at the brand name of the commodity, rather than its country of origin. At the right side of Figure 1, if the destination region is equal to the source region, the commodity is meant to be domestically produced. Otherwise, it will be an imported commodity. The notations for Figure 1 are given in section 2.1, and subscript T denotes the number of regions in the model. Our demand structure is shown in Figure 2, whose notations are the same as in section 2.1. The superscript N is the total number of commodities. A Cobb-Douglas (C- D) formulation is specified for the top nest, and each region has one representative consumer, whose welfare level represents the welfare level for the region. The household’s utility level will depend on the consumed amounts of the composite goods. 27 Figure 2 shows two levels of consumer decision making: The first stage of the C- D nests will determine the expenditure shares for each of the composite commodities. At the second stage, the brand name (or firm) for each commodity will be identified.l8 Mathematically, consumer preferences at the top nest will be defined as a C-D utility function of composite demand for all final commodities (both imported and domestic), assuming constant expenditure shares ( 5 i ) : N . i , N . ur = “d;6'*S§', where 25', + 5i: 1.19 (15) i=1 i=1 In equation (15), savings will be treated as one of the consumed commodities. Equation (15) shows that regional utility will be the product of consumed commodity aggregates and savings, weighted by the expenditure shares. The second level of the utility fimction determines the optimal composition of the consumption aggregates in terms of regional origin. For the perfectly-competitive sectors, we have : (SC—y 6%C—l di=‘1’{2dfv ‘ , (16) 3:] where O' c is the elasticity of substitution between traded commodities for consumers, and ‘P is a scale parameter with positive value. The imperfectly-competitive sectors will '8 Each fu'm is assumed to produce only one brand of product. ’9 We add savings to the utility function, in order to keep as many properties of the GTAP model as possible. More importantly, keeping the data for savings in our model minimizes the modification of the data base, without changing the basic structure of the data. 28 Figure 2. The Structure of Household Demand Regional Utility ur [ Cobb-Douglasj d1 d3 -- - - d1 ------ C.E.S C E S have ; SECIOT region; compo» aggreg; formuh Where / eqttatior Where p 29 have additional components : The number of firms operating in region s’s production sector i, ni, and region r’s market share for good i from region s, (p1, .20 I a, 1 Oc_ UC- r - A. i . - 1 dr = W Z] nfg’k‘P sr*dsr . (17) S: The top nest (eq. (15)) transforms composite commodity consumption into the regional utility level. The second level nest (eq. (16)) will identify the sources of composite consumption. For this transformation, we need the composite price index of aggregated good i in region r, p2,, . This price index will be aggregated with a C.E.S. formulation: 1 i T i lfl‘l/l—O pa = i 2 pm i , (18) Ls=1 where pi is the consumer price for good i from region s in region r. A similar csr equation will be defined for savings (capital good). I sav T "0 1%” p, = . 2 pi?” J , (19) 5:1 where psav is region r’s price of capital goods from region s. Sf 2° This is a typical method of adding firm-level product differentiation into CGE model, used by trade modelers, such as Brown (1992), Mercenier (1994), Mercenier/Schmitt (1992), and Nguyen/Wigle (1992). no no no re (1" the or Whe e ] 30 The household’s demand can be summarized as follows: consumer prices are aggregated into the composite price index through eq. (18), which is the basis for deriving the conditional demands for composite commodities, in eq. (17). The information about the consumption of the composite goods will calculate the regional utility level, via eq. (15), weighted with the expenditure shares for composite commodities, which are aggregated over all sources. The change of utility will be used to compute the regional equivalent variation (EV) as in GTAP: EVrz Y0*{ur —0ur},21 (20) ur where Y0 is the regional income level before the policy, and u]? and u? denote the utility level after policy and before policy, respectively. 2.4. Production Sector In our model, some of the production sectors are assumed to be perfectly competitive (PCM) and the rest are imperfectly competitive (IMC). One IMC sector is chemicals, plastic, resources, and resource refinery (aggregated as RPR in this paper). 2' In a non-linear CGE model, EV is defined as (Vb-V,)'Pb , where V is the indirect utility level, and P is the price level. The subscripts b and r imply base case and revised case, respectively. A linear CGE model cannot calculate E V with this formula, since the utility variable in eq. (20) is the level of utility, and this variable is the percentage change of the price level in the linearized version. But eq. (20) gives the EV for linear modeling. 31 The other IMC sector is transportation and machinery equipment (TME). This classification is based on the size of scale economies, studied by Prattem. In the PCM sectors, the producer’s price is equal to marginal costs. It is assumed that the perfectly- competitive firms operate with constant-retums-to-scale technologies in production. All firms (including both PCM and [MC firms) use capital, labor, and intermediate goods as their inputs for production. Firms employ labor and capital as primary production factors. Both labor and capital are assumed to be perfectly mobile within the region, but immobile between regions. The [MC firms have fixed costs, in addition to the variable inputs, and thus, their technology exhibits increasing returns to scale. Fixed costs will be composed of labor and capital, i.e., parts of the labor and capital employed will be regarded as fixed costs.22 The [MC sectors are characterized by free entry and exit. No net profits will exist in the [MC model. Thus, we can think of these firms as monopolistically competitive. According to Krugman (1979, 1980), a Chamberlin approach was suggested to be useful here, in that the equilibrium of the model is unique. Each firm can ignore the effects of its strategic actions on other firms’ behavior, eliminating the indeterminacies of oligopoly. That is, if the numbers of firms are large in a monopolistic competition model, firms will be hardly affected by one firrn’s price change. In addition to the detenninacy of the model, Charmberlin models can be easily modified to reflect firm- level product differentiation. 22 Details about fixed costs will be given in the section for market-clearing conditions. 32 Each industry in the imperfectly-competitive sectors has N firms per region, whose numbers are exogenously given for the initial equilibrium. More description about the number of firms will be given in the section 3.1. The variable for the number of firms will be endogenously determined as the new equilibrium is calculated, because of free entry and exit. Each firm in an industry has the same technology and the same pricing rule. And each industry is assumed to produce N varieties of commodities. That is, each firm is assumed to produce exactly one variety. If a new free trade area were to be formed in the Pacific-rim region, the demand for each variety would increase, since price would go down due to the elimination of tariffs, as. long as the traded commodities are normal goods. Responding to the increased demand, firms increase their production, which decreases the average total costs in the imperfectly-competitive industries. Then, they will move downward along the curve for their average total costs, exploiting scale economies. On the other hand, the number of firms should be interpreted with caution. If the number of firms decreases, then existing firms can exploit scale economies. But the reduction of the variety of goods entails a welfare loss, as shown in the functions of household’s utility and the aggregation of commodities (eqs. (15) and (17) respectively). Figure 3 shows the production structure for the imperfectly-competitive sectors. Commodities at the firm level will be aggregated into a composite commodity with a C.E.S. formulation. Primary production factors will be aggregated into 23fixed value added and variable value added, once again using a C.E.S. equation. In addition, the top 23 The shares of IMPC sectors in regional output are very wide by region. Taiwan/Singapore have the highest share of 35%. Korea and Malaysia take the second highest shares (28%). Philippines and Thailand have the lowest (l3%-14%). The US. and Japan are in the middle (21%-24%). 33 of the production structure in the [MC sectors will combine variable value added and composite intermediate goods, using a fixed-coefficient (Leontief) technology. Solid arrow lines and dotted arrow lines indicate intermediate goods and endowment factors, respectively. Figure 3 summarizes how this model is different from the GTAP model. First, our model extends the GTAP model to have imperfect competition, by incorporating fixed costs. Second, for the intermediate goods, firm-level product differentiation is specified, as in consumer’s demand structure, to replace the “Armington” assumption. Third, primary production factors are modified to have fixed costs in the imperfectly- competitive sectors. K and L in the Figure 3 are capital and labor, respectively. VAfi ( VAiii) is fixed (variable) value added for the production sector i in region r. 2;}. is the conditional demand of the production sector i in region r for intermediate good j from region s. The demand equations for producers will be similar to those for consumers, except at the top nest of production. The top nest has a fixed-coefficient technology, such that ‘1’, = writ-end q; = zfifori = 1,2, ..... , N. (21) K. Fig Figure 3. mi r i l K; L; K; L; 34 OUTPUT of good i in region r q. l [ C.E.S. J // l \ 2y .2?" """ 29/1 @fig\ Gigi 2i 2i 21 er ZZr ZTr I I l 11'] 113 ---- n7 u- for pe for iml. in regi The co Where 1 solutes 35 Composite intermediate goods will be defined as follows : T o—% %—l zii =9 225 , 3:1 for perfectly-competitive sectors, and o 1% % — I z{’=¢ myrrh»: , (22) for imperfectly-competitive sectors. (1) is a scale parameter, and 5,11 is firm’ i 5 share in region r for good j from region s. The composite prices will be - T -- 1—0 I l I pfi = L: p}, J . (23) where pjffv is firm i’s price in region r for intermediate good j from region 5. Total variable costs, Ci: , will be the sum of variable value added and intermediate demand multiplied by producer’s costs for the intermediate demand from all SOUI'CCS. Chi: lepfsr‘zs Zin‘J' +VAri (24) s=lj=l Eq.t. be or where . The it" added Total c (36), u‘ Dividir Equatio llle Imp, 36 Eq. (24) represents the cost-minimizing input demands for a given output, q: , and it can be written as follows : A-V * ”ii fl : 17+ V- (25) Cr: ‘1, pfgr Zsr VArz’ s=lj=l where 0;: is variable costs per unit for producing q',. The total costs of producing good i in region r, CL. , will be the sum of fixed value added and total variable costs. Cf,- = VArfi + CZ. (26) Total costs is the product of average total costs, cg, times output, q; , and rewriting eq. (26), using eq. (25), off * qi = VAfI+cIIV = VA/wdf’tqi. (27) Dividing eq. (27) by q i. , it becomes AT _ AV VA rji (28) Equation (28) demonstrates the scale economies, since average total costs will decline for the imperfectly-competitive sectors as output increases, given constant fixed value added and constant average variable costs in the short term. $511111 equal imper the rel variabf averag " [sing I lUSt the lhe ran lllllCl] I rate. ] 37 With monopolistic competition, net profits should be zero, since the model assumes free entry and exit. The zero-profit condition requires that average total costs be equal to the price that producers receive from selling a unit of their product, pf”, for the imperfectly-competitive sectors. i _ AT 17,, - cri~ (29) The fixed value added of [MC sector i in region r, VA [ , will be calculated from the relation of average variable costs, Cri and average total costs, cm" The ratio of variable costs to total costs, 9 ’, can be written as the ratio of average variable costs to average total costs. CAthi CAV i _ c" __r _ ri o, — AT — —,,T. (30) at: C - Cri q:- r1 Using the zero-profit condition for [MC sectors, eq. (29), the variable cost ratio will be just the ratio of average costs to supply prices. AV 0n o,‘ = —.. (31) psr The ratio of marginal costs to supply prices is the inverse of the [MC firrn’s markup rate, which will be discussed in section 2.4. The Lerner formula says that the optimal markup rate, M f. , is the ratio of total perceived demand elasticity to total perceived demand 38 elasticity minus one. Then, the variable cost share equation will be written as the following: . _1 . AV 1 I . c ~ I E E ‘l 9; = "I- : I =l l'rl = rt = l—lr (32) 173,. Mr Er"l Er Er Substituting eqs. (25) and (27) into eq. (32), AV AV . T N Pji*zji+VAv' . 9i _ CrI _ Cri "1i _ s=tj=t 3’ 3’ " _ Vi‘VArfI 33 r‘Ar‘ AT,,-“TN.,... ‘ ,-a () cri cri qr lelpir*z£+VA:l-+VA£ Vr 3:}: . r N ~~ .. where V’r = 2' 21p::*21;+ VA; + VAfl , which says that the total costs of producing s: j: IMC good i in region r will be the total revenue of IMC firms, Vi... Substituting eq. (32) into eq. (33), and rewriting eq. (33) becomes . - I . 1 , VArfI = {l-Gii‘V’r = {Iii-FEW; = BTW}. (34) That is, fixed value added will be the product of the inverse of the total perceived demand elasticity and the total revenue of IMC firms. Currently, engineering information for fixed costs is not available at levels of aggregation that are sufficiently high to be used in nationwide CGE modeling. The eq. (34) will be used to calibrate fixed value added for the IMC model in this paper. As the total perceived demand elasticity goes up, fixed value added will be lower, given the market value of firm’s output. Since fixed value added is a part of total value added, the share of fixed value added to total value added cannot be greater than one. During simulations, this point will be observed carefully, and 39 the elasticity of substitution will be calibrated, such that the fixed share is less than one, as shown the section 3.1. This model does not need extra equations, such as equations (30) - (34), for the PCM sectors, since there will be no fixed factors, and average total costs will be the same as marginal costs. Therefore, the PCM firms will have constant-returns-to-scale technology. As shown in Figure 3, the fixed value added and variable value added will be aggregated in a C.E.S. formulation, in the same way that GTAP specifies the primary factors for the competitive sector. But we add fixed primary production factors, and thus, the pricing rule will be modified to reflect that. We assume that primary factor markets are perfectly competitive, such that the price of primary factors (labor and capital) is the same for competitive and imperfectly competitive sectors. 2.5. Total Perceived Demand Elasticities The pricing rule for the monopolistically-competitive firms can be specified with either the Lerner formula or the Eastman-Stykolt hypothesis (ESH). The ESH was used by Cox and Harris ( 1985, 1986, and 1992), and Nguyen and Wigle (1992). The ESH assumes that the firm sets its price equal to the price of the 40 import-competing good, inclusive of the domestic tariff, such that domestic price = the world price of import * ( 1 + tariff). This is a less-aggressive pricing policy. This pricing rule has been supported in Canadian industrial organization studies.24 ESH is a collusive price-setting rule, in the sense that the prices that domestic firms set would be world prices plus domestic tariffs of the imports, without the large shifts of demands into imports, since the price of the domestic good is closely linked to the price of imported good. The Lerner rule is collusive and non-aggressive, since monopolistically-competitive firms establish a market niche for their product and mark up their prices over the marginal costs of their products, in order to maximize their profits, rather than increase their market shares, by setting prices lower than those of competing goods. The Lerner optimal markup rule is based on a microeconomic foundation. On the other hand, the ESH has no theoretic basis. A serious problem can be raised with the ESH pricing rule. When the ESH rule is used as the pricing rule for monopolistic firms, the welfare effects of FTA may be overestimated,25 because of the direct linkages between tariff cuts and domestic prices. Thus, our model will use the monopolistic pricing rule of Lerner, in order to provide a conservative evaluation of the benefits of new FTA. The Lerner formula for the optimal pricing rule for a monopolistically- competitive firm is given in eq. (35). 2“ See Cox and Harris (1986), p. 382 and Karikari (1988) for evidence supporting ESH. 25 Harris / Cox was criticized by Nguyen and Wigle (1992) for overestirnating the welfare effects of Canada-U.S.A. FT A. See also Sobarzo (1991). 41 _,.___., 35 E ( ) where c}: is marginal cost of producing good i in region r, and E 1r is the value of the perceived total demand elasticity and its value will be greater than one, since the supply price will be greater than or equal to marginal costs. pi Defining the markup rate as M i: = i eq. (35) will be transformed to the M ’ Cri following equation. Mi = —E.i——. <36) E'r-l where the markup rate is greater than one, since the total demand elasticity is greater than one. From eq. (3 6), we can see that markup rate will go down if the perceived demand elasticity increases. Then, lower welfare gains are expected, since the model will quickly approach the competitive position, yielding small efficiency gains. Another reason that lower welfare gains may be generated by higher elasticities can be seen in eq. (34) : If the perceived total demand elasticity increases, the fixed value added will be lower. Smaller fixed value added will be related to smaller welfare gains from removing tariffs and non- tariff barriers, which was discussed in the section for the production side. Firms will set a markup above marginal cost which is inversely related to the absolute value of the elasticity of the firrn’s total perceived demand elasticity. That is, if a firm faces a more elastic demand curve, then the firm will have low markup rates, and thus will lower its 42 supply price. Combined with the increasing returns to scale and the zero-profit condition, lower markup will bring smaller changes of welfare with a formation of a FTA. The perceived total demand elasticity will be derived from the perceived demand elasticity, 11 slr , weighted with market shares, (psir , as shown below : "1 E; = my ns’, (37) szl As tariffs are removed, region s’s market share for good i in region r, (1);], , increases, as long as region r and s are members of the new FTA, due to trade creation effects of FTA. Eq. (3 7) implies that, as market shares increase with the new FTA, a firm’s total perceived demand elasticity will be increased. Then, markup rates will be decreased from eq. (36). That is, the sale share, (Dsir, will be negatively related to the markup rate. Total perceived demand elasticities calculated from eq. (3 7) will be used for the calculation of optimal prices for producers in eq. (3 5), and fixed value added in eq. (34). Increasing total perceived demand elasticities imply that markets for goods are changing to be more competitive. Thus, producer’s prices move closer to marginal costs, as shown in eq. (3 5), and fixed costs become smaller, such that new firms can be established with lower burden of fixed costs. The IMC firms will increase their sales as trade barriers are removed, and decreasing average total costs will reinforce this, since unit average costs will decrease when output increases. More sales will increase the total perceived elasticity in eq. (3 7), 43 and then, the markup rate in eq. (36) will decrease. Then, producers will lower the price of their products, and consumers enjoy the lower price, which will increase real income and regional utility. The perceived demand elasticity, 11S: , can be defined in several ways, depending on the IMC firm’s expectations about rival firm’s behavior. In recent CGE modeling for imperfectly-competitive models, the Coumot conjecture has been used widely, for example, Norman (1990) and Harrison, Rutherford and Tarr (1995). But we add the Bertrand conjecture to our model. The first approach is to assume that a rival firm’s quantity will be fixed, but rivals adjust their prices to clear the markets for differentiated products. The second approach is to assume that firms will change their output, while leaving their prices unchanged. In this paper, simulations will be performed under both of the two approaches discussed here. The derivations for the perceived demand elasticities will be to differentiate the conditional demand with respect to price.26 Under the Bertrand conjecture, if we set the changes of other prices to zero (except the price concerned), we will have the following equation : . (I) i nsBi=°-{°-1}*{—S,-’la (38) where 0’ is the elasticity of substitution, N; is the number of firms in the imperfectly- competitive sector i in region s, and the superscript B in the perceived demand elasticity represents Bertrand. Alternatively, the Coumot perceived demand elasticity will be 26 Detailed derivations for the elasticity of substitution are given at Hertel (1992). 44 derived, if the changes of other demands are set equal to zero, except for the demand concerned. The Coumot perceived demand elasticity will be 0' 1+{6-4lflnd/Nl’ n3= on where C is used to denote Coumot. It can be said that perceived demand elasticities will increase, as the elasticities increase, when the market shares and the number of imperfectly-competitive firms remain constant, from eqs. (3 8) and (39). As the number of IMC firms increases, the demand elasticity will go up in eqs. (38) and (39).27 Hertel (1992) showed that the Coumot perceived elasticity will be lower than the alternative perceived elasticity, and the associated markup will be larger, with the same elasticity of substitution. Thus, it is expected that the effect of welfare may be overestimated, if IMC firms are assumed to operate under the Coumot conjecture. This overestimation may lead to incorrect results. If we take a conservative position in evaluating the welfare effects of a new FTA, the Bertrand perceived elasticity will be recommended for IMC firm’s conjecture. This point will be studied in chapter IV, by performing simulations under the two conjectures. 27 The fact that the numbers of IMC fums increase implies that more varieties of goods are available. Following Spence-Dixit—Stiglitz’s “love of variety” preference, consumption per variety will be smaller, in order to maximize utility, given budget constraint. On the other hand, demand elasticity will increase if the number of firms increases, as explained above. That is, consumption will be negatively related to the demand elasticity. This was assumed in ngman’s simple model (1979, and footnote 3 in 1980). But this assumption is not required, since eqs. (32) and (33) have a negative relationship between consumption and demand elasticity. 45 2.6. Market-Clearing Conditions The primary production factors are labor (L) and capital (K), each of which is perfectly mobile within each region, and immobile between regions. The immobility assumption rules out migration and international capital flows in a static model like this paper. The market-clearing conditions for the factors for each region are : h, h, L.= Z Lr}?+ z N. Lri+h€§4CNr LII, j ePC M lie/MC (40) K.= z K"+ 2 Nf*Krt+hz Nf*K,’I..’ jePCM 'j helMC e]. C where L, ( K r ) denotes the total supply of labor (capital) in region r. L‘r’j (K2. )is labor (capital) per firm for competitive sector j in region r, 28 2),, ( K3,), ) is variable labor (capital) per firm for the imperfectly-competitive sectors, and L; ( K J: j is fixed labor (capital) per firm for the imperfectly-competitive sectors. N ',' is the number of IMC firms in production sector h in region r. Equation (40) shows how the endowments in each region are allocated between perfectly-competitive and imperfectly-competitive sectors. As new firms are established, additional variable inputs and fixed inputs are required for the imperfectly-competitive firms. For each region in the model, the domestically-produced commodities, q; , should be equal to the sum of region r’s sales of commodity i, such that 2’ Since PCM sectors have no fixed factors, the percentage changes of primary factors for perfectly- competitive sectors are represented with only variable primary production factors. 46 i T i qr=§ssfl (41) where Sisr is region s’s sale of commodity i to region r. Total imports of each commodity should satisfy both the final demand for that good by private households and the intermediate demand by production sectors. Imports (or the use of domestic goods) by source will be equal to the sum of all the domestic demands for the imported good in each region. The equilibrium condition for imports by source will be . , .. 29 ss'r = d s; + 225‘; ' (42) '=l N The market-clearing conditions apply for perfectly-competitive sectors and imperfectly- competitive firms. If r = s, eq. (41) will be applicable to domestic sales of domestically- produced commodities. The rest of domestic output will be exported and the market for good i produced by region r will be cleared according to eqs. (41) and (42). 29 3:, is region s’s exports of good ito region r. But this can also be interpreted as region r’s imports of good i from region s. 47 2.7. Price Linkages The last part of this chapter describes how prices are connected, as transactions of goods proceed. The price-linkage system in the GTAP model should be modified, so that the modified GTAP data are compatible with the model in this paper. First, the supplier’s price will be equal to marginal costs, 0,34,- , for the competitive sectors, and the sum of marginal costs and markups for imperfectly-competitive sectors, such that i_ M pr—Cri’ for the competitive sectors, and i M i pr = Crt . Mr ’ for the imperfectly-competitive sectors. (43) As explained in section 2.2, our model uses a simplified version of GTAP to add imperfect competition into the model. One of the simplifications is the omission of the world transportation sector, which links the PCB. and C.I.F. prices. Without this linkage, F.O.B. prices will be interpreted to be equal to C.l.F. prices. Thus, we introduce world prices of good i from region r, pf”, for F .O.B. prices. The world price will be the supplier price, psir, divided by export taxes. pi_pslr wr _ ' Tx'sr , (44) 48 where Txisr is export taxes on the exported good i of region s to region r, and all tax rates in this paper should be interpreted as (1 + tax rate). The next equation links domestic and world prices : i i . pmsr : pwr * ngr’ (45) where pmlsr denotes region r’s domestic market price for imported good i from region s, and Tmisr is region r’s import tariff on imported good i from region 5. With the modification of the data, which will be described in section 3.1, Tmis‘r will be zero for a single country, if r = 5, since there is no re-export in our model. But if several regions are aggregated into one region, for example, the rest of world (ROW) in Table 1, T min may have positive values, even if r = s. The consumer’s price will be the product of the imported price of good i, pmisr , and the consumption tax on imported good i of region s to region r, Tciri i i ' pcsr = pmsr * T0517 ° (46) The gross-of-tax price of commodity i equals the market price of imported commodity i plus the consumption tax, when region r differs from region s. If r = s, Tcir is the output tax on a domestically-produced commodity. Similarly, the firm’s prices for intermediate goods will be defined to include the tax on intermediate goods, T13; : 49 fl _ i at: 'i pfsr _pmsr Tfisr’ (47) where p g r is the after-tax price of good j from region s for production sector i in region r. As for the consumer’s price, if r = s, p if: r is the firm’s intermediate price for domestically-produced goods, including the tax on the use of domestically-produced goods as intermediate goods. 50 CHAPTER III DATA, PARAMETERS, AND SIMULATIONS 3.1. The Modification of the GTAP Data and Parameters The GTAP data base draws heavily from the SALTER-III data base.30 In particular, the GTAP data base uses regional input-output matrices taken from the SALTER-III data base, and international trade and protection data were incorporated into the GTAP data base. The 1994 version of the GTAP data base used in this paper comprises 26 disaggregated regions and 37 disaggregated sectors. The data in the GTAP can be grouped into two categories : Data for domestically- produced goods, and data for intemationally-traded goods. Our model does not classify the geographical origins of products, that is, whether the good concerned is produced domestically or imported, since products are differentiated at the firm level. The original GTAP data base was designed for competitive models. Thus, it is necessary to modify the data base suitable for our model in this paper. 3° See Jomini, et a1. (1991). 51 We can summarize how the data base was modified and aggregated to be adapted to the model : First, re-export through Hong Kong will just be added to domestic uses of domestically-produced commodities. Second, some of the coefficients in the modified GTAP data base will be modified to be workable under the model, using the GTAP data aggregation program. For example, the GTAP data base does not have the sources of imported goods for final consumption and intermediate use, which are needed under the IMC model in this paper to account for the firm-level product differentiation. The sources of imports will be identified by modifying the aggregation program and running it with the SALTER-III data base. Third, the data for the domestic consumption of domestically-produced commodities will be added into the consumption components with the same origin and destination, and the data for domestically-produced production factors will be treated similarly. Fourth, we drop land for agriculture as a primary production factor, but the data for land will be added into the data for capital. The preference and technology parameters in GTAP data base are taken from the SALTER data set. These parameters can be aggregated, according to the aggregation of the data base. For a comparison between the GTAP model and our model, simulations will be performed under both the GTAP model and the model with imperfect competition. We will use the original GTAP data base and parameters for the simulation of the GTAP model. We will take the elasticities of substitution for imports from the GTAP parameter set for the perfectly-competitive sectors in our model. For the imperfectly-competitive sectors, a calibration procedure will be required. This procedure is described below. In addition, information about the number of firms will be needed for 52 the IMC sectors. We will follow Nguyen and Wigle (1992), by assigning some positive numbers, for example, 100, to each of the imperfectly-competitive sectors,31 and conducting a sensitivity test by assigning different numbers, for example, 25, 50, 200 and 1000. Table 4 summarizes the values of the elasticities of substitution used by other trade CGE modelers, and those taken for this paper. For easy comparison and the adjustment of differences for aggregated commodities, elasticities are given for 11 production sectors. We have a wide range of the parameters used by modelers. Mercenier and Schmitt used 2 - 4 for the elasticities for competitive sectors, while values of 5 - 10 were used for imperfectly-competitive sectors. But Brown and Stern (1989b) used a high elasticity of 15. This high elasticity was required by Brown and Stern, in order for the fixed value-added shares to be lower than one. As described above, GTAP has two sets of elasticities for traded commodities. The numbers on the left side of the GTAP parameter column are Armington elasticities (AE), which are used for dividing aggregated commodities into domestically-produced goods and imported goods. The right side contains the elasticities for imported commodities (EM). Only EM elasticities will be used in order to identify the sources of aggregated imports. Simulations using the GTAP model will be performed with the GTAP parameters. These numbers are moderate, compared to those in Mercenier/Schmitt and Brown/Stem. For the model with product differentiation at the firm level, we will use the elasticity of substitution for 3' One hundred firms per IMC industry may seem to be too many firms. But this can be a reasonable number of firms, since our model has high degree of aggregation, which is five sector per economy and two of them are IMC sectors. Nguyen and Wigle have a similar degree of aggregation and use 100 firms as the initial equilibrium number of firms per IMC sector. 53 imported goods for the base case. It is reasonable to compare results simulated from two models with the same elasticities. For the competitive sectors, there is no strict restriction on choosing the values of elasticities. Our model will take the elasticities for competitive sectors from the GTAP database. But the elasticities for the imperfectly-competitive sectors must be calibrated, such that the fixed value-added shares be less than one. Since the numerical results of the models will be interpreted in light of their chosen parameters and data, sensitivity tests should be conducted to check how robust the results are to the different choices of parameters. We will perform sensitivity tests with different sets of elasticities of substitution, in order to study the effects of the different magnitudes of the elasticities on the variables concerned, such as regional utility level and income. Alternative sets of elasticities will be taken from the calibrations of different fixed value-added shares. Sections 2.4 and 2.5 deal with the relationship between elasticities and fixed value added. As the elasticities of substitution get larger, fixed value-added shares will be decreased. 99% in the second column of the elasticities for this paper implies that the elasticities are calibrated, so that the fixed value-added ratios to the total value added are maintained below 99%.32 Even though we set the fixed value-added ratios below 99%, only a few of the 26 IMC sectors (2 IMC sectors * 13 regions in this model) will have fixed value-added shares near the 99% ratio, and the other sectors will have shares far below 90%. This happens, because 32 Note that the elasticities of substitution in the EM column are higher than 99% elasticities. Thus, we can use these elasticities for IMC sectors in this paper. 54 common elasticities will be used for all the regions, and the fixed value-added shares will depend on market shares and the number of firms, in addition to the elasticities of substitution. Later, we will study how the welfare effects change when changing the elasticities of substitution, and compare it with the results for our base-case parameters. The same calibration will be done for the 59% ratio. Another sensitivity test will be done by assigning different numbers for the [MC sectors, to determine how sensitive our results are to varying parameter values. Initially, it was assumed that each IMC production sector have 100 monopolistically-competitive firms. Alternative numbers for IMC sectors will be 25 , 50, 200, and 1000 to each IMC sector per region. Simulations will be conducted with these alternative nmnbers of firms, and compared to assess the robustness of model. 3.2. The Simulation Method Current CGE models can be grouped into the linearization school and the levels school.33 The levels approach is descended from the work of Scarf (1973). This approach solves non-linear general equilibrium problems, by furnishing subroutines of explicit algebraic formulas for indicating the levels of variables. The linearization school solves the problem by inverting the matrix of linearized equations. This is the Norwegian/Australian approach to CGE modeling, which builds on 33 See Hertel, Horridge, and Pearson (1993) for the grouping of the CGE models. 55 the Johansen (1960) approach, a method of solving systems of linearized equations by inverting a single matrix. The original Johansen approach did not involve updating, so that there could be substantial approximation errors for large perturbations. These linearization errors are likely to be significantly diminished by adopting a multi-step solution procedure and updating method (enhanced solution method) of Euler or Gragg. Utilizing the Johansen approach with an updating procedure, the Impact Project in Australia made a notable contribution to CGE modeling. Hertel et al. (1993) demonstrate that the linearized and non-linear schools of CGE modeling can be reconciled. They report reasonably similar simulation results from both linearized and non-linear approaches, as long as the data base is updated in the linearized case. All variables in the model used here are denoted in terms of proportional changes, which can be used to update coefficients read from the data source file. The linearization of the equations produces the relationships of percentage changes of variables, coefficients, and parameters. The initial data base will be systematically updated, and the updated data bases will be saved for the next-stage simulations, in order to provide more accurate solutions. With a relevant shock, coefficients will be updated as long as the iterative solution process continues. This can be explained by comparing the original Johansen method with Euler’s method. As shown in Figure 4, Johansen’s method is to calculate the direction of change (dy / dx ) at the initial point A, and to move from origin A to estimated point B(J) along the tangent line to the curve at A, while changing x from X(0) to X ( I ). But point B is the point to reach. Therefore, {Y (.0 - Y(1)} will be the approximation error associated with 56 Figure 4. IllustraticnnfErrlerMethod Y (0 Y(E2) Y(I) Y(0) kw Source: Harrison, Jill, and Ken Pearson (1993), QEMEACKDncumentatinn, P-2-24- 57 the Johansen method. One way of achieving a more accurate solution may be to divide the interval, {X(I) - X (0)}, into two pieces. First, move to point A, and, then, recalculate the slope and move along the line {C(2), 8(2)}, starting at C(2). The estimation error will be {Y(EZ) - Y(1)}, which is smaller than the Johansen error, {Y(.D - Y(1)}. Euler’s method is to calculate the derivatives, ( dy / dx ), at each sub-interval, to move that direction each time of the calculation of the derivative for a short distance. The above figure explains a two-stage Euler’s method. If the distance, {X ( 1 ) - X (0)}, is divided into N times, and if the direction in which to move is recalculated N times, we will have an N- stage Euler’s method. If N is sufficiently large, the solution can be close to the true solution Y(I). That is, performing the Johansen method multiple times will bring accurate solutions. The most important advantage of the linearized modeling is that it becomes simpler to formulate and modify models. Furthermore, benchmarking is also easy with the linearization method. In order to benchmark the data base to the linearized model, it is simply necessary to check that the model data base satisfies accounting identities, whereas non-linear CGE modeling requires that the calibration procedure correctly represent the systems of equations that are used in the later simulations. Therefore, we will solve the model in this paper, taking the linearization approach. The model will be written with GEMPACK (General Equilibrium Modeling PACKage), which is a suite of general-purpose economic modeling sofiware. A recent 58 version of GEMPACK can solve the non-linear problem, but it takes more computer time than the linearized version, with the same results. To save computer time, the non-linear equations will be linearized. In the linearized version, the variables should be interpreted as the percentage changes of the levels variables. 59 CHAPTER IV INTERPRETATION OF RESULTS This section reports on the results of F TA simulations with the CGE model. In the first case, we will discuss the welfare effects of APEC, obtained from the simulation of the perfectly-competitive GTAP model. Next, we will present the simulation results taken from our model with firm-level product differentiation, under both conjectures on rival firms’ behavior, described in the section for the production side. Then, we will report the results with different sets of parameters for the sensitivity tests. It was assumed that a F TA is created by removing all nominal tariffs existing between all member regions. To study the economic welfare changes of Asia Pacific Economic Cooperation and AF -1 1 F TA, we will compare the change of welfare after establishing new F TA. It will be interpreted that the member countries of a new FTA will be willing to create a free trade area, if welfare is improved in all of the member countries after a FTA simulation. Formation of a new F TA gives different welfare consequences to the regions involved in the FTA, depending on the net effect of trade creation and trade diversion. Trade creation means that a country will begin to import from the other member countries the goods which were previously produced domestically at higher costs. This happens, 60 due to the elimination of tariffs. This improves welfare, because of enhanced production efficiency. That is, the resources that were previously used for the good will be transferred to other sectors, thus raising the efficiency of production by using resources more efficiently. On the other hand, some countries may experience welfare losses due to trade diversion, which means that a country switches the source of imports from a more efficient producer outside the F TA to a less efficient producer in the F TA. For example, assume that, before NAFTA, the US. imports microwave ovens at lowest prices after import tariffs from Korea, which was the most efficient producer in the world. After NAFTA, U.S. imports the goods from Mexico, since Mexican producers can supply the goods with lower prices than Korean firms, because of the elimination of tariffs within NAFTA. The production efficiency gain will be enhanced with the introduction of imperfect competition into the competitive GTAP model, since the elimination of import tariffs will provide IMC firms with a larger market for their products. With a larger market, they will lower their markup, and they can produce goods at lower costs with the higher level of outputs. In addition to welfare changes, the model produces measures of the changes of regional income, the overall price indices, and firm outputs for selected scenarios. But our main concern is to look at the welfare changes that result from eliminating tariffs and NTBs in the region. Each model will have ten possible scenarios of grouping regions for a new FTA. Each of the ten scenarios will be simulated under all the combinations of 61 two alternative firm’s conjectures (Bertrand and Coumot), three different sets of elasticities (GTAP parameters, 99%, and 59%), and five different numbers of firms (100, 25, 50, 200, 1000). Our base case for our IMC model will be the simulation with GTAP parameters and 100 firms operating in each IMC sector. Possible groupings of regions for new F TAs in the Asia-Pacific region will be : (1) APEC : All regions in the Pacific-Rim are assmned to participate in Asia- Pacific FTA, which was discussed in Seattle, Washington, November 1993, and Bogor, Indonesia, November 1994. In our data aggregation, all the regions except ROW will be APEC members. (2) APEC - Canada (3) APEC - Mexico (4) APEC - Thailand Groupings from scenarios (2) - (4) will be the cases in which only one country withdraws from the full size APEC. Canada and Mexico are chosen for exclusion from the FTA in simulations (2) and (3) because they are members of the North America FTA, so that they have access to the US. market. If a new F TA is formed with the US, Canada and Mexico will compete with other new nations in the North American market. Thailand is chosen because it is a traditional agricultural country and its economy’s international trade is low relative to its GNP, which only leaves a small amount of room for improving its welfare with lower tariffs and non—tariff barriers. (5) APEC - CM : Canada and Mexico will not participate in the APEC. 62 (6) APEC - CMT : Canada, Mexico and Thailand will not involve in the APEC. An Asia-based FTA (AF-11) will be comprised of four Asian NIE economies (Hong Kong, Korea, Singapore, and Taiwan), four ASEAN economies (Indonesia, Malaysia, Philippine, and Thailand), China, and two Oceanic countries (Australia and New Zealand). (7) AF-ll + Japan : Japan joins AF-l l. (8) AF-ll + U.S. : U.S. joins the AF-l l. (9) AF -1 l : The formation of the full APEC in the area may be too extensive. In this case, only AF -1 1 members are assumed to participate to form a new FTA. (10) AF-ll - Thailand : Thailand withdraws from the AF -1 1. These scenarios are designed to study the consequences of removing import barriers within regions involved in each scenario, scrutinizing carefully what happens to welfare for the possible member regions. The scenarios are ordered according to the number of countries involved in new FTA (i.e. APEC has the most countries, and AF-l l-Thailand has the fewest). The model is solved using a computer with 75 Megahertz Pentium processor and 32 Megabytes of memory. With the high capacity of memory, the model can be solved in around 26 minutes for one scenario of a new FTA in the Asia-Pacific region. 63 4.1. The Simulation of GTAP model Table 5 reports the welfare effects, taken from the simulations of the perfectly- competitive GTAP, with GTAP parameters. Note that it is not necessary to specify the number of firms here, since the GTAP is assumed to have perfect competition for all of the production sectors. The first five groupings are in the top of the table, and next five groupings are in the bottom. The first column displays the welfare effects under the full APEC. Welfare is measured by percentage changes in the utility of the representative consumer. It is found that four countries will suffer a deterioration of their welfare as a result of APEC. Two ASEAN countries and two NAFTA countries seem not to be positive to the new APEC, which implies that the formation of APEC is not likely, if we are guided by the results of the competitive GTAP model. Canada’s withdrawal from APEC will worsen its welfare, while no significant effects occur to the other countries. Canada will have negative benefit changes in 9 cases out of 10 groupings, which implies that Canada will be reluctant to form a new FTA in the Pacific basin, from the economic point of view. Mexico has similar results as Canada’s withdrawal. An interesting aspect for Mexico is that Mexico will have a possibility of improving its welfare with the variations of AF-1 1. As shown in Table 3a, Canada and Mexico have traded with Asian nations, except Japan, which implies that they are not likely to collect gains from a new FTA. And these two countries export large portions of their total exports into the US. They may lose the North American market, 64 when joining the APEC. The GTAP model also simulates this assertion (The column of APEC in Table 5 has negative welfare changes for Canada and Mexico). Evidently, Thailand is a loser for all 10 groupings of countries within the GTAP model. Thailand is a traditional agricultural country, and Thailand’s exports are highly concentrated to a few countries, such as US, Japan, and Taiwan, with low levels of trade with Asian neighbors, such as Indonesia, Philippines, and Korea.34 Under APEC and AF- 11, Thailand loses 5 % of its welfare. It is expected that Thailand will minimize its welfare losses, if it does not involve in the APEC. Reading two columns of APEC - CM and APEC - CMT, APEC does not seem to be supported in the perfectly-competitive GTAP model, since Malaysia worsens its welfare, in addition to three nations’ objection to APEC. Japan improves its welfare when it joins AF -1 1, while the US. will worsen. On the contrary, it is expected that US. benefits, when US. joins the AF -1 1. Japan and US. trade substantial volumes of goods with Asian nations. Under the competitive GTAP model, it is inferred that US. and Japan compete for the Asia market. Under AF- 11+Japan, the US. is estimated to export less manufactured goods to Korea, Malaysia, Philippines, and Thailand, while Japan will increase its LMF and RPR commodities to these Asian nations. Especially, Korea imports more RPR (77%) and LMF (91%). On the contrary, under AF -1 1+U.S.A., the opposite case happens. Only Taiwan/Singapore remain stable in trade values under two free trade scenarios. It is noticeable that 3’ Thailand’s biggest trade partner among APEC member nations is Japan, (The total value of Thailand’s trade with Japan is US. 3 18730 million). The US. and Taiwan/Singapore are the second and third trade partners, respectively. Trade between Thailand and other nations in the possible APEC is less than US. $ 3 billion. 65 Australia/New Zealand face a substantial decline in their economic well-being, from a 2.39 % gain under AF-ll + Japan to a 0.06 % gain under AF-ll + US. It can be guessed from this point that Australia/New Zealand compete with the US. in the Asia market, especially for agriculture and food products. When AF-ll opens free trade with the US, there will be decreased exports of agricultural products from Australia/New Zealand to Korea and Taiwan/Singapore by 75%. Under AF -1 1+Japan, Australia/New Zealand is expected to export more agricultural products to Japan by four times. The formation of AF-ll is anticipated to have trouble within its member nations, since Thailand has negative welfare changes, regardless of the grouping of regions. The column of AF -1 1 - Thailand shows that all countries in AF-ll are winners, if they form a F TA of AF -1 1 without Thailand. It is still expected that either Japan or the US. will join the AF -1 1 without Thailand, but not 131th Japan and the US, since Malaysia will not be better off, as shown in the column of APEC - CMT. From Table 5, we see that Korea is expected to collect the highest gains, ranging from 6.6 % to 8.9 %, throughout all of the ten scenarios. Following Korea, Taiwan/Singapore seems to be the second biggest winner. These nations have high ratios of total values of exports and imports to gross domestic output, especially with US, Japan, and ASEAN nations.35 Under APEC, the light manufacturing sector (LMF) in Korea and Taiwan/ Singapore are simulated to increase their exports to Japan by factors of nearly three and nearly two, respectively. Similar trade patterns are found for the US. 3’ See Sakong (I993). 66 Resource-rich ASEAN nations export resources and intermediate goods to the two NIE nations, and import manufactured goods from the NlEs. Japan and Australia/New Zealand have wide ranges of welfare changes, depending on the combinations of nations for a F TA. Japan experiences positive welfare gains when it participates in any form of free trade area, while losing welfare, otherwise. This means that Japanese firms are taking big trade creation with FTA. Under APEC, Japan is expected to increase its exports of manufactured commodities by 22-24% (RPR and TME sectors) and 55% (LMF). Other nations than these countries are expected to be affected by less than 1% of welfare. For example, the US. is anticipated to have negligible welfare changes, ranging from - 0.0309% to 0.067%. This can be explained with that the US. has a big domestic market in its territory, that is, 94.4% of the goods made in America are sold in the US. market for domestic uses. Table 6 reports equivalent variations simulated from 10 scenarios of APEC and AF -1 1. The equivalent variations are measured in US. $ million. Equivalent variations are directly related to the percentage changes of regional utility and the initial GNP levels, as shown in eq. (14). Thailand loses by more than 5 and 4 billion US dollars under the full-size APEC and AF-l 1, respectively. The welfare of the rest of world (ROW) declines by US. $31 billion under the APEC scenario. This is due to the trade diversion under a FTA in the region. Korea experiences the highest welfare improvement, of $23 billion, under AF -1 l + US. It can be inferred that Korea would prefer to become involved in a free trade agreement with the US, since Korea collects a relatively lower welfare gain under AF -1 l + Japan than AF-ll + US. Japan collects the 67 highest gain of $83 billion, if the full-size APEC is realized. This means that Japanese firms could enjoy the highest degree of trade creation effects under the largest FTA (which is APEC) in the Pacific-rim region. We can see that Japan and the US. compete in the Pacific-rim region, by noting that Japan and the US. each experience welfare declines, under the scenario in which one of them is not involved in the new F TA, and the other does participate, such as AF-ll + Japan and AF-ll + U.S. Tables (7) and (8) present the changes of income and prices, respectively, obtained from the simulation of the GTAP model. Some countries, such as Australia/New Zealand and Taiwan/ Singapore, are expected to have positive changes of the general price level with the elimination of tariffs in the APEC region. This happens because of the increases in the demand for goods and services. Canada and Thailand have large declines of prices with a FTA in the Pacific-rim region. For example, Australia/New Zealand and Taiwan/ Singapore have positive changes of final consumption for services under full-size APEC, while composite price of services will go up in the nations. On the other hand, Canada and Thailand are expected to have the opposite results. 68 4.2. ImperfectIy-Competitive Model The model with imperfect competition and scale economies will be simulated with base case parameters of GTAP elasticities and the assumption of 100 firms operating in each of [MC sectors. Tables 9 and 10 summarize the changes of welfare from 10 groupings for new FTAs under Coumot and Bertrand, respectively. The results from GTAP and the IMC models differ significantly. First, it should be noted that welfare changes under IMC model are significantly larger than under the competitive model, as reported in Table 5. For example, some regions are expected to have two-digit welfare gains : Korea improves welfare by at least 13% under all of ten of the scenarios of the IMC model. Taiwan/Singapore gains by more than 11% of welfare under 7 scenarios out of 10 groupings of countries / regions. (The GTAP model gives 9% at most.) Significantly bigger welfare gains were obtained from the IMC model in most cases.36 The IMC model is defined to have a fixed factor, and the removal of protection will result in the firm producing at lower unit cost with higher levels of outputs. And more competition will drive firms to lower markups over marginal costs. Thus, the scale economies and efficiency gains will be realized. Second, the smaller economies such as Korea and Taiwan/Singapore, are affected more, when we move from GTAP to the IMC model. On the other hand, large economies (US. and Japan) are affected relatively less, even though their welfare changes are 3" Under AF-ll and some of its variations, the differences between the welfare changes with the PCM model and the welfare changes with the IMC model are small for Canada, Indonesia, and the US. 69 increased a little under the IMC model, relative to the welfare changes of the GTAP model. This is explained by the fact that bigger countries have already exploited scale economies relatively more,37 even without the policy change. Third, China/Hong Kong, and Australia/New Zealand are to collect substantially higher welfare gains for the new FTA, even though Korea and Taiwan/ Singapore are still the biggest winners. Under the GTAP model, those regions have moderately low welfare gains. Relatively, Japan and the US. are estimated to remain at the similar levels of welfare under both GTAP and the IMC models. Fourth, Malaysia will change to be a strong supporter of the formation of a FTA at any level. Under the GTAP model, Malaysia experiences welfare losses with any scenarios of APEC, but positive gains at all variations of AF-1 1. The IMC model predicts positive welfare gains for all of 10 scenarios. Fifth, all of ASEAN and the NIEs except Thailand prefer to establish a FTA with the US. rather than with Japan, under the competitive GTAP model, since they will experience higher benefits under AF-ll + U.S. than AF-ll + Japan, if they are supposed to have free trade with only one of Japan and the US. This also applies to the predictions from the [MC model, except China/Hong Kong. China/Hong Kong will have higher welfare gains under AF -1 1 + Japan than AF -1 1 + US. This can be explained with one of common aspects of these economies, that is, most countries in the region export more to the US. than to Japan, but they import more from Japan than from the US. Thus more trade creation effects will be realized, when these nations are involved in a FTA with the 37 See p. 13 in Brown and Stern (19893), for the smaller level of scale economies in the US. US. US. lapa S 56 trade APE relat vvelf size pred seen Can: invo in th num AF. POSII 70 US. China is different, because of higher trade dependence with Japan rather than the US. According to Table 3a, the total value of imports and exports between China and Japan was US. $ 58089.3 million, and that of trade between China and the US. was US. $ 56376.5 million. But other nations, such as Korea and Taiwan/Singapore, have higher trade with the US. than Japan. Sixth, Indonesia experiences relatively variable welfare changes under each of the APEC scenarios and AF -1 1 scenarios of the GTAP model, while the IMC model predicts relatively stable welfare changes. But the Philippines is expected to have relatively stable welfare changes under both GTAP and the IMC model. From the results of both GTAP and the IMC model, we can conclude that the full- size APEC is not likely to be established in the Pacific-Rim area. But the IMC model predicts a higher possibility for establishing the APEC, since only Mexico and Thailand seem to experience welfare losses under the IMC model. The GTAP model suggests that Canada and Malaysia, in addition to Mexico and Thailand, would not want to get involved in the F TA. Thailand’s welfare losses are lower, when she does not participate in the APEC and AF —1 1, than with alternative choices. This is learned by comparing numbers in the columns of APEC and APEC - Thailand, and the columns of AF -1 1 and AF-ll - Thailand. The opposite case happens to Canada. That is, Canada is expected to be better off (minimize welfare losses), if Canada participates in the APEC. From Table 5, Canada’s welfare loss is 1.43% under APEC-Canada, while their welfare loss will be decreased to 0.6% under the full-size APEC. Canada under IMC model collects even positive welfare gains. AF -1 1 seems not to be realized under either the GTAP or the [MC models. l - Thailai since all gains unI APEC-C in the Al Asian l\'l China’li- does not 11 regioi the US. I COmpetit PYEVious firms is j markup . trade lib Illis wII ,,\ Compa _expeqed t gimme Wile ere e 71 models, because of Thailand’s losses for the scenario. But both models show that AF -1 1 - Thailand is an economically-desirable scenario for a new FT A in the Pacific basin, since all member nations under AF -1 l-Thailand are estimated to have positive welfare gains under the GTAP and IMC model. Under the [MC model, a candidate for a wide-range F TA in the area will be APEC-CMT, which contains both Japan and the U.S. But it is not likely that the regions in the AF -1 1 could agree to open their markets with only one of Japan and U.S., since Asian NIEs prefer to establish a F TA with U.S., but Australia/New Zealand and China/Bong Kong want a FTA with Japan rather than the U.S.38 The competitive model does not support the FTA grouping of APEC - CMT, but it seems to be possible that AF- 11 regions excluding Thailand might agree to eliminate import tariffs with either Japan or the U.S. Table 10 presents the changes of welfare from the simulations of the imperfectly- competitive model with our base-case parameters and the Bertrand conjecture. In the previous section, it was described that the perceived demand elasticity of the Bertrand firms is larger than that of the Coumot firms. As a result, Coumot firms will have higher markup than Bertrand firms. Therefore, generally, the economic efficiency gains from trade liberalization will be larger under Coumot approach than the alternative approach. This will be clear, by comparing Table 9 and Table 10.39 All of the discussions above 38 Compare utility changes under AF-11+Japan and AF-11+USA in Table 9. Taiwan/Singapore are expected to have 12.4% under a FTA with the U.S. but their welfare declines to 6.13% when only Japan joins AF -1 1. On the contrary, Australia/New Zealand are estimated to increase welfare by almost three times, when they establish a FTA with Japan (6.87%) rather than the U.S. (2.39%). 39 There exist two exceptions for the scenario of AF -1 l, but they are not substantial. aboul differ \xith ' SCCIO' full-s marki of IO Comb and 1( Welfar examp Percen Similar utility . reducu' dlSCuss finns’ ; 72 about Table 9 apply also to Table 10. Our IMC model does not show substantial differences of the changes of welfare under Coumot and Bertrand conjectures, simulated with GTAP parameters and the assumption of 100 firms actively operating in the IMC sectors of each region. Most regions in our model are expected to have decreasing markup rates under the full-size APEC. But the reductions are very moderate, in the ranges of -.003% to -0.32%. The markup rates of Taiwan/ Singapore are not affected at all, while ROW has positive markup rates. Tables 11 and 12 summarize the estimated income changes with the simulations of 10 FTA scenarios, under our base-case parameters and two alternative conjectures. Combining the income changes of Tables 11 and 12 with the welfare changes in Tables 9 and lo, it can be seen that the percentage changes of income are positive, if those of welfare are positive.40 Different patterns can be found for regions in the model. For example, Taiwan/Singapore has higher income changes than welfare changes by 3-4 percentage points, in the ten groupings of FTA in the area. China/Hong Kong has a similar pattern. On the contrary, in Korea and Indonesia, the percentage changes of utility are higher than those of income, for 9 out of 10 groupings. This is due to price reductions in these countries, as shown in Tables 13 and 14. The previous page had the discussion about the relative sizes of welfare changes under imperfectly-competitive firms’ alternative conjectures: Coumot and Bertrand. The estimated income changes are ‘0 One component of AF-ll scenario has a opposite direction for U.S. under the Bertrand conjecture in Table 12. But this is negligible. larger, Wilt" exception t The under the (I the main el member co imports wi for consun‘ production linked to a This is foui Thailand, ln s such as Au incl-eases 0 among fim of these re; thus emPlc the Wage r. This affeCI c08t inCrea domestic i; 73 larger, when employing the Coumot assumption than the Bertrand assumption, with the exception of Australia/New Zealand in the AF -1 1 scenario. The estimated changes of consumer price indices are reported for the base case under the Coumot conjecture (Table 13) and the Bertrand conjecture (Table 14). One of the main effects of eliminating import tariffs is the price reduction for imports from the member countries. As described in the section on price linkages, section 2.7, the price of imports will be just the world price, if no tariffs are charged. Then, the prices of imports for consumers and producers will be lowered, and firms will be able to reduce their production costs, because of lower costs for intermediate goods. Lower costs will be linked to a lower supplier’s price. Therefore, overall consumer price indices will decline. This is found in some countries, for example, Indonesia, Japan, Korea, Mexico, and Thailand. In spite of this import price reduction, with the removal of tariffs, some countries, such as Australia/New Zealand, China/Hong Kong, and Taiwan/Singapore, experience increases of their consumer price indices. This can be explained by the competition among firms for hiring labor and capital. With a formation of a new FTA, domestic firms of these regions will increase output and (or) new firms will enter the production sector, thus employing more labor and capital. If labor is demanded more in an economy, then the wage rate will be increased, which will induce substitution between labor and capital. This affects the production costs for their products. In a general equilibrium model, the cost increase will be spread over the whole economy. The rise in the wage increases domestic income, since national income consists of the sum of the returns to the 74 production factors and tax/tariff revenues. As shown in Tables 11 and 12, countries with positive price changes have relatively higher income increases, as in Australia/New Zealand. On the other hand, the rise in income will increase demands, and thus increase prices. Tables 15 - 18 contain the percentage changes of the numbers of IMC firms. Tables 15 and 17 report the changes of the numbers of resources, plastic, and refinery (RPR) firms under the Coumot conjecture and base-line parameters. Tables 16 and 18 are under the Bertrand conjecture with base-case parameters. There is no pattern of changes, but it can be said that if the numbers of firms decrease, then the [MC sector can exploit scale economies, but the economies sacrifice the diversity of goods. Tables 19 - 28 are presented for conducting sensitivity tests for chosen parameters of the elasticities of substitution and the number of firms. First, the elasticities of substitution are altered to produce a 5 9% fixed value-added share, shown in Table 4, while we leave the number of firms unchanged. The 5 9% elasticities are larger than the GTAP elasticities. Each table reports the changes of welfare for 10 FTA scenarios under Coumot (Table 19) and Bertrand (Table 20). Both tables predict the same directions of welfare changes, with no exception. In most cases, the welfare changes are estimated to be lower in the simulations with 5 9% elasticities than those with the GTAP elasticities, because of the higher elasticities of substitution for the 5 9% fixed value-added share. The next test will be to assume lower elasticities of substitution, while still keeping constant the number of firms. We find a set of elasticities, such that the common elasticities assign the fixed value-added share to be less than 99%, which is the highest share. lower case. 2 the 53: only C that 0 Chang Other with more 1001 75 share, since the fixed value-added share cannot be larger than one. That is, we assume lower elasticities of substitution than GTAP. It is simulated, similar to that of the 5 9% case, as reported in Tables 21 and 22. It is found that the patterns of welfare changes are the same as in Tables 19 and 20 for Coumot and Bertrand conjectures, respectively. The only differences are lower changes of welfare, as expected. Thus, it can be concluded that our results are robust to the choices of the elasticities of substitution. Another question may be raised for the number of firms. The first test is to change the number of firms from 100 (central case) to 25, keeping the GTAP elasticities. The simulation results under alternative conjectures are reported in Tables 23 and 24. Other simulations were performed for 50 firms, in order to study how robust the model is with respect to the number of firms (reported in Tables 25 and 26). The welfare changes move in the same direction and are of about the same magnitude as in the simulations of 100 firms. As expected, the welfare changes are larger in the simulations with 25 firms than in those with 100 firms, since lowering the numbers of firms in the model reduces the perceived demand elasticities. And the simulations with Cournot give higher welfare changes than Bertrand. It is also confirmed that the results with 25 firms are not substantially different from those with 50 firms, by comparing Tables 23/24 and 25/26. Increasing the numbers of IMC firms from 100, to 200 and to 1,000,“ presents the same patterns of welfare changes as the simulations of 50 firms and 100 firms. Tables 27 and 28 report the simulation results of 200 firms under Coumot and Bertrand, respectively. It is confirmed again that the IMC model used in this paper is stable for the parameters. 4' Simulations with 1,000 firms are not reported in this paper. 76 Simulations are performed for three different numbers of firms with 5 9% and 99% elasticities, and we have similar results, as described above. Tables 29-38 summarize the percentage changes of sectoral output. The first five tables (Tables 29-33) are estimated production output changes by sectors and regions in the model, obtained from the simulation of the competitive GTAP model. The output changes for the IMC model are reported in Tables 34-38. Generally, the output changes in the IMC model are larger than in the GTAP model. 77 CHAPTER V CONCLUSION The main results of the paper can be summarized as follows : (1) The groupings of regions seem to be important for a formation of free trade area in the Pacific-Rim region, which contain countries with diverse backgrounds. (2) The introduction of imperfect competition into the model projects large discrepancies between the simulations from the GTAP model and the IMC model. (3) No substantial differences are raised from alternative assumptions about the firm’s expectations regarding on rival firm’s behavior. (4) The IMC model in this paper is very robust in the choices of parameters. Both GTAP and the IMC models predict no formation of a free-trade area for the entire Asia-Pacific region. The GTAP model is more pessimistic about APEC than is the [MC model, since more countries are losers in the GTAP simulations. On the contrary, the IMC model points out the possibilities of a variation of APEC, by excluding a couple of countries. For example, a small-size APEC may be formed by excluding Canada, Mexico, and Thailand. An Asia-based FTA, AF-l 1, is demonstrated as a candidate for an alternative free-trade area under imperfectly-competitive CGE modeling, at the expense of Thailand, since the [MC model gives smaller losses to Thailand than does the GTAP 78 model. Negative welfare changes for Thailand seem to come fiom her industry structure. Thailand’s trade dependence (measured by the ratio of the total value of exports and imports to the gross domestic products) is the lowest among the AF -1 1 regions.42 Thus, they have a small mechanism for improving welfare, with the introduction of scale economies, and firm-level product differentiation. Another point to notice is that the IMC model simulates substantially larger welfare changes than the GTAP model. This makes big winners of the NIE countries, for example, Korea, Taiwan/Singapore, and China/Hong Kong. These economies will be active supporters of a F TA in the region under the IMC model. In simulations with imperfect competition, the specifications of a firm’s conjecture on its rival’s behavior do not seem to be important in evaluating the welfare changes in our model. In most cases, the Coumot assumption presents larger values for the welfare changes, but the differences are negligible. In our model, the GTAP elasticities are used as a central case, and the sensitivity tests confirm the stability of the model with respect to the parameters. A similar conclusion was reached from sensitivity tests with respect to the number of firms. A couple of qualifications should be pointed out. First, in this model, the complete elimination of import tariffs and non-tariff barriers means the formation of free trade area. But the welfare changes should be interpreted as an upper bound for the economic benefits that the model predicts, because NTBs are not likely to be removed completely, taking various forms of security regulations and government procurement practices. Second, the benefits of scale economies cannot be fully captured by a static ‘2 See Shibusawa et al. (1992), for detailed description. CGErnod dynamic r HAmm 79 CGE model, since the regional economies will be growing with a new FTA. Thus, a dynamic modeling is suggested, for full estimation of the welfare effects under a new FTA in the Pacific basin. Table 1. Members of APEC Regions in GTAP Mappings of regions in our study Australia (1939)* Australia Australand (ANZ) Brunei (1989) Rest of World (ROW) Canada (1889) Canada Canada (CND) Chile (1994) Rest of World (ROW) China (1991) China China (CHK) Hong Kong (1991) Hong Kong China (CHK) Indonesia (1989) Indonesia Indonesia (IND) Japan (1989) Japan Japan (JPN) Korea (1989) Korea Korea (KOR) Malaysia (1989) Malaysia Malaysia (MLS) Mexico (1993) Mexico Mexico (MXC) New Zealand (1989) New Zealand Australand (ANZ) Papua New Guinea (1993): Rest of World (ROW) Philippines (1989) Philippines Philippines (PHL) Singapore (1989) Singapore Singapore (TSP) Taiwan (1991) Taiwan Taiwan (TSP) Thailand (1989) Thailand Thailand (THL) United States (1989) United States United States (U.S.A.) Argentina Rest of World (ROW) Brazil Rest of World (ROW) EEurope and Soviet# Rest of World (ROW) EEC Rest of World (ROW) MEast and NAfrica& Rest of World (ROW) Other Latin America Rest of World (ROW) Other Regionso/o Rest of World (ROW) South Asia Rest of World (ROW) SS Africa@ Rest of World (ROW) ’. The numbers after the countries in the APEC column indicate the years in which they joined the APEC. #. EEurope and Soviet : Eastern Europe and Former Soviet Union &. MEast and NAfrica : Middle East and North Africa %. Other Regions : Regions not elsewhere classified @. SS Africa : Sub-Saharan Africa 81 Table 2. ListsnfhdnstdeaLCommodifiesaniMappingsinmSmdx Listings of Industries in GTAP Mappings of Industries in This Paper Paddy Rice Wheat Grains (except rice and wheat) Non-grain Crops Wool Other Livestock Forestry Fishery Coal Oil Gas Other Minerals Processed Rice Meat Product Milk Products Other Food Products Beverages and Tobacco Textile Wearing Apparel Leather, etc. Lumber and Wood Pulp, Paper, etc. Petroleum and Coal Products Chemicals, Rubber, and Plastics Non-Metallic Mineral products Agriculture (AGR) Agriculture (AGR) Agriculture (AGR) Agriculture (AGR) Agriculture (AGR) Agriculture (AGR) Agriculture (AGR) Agriculture (AGR) Resource, Chemical, and Refinery (RPR) Resource, Chemical, and Refinery (RPR) Resource, Chemical, and Refinery (RPR) Resource, Chemical, and Refinery (RPR) Light Manufacturing (LMF) Light Manufacturing (LMF) Light Manufacturing (LMF) Light Manufacturing (LMF) Light Manufacturing (LMF) Light Manufacturing (LMF) Light Manufacturing (LMF) Light Manufacturing (LMF) Light Manufacturing (LMF) Light Manufacturing (LMF) Resource, Chemical, and Refinery (RPR) Resource, Chemical, and Refinery (RPR) Resource, Chemical, and Refinery (RPR) Primary Ferrous Metals Resource, Chemical, and Refinery (RPR) Non-ferrous Metals Resource, Chemical, and Refinery (RPR) Fabricated Metal Products Transportation, Machinery, and Equipment (TME) Transport Industries Transportation, Machinery, and Equipment (TME) Machinery and Equipment Transportation, Machinery, and Equipment (TME) Other Manufacturing Light Manufacturing (LMF) Electricity, Water and Gas Services (SVS) Construction Services (SVS) Trade and Transport Services (SVS) Other Services (private) Services (SVS) Other Services (government) Services (SVS) Ownership of Dwellings Services (SVS) 82 Table 33 WW SouroeDestination Astral/NZeal China/HK Canada Indonesia Japan Korea Malaysia AstralialNZeala nd 537969 3569.03 784.889 1652.21 17874.7 3712.23 1215.74 China/Hang Kong 3706.4 1.01 E+06 2844.13 1297.21 20976.9 5567.54 1558.87 Canada 923.93 2861.38 959712 463.771 8760.17 1850.56 278.756 Indonesia 1054.16 2706.8 282.268 188613 13213 2331.5371 1 496.969025 Japan 9275.45 371 12.4 8909.02 6585.65 6.55E+06 21358.5 8589.32 Korea 1701.56 8946.52 1861.29 2088.31 15581.9 571107 1221.73 Malaysia 790.91 2374.72 435.93 648.49 7216.59 1630.69 83768.8 Mexico 107.83 387.28 2409.62 85.11 2459.61 257.1 11.864 Philippines 164.38 696.48 227.32 100.54 3370.02 405.313 173.931 ROW 16866.2 38678.7 19998.7 9384.21 1 12618 27374 6217.39 Thailand 650.08 2855.25 470.44 380.20 7519.96 711.38 862.705 Taiwan/Singapore 4439.58 24616.8 2588.71 3269.84 19460 4409.85 10261 .2 USA 14332.5 21534.7 81040.9 4289.01 75880.1 20845.6 5013.89 SouroeDestination Mexico Philippines ROW Thailand Taiwan/Sing USA Astralia/NZealand 295.303 586.356 17665.7 887.789 3877.67 5453.69 China/Hang Kong 1233.91 1030.7 51788.4 2017.26 7549.55 34841.8 Canada 1096.27 221.113 21324.3 380.585 1319.91 103615 Indonesia 108.17 202.31 7335.14 373.82 4463.18 4459.33 Japan 4406.93 3456. 36 127620 1 1210.1 392 50 103244 Korea 945.49 813.66 33670.2 1790.99 6346.63 19091.9 Malaysia 166.255 389.247 9798.83 1402.82 1 1334 8107.88 Mexico 472546 10.846 10548.6 70.779 141.98 38431 Philippines 37.899 81 582.8 3034.9 237.493 683.584 4509.97 ROW 13462.3 4583.76 1.81 E+07 1 1358.6 32305.9 201742 Thailand 137.463 114.713 12049.3 165685 3415.49 7885.96 Taiwan/Singapore 31 1 .818 1653.97 47897.4 6077.78 475198 40618.5 USA 42577.2 2834.29 257457 5005.85 29164.4 9.49E+06 83 Table 3b WWW Mexico Philippines ROW Thailand Taiwan/SP USA World 54922.3 13641.9 1400700 37053 173783 559976 3205710 Mexico ROW Thailand Taiwan/SP USA * Total of Shares may not be one, due to rounding errors. 84 Table 4 BarametetszLtheElasticinmeuhstinrtion Mercenier- Brown- GTAP Model with Firm-Level Product Schmitt Stern Parameter (AEI (EM) I EM) (99 96) (59 %) Agriculture 2 15 2.48 4.72 4.72 4.72 4.72 Forestry, Fishery 2 15 2.48 4.72 4.72 4.72 4.72 Processed Food 4 15 2.53 5.82 5.82 5.82 5.82 Lumber 4 15 2.53 5.82 5.82 5.82 5.82 Paper Product 4 15 2.53 5.82 5.82 5.82 5.82 Textile. Leather 4 15 2.53 5.82 5.82 5.82 5.82 Chemical 5 15 2.32 4.81 4.81 4.40 7.30 Resource 5 15 2.32 4.81 4.81 4.40 7.70 Machinery 10 15 l 3.42 6.91 6.91 5.10 8.10 Vehicle 10 15 3.43 6.91 6.91 5.10 8.10 Service 2 15 1.94 3.92 3.92 3.92 3.92 AE : The elasticity of substitution for Armington specification in GTAP. EM : The elasticity of substitution between imported goods in GTAP. Table 5 1 .09524 1.15251 o .'. 1 .09464 'an 1.09273 1.15233 AstralialNZealand China/Hang Kong 0.834849 0.808756 0.832409 0.863859 0.805972 Canada -0.60126 -1 .43455 -0.57617 -0.61575 -1 .39339 Indonesia 0.63141 1 0.65044 0.632843 0.560754 0.652738 Japan 2.667 2.55356 2.66271 2.5577 2.54889 Korea 7.45897 7.35196 7.39872 6.78335 7.29244 Malaysia -0.80981 -0.77479 -0.80853 -1 .55657 -0.7701 1 Mexico -0.81881 -0.80155 -1.01526 -0.8348 -0.924 Philippines 0.654334 0.632357 0.669579 0.653426 6.49E-01 ROW -0.34027 -0.32276 -0.32913 -0.32716 -0.31 176 Thailand 4.98103 4.96538 4.98248 -2.6921 1 4.97238 Taiwan/Singapore 4.12822 4.1 1727 4.15572 3.83373 4.14618 USA 0.0466 0.0562 0.02.7 0.0447 0.0333 APEC-CMT AF-1 1+Japan AF-1 1+USA AF -1 1 AF-1 1-Thailand AstralialNZealand 1 .15005 2.38917 0.0631 0.379166 0.295108 China/Hang Kong 0.833804 1.09651 1 .78369 1.21916 1 . 16231 Canada -1 .39381 -0.09.79 0.504199 -0.0124 -0.0226 Indonesia 0.581256 0.935309 1.79841 1.61535 1.47456 Japan 2.4382 0.404067 -0.4201 -0.24419 -0.20901 Korea 6.62229 6.64 8.94003 8.15 7.14 Malaysia -1 .52679 0.841 509 1 .74271 2.05252 0.807879 Mexico -0.934 0.0432 0.738054 0.041 5 0.0292 Philippines 0.647602 0.781469 1 .2795 0.410352 0.248741 ROW -0.29885 -0.138 -0.0846 -0.0464 -0.0444 Thailand -2.5800 4.0800 4.16084 4.1600 -0.460 Taiwan/Singapore 3.84577 2.07807 6.38234 3.04665 2.1 897 USA 0.0309 -0.159 0.0685 -0.0312 -0.0279 Table 6 3214.33 U h. ' ;1:J:{o!m_;1:1deltmdflialnm—Jfldédl AstralialNZealand 3381 .45 3.21 E+03 3207 3380.94 China/Hang Kong 3868.25 3747.83 3856.99 4002.09 3734.99 Canada -3177.07 -7610 -3044.12 -3250 -7390 Indonesia 751 .786 774.373 753.485 667.896 777.1 Japan 83351.4 79850.5 83219.1 79978 79706.4 Korea 19518.2 19248.4 19366.2 17808 19098.1 Malaysia -500.481 478.752 499.684 -965.626 475.85 Mexico -2414.12 -2360 -2996.28 -2460 -2730 Philippines 319.063 308.383 326.473 318.629 316.47 ROW -31 895.3 -30300 -30800 -30700 -29200 Thailand -5061.1 3 -5044.81 -5062.63 -2703.27 -5050 Taiwan/Singapore 9723.47 9698.1 9 9786.92 9042.72 9764.97 USA 2450 2960 1420 2350 1 750 Am PIP-11 AW AstralialNZealand 3374.29 6967.13 186.062 1 1 16.79 869.597 China/Hang Kong 3863.42 5073.93 8225.96 5638.18 5376.75 Canada -7394.44 -516 2649.51 -65.5 -119 Indonesia 692.248 1111.96 2128.92 1913.97 1748.37 Japan 76286.4 12770.3 -1 3331 .5 -7742.48 6625.74 Korea 17398.8 17400 23230.3 21300 18700 Malaysia -947.009 515.841 1063.43 1250.61 495.334 Mexico -2760 127 2159.14 122 85.9 Philippines 315.802 380.832 621.974 200.345 121.554 ROW -28007.4 -12900 -7920 4340 4150 Thailand -2590 41 30 4209.72 4200 456 Taiwan/Singapore 9070.65 4943.69 14871.3 7213.61 5206.38 USA 1624.93 -8340 3600 -1640 -1470 Table 7 ;.'J=( ChangesoflncomelfilARModeLfiIAKEarameter] -1:J:10!m_;1:1 :10! .'r ms flifilfl'film-flflifid r'. AstralialNZealand 3.17905 3.38667 3.1 5 3.09299 3.35331 Chinleong Kong 0.669072 0.715333 0.655933 0.748983 0.701 557 Canada -2.73135 -5.25 -2.69884 -2.82 -5. 1 7 Indonesia —0.0497 0.0112 -0.0451 -0.23953 0.0176 Japan 4.14662 4.12974 4.12142 3.78829 4.10207 Korea 5.97049 5.81834 5.86138 4.92166 5.70996 Malaysia -2. 16798 -2.09749 -2.17878 -3.22788 -2.10401 Mexico -3.04003 -2.99 4.57188 -3.16 4.28 Philippines 1.49538 1.48895 1.51391 1.48863 1.51044 ROW -2.25 -2.19 -2.21 -2.25 -2.16 Thailand -9.07509 -9.01221 -9.0756 -5.93647 -9.01 Taiwan/Singapore 4.48203 4.49909 4.50853 3.8428 4.52701 USA -0.899 -0.974 -1.05 -0.998 -1.15 M AF-‘l HUSH—A1211 W AstralialNZeaIand 3.2676 8.22548 0.4429 1 .52026 1 .1356 China/Hang Kong 0.778462 4.23593 3.91775 3.38319 3.32699 Canada -5.22092 -0.852 1 .68222 -0.283 -0.351 Indonesia -0.17425 1 .73504 2.48541 2.63958 2.2866 Japan 3.74118 0.630101 -1.16252 -0.71106 -0.64475 Korea 4.66715 6.0600 9.45771 8.2200 6.5600 Malaysia -3.17875 0.807745 2.25645 2.88128 1.01787 Mexico 4.3800 -0.473 3.23177 0128 -0.215 Philippines 1.50064 4.36056 4.12017 2.2759 1.88826 ROW -2.16143 -0.816 -0.353 -0.272 -0.318 Thailand -5.7200 -6. 1200 -6.62989 -6.2700 -1 .0300 Taiwan/Singapore 3.8769 4.06948 8. 87062 5. 79031 4.06819 USA -1.2443 -1.3000 0.628 -0.432 -0.474 Table 8 AstralialNZealand China/Hong Kong Canada Indonesia Japan Korea Malaysia Mexico Philippines ROW Thailand Taiwan/Singapore USA 2.35579 -0.0114 -2.70323 -1.03374 2.84788 -3.04401 -2.85483 -3.1311 0.0629 -2.16399 -1 0.771 9 1.2506 -1 .03443 2.55914 0.10888 -5.05577 0.955815 2.95425 -3.10439 -2.7705 -3.07538 0.0898 -2.1 134 -1 0.6824 1.28816 -1 .08953 2‘9: 2.18972 -0.0946 -2.73064 -1 .16018 2.84345 -2.93212 -2.88612 4.30981 -0.0846 -2.14989 -1 0.8352 1.27038 -1.22584 WWW 2.27046 0.0942 -2.78846 -1 .2191 2.50441 -3.62438 -3.7657 -3.24799 0.1 13568 -2. 17501 -5.3235 0.649284 -1.13394 ; flifi.'.mm-;1flifile-;‘:J=IOEOI.'.- 2.5259 0.0956 4.9834 -0.94722 2.92627 -3.17473 -2.78242 4.0209 0.103014 -2.07976 -10.6603 1 .30329 -1.25178 ; 31.5%.‘iI-3fiI-Im—flfil'lfl‘ AstralialNZealand China/Hang Kong Canada Indonesia Japan Korea Malaysia Mexico Philippines ROW Thailand Taiwan/Singapore USA 2.4408 0.198451 -5.03483 -1.13479 2.5804 -3.75757 -3.7066 4.1 1777 0.151037 -2.09097 -5.1364 0.692304 -1.35177 7.31129 4.42095 -0.87691 1.24453 0.510181 -1.60677 0.243335 -0.50196 3.82214 -0.78196 -7.12167 3.70734 -1 .24732 0.101917 3.30823 1.54749 1.51025 -1.07637 0.555099 1.57904 3.03958 3.14269 -0.33497 -6.60611 5.31636 0.615293 1 .27365 3.20935 -0.30651 2.12237 -0.66623 0.43887 2.35856 -0.14684 2.19128 -0.26443 -5.50071 5.4598 -0.43206 0.917786 3.22783 -0.37207 1.78646 -0.60456 -0.57564 0.726593 -0.23146 1.97532 -0.31 166 -0.9701 1 3.82169 -0.4733 Table 9 ~ fli_;lflio!m_;lfl 10!.Vim5'o-Jflialifilm—Jflifld 1. WW AstralialNZealand 4.49704 4.58664 4.47381 4.42688 4.56348 China/Hang Kong 6.00049 5.89338 5.96754 6. 05201 5.85961 Canada 0.48985 -0.42194 0.489341 0.480036 -0.41 796 Indonesia 1 .19394 1 .15865 1 . 16547 1.04842 1 .13022 Japan 2.90069 2.71 142 2.88436 2.7909 2.69439 Korea 16.337 16.0624 16.1474 14.7306 15.8729 Malaysia 2.40131 2.37156 2.3588 1 .1242 2.33357 Mexico -0.51397 -0.50714 -0.29923 -0.52548 -0.224 Philippines 3.10027 3.05301 3.10062 3.15258 3.05278 ROW -0.221 -0.206 -0.21934 -0.22426 -0.204 Thailand -2.21026 -2.23822 -2.24935 -1 .37144 -2.28225 Taiwan/Singapore 1 1.758 1 1.678 1 1.7543 1 1.1248 1 1.6758 USA 0.375239 0.392132 0.348812 0.368267 0.36312 - + - + - - - Astralia/NZealand 4.49444 6. 86767 2. 39464 2. 37631 2.24773 China/Hang Kong 5.91097 7.03539 5.52615 4.65433 4.82647 Canada -0.42102 -0.0710 0.796 -0.0300 -0.0367 Indonesia 0.98347 1 .33381 1 .45237 1 .04307 0.891 183 Japan 2.58461 -O.16399 -0.41048 -0.34913 -O.33053 Korea 14.2511 14.7 16.5083 15.3 13.2 Malaysia 1 .04371 3. 06232 3.51007 3.05214 1 .33268 Mexico -0.237 -0.0451 1.1500 -0.0262 -0.0405 Philippines 3.10549 3. 56301 3.17485 2.4757 2.62295 ROW -0.20717 -0.0743 -0.0768 -0.0616 -0.0693 Thailand -1 .3300 -1 .2300 -2.45463 -2.5100 -0.1 12 Taiwan/Singapore 1 1 .0369 6.12743 12.3932 5.7531 1 4.67124 USA 0.35635 -0.0859 0.0926 0.0497 -0.0534 Table 10 WWWM - dio-Jfliolm _;1 2:10! .1 WAQdOBIMFfiF—fl :JiOEOl 1'. AstralialNZealand 4.45665 4. 54726 4.4335 4. 38805 4.52418 China/Hong Kong 5.91205 5.80821 5.87994 5.96533 5.77515 Canada 0.440224 -0.433144 0.440479 0.430651 -0.42891 Indonesia 1.1625 1.12758 1.13412 1.01792 1.09925 Japan 2.84955 2.65995 2.83306 2.74106 2.64286 Korea 16.1327 15.8621 15.9446 14.5357 15.6742 Malaysia 2.3363 2.30736 2.29391 1 .06659 2.26944 Mexico -0.54283 -0.535109 -0.30469 -0.55413 -0.230 Philippines 3.04707 3.00158 3.048 3.10028 3.00189 ROW -0.206 -0.191541 -0.20474 -0.20965 -0.190 Thailand -2.27469 -2.30127 -2.31 325 -1 .38569 -2. 34477 Taiwan/Singapore 1 1.6667 1 1.5882 1 1 .663 1 1.0402 1 1.586 USA 0.331201 0.349994 0.305912 0.324593 0.322042 -- .. - s - +l ~r- AF-‘Ii AF-fi-W AstralialNZealand 4.45679 6.8471 8 2.37777 2.38286 2.25664 China/Hong Kong 5. 82828 6.96043 5.46217 4.60672 4.78141 Canada -0.43196 -0.0709 0.779773 -0.0306 -0. 0374 Indonesia 0.953465 1 .30837 1 .43432 1 .02863 0.877646 Japan 2. 53439 -0.17977 -0.4019 -0.34674 —0.32858 Korea 14.0617 14.5736 16.3499 15.2215 13.1303 Malaysia 0.987032 3.01742 3.47505 3.03381 1 .31947 Mexico -0.243 -0.0444 1.13946 -0.0275 -0.0419 Philippines 3. 05553 3.52636 3.14338 2.45163 2.59947 ROW -0.19436 -0.0724 -0.0749 -0.0618 -0. 0697 Thailand -1 .34447 -1 .2861 -2.49456 -2.53592 -0.1 19 Taiwan/Singapore 10.9539 6.07891 12.3223 5.73574 4.65529 USA 0.315646 -0.0849 0.0790 -0.0490 -0.0529 Table 11 AstralialNZealand China/Hang Kong Canada Indonesia Japan Korea Malaysia Mexico Philippines ROW Thailand Taiwan/Singapore USA 10.622 9.30647 0.542054 0.646982 2.79636 15.133 2.25921 -1 .22599 5.73075 -0.80206 -5. 12095 15.4846 0.370991 10.8876 9.20752 -1 .3771 1 0.604268 2.66742 14.7489 2.23801 —1 .1999 5.69635 -0.75337 -5. 1 3485 15.3943 0.405099 10.5448 9.23246 0.529442 0.591209 2.75206 14.889 2.19762 -1 . 14279 5.70819 ~0.80703 -5.18164 15.4554 0.284526 3 :liolm _;1:J=(O!.'. WASiBIMW—Jflifidl 10.4239 9.34472 0.508991 0.395444 2.56054 13.5481 0.717248 -1 .2638 5.87801 -0.83138 -2.50306 14.4033 0.331727 10.8101 9.13227 -1.37837 0.548593 2.62114 14.505 2.18182 -0.936 5.6700 -0.75762 -5.18878 15.3669 0.310904 AstralialNZealand China/ng Kong Canada Indonesia Japan Korea Malaysia Mexico Philippines ROW Thailand Taiwan/Singapore USA 10.6149 9.17121 -1.4036 0.295364 2.38629 12.9014 0.624632 -0.986 5.82121 -0.787 -2.43033 14.277 0.272372 17.36 12.5419 —0.305 1.86499 —0.85396 15.2924 3.51331 -0.313 9.65264 -0.183 -2.39443 10.2559 -0.408 ; fldilo'lt'rI-Jfil-Im—AlflllRfl—J - 5.74622 9.5494 1.87141 1.08791 -0.82119 15.5236 3.9 3.05386 6.28838 -0.268 4.73061 16.4192 0.0162 6.17023 8.74934 -0.219 1.02651 -0.66597 15.595 3.63244 -0.269 6.18466 -0.199 4.13152 9.41615 -0.315 5.82 9.03124 -0.239 0.769277 -0.637 13.3394 1.5200 -0.309 6.66172 -0.231 -0.170 7.61408 -0.333 Table 12 ' :J iO—;1:J:(O!m_;l1:10!.'.mfo?o-;1:J=IOBIIT?11ETPF_J:J=IOEOIJ. ChangemflmflMCMmlelflAflfimmflmmJflm AstralialNZealand 10.5496 10.817 10.4725 10. 3548 10.7396 China/Hang Kong 9.18027 9.0867 9.10746 9.22161 9.01255 Canada 0.438899 -1 .40142 0.428078 0.406331 -1 .40208 Indonesia 0.599181 0.556904 0.543439 0.34915 0.501241 Japan 2.69758 2.56776 2.65297 2.46415 2.52133 Korea 14.858 14.4796 14.6159 13.2834 14.2376 Malaysia 2.18827 2.16785 2.12675 0.655244 2.11167 Mexico -1 .29595 -1 .26746 -1 .15939 -1 .33342 -0.952 Philippines 5.62726 5.59629 5.6058 5.7763 5.5700 ROW -0.75495 -0.70959 -0.76198 -0.78608 -0.71584 Thailand -5.20533 -5.21757 -5.26542 -2.51875 -5.27097 Taiwan/Singapore 15.3534 15.2655 15.3242 14.2823 15.2381 USA 0.258281 0.297009 0.174727 0.219797 0.20547 AEEg£MI 55-1 1+.Ianan 55-1 1+; 135 Apr 1 Air-rum AstralialNZealand 10.5477 1 7.3281 5.72631 6. 1 9974 5. 85552 China/Hong Kong 9.0547 12.4397 9.46605 8.69574 8.98218 Canada -1.42736 -0.303 1 .83636 -0.220 -0.241 Indonesia 0.249566 1 .82807 1 .06199 1 .00818 0.752493 Japan 2.28888 -0.88495 -0.80518 -0.66117 -0.633 Korea 12.6444 15.1 15.3158 15.50 13.2 Malaysia 0.563496 3.46487 3.86255 3.61462 1.5100 Mexico -1.0000 -0.310 3.02642 -0.273 -0.313 Philippines 5. 72399 9.59013 6.23382 6.14978 6.6286 ROW -0. 74698 -0.176 -0.263 -0.200 -0.232 Thailand -2.4500 -2.4600 4.78083 4.1700 -0.177 Taiwan/Singapore 14.1584 10.1934 16.3263 9.40288 7.60181 USA 0.167747 -0.404 -0.0202 -0.313 -0.332 Table 13 WWW APEC - :rdame-nascenmmnaaomMM-naaonr. AstralialNZealand 5.86329 6.0259 5.81294 5.74463 5.97524 China/Hang Kong 3.1200 3.13159 3.0800 3.1100 3.09 Canada 0.0522 0.95923 0.0401 0.0290 0.96446 Indonesia 054076 -0.54825 0.56793 06466 057536 Japan 0101 19 -0.0424 -0. 12838 022379 -0.0709 Korea -1.08271 -1.17872 -1.12989 4.06408 -1.22618 Malaysia -0.13881 -0.13048 -0.15751 -0.40254 -0.14832 Mexico —0.71568 -0.69626 -0.84603 074221 071294 Philippines 2.5600 2.5700 2.54 2.65654 2.5519 ROW -0.5822 -0.54865 -0.58883 -0.60833 -0.55478 Thailand -2.9805 -2.96624 -3.00389 -1.14757 -2.97778 Taiwan/Singapore 3.32794 3.32061 3.30526 2.94483 3.29816 USA -0.00416 0.0130 -0.0640 -0.0363 00520 W Air-mm AF-u AF-u-Ihailand AstralialNZealand 5.85835 9.83963 3.27349 3.70597 3.49472 China/Hang Kong 3.081 5.16466 3.8169 3.91802 4.02024 Canada -0.98675 -0.23408 1.06737 018915 020269 Indonesia -0.68178 0.525102 -0.35907 -0.0160 -0.12041 Japan -0.19281 -0.69014 -0.41236 -0.31789 -0.30763 Korea -1.21283 0.486538 -0.89962 0.236 0.125 Malaysia -0.4148 0.437581 0.376898 0.563107 0.186485 Mexico —0.75098 -0.26849 1.88248 -0.24307 -0.26881 Philippines 2.64751 5.92698 3.03618 3.65951 3.9917 ROW 058087 -0.108 -0.19123 -0.13772 -0.16148 Thailand 4.11522 4.17724 -2.33718 -1.6700 -0.0587 Taiwan/Singapore 2.91213 3.89073 3.57316 3.46437 2.81175 USA 00836 -0.32287 -0.0764 -0.26571 -0.28015 Table 14 WW ”EEC EEEIE | EEEDII . EEEC-Il 'l | EEED ill AstralialNZealand 5.83193 5.99524 5.78155 5.71475 5.94455 China/Hang Kong 3.0800 3.09474 3.0400 3.0700 3.0600 Canada -0.00151 -0.97246 -0.0126 -0.0244 -0.97734 Indonesia -0.55736 -0.56477 -0.58459 -0.66268 -0.591 97 Japan -0. 14769 -0.0895 -0.17505 -0.26933 -0.1 181 Korea -1.15025 4.24498 -1.19717 -1.13156 -1.29221 Malaysia -0.14673 -0.1384 -0.16548 -0.40907 -0.1563 Mexico -0.75748 -0.73651 -0.85722 -0.78391 -0.72388 Philippines 2.5100 2. 5300 2.4900 2.60746 2. 50396 ROW -0.55023 -0.51 89 -0.55824 —0.57749 -0.52635 Thailand -3.00609 -2.99148 -3.02942 -1 .15031 -3.003 Taiwan/Singapore 3.28902 3.28236 3.26631 2. 90856 3.25989 USA -0.0728 -0.0530 -0.13092 -0.1046 -0.1 1636 APEC-CMT AF-r “Jam—AM AF-1 1 AF-1 1-111ailand. AstralialNZealand 5.82917 9.82737 3.26876 3.72589 3.51721 China/Hang Kong 3.04563 5.13646 3.79544 3.90847 4.01236 Canada -0. 99969 -0.23259 1 .04836 -0. 18993 -0.20343 Indonesia -0.69785 0.512934 -0.36731 -0.0204 -0.12422 Japan -0.23904 -0.70514 -0.40482 -0.31542 -0.3055 Korea -1 .27875 0.432258 -0.94744 0.206 0.0969 Malaysia -0.42139 0.431674 0.372654 0.561751 0.186022 Mexico -0.76216 -0.26568 1.86545 -0.24515 -0.2709 Philippines 2.60043 5.90109 3.01262 3.64754 3.98068 ROW -0.55355 -0.103 -0.18815 -0.13796 -0.16203 Thailand -1.1 185 -1 .19456 -2.35137 -1 .6800 -0.0588 Taiwan/Singapore 2.87654 3.8755 3. 54985 3.46528 2.81253 USA -0.14761 -0.31918 -0.0991 -0.26456 -0.2791 Table 15 WW [1W _— _ . _— AstralialNZealand -7.46371 -7.6529 -7.41925 -7.58895 -7.60875 China/Hang Kong -3.66832 -3.8993 -3.62715 -3.63564 -3.85867 Canada -8.3103 -8.17659 -8.24457 -8.87171 -8.11076 Indonesia 7.73532 7.96033 7.72998 7.60413 7.94839 Japan -1 .48124 1.85829 -1.51882 -1 .56496 1.85005 Korea 6.39995 -3.87069 617353 -6.44752 -3.47313 Malaysia 1 .42824 1 .49626 1 .48049 1 .35684 1.5481 Mexico 19.3115 19.7162 19.1267 20.1892 19.5276 Philippines 1.5228 1.59856 1.51633 1.41028 1.59288 ROW 7.8881 8.187 7.94342 8.08009 8.23581 Thailand -2.2145 -1 .9748 -2.12145 -2.42588 —1 .88211 Taiwan/Singapore -26. 149 -25.5485 -25.9272 -24.5793 -25.337 USA 0.577603 0.61402 0.636845 0.705954 0.663703 Am AF-11+ iii-=11 ABM-W AstralialNZealand -7.73861 -1 2.9023 -1 .3796 -1 .23539 -1 .2995 China/Hang Kong -3.83499 -9.1 1724 1 1 .4736 1 7.4202 17.1838 Canada -8.67627 42.5493 -7 12144 -7.31364 -8.30967 Indonesia 7.81057 2.10213 10.9561 11.4286 10.3984 Japan 1.82586 0.267275 0.181898 0.117225 0.0851 Korea -3.42252 -0.61958 1 .92269 -0.46436 -0.42983 Malaysia 1 .4747 -2.68669 -0.50829 -1 .17519 -1 .32273 Mexico 20.4021 1 8.9386 24.6701 26.3781 26.1798 Philippines 1.4761 2.29254 0.444121 0.328443 0.296126 ROW 8.42139 2.56333 -1 .20173 -0.92893 -0.72454 Thailand -2.09672 -2.83922 -1 .1 1373 -1 .64678 -1 .62453 Taiwan/Singapore -23.7215 -29.0331 -22.5088 -23.5375 -21.6409 USA 0.782346 -3.31977 -1.38762 -2.06734 -1 .92218 96 Table 16 W W AP—E-C APEC-Canada milled APEC-Thailand APEC-CM AstralialNZealand 3.6767 3.91054 3.8800 7.65267 4.08089 China/Hang Kong -0.75045 -0.66464 0.801038 -0.8563 0.700012 Canada 3. 971 86 3.6500 -7. 12081 3.9700 -5.9300 Indonesia -2.69078 -2.75774 -2.70388 -3.91476 -2.771 36 Japan 55.8728 54.9527 56.0795 50.7781 55.1222 Korea 0.289584 0.28326 0.289566 0.299561 0.283666 Malaysia «0.55759 -0.49744 -0.47949 -0.45786 -0.42289 Mexico 4 3.8366 —0.138 43.6809 5.1700 -0.136 Philippines 35.1282 34.9845 35.3172 -2.43594 35.1348 ROW 4.56957 4.5200 4.5600 4.2200 4.5100 Thailand -6.86018 -6.46876 -6.74771 -6.37257 -6.3700 Taiwan/Singapore -0.54868 -0.87065 -0.58629 -0.63629 -0.9016 USA -9.1600 -9.8100 90400 91700 97300 AWTFJWA—m AstralialNZealand 8.10214 -2.57847 3.50086 -0.77071 0.287075 China/Hang Kong 0.714323 0.172296 4 .83051 0.129047 0.138494 Canada -5.88688 -0.0670 2.25191 0.204 0.251 Indonesia -3.99827 -7.79629 -3.44375 4.471 36 642673 Japan 50.0317 29.7844 43.5472 21.429 8.51267 Korea 0.292912 0.0929 0.134052 0.0297 0.0293 Malaysia -0.32537 -0.95074 -0.60895 -0.55423 -0.48979 Mexico 5.0100 4 7.6 -9.43693 -9.2900 0.531 Philippines -2.25917 16.4474 21 .4682 10.6952 -3.61 105 ROW 4.16167 0479 4.2900 0.162 0.725 Thailand -5.8700 4 .9200 -0. 32886 5.8100 3.1 300 Taiwan/Singapore -0.99204 0.193704 0.302447 0.0855 0.0723 USA -9.75206 4.2100 -0.401 -0.678 -0.572 Table 17 W W APEC APEC-Canada AP-EC-Mexico APEC-Thailand APEC-CM AstralialNZealand -7.49132 -7.68439 -7.4478 -7.6212 -7.64115 China/Hang Kong -3.75798 -3.99047 -3.71724 -3.73084 -3.95009 Canada -8. 36953 -8.23857 -8. 30482 -8. 93824 -8.17367 Indonesia 7.36004 7.57987 7.35369 7.22641 7.5674 Japan 4.38569 1.8621 4.42611 4.47193 1.85225 Korea -6.20843 -3.82752 -5.98288 -6.25789 -3.42849 Malaysia 1.56771 1.63159 1.61822 1.48961 1.68176 Mexico 19.5844 19.9878 19.3994 20.4634 19.7989 Philippines 1.60248 1.67806 1.59531 1.48417 1.67156 ROW 8.20173 8.50305 8.25779 8.38811 8.5523 Thailand -2.28825 -2.04851 -2.19494 -2.49389 4 .95556 Taiwan/Singapore -26.191 1 -25.5916 -25.9687 -24.61 1 1 -25.3793 USA 0.696655 0.729277 0.754516 0.815932 0.777649 m—WfiEp—aF—KFM‘ AstralialNZealand -7.77571 42.9701 4.41785 4.30671 4.37703 China/Hang Kong -3.93187 -9.25997 11.1951 16.9746 16.7231 Canada -8.74652 42.6516 -7.18551 -7.38597 -8.38972 Indonesia 7.42703 1 .67397 10. 5208 10.9598 9.92849 Japan 1.82677 0.281578 0.220871 0.132008 0.0986 Korea -3.37808 -0.60314 1.99135 -0.45643 -0.422 Malaysia 1.60153 -2.5633 -0.38373 4.06758 4.22315 Mexico 20.6747 19.1964 24.8861 26.5503 26.35 Philippines 1.54891 2.34986 0.460614 0.348121 0.314827 ROW 8.7323 2.6621 4.22117 -0.94278 -0.73762 Thailand -2. 16438 -2.89875 4 .17768 4 .70253 4 .67378 Taiwan/Singapore -23.7531 -29.1039 -22.588 -23.6281 -21 .7229 USA 0.887148 -3.21977 4.28666 4.98809 4.85246 Table 18 WW [1W APEC APEC-Canada Apia—Mexico APEC-Thailand APEC-CM AstralialNZealand 3.78892 4.01959 3.99219 7.75596 4.19041 China/Hang Kong -0.67361 -0.591881 0.806922 -0.78208 0.703732 Canada 4.18854 3.8629 -7.0703 4.18679 -5.8778 Indonesia -2.61453 -2.68383 -2.629 -3.84349 -2.69874 Japan 55.7447 54.8184 55.9516 50.6576 54.9885 Korea 0.281174 0.275567 0.281398 0.291046 0.276148 Malaysia -0.58005 -0.517265 -0.49872 -0.47764 -0.43946 Mexico 4 3.709 4 3.6437 4 3.5543 5.26795 4 3.4922 Philippines 35.2507 35.1004 35.4392 «2.40058 35.2506 ROW 4.59376 4.5437 4.58794 4.24558 4.53996 Thailand 6.99285 -6.60477 —6.88044 -6.50026 -6.50442 Taiwan/Singapore -0.46611 -0.79722 -0.50779 -0.55681 -0.83224 USA -8.78724 -9.44516 -8.66814 -8.80287 —9.37486 AWM‘ AstralialNZealand 8.20237 -2.48532 3.5477 -0.72107 0.320732 China/Hang Kong 0.717515 0.179088 4.81117 0.138704 0.147623 Canada -5.83273 -0.0575 2.32648 0.215312 0.263004 Indonesia -3.9308 -7.7388 -3.36632 4.39711 -6.36248 Japan 49.9057 29.6965 43.4164 21.3436 8.43157 Korea 0.285273 0.101215 0.142341 0.0406 0.0395 Malaysia 033925 —0.94928 -0.60841 -0.55773 -0.49177 Mexico 5.10552 4 7.5001 -9.31729 -9.18535 0.599072 Philippines -2.22468 16.5287 21 .5023 10.7267 -3.6031 8 ROW 4.19358 -0.48514 4.33442 0.137111 0.699784 Thailand -6.00144 -2.02644 -0.48927 5.69341 3.02682 Taiwan/Singapore -0.92581 0.203267 0.342111 0.0944 0.0803 USA -9.39818 4.19921 -0.27478 -0.67933 -0.57366 Table 19 WW APEC APEC-Canada APEC-Mexico APEC-Thailand APEC-CM AstralialNZealand 3.9938 4.07993 3.97231 3.92193 4.05845 China/Hang Kong 5.0287 4.9245 4.99831 5.05153 4.89312 Canada 0.350274 -0.49384 0.348636 0.334991 -0.48809 Indonesia 1 .34924 1 .3204 1 .321 52 1 .19604 1 .29269 Japan 3.1891 3.01439 3.17581 3.08491 3.00032 Korea 15.1395 14.89 14.9605 13.6021 14.7115 Malaysia 2.06251 2.03423 2.02529 0.820514 2.00094 Mexico -0. 58445 -0.58085 -0. 38304 -0.60468 —0.308 Philippines 2.91358 2.85723 2.91213 2.91475 2.85505 ROW -0.270 -0.253 -0.26692 -0.27084 —0.250 Thailand -2.38877 -2.4214 -2.43189 4.60337 -2.46966 Taiwan/Singapore 1 0.5783 1 0.5088 10.5754 9.99281 1 0.5074 USA 0.29109 0.293921 0.263678 0.280214 0.26369 ARM-1 1+Japan - + - - - a AstralialNZealand 3.98746 6.2681 2.03851 2.01063 1 .87185 China/Hang Kong 4.91555 5.96779 4.72372 3.83047 3.93933 Canada -0.49416 -0.0966 0.827918 -0.0358 -0.0447 Indonesia 1.13793 1 .34427 1 .55392 1 . 14517 0.970471 Japan 2.89577 0.00788 -0.42149 —0.34637 -0.32517 Korea 13.1601 13.6559 15.4212 14.3281 12.3098 Malaysia 0.744224 2.58935 3.24544 2.79559 1 .02285 Mexico -0.323 -0.0562 1.13769 -0.021 1 -0.0361 Philippines 2.85624 3.26004 2.98029 2.1899 2.22573 ROW -0.25185 ~0.102 -0.0845 -0.0682 -0.0761 Thailand -1 .56274 -1 .64478 -2.90837 -3.261 1 5 -0.261 Taiwan/Singapore 9.91 592 5.53141 1 1 .4126 5.35884 4.30098 USA 0.252896 -0.113 0.103 ~0.0569 -0.0604 Table 20 100 APEC APEC-Canada APEC-Mexico APEC-Thailand APEC-CM AstralialNZealand 3.94945 4.03737 3.92793 3.87991 4.01588 ChinaIHong Kong 4.9345 4.83525 4.9053 4.95965 4.80503 Canada 0.279449 -0.50419 0.278661 0.264453 04983 Indonesia 1.30611 1.27723 1.2784 1.15414 1.24953 Japan 3.12322 2.94769 3.10973 3.021 2.93353 Korea 14.8983 14.6535 14.7211 13.3723 14.4767 Malaysia 1 .97933 1.95178 1.94208 0.749601 1 .91839 Mexico -0.62737 -0.62251 -0.38621 -0.64713 0311 Philippines 2.84179 2.78813 2.8411 2.84477 2.78668 ROW -0.243 -0.228 -0.241 -0.24506 -0.227 Thailand -2.4742 -2.50512 -2.51656 4.60277 -2.55258 Taiwan/Singapore 10.4864 10.419 10.4834 9.90974 10.4173 USA 0.234488 0.240228 0.208606 0.224157 0.211417 mm m Ila-11:11:15 W - - AstralialNZealand 3.94727 6.25444 2.02101 2.0319 1.89663 China/Hang Kong 4.82994 5.88334 4.65724 3.7872 3.89966 Canada -0.50453 -0.0937 0.805661 -0.0356 -0.0446 Indonesia 1 .09605 1 .30663 1.5283 1 .1225 0.949317 Japan 2.83095 -0.0138 -0.40641 -0.34114 -0.32089 Korea 12.9366 13.5 15.2314 14.2 12.2 Malaysia 0.674053 2.52459 3.1 9835 2. 76777 1 .0061 Mexico -0.327 -0.0530 1.1243 -0.0221 -0.0373 Philippines 2.78979 3.20859 2.93503 2.15415 2.19201 ROW 022944 -0.0962 -0.0792 -0.0670 -0.0754 Thailand -1 .5600 -1 .7200 296538 -3.3000 -0.262 Taiwan/Singapore 9.83469 5.48057 1 1.3376 5.34857 4.29264 USA 0.201217 -0.109 0.0848 -0.0550 -0.0587 101 Tab1e21 1.-I°‘ l r- 4‘ .11 u U!" 000‘ {6.11“ on". 1. ;- ;- _:.:..-_ ;- _...;'. ;° -|.::.. ;- _... AstralialNZealand 4.7741 4.86428 4.74947 4.70404 4.83981 China/Hang Kong 6.57324 6.46797 6.541 15 6.63899 6.43516 Canada 0.721841 -0.23017 0.720614 0.716495 -0.23045 Indonesia 1 .1651 1 .12652 1 .13721 1.02585 1 .09883 Japan 2.40514 2.19367 2.38249 2.28027 2.17072 Korea 17.4349 17.139 17.2388 15.7536 16.943 Malaysia 2.89234 2.86251 2.84597 1.53546 2.821 13 Mexico -0.39428 -0.37855 -0.10395 -0.401 16 -0.0374 Philippines 3.15863 3.12222 3.16055 3.25094 3.1200 ROW -0.11447 -0.10062 -0.1146 -0.11916 -0.10047 Thailand -2.30564 -2.32528 -2.34494 4 .09271 -2.36908 Taiwan/Singapore 12.5009 12.4065 12.4946 1 1 .8378 12.4019 USA 0.550143 0.578771 0.523607 0.544539 0.550352 Triple-(2m mmm‘ AstralialNZealand 4.771 7.20459 2.63343 2.62068 2.5026 China/Hang Kong 6.50102 7.51939 5.87937 4.98267 5.18723 Canada -0.23098 -0.0231 0.769106 -0.0114 -0.0162 Indonesia 0.958418 1.28935 1 .37925 0.942405 0.802315 Japan 2.04618 -0.3013 -0.34006 -0.30812 -0.296 Korea 15.2436 15.8228 17.4248 16.1793 14.019 Malaysia 1 .45134 3.50037 3.7747 3.30141 1 .5600 Mexico -0.0476 -0.0120 1.12273 -0.0195 -0.0320 Philippines 3.2169 3.64534 3.23381 2.61729 2.85883 ROW -0.10519 -0.0193 -0.0452 -0.0383 -0.0451 Thailand 4.04913 4.15129 -2.01788 4.71834 -0.0110 Taiwan/Singapore 1 1 .7328 6.49761 12.7807 5.791 99 4.77999 USA 0.545084 -0.0352 0.0987 -0.0280 -0.0327 Table 22 102 APEC-Mexico APEC-Thailand 4.66862 6.55034 0.678044 0.998675 2.24E+00 15.5946 1 .48E+00 -0.42481 3.20727 -1 .14E-01 -1 .1 1 E+00 1 1.7613 0.512428 APEC-CM 4.8042 6.34827 -0.24144 1.07145 2.13E+00 16.7805 2.76E+00 4.31 E-02 3.08154 -9.63E-02 -2.42E+00 12.3217 0.5211 l -1£IBIMIF:T.T~' AP§C APEC-Canada AstralialNZealand 4.73724 4.82854 4.71272 China/Hang Kong 6.48292 6.38042 6.45148 Canada 0.6831 15 -0.2415 0.682568 Indonesia 1.13712 1.09903 1.10934 Japan 2.36736 2.16E+00 2.34449 Korea 17.2668 16.975 17.0723 Malaysia 2.83075 2.801 91 2.78E+00 Mexico -0.41817 -0.40154 -0.10986 Philippines 3.11417 3.07948 3.11661 ROW 4 .09E-01 -9.60E-02 4 .10E-01 Thailand ~2.36E+00 -2.3804 -2.40E+00 Taiwan/Singapore 12.419 12.3262 12.4128 USA 0.517632 0.548439 0.492221 - .. - mar-Jaimie!- AstralialNZealand 4.73683 7.18659 2.61267 China/Hang Kong 6.41583 7.44373 5.81 188 Canada -0.24186 -2.49E-02 0.756514 Indonesia 0.931889 1 .26831 1 .36376 Japan 2.00884 -0.31425 -0.3354 Korea 15.09 15.6947 17.2999 Malaysia 1 .39728 3.45786 3.74146 Mexico -5. 32E-02 4 .28E-02 1 .1 1476 Philippines 3.17547 3.61235 3.20641 ROW -0. 10139 4 .97E-02 4.56E-02 Thailand 4 .06803 4 . 19705 -2.05079 Taiwan/Singapore 11.6581 6.45211 12.7196 USA 0.51625 -3.64E-02 8.77E-02 2.61748 4.92809 4 .28E-02 0.92984 -0.3069 16.0924 3.2803 -2.10E-02 2.59478 -3.93E-02 4 .74386 5.76751 -2.86E-02 2.50131 5.1347 4 .75E-02 0.790506 -0.2955 13.9439 1 .54687 -3.35E-02 2.83661 4.62E-02 -2.00E—02 4.75926 -3.33E-02 Table 23 WW APEC APEC-Canada APEC-Mexico APEC-Thailand APEC-CM AstralialNZealand 4.63431 4.7197 4.61085 4.5581 4.69631 China/Hang Kong 6.32551 6.20679 6.28992 6.36944 6.17044 Canada 0.661265 -0.381 15 0.658022 0.650424 -0.3781 1 Indonesia 1.30838 1.27185 1.27967 1.15958 1.24324 Japan 3.08253 2.89391 3.06646 2.96816 2.87688 Korea 17.0573 16.7683 16.8621 15.4163 16.5732 Malaysia 2.62565 2.59297 2.58284 1 .32252 2.55484 Mexico -0.41269 —0.4091 -0.28078 —0.4252 -0.206 Philippines 3.28701 3.23317 3.28537 3.33556 3.2300 ROW -0.26593 -0.24669 -0.26189 -0.2669 -0.24253 Thailand 4 .9818 -2.01493 -2.02298 4.31273 -2.06098 Taiwan/Singapore 12.0652 11.979 12.0615 11.4092 11.9768 USA 0.529494 0.53869 0.498751 0.520907 0.505651 . iiOEOIM-Jall . . - AF41_‘ AstralialNZealand 4.62118 6.92265 2.45784 2.35767 2.22067 China/Hang Kong 6.21384 7.30096 5.76116 4.82905 4.98973 Canada -0.38131 -0.0708 0.849626 -0.0284 -0.0350 Indonesia 1.0931 1.42037 1.51496 1.09286 0.937589 Japan 2.76256 -0.10603 -0.43817 -0.35732 -0.337 Korea 14.9169 15.2755 17.0542 15.6712 13.5218 Malaysia 1.23866 3.21473 3.62724 3.11441 1.3800 Mexico 0219 -0.0463 1.18266 —0.0220 -0.0361 Philippines 3.27975 3.691 16 3.28464 2.56029 2. 70445 ROW -0.24364 -0.0806 -0.0828 -0.060 -0.0687 Thailand 4.27151 4.04974 -2.31565 -2.40117 -0.0867 Taiwan/Singapore 11.3147 6.29197 12.6229 5.81047 4.72447 USA 0.497229 -0.0886 0.141 -0.0523 -0.0556 Table 24 APEC APEC-CanadL APEC-Mexico APEC-Tha_iland APEC-CM AstralialNZealand 4.483 4.57275 4.46E+00 4.41323 4. 54963 China/Hang Kong 5.97516 5.86904 5.9425 6.02709 5.83547 Canada 0.47197 -0.427 0.471742 0.462 0423 Indonesia 1.18326 1.14813 1.15482 1.038 1.11974 Japan 2.88161 2.69221 2.86521 2.77214 2.67517 Korea 16.2676 15.994 16.0784 14.6637 15.8049 Malaysia 2.37981 2.3503 2.33735 1.10469 2.31233 Mexico -0.523 -0.516 -0.30163 -0.535 —0.227 Philippines 3.08384 3.03698 3.08439 3.1 3629 3.03691 ROW —0.21502 -0.200 -0.214 -0.219 -0.198 Thailand -2.2294 -2.25708 -2.26841 4 .3756 -2.3000 Taiwan/Singapore 1 1.7271 1 1.6475 1 1.7234 11.096 11.6452 USA 0.358 0.375 0.332 0.351 0.346 - - :(ololr. - ~1 - +Il.- AF-‘I‘l ABM-W AstralialNZealand 4.48103 6.8590 2.39113 2.38037 2.25251 China/Hang Kong 5.88719 7.01514 5.5098 4.64374 4.81661 Canada -0.42564 -0.071 1 0.789926 -0.0302 -0.0369 Indonesia 0.973251 1 .32548 1 .44652 1 .03894 0.887287 Japan 2.56573 -0. 16923 -0.40707 -0.34825 -0.32977 Korea 14.1856 14.7 16.4542 15.3 13.2 Malaysia 1 .02444 3.04956 3.49968 3.049 1 .33012 Mexico -0.239 —0.0450 1.14592 -0.0268 -0.041 1 Philippines 3.08975 3.5534 3.16656 2.47022 2.61754 ROW -0.20202 -0.0737 -0.0761 -0.0617 -0.0694 Thailand 4.3300 4 .2500 -2.466 -2.5100 41.3 Taiwan/Singapore 1 1 .0084 6.1 1309 12.3704 5.7512 4.66853 USA 0.3396 -0.0855 0.0877 -0.0493 -0.0531 Table 25 WW APEC AstralialNZealand 4.54043 China/Hang Kong 6.101 1 1 Canada 0.547833 Indonesia 1.23183 Japan 2.96141 Korea 16.5714 Malaysia 2.47546 Mexico -0.47984 Philippines 3.16096 ROW -0.23762 Thailand -2.13693 Taiwan/Singapore 1 1.8558 USA 0.426898 4.62864 5.99013 —0.40849 1.19614 2.77247 16.2921 2.44476 -0.47407 3.1115 -0.22095 -2.16655 11.7738 0.441268 4.51709 6.06732 0.546404 1 .20329 2.9452 16.38 2.43286 -0.29294 3.16066 -0.23514 -2.17672 1 1 .8521 0.399055 APEC-Canada APEC-Mexico APEC-Thailand 4.46819 6.15023 0.537675 1.08519 2.85002 14.9538 1.18955 -0.49167 3.21199 -0.24007 4.35452 11.215 0.419411 APEC-CM 4.60538 5.95555 -0.40481 1.16764 2.75548 16.1008 2.40673 -0.218 3.1100 —0.2183 -2.21128 11.7716 0.410946 AstralialNZealand China/Hang Kong Canada Indonesia Japan Korea Malaysia Mexico Philippines ROW Thailand Taiwan/Singapore USA 6.00433 -0.40791 1.01972 2.64413 14.4679 1.10794 -0.231 3.16198 —0.22087 4 .3100 11.125 0.403643 6.88453 7.11756 -0.0712 1.36251 -0. 14432 14.9 3.1 1179 -0.0457 3.60366 -0.0767 4 .1700 6.17942 -0.0870 2.41281 5.59767 0.814024 1.47323 -0.42016 16.6868 3.54834 1.16105 3.21007 -0.0790 -2.41043 12.466 0.109 2.3666 4.70583 -0.0294 1.05962 -0.35189 15.4 3.07154 -0.0247 2.50222 —0.061 7 -2.4700 5.7691 -0.0506 2.2352 4.87421 -0.0361 0.90652 -0.333 13.3 1 .3500 -0.0390 2.6483 -0.0690 -0.105 4.68594 -0.0542 Table 26 W APEC APEC-Canada APEC-Mexico APEC-Thailand APEC-CM AstralialNZealand 4.4653 4.55563 4.44211 4.44211 4.53254 China/Hang Kong 5.93272 5.8281 3 5.90041 5.90041 5.79491 Canada 0.450664 -0.43103 0.450766 0.450766 -0.42685 Indonesia 1.16933 1.13435 1.14093 1.14093 1.10598 Japan 2.85999 2.67047 2.84355 2.84355 2.6534 Korea 16.177 15.9053 15.9885 15.9885 15.717 Malaysia 2.35064 2.32151 2.30822 2.30822 2.28357 Mexico -0.53632 -0.52883 0.30369 -0.30369 -0.229 Philippines 3.05915 3.01321 3.05994 3.05994 3.0100 ROW -0.209 -0.194 -0.20768 -0.20768 -0.193 Thailand -2.2598 -2.28675 -2.29851 -2.29851 -2.3304 Taiwan/Singapore 1 1.6866 11.6077 1 1.6828 11.6828 1 1.6054 USA 0.339896 0.358165 0.314329 0.314329 0.329956 WWW‘ WWW—6785109 2.38214 TM China/Hang Kong 5.84761 6.97834 5.47781 4.61889 4.79298 Canada -0.42991 -0.0710 0.7831 17 -0.0305 -0.0372 Indonesia 0.95997 1 .31402 1 .43834 1 .03204 0.880836 Japan 2.54461 -0. 17633 -0.4036 -0.34723 -0.329 Korea 14.1023 14.6 16.3842 15.2 13.2 Malaysia 0.999358 3.02803 3.48319 3.03882 1 .3200 Mexico -0.242 -0.0446 1.14158 -0.0273 -0.0416 Philippines 3.06678 3. 53524 3.15101 2.45778 2.60543 ROW -0.19689 -0.0728 -0.0753 -0.0618 -0.0696 Thailand 4 .3400 4 .2700 -2.48516 ~2.5300 -0.1 17 Taiwan/Singapore 10.9718 6.09016 12.3382 5.74083 4.65964 USA 0.32347 -0.0851 0.0818 -0.0491 -0.0529 Table 27 MMWWWQQ] AP§C APEC-Can_gd_a APEC-Mexico APErC-Thaiind APEC-CM AstralialNZealand 4.47107 4.56128 4.44786 4.40186 4. 5382 China/Hang Kong 5.94275 5.83763 5.91032 5.99537 5.80437 Canada 0.459873 -0.42857 0.459822 0.45021 -0.42443 Indonesia 1.17472 1.13965 1.14631 1.02977 1.11129 Japan 2.86998 2.68054 2.85356 2.76098 2.66347 Korea 16.2095 15.9375 16.0209 14.6092 15.749 Malaysia 2.36082 2.33157 2.31839 1.08831 2.29364 Mexico -0.53163 -0.52424 -0. 3024 -0. 54301 -0.228 Philippines 3.06662 3.02048 3.06735 3.1 1949 3.02057 ROW -0.212 -0.197 -0.21082 -0.21573 -0.196 Thailand -2.25148 -2.27857 -2.29025 4 .38095 -2.32227 Taiwan/Singapore 1 1.6992 1 1.6201 1 1.6956 1 1.0702 1 1.618 USA 0.348942 0.367023 0.323223 0.342196 0.338675 ' “It'll-Jal- AstralialNZealand China/Hang Kong Canada Indonesia Japan Korea Malaysia Mexico Philippines ROW Thailand Taiwan/Singapore USA 4.47019 5.85679 -0.42749 0.96512 2.55448 14.1332 1.00842 -0.240 3.07386 -0.19973 4.3400 10.9833 0.332141 ‘Valfllfl'film 6.8539 6.98566 -0.0710 1.31788 -0.17335 14.6 3.03334 -0.0447 3.53894 -0.0732 4.2700 6.09563 -0.0853 2.38272 5.48336 0.786069 1.441 17 -0.40541 16.4093 3.4876 1 . 14349 3.15438 -0.0757 -2.48061 12.3468 0.0844 2.3786 4.621 39 -0.0304 1 .03395 -O.34769 1 5.3 3.0395 -0.0270 2.45959 -0.0617 -2.5300 5.73999 -0.0493 2.2514 4.79502 -0.0371 0.882588 -0.32935 13.2 1.32324 -0.0413 2.60714 -0.0695 -0.117 4.65932 -0.0531 Table 28 CWWWZQQ} APEC AstralialNZealand 4.45238 China/Hang Kong 5.90188 Canada 0.435056 Indonesia 1.15912 Japan 2.84441 Korea 16.1109 Malaysia 2.32919 Mexico -0.54605 Philippines 3.04108 ROW 0204 Thailand -2.28206 Taiwan/Singapore 1 1.6568 USA 0.326927 APEC-Canada 4.54313 5.79834 -0.43418 1 . 12424 2.65478 1 5.8407 2.30034 -0.53822 2.99582 -0.190 -2. 30848 1 1.5785 0.345967 4.42919 5.86981 0.435394 1.13075 2.82792 15.9229 2.28681 -0.30518 3.04209 -0.20329 -2.32054 1 1.6531 0.301764 APEC-Mexico APEC-Thailand 4.38396 5.95535 0.425516 1.01465 2.73608 14.515 1.06038 -0.55732 3.09444 -0.20819 4.38727 1 1.031 1 0.320367 APEC-CM 4.52006 5.76543 -0.42991 1.09592 2.63766 15.653 2.26243 -0.230 2.9962 -0.189 -2.35192 11.5763 0.318141 AstralialNZealand China/Hang Kong Canada Indonesia Japan Korea Malaysia Mexico Philippines ROW Thailand Taiwan/Singapore USA 4.45284 5.81879 -0.43297 0.950248 2.52936 14.0416 0.980927 -0.243 3.04998 -0.1931 4.3500 10.945 0.312 6.95152 -0.0709 1.30556 -0.18145 14.6 3.01214 -0.0443 3.52194 -0.0721 4 .2900 6.07331 -0.0848 2.38E+00 5.45444 0.778115 1.43231 -0.40106 16.3329 3.47101 1.13839 3.13958 -0.0747 -2.49922 12.3144 0.0775 2.38328 4.60073 -0.0307 1.02694 -0.3465 15.2 3.03132 -0.0277 2.44858 -0.0618 -2.5400 5.73317 -0.0490 2.25733 4.77567 -0.0374 0.876066 -0.32839 13.1 1 .31773 -0.042 2.59653 -0.0697 -0. 120 4.65309 -0.0528 Table 29 MW AP§C APEC-Canada APEC-Mexico APEC-Thailand APEC-CM AstralialNZealand 033686 -0.373581 -0.329 -0.36119 -0.38183 China/Hang Kong 2.74206 3.55505 2.73512 2.79639 3.62995 Canada 16.2521 16.0 15.8183 16.9 16.0 Indonesia 4.06855 4.12704 -3.86368 4.32542 4.09312 Japan 4 9.1936 4 9.2751 48.973 4 9.4688 4 9.2847 Korea 0.233569 0.214023 0.240664 0.182538 0.208123 Malaysia 074569 —0.844963 -0.72551 -0.89626 -0.86944 Mexico 2.68883 2.8600 2.61183 3.2500 2.9100 Philippines -5. 1262 -5.5466 -5. 14682 -5.40556 -5.59394 ROW -2.71 373 -2.6700 -2.6200 -3.3200 —2.6600 Thailand 4.61753 4.98571 4.75687 4.29128 4.9200 Taiwan/Singapore 0.510555 0.467829 0.526061 0.399006 0.454932 USA 0.0163 0.647 0.00523 -0.0102 0.619 -- claim-Jal- AE-ir W AstralialNZealand -0.40562 -0.86 -0.18334 0294 -0.287 China/Hang Kong 3.69078 20.295 -0.344 1 .5300 1 .57868 Canada 16.6253 9.8200 1 64769 4 .1800 -0.616 Indonesia 4.35816 -7.82266 3.2000 2. 75467 2.54942 Japan 4 9.5669 -21.1095 -5.9800 -0.853 4 .6600 Korea 0.158237 0.130 0.110 0.0841 0.0610 Malaysia 4 .01736 -2.5700 4 .05371 4 .4900 4 .64905 Mexico 3.4600 9.2600 1 .9300 3.0100 3.8300 Philippines -5.86793 4 2.3569 -3.76656 -6.58494 -7. 0700 ROW -3.27425 -6.3900 -2.5800 -2.7200 -3.5700 Thailand 4.5900 9.53E-01 7.22482 9.4500 8.6200 Taiwan/Singapore 0.345887 0.2831 37 0.240 0. 1 83826 0.1 3341 8 USA 0.589732 0.192 0.171 0.137 0.118 Table 30 WW APEC APEC-Canada APEC-Mexico APEC-Thailand APEC-CM AstralialNZealand 19.5364 4.1456 19.7681 19.729 -0.81265 China/Hang Kong -6.01974 -2.07647 -5.99794 -5.7623 -2. 146 Canada 2. 36976 3.63556 2.2769 2. 30297 3.55603 Indonesia 4.18768 -6.1 1 586 4.08203 4.36205 -5.84099 Japan 0.343389 0.314652 0.333639 0.268363 0.305978 Korea 0.789275 0.742453 0.768332 0.747767 0.72201 1 Malaysia -5.69663 -5.61488 -5.57535 -5.671 33 -5.49007 Mexico 0.71 5796 0.539262 0.606487 1 . 1 5251 0.42845 Philippines 4.13454 4.11318 4.14955 4.27605 4.1299 ROW 49.6055 50.37 49.4309 49.8035 50.1689 Thailand 0.621038 0.569067 0.603406 0.48535 0.553378 Taiwan/Singapore 0.618868 0.545874 0.61 1423 0.601461 0.539131 USA 41 .3389 -39.0085 41 .3293 41 .2724 -39.0061 m AF-11+.Ianan_ AF-J'l Am AstralialNZealand -0. 70766 -1 .696 4 .49283 -0.48239 -0.441 85 China/Hang Kong 4 .93738 4 .62481 -3.90972 -1 .1 5796 4 .08986 Canada 3.53511 0.348507 1.36896 0.351149 0.334115 Indonesia -5.92299 1 . 16731 3.44546 0.351426 0.409748 Japan 0.232637 0.190432 0.161 114 0.123638 0.0897 Korea 0.682024 0.061 5 0.662626 0.300782 0.289106 Malaysia -5.46489 -1 .29416 4.23559 -2.42532 -2.40291 Mexico 0.867537 4.21383 2.06392 —2.90743 -2.5178 Philippines 4 .2751 5 -2.20526 -3.85239 -3.19862 —3.26063 ROW 50. 3584 50.6803 57.2288 61 .084 60.0743 Thailand 0.420737 0.344408 0.291 384 0.223606 0. 162289 Taiwan/Singapore 0.52197 0.120037 0.206785 0.124613 0.100321 USA -38.9199 42.0868 -0.10402 -0.18545 -0.27577 Table 31 WWI] APEC APEC-Canada AstralialNZealand 1.7496 1.33021 China/Hang Kong 1.52408 1.5252 Canada 5.34324 5.44091 Indonesia 0.461691 0.423054 Japan 0.260927 0.22566 Korea -38.5787 -38.2208 Malaysia 69.3196 68.5623 Mexico 0.705719 0.854787 Philippines -1 9.691 5 -1 9.3075 ROW 0.603919 0.55338 Thailand 0.736246 0.662173 Taiwan/Singapore 40.3279 40.2453 USA 2.76923 2.68795 1.7775 1.47712 5.311 15 0.475712 0.403673 -39.4608 69.8226 0.541806 -20.1583 0.622261 0.719643 4 0.2236 2.41627 APEC-Mexico APEC-Thailand Ach-CM 1 .89983 1 .33427 1.44844 1.51891 5.38637 5.47915 0.360818 0.411391 0.240873 0.217234 -38.1008 -38.081 67.3997 68.2676 0.571859 0.912488 48.1943 49.1919 0.471972 0.538124 0.897164 0.636546 40.476 40.1876 1.08016 2.59143 AstralialNZealand China/Hang Kong Canada Indonesia Japan Korea Malaysia Mexico Philippines ROW Thailand Taiwan/Singapore USA 1 .48202 1 .43843 5.52051 0.312783 0.200562 -37.6266 66.3554 0.764022 4 7.6884 0.409139 0.80042 40.3254 0.877731 ; UKEOIJII-AEIIWJIIIIIN' 0.0754 1 .57308 1 .70013 0.256039 -0.26173 -28.1 60.792 0.70737 48.4872 0.334914 -0.49647 -2.281 1 1 .59728 4.09901 0.386 —0.73181 0.217 -0.0305 -35.7 67.0416 0.266 4 7.8247 0.283352 -0.2227 -5.7500 1.94439 0.284 -0.37148 0.166232 0.0155 -32.0 60.627 1.0300 46.8872 0.2.7 -0.73114 -2.16677 1.14604 -0.737 0.275783 -0.23044 0.120649 0.0448 -31 .3 58.308 0.82348 4 5.4628 0.157816 -0.45946 4 .43402 —0.38595 Table 32 CW APEC APEC-Canada AstralialNZealand 5.3714 5.36921 China/Hang Kong 3.20963 3.47029 Canada 0.554151 0.507777 Indonesia -0.29299 -0.350149 Japan 3.74107 4.0284 Korea 4 .344 4 .32976 Malaysia 0.412845 0.470897 Mexico -2.04835 -2.13235 Philippines 0.368094 0.33729 ROW -0.38595 -0.479465 Thailand 0.940556 1 .25329 Taiwan/Singapore -7.09504 -7.26292 USA 5.26142 5.24887 5.41352 3.33731 0.570982 0.562569 0.738075 4.10391 1.86819 40.0413 0.379275 -0.38979 0.823698 6.96556 5.35173 APEC-Mexico APEC-Thailand Ange-CM 4.63021 5.39554 5.09765 3.53087 0.433077 0.493778 —0.31583 0.479939 3.84729 0.81 3606 4.19546 4.08896 0.311538 1.70578 -2.11253 -8.86107 0.287671 0.327992 -0.47959 -0. 52063 1.86163 1.33883 -7.4332 -7.26901 4.29928 5.24183 AstralialNZealand China/Hang Kong Canada Indonesia Japan Korea Malaysia Mexico Philippines ROW Thailand Taiwan/Singapore USA 4.62902 5.4455 0.375423 0.453309 0.859059 -1 .01508 1.75 -8.91658 0.249374 -0.61 196 2.25463 -7.60336 4.26831 0.652 3.62587 0.307314 0.166792 -0.52385 -0.546 -0.0454 0.501207 0.204133 4.72398 8.17759 4 3.8943 3.34537 0.260002 -0.139 2.3400 -0.512 4.80108 2.2400 0.172706 4.12921 2.08082 -5.9000 4.11491 5.1200 0.199523 0.121083 -0.314 —0.475 0.0879 0.129 0.132533 4.1400 4.76015 -9.34032 4.79195 -0.283 4.98211 0.14481 0.10518 -0.307 -0.455 0.107543 0.212012 0.0962 4 .171 6.2377 -9.80557 3.46195 Table 33 113 WWW APEfC Ach-Canada AP§CoMexico flC-Thflnd APEC-CM , AstralialNZealand 36.5368 36.519 36.9808 3.26E+01 36.656 China/Hong Kong 0.37954 0.347778 0.391068 0.296616 0.33819 Canada 0.372519 0.348127 0.35852 0.345713 0.332675 Indonesia 4.30285 4.31216 4.30022 4.29137 4.26808 Japan 4.64154 4.59313 4.64085 4.58908 4.57536 Korea 0.0970 0.140428 0.0886 0.137593 0.130282 Malaysia -0.24831 -0.15672 -0.12385 -0.15436 -0.0712 Mexico 0.311779 0.285688 0.321249 0.24366 0.277812 Philippines 4.26299 4.31871 4.21846 1.33951 4.34468 ROW 4.85256 4.43295 -5.03753 -5.45988 -4. 1 8837 Thailand 2.28434 2.10086 2.20918 -2.63816 1.99375 Taiwan/Singapore -5.90118 -5.89821 -5.7675 5.79241 -5.8515 USA 30.5 30.545 30.5 -0.681 30.4384 Am KF—ihm AF-‘I‘IHISL AF4‘I - - AstralialNZealand 32.7 35.7 37.0334 27.4 16. 5 China/Hang Kong 0.257128 0.210481 0.178 0.137 0.0992 Canada 0.306406 0.164522 0.217797 0.121401 0.102463 Indonesia 4.25398 -0.44442 -0.691 -0.33222 -0.3061 1 Japan 4 .52263 -0.75084 4 .0400 -0.620 0569 Korea 0.16921 0.0866 0.207 0.135 0.138 Malaysia 0.0181 -0.247 -0.31592 -0.202 -0.12991 Mexico 0.211222 0.172903 0.146 0.112 0.0815 Philippines 1 .24522 -2.26933 4 .60787 4.78782 0.322 ROW -5. 14848 8.3065 -0.87621 4.8800 -0.46645 Thailand -2.51173 -3.08537 1.28748 4.83053 4 .4700 Taiwan/Singapore 5.5856 -7.66869 -4.3200 -3.41751 1.55722 USA -0.55259 24.9735 23.8628 18.2267 0.0188 114 Table 34 WW1. W199] APEC APEC-Canada APEC-Mexico APEC-Thailand APEC-CM AstralialNZealand -3.08854 -3.1531 9 -3.0900 -3. 1 000 -3. 1 5499 China/Hang Kong 6.691 13 8.0903 6.69931 6.52541 8.09745 Canada 24.6078 24.4 24.5493 25.1 24.3 Indonesia 8.15101 7.88411 8.18839 7.92583 7.92124 Japan 16.5371 16.2324 16.5722 16.5215 16.2678 Korea 0.270 0.257242 0.251 0.185505 0.239457 Malaysia 4 .3265 4 .41033 4 .3421 1 4 .53039 4 .4300 Mexico -0.48495 -0.285 -0.443 0.182 -0.242 Philippines 4.59948 -5.09035 4.71299 -5.09531 -5.20333 ROW 13.0414 13.1 13.1 12.4 13.1 Thailand 41.1251 41.2182 41.0729 41.0027 41.2 Taiwan/Singapore 0.270155 0.257242 0.251281 0.185505 0.239457 USA -0. 369 0.453 -0.375 -0. 387 0.437 - - only. - FT'oTTi—Jfillllci- AF-ji AF-‘l 1.1mm; AstralialNZealand -3.1700 4. 5800 -2.67792 -2.9200 -2.9200 China/Hang Kong 7.95016 40.0834 -0.856 1.1200 0.865 Canada 24.8332 19.5 1 1.9445 9.7800 10.2 Indonesia 7.69135 2.07422 13.7 12.9759 12.7917 Japan 16.2422 10.405 32.1 37.7 37.3 Korea 0.155263 -0.275 0.0761 —0.0750 -0.177 Malaysia 4 .6300 -3.8800 4 .66482 -2.3900 -2.74263 Mexico 0.431 8.4200 -2.0000 0.404 1 .8200 Philippines -5.70782 4 2.7558 -2.17304 -6.09343 -7. 1600 ROW 12.452 8.0400 13.7 12.7 11.6 Thailand 41.0 34.7 43.4129 43.0 41.8 Taiwan/Singapore 0.155263 -0.27533 0.0761 ~0.0750 -0.17665 USA 0.41707 0.0581 0.00340 0.0629 0.0433 115 Table 35 W W APEC APEC-Canada APEC-Mexico APEC-Thailand APEC-CM AstralialNZealand 45.4249 0.846709 45.3 45.5726 0.766894 China/Hong Kong 4 0.5355 4.50551 4 0.5421 4 0.3744 4.64948 Canada 0.862635 1 .5200 0.807221 0.770 1 .5100 Indonesia -2.44688 4.23072 -2.27793 -2.50086 -3.83234 Japan 0.270155 0.257242 0.251281 0.185505 0.239457 Korea 0.915126 0.887169 0.898189 0.904711 0.870591 Malaysia -7.63292 -7.56813 -7.54496 -7.73021 -7.48131 Mexico 2.87157 2.6600 2.69568 3.2500 2.4800 Philippines 3. 32543 3.36562 3.36378 3.18899 3.40359 ROW 27.7738 28.1 27.5 28.5 27.9 Thailand 0.270155 0.257242 0.251281 0.185505 2.39E-01 Taiwan/Singapore 1 .07155 1 .0192 1 .06531 1 .05208 1 .01355 USA -71.7 -69.3 -71.7 -71.7 -69.2 J - + Fm—JfiIIlRi‘ AstralialNZealand China/Hang Kong Canada Indonesia Japan Korea Malaysia Mexico Philippines ROW Thailand Taiwan/Singapore USA 0.787741 4 .55872 1 .49433 -3.77475 0.1 55263 0.860074 -7.57422 2. 8600 3.26482 28.5666 0. 155 0.994244 -69.1977 2.66794 4 .05557 0.241 -0.59612 -0.27533 —0.451 2.90739 -7.4700 -0.83091 27.4 -0.275 0.421033 4 9.4 2.34148 -2.84189 0.702079 2.72249 0.0761 0.698002 -6.05965 3.94372 1.01292 30.7 0.0761 0.314077 1.8400 1 .9300 2.80035 4.00931 0.0901 -0.41182 -0.177 -0.108 0.56538 —6.7100 -0.0427 30.5 -0.177 0.168201 1.7800 116 Table 36 W W APEC APEC-Canada APEC-Mexico APEC-Thailand APEC-CM AstralialNZealand 6.68999 5.47986 6.6500 6.74034 5.45258 China/Hang Kong 4.16506 4.10649 4. 15105 3.99067 4.09227 Canada 10.2292 10.4 10.2841 10.4 10.5 Indonesia 0.270155 0.257242 0.251281 0.185505 0.239457 Japan 3.6338 3.62654 3.62144 3.42541 3.61432 Korea -86.4097 -86. 1432 —86.0426 -82.7839 -85.7759 Malaysia 135.54 134.205 134.664 129.706 133.344 Mexico 20.6427 20.5 20.5747 19.4 20.4 Philippines 4 0.4939 4 0.0201 40.3179 -9.09315 -9.85745 ROW 0.270155 0.257 0.251 0.186 0.239 Thailand 1.13274 1.08728 1.12824 1.51434 1.0800 Taiwan/Singapore 4 0.5162 4 0.458 4 0.5633 4 1.2264 4 0.4965 USA 4 .9000 -2.0800 -2. 1 500 -6. 5800 -2.2900 m AF-‘l‘l+ 2.: - M—Jal_n:simmm AstralialNZealand 5.51239 4.79239 -2.02561 4.43097 4.3421 China/Hong Kong 3.913 2.49114 0.127518 0.0546 0.0594 Canada 10.6192 2.9000 4.39869 4 .1000 -0.875 Indonesia 0. 1 55263 -0.27533 0.0761 -0.0750 -0. 1 77 Japan 3.40424 2.17171 3. 33558 2.34246 2.18677 Korea -82.1123 -67.6 -85.3 -69.9 -64.9 Malaysia 127.43 121.308 131.383 117.031 108.154 Mexico 19.1 18.1 19.3508 17.0 15.4 Philippines -8.428 4 4.8004 -8.7773 4 1 .7241 4 0.4638 ROW 0.155263 -0.275 0.0761 -0.0750 -0. 177 Thailand 1 .4600 -0.987 -0.0986 -0.965 -0.608 Taiwan/Singapore 41.1908 2.20385 4.46939 1.2465 2.721 1 1 USA -7.041 -0.0660 4 .6900 -0.401 -5.1100 117 Table 37 WWI. WM APEC APEC-Canada APEC-Mexico APEC-Thailand APEC-CM AstralialNZealand 6.25714 6.26476 6. 3200 5.47127 6.31726 China/Hang Kong 15.2081 15.4153 15.428 18.1983 15.6019 Canada 0.270155 0.257 0.251281 0.186 0.239 Indonesia -0. 36802 -0.38745 0.509902 -0.38934 0.461 173 Japan 2.67324 2.91854 0.507001 2.72673 0.378784 Korea -3.58833 -3.59664 -0.56799 -3.41913 -0.65686 Malaysia 1.18949 1.22219 0.503378 1.07778 0.452635 Mexico 7.64332 7.2300 -7.41584 7.6400 -6. 1600 Philippines 0.270155 0.257242 0.251281 0.185505 0.239457 ROW 4 .67357 4 .7200 4 .6900 4 .8400 4 .7300 Thailand 7.46678 7. 87705 7.53109 8. 9937 7.9400 Taiwan/Singapore 45.8328 46.1374 45.9224 46.6055 46.2206 USA 13.7 13.4 13.7 12.5 13.3 ; flddfiM-flfil- : .. -V - HI»; AF-H W AstralialNZealand 5.51104 1.62107 3.12137 1.9300 0.948647 China/Hang Kong 18.6256 8.12231 13.3243 7.16325 6.46991 Canada 0.155263 -0.275 0.0761 -0.0750 -0.177 Indonesia 0.438045 0.0896 -0.4425 0.104 0.0822 Japan 0.411471 -0.13957 2.99185 -0.16793 -0.1389 Korea -0.60175 -0.460 -0.36161 -0.584 -0.527 Malaysia 0.466306 0.15132 4.01898 0.119381 0.125126 Mexico -6.1100 -0.0739 3.51424 0.196 0.238 Philippines 0.155263 -0.27533 0.0761 -0.0750 -0.17665 ROW 4 .8991 3 -3.8400 -2.0700 -2.7000 -3.0200 Thailand 9.4700 22.20 8.49137 16.2 19.2 Taiwan/Singapore 4 6.9968 -26.8726 4 3.4702 -20.0036 -21.4509 USA 12.111 9.1800 10.7 8.0400 6.4000 118 Table 38 W W APEC APEC-Canada [WC-Mexico APEC-Thailand W AstralialNZealand 93.6858 92.1304 93.9 87.6 92.347 China/Hang Kong 0.270155 0.257242 0.251281 0.185505 0.239457 Canada 0.352864 0.343 0.343914 0.329 0.334 Indonesia -0.69615 -0.70757 -0.70514 -0.70452 -0.71586 Japan 4 .72378 4.69977 4 .72839 4 .65559 4 .70372 Korea 0.168 0.169604 0.170 0.183193 0.171974 Malaysia -0.71201 -0.64052 -0.62515 -0.60381 -0.557 Mexico 0.270155 0.257 0.251 0.186 0.239 Philippines 4.08645 4.12909 4.10217 1.63566 4.14764 ROW 1 .947 2.4600 2.2300 -3.5600 2.7400 Thailand -8.04237 -8.27427 -8.34665 -3.82842 —8. 5500 Taiwan/Singapore -3.37086 -3.37829 -3.26216 3.75084 -3.28155 USA 63.4 63.1 63.5 4.17 63.1 ‘ :J=(0£OIM_;IaII . .: AstralialNZealand 86.3 63.8 70.6169 39.9 25.8 China/Hang Kong 0.155263 -0.27533 0.0761 -0.0750 -0. 177 Canada 0.310301 0.159 0.221 0.148 0.123 Indonesia -0.72224 0.325 -0.0354 0.208 0.233 Japan 4.63472 -0.60556 4 .1200 -0.716 -0.619 Korea 0.186456 0.0332 0.0660 -0.00705 -0.00394 Malaysia -0.451 -0.989 -0.66742 -0.594 -0.52382 Mexico 0.155 -0.275 0.0761 -0.0750 -0. 177 Philippines 1.5762 -3.01989 4 .4200 -2.0500 0.510 ROW -3.4171 29.4 7.7100 22.8 2.0400 Thailand -3.7400 4 6.9 -7.78639 4 4.5 -2.6400 Taiwan/Singapore 3.64185 -7.42195 4.1100 -2.0800 0.384446 USA -3.9278 41.9 38.5 23.8 -3.6500 119 APPENDIX ! This is written for the model under the Coumot conjecture only I ! This appendix contains a part of the program used for the study in this paper ! I The non-linear equations in this dissertation were linearized, and ! linearized eqautions are written in this appendix ! ! Some statements, such as “UPDATED” and “READ” are omitted ! ! Quantities are specified, according to the order of index I ! after variables and coefficients I ! Variables and Coefficients are reogranized for readers to understand easily ! FILE DSET # File with set specification #; FILE DATA # The file containing all base data for the economy. # ; FILE (TEXT) PARM # The parameter file # ; SET REG # Regions in the model # MAXIMUM SIZE 13 READ ELEMENTS FROM FILE DSET HEADER "REG"; SET NSAV_COMM # NON-SAVINGS COMMODITIES # MAXIMUM SIZE 10 READ ELEMENTS FROM FILE DSET HEADER "NSAV"; SET PROD_COMM # PRODUCED COMMODITIES # MAXIMUM SIZE 8 READ ELEMENTS FROM FILE DSET HEADER "PROD"; SET PCM_COMM # PERFECTLY COMPETITIVE COMMODITIES # 120 MAXIMUM SIZE 5 READ ELEMENTS FROM FILE DSET HEADER "PCMP"; SET NIMC_TRAD # TRAD_COMM - IMC_COMM # MAXIMUM SIZE 4 READ ELEMENTS FROM FILE DSET HEADER "NIMP"; SET TRAD_COMM # TRADED COMMODITIES # MAXIMUM SIZE 7 READ ELEMENTS FROM FILE DSET HEADER "TRAD"; SET IMC_COMM # IMPERFECTLY COMPETITIVE COMMODITIES # MAXIMUM SIZE 4 READ ELEMENTS FROM FILE DSET HEADER "IMPC"; SET ENDW_COMM # ENDOWMENT COMMODITIES # MAXIMUM SIZE 2 READ ELEMENTS FROM FILE DSET HEADER "ENDW"; SET DEMD_COMM # DEMANDED COMMODITIES # MAXIMUM SIZE 9 READ ELEMENTS FROM FILE DSET HEADER "DEMD"; SET CGDS_COMM # CAPITAL GOODS COMMODITIES # MAXIMUM SIZE 1 READ ELEMENTS FROM FILE DSET HEADER "CGDS"; SUBSET DEMD_COMM IS SUBSET OF NSAV_COMM ; SUBSET PROD_COMM IS SUBSET OF NSAV_COMM ; SUBSET PCM_COMM IS SUBSET OF PROD_COMM ; SUBSET IMC_COMM IS SUBSET OF PROD_COMM ; SUBSET TRAD_COMM IS SUBSET OF DEMD_COMM ; SUBSET TRAD_COMM IS SUBSET OF PROD_COMM ; SUBSET IMC_COMM IS SUBSET OF TRAD_COMM ; SUBSET NIMC_TRAD IS SUBSET OF TRAD_COMM ; SUBSET ENDW_COMM IS SUBSET OF DEMD_COMM ; 121 SUBSET CGDS_COMM IS SUBSET OF NSAV_COMM ; ! Define Variables ! qvap (i,r : PCM_COMM, REG) : value-added in PCM_COMM industry i of region r qua (i,r :IMC_COMM, REG) : variable value-added in sector i of region r fqva (i,r : IMC_COMM, REG) : fixed value-added in sector i of region r qxs (i,r,s : TRAD_COMM, REG, REG) : exports of commodity i from r to region s qfep (i,j,r : ENDW_COMM, PCM_COMM, REG) : IMC firm’s demand for dowment i for use in j in region r qfem (i,j,r : ENDW_COMM, IMC_COMM, REG) : PCM firm’s demand for endowment i for use in j in region r qf (i,j,r : TRAD_COMM, PROD_COMM, REG) : demand for traded composit commodity i for use in j in region r qfsp (i,j,r,s : IMC_TRAD, PROD_COMM, REG, REG) : PCM firm’s demand for commodity i from r for use in j in region s qfsm (i,j,r,s : IMC_COMM, PROD_COMM, REG, REG) : IMC firm’s demand for commodity i from r for use in j in region s qc (i,r, : TRAD_COMM, REG) : household demand for composite commodity i in region r qcsp (i,r,s : NIMC_TRAD, REG, REG) : hhld demand for NIMC_COMM commodity i from r in region s 122 qcsm (i,r,s : IMC_COMM, REG, REG) : hhld demand for IMC_COMM commodity i from r in region s globalcgds : global supply of capital goods qsave (r, : REG) : region r’s demand for save : walras_dem : demand in the ommitted market--global demand for save walas_sup : supply in the omitted market «global supply of cgds composite mkr (i,r, : IMC_COMM, REG) : markup in industry i of region r ela (i,r : IMC_COMM, REG) : total perceived d. elast. facing producers of i in r elas (i,r,s : IMC_COMM, REG, REG) : perceived d. e]. facing sales of IMC_COMM i from r into s avc (i,r : IMC_COMM, REG) : average variable cost in the production of i in r atc (i,r : IMC_COMM, REG) : average total cost for i in r pvap (i,r : PCM_COMM, REG) : price of value-added in PCM_COMM industry i of region r pvam (i,r : IMC_COMM, REG) : price of value-added in IMC_COMM industry i of region r ps (i,r : NSAV_COMM, REG) : supply price of commodity i in region r pm (i,r, : TRAD_COMM, REG) : market price for traded commodity i in region r pfe (i,j,r : ENDW_COMM, PROD_COMM, REG) : demand price for endowment commodity i in j of region r pf (i,j,r : TRAD_COMM, PROD_COMM, REG) : composite price for traded comm i for use in j in region r # ; 123 pfs (i,j,r,s : TRAD_COMM, PROD_COMM, REG, REG) : agents' price of commodity i fi'om r for use in j in region 5 pc (i,r : TRAD_COMM, REG) : household price for traded commodity i in region r pcs (i,r,s : TRAD_COMM, REG, REG) : agents' price for commodity i from r in region s pme (i,r : ENDW_COMM, REG) : domestic price for primary factor i in region r pms (i,r,s : TRAD_COMM, REG, REG) : domestic price for good i supplied from r to region s pw (i,r,s : TRAD_COMM, REG, REG) : world price of commodity i supplied from r to s pcgds : price of capital goods supplied to savers walaslack : slack variable associated with walras law -- normally endogenous EV (r : REG) : Equivalent Variation ! This variable reduces the accuracies of solutions. Unless the calculation of EV(r) is necessary, this variable and relevant equation were removed from the file ! to (i,r : ENDW_COMM, REG) : income tax on endowment commodity i in region r tf (i,j,r : ENDW_COMM, PROD_COMM, REG) : tax on primary factor i used by j in region r tcs (i,r,s : TRAD_COMM, REG, REG) : tax on i purchased by hhlds in r from s tfs (i,j,r,s : TRAD_COMM, PROD_COMM, REG, REG) : tax on i purchased by j in r from s txs (i,r,s : TRAD_COMM, REG, REG) : combined tax in r on good i bound for region s tms (i,r,s : TRAD_COMM, REG, REG) : import tax in s on good i imported from region r u (r : REG) : aggregate utility of household in region r 124 y (r : REG) : household income in region r gp (r : REG) : general price index for region r qo (i,r : NSAV_COMM, REG) : industry output of commodity i in region r n (i,r : IMC_COMM, REG) : number of firms active in sector i of region r ! Define Coefficients ! VOA (i,r : NSAV_COMM, REG) : value of commodity i output in region r VXA (i,r,s : TRAD_COMM, REG, REG) : value of exports of commodity i from region r to s VFA (ij,r : DEMD_COMM, PROD_COMM, REG) : producer expenditure on i by industry j, region r valued at agent's prices VCA (i,r : TRAD_COMM, REG) : household expenditure on commodity i in region r valued at agent's prices VFAS (i,j,r,s : TRAD_COMM, PROD_COMM, REG, REG) : purchases of commodity i from r for use in j in region s VCAS (i,r,s : TRAD_COMM, REG, REG) : household expenditure on i from r in s VOM (i,r : ENDW_COMM, REG) : value at market prices of commodity i in region r VFM (i,j,r : ENDW_COMM, PROD_COMM, REG) : producer expenditure on i in industry j, region r valued at domestic market prices 125 VFMS (i,j,r,s : TRAD_COMM, PROD_COMM, REG, REG) : purchases of commodity i from r for use in j in region s VCMS (i,r,s : TRAD_COMM, REG., REG) : household expenditure on i from r in s VIWS (i,r,s : TRAD_COMM, REG, REG) : imports of commodity i from region r to s valued cif (tradeables only) SAVE (r: REG) : regional savings INCOME(r : REG) : level of income in region r INCOME(r) = sum(i,TRAD_COMM, VCA(i,r)) + SAVE(r) VIMS (i,r,s : TRAD_COMM, REG, REG) : value of imports of commodity i from r in s at domestic market prices VIMS(i,r,s) = sum(j,PROD_COMM, VFMS(i,j,r,s)) + VCMS(i,r,s) VCGDS : value of world capital goods VCGDS = sum(r,REG, sum(k,CGDS_COMM, VOA(k,r))) VSAVE : The value of global savings VSAVE = sum(r,REG, SAVE(r)) SALSHR (i,r,s : IMC_COMM, REG, REG) : The share of sales, by source, in total sales of i from r to s, at agent's prices SALSHR(i,r,s) = VXA(i,r,s) / VOA(i,r) VIMSHR (i,r,s : TRAD_COMM, REG, REG) : The share of demand by source in the composite demand for region 3 as a whole, at market prices 126 VIMSHR(i,r,s) = VIMS(i,r,s) / sum(k,REG, VIMS(i,k,s)) VA (i,r : PROD_COMM, REG) : Value-added in sector i of region r VA(i,r) = sum(k,ENDW_COMM, VFA(k,i,r)) CASHRS (i,r,s : TRAD_COMM, REG, REG) : The share of demand by source in the commodity i evaluated at agents' prices CASHRS(i,r,s) = VCAS(i,r,s) / sum(k,REG, VCAS(i,k,s)) FASHRS (i,j,r,s : TRAD_COMM, PROD_COMM, REG, REG) : The share of demand by source in the commodity j evaluated at agents' prices FASHRS(i,j,r,s) = VFAS(i,j,r,s) / sum(k,REG, VF AS(ij,k,s)) PELAS (i,r,s : IMC_COMM, REG, REG) : perceived demand elasticity PELAS(i,r,s) = SIGMAS(i,r) / [l + {{SIGMAS(i,r) - 1} VIMSHR(i,r,s)/ N_L(i,r)}] TELA (i,r : IMC_COMM, REG) : total demand elasticity TELA(i,r) = sum(s,REG, SALSHR(i,r,s) * PELAS(i,r,s)) FVA (i,r : IMC_COMM, REG) : Fixed value-added in sector i of region r FVA(i,r) = [1 / TELA(i,r)] * VOA(i,r) VVA (i,r : IMC_COMM, REG) : Variable value-added in sector i of region r VVA(i,r) = VA(i,r) - FVA(i,r) SHR_FVA(i,r : IMC_COMM, REG) : Variable value-added in sector i of region r SHR_FVA(i,r) = FVA(i,r) / VA(i,r) VC (i,r : IMC_COMM, REG) : variable cost in the production of i in region r 127 VC(i,r) = VOA(i,r) - FVA(i,r) SHRPELAS (i,r,s : IMC_COMM, REG, REG) : share of PELAS, weighted with SALSHR SHRPELAS(i,r,s) = SALSHR(i,r,s) * PELAS(i,r,s) / TELA(i,r) ESHR (i,j,r : ENDW_COMM, PROD_COMM, REG) : the share in sector j value-added of ENDW_COMM i ESHR(i,j,r) = VF A(i,j,r)/V AG,r) ALPHA (i,r,s : IMC_COMM, REG, REG) : The multiplier in the perceived demand ALPHA(i,r,s) = [l-SIGMAS(i,r)] A 2 * VIMSHR(i,r,s) / {(SIGMAS(i,r) -1) * VIMSHR(i,r,s) + N_L(i,r)} ! Define Parameters ! SIGMAS (i,r : TRAD_COMM, REG) : elasticity of substitution ESUBVA(i : PROD_COMM) : elasticity of substitution between primary endowments N_L (i,r : IMC_COMM, REG) : number of IMC_COMM firm i in region r INC(r : REG) : = INCOME(r) I Define Equations ! EQUATION (i,j,r : ENDW_COMM, PCM_COMM, REG) : calculate qfep(i,j,r) qfep(i,j,r) = qvap(j,r) + ESUBVA(i) * [pvap(j,r) - pfe(i,j,r)] 128 EQUATION (i,j,r : ENDW_COMM, IMC_COMM, REG) : calculate qfem(i,j,r) qfem(i.j,r) = “VVA(i,r)/VA(i,r)] * qua0,r)} + {[FVAG,r)/VA(j,r)] * fqva(j,r)} + ESUBVA(i) * [pvam(j,r) - pfe(ij,r)] EQUATION (i,r : PCM_COMM, REG) : calculate pvap pvap(j,r) = sum(e,ENDW_COMM, ESHR(e,j,r) * pfe(e,j,r)) EQUATION (i,r : IMC_COMM, REG) : calculate pvam pvam(j,r) = sum(e,ENDW_COMM, ESHR(e,j,r) * pfe(e,j,r)) EQUATION (i,r : IMC_COMM, REG) : top nest of IMC_COMM production function qua(j,r) = qo(j,r) EQUATION (j,r : IMC_COMM, REG) : determination of fqva(j,r) fqva(j,r) = n(j,r) EQUATION (i,r : PCM_COMM, REG) : top nest of PCM_COMM production function qvaptiJ) = qotiJ) EQUATION (i,j,r : TRAD_COMM, PCM_COMM, REG) : top nest of PCM_COMM production function qf(i,i,r) = (100;) EQUATION (i,j,r : TRAD_COMM, IMC_COMM, REG) : top nest of IMC_COMM production function qf(i.i,r) = C100 ,r) EQUATION (i,j,r,s :NIMC_TRAD, PROD_COMM, REG, REG) : calculate qfsp(i,j,r,s) qfsp(i,j,r,s) = qf(i,j,s) - SIGMAS(i,s) * [pfs(i,j,r,s) - pf(i,j,s)] EQUATION (i,j,r,s : IMC_COMM, PROD_COMM, REG, REG) : calculate qfsm(i,j,r,s) 129 qfsm(i,j,r,s) = qf(ij,s) - SIGMAS(i,s) * [pfs(i,j,r,s) - pf(i,j,s)] - sum(l,REG,FASHRS(i,j,l,s)*n(i,l)) EQUATION (i,j,r : TRAD_COMM, PROD_COMM, REG) : calculate pf(i,j,r) pf(i,j,r) = sum(c,REG, FASHRS(i,j,c,r) "‘ pfs(i,j,c,r)) EQUATION (i,s : TRAD_COMM, REG) : calculate pm(i,s) pm(i,s) = [sum(f,REG, VIMSHR(i,f,s) * pms(i,f,s))] EQUATION (i,r : IMC_COMM, REG) : calculate ps(j,r) ps(j,r) = avc(j,r) + mkr(j,r) EQUATION (i,r : IMC_COMM, REG) : calculate avc(j,r) VC(i,r) * avc(j,r) = sum(i,TRAD_COMM, VFA(ij,r) * pf(i,j,r)) + VVA(i,r) * pvam(j,r) EQUATION (i,r : IMC_COMM, REG) : calculate atc(j,r) VOA(j,r) * atc(j,r) = sum(i,TRAD_COMM, VF A(i,j,r) * pf(i,j,r)) + VA(i,r) * pvam(j,r) EQUATION (j,r : IMC_COMM, REG) : calculate IMC_COMM ps(j,r) VOA(j,r) * ps(j,r) = VOA(j,r) * atc(j,r) - FVA(i,r) * [qo(j,r) - n(j,r)] EQUATION (i,r :PCM_COMM, REG) : calculate PCM_COMM ps(j,r) VOA(j,r) * ps(j,r) = sum(i,TRAD_COMM, VFA(ij,r) * pflij,r)) + VA(i,r) * pvap(j,r) EQUATION (i,r : TRAD_COMM, REG) : calculate qc(i,r) qc(i,r) = W) - pC(i,r) EQUATION (r : REG) : calculate qsave(r) 130 qsave(r) =y(r) - pcgds EQUATION (i,r : TRAD_COMM, REG) : calculate pc(i,r) pc(i,r) = [sum(p,REG, CASHRS(i,p,r) * pcs(i,p,r))] EQUATION (i,r,s : NIMC_TRAD, REG, REG) : calculate qcsp(i,r,s) qcsp(i,r,s) = qc(i,s) — SIGMAS(i,s) * [pcs(i,r,s) - pc(i,s)] EQUATION (i,r,s : IMC_COMM, REG, REG) : calculate qcsm(i,r,s) qcsm(i,r,s) = qc(i,s) - SIGMAS(i,s) * [pcs(i,r,s) - pc(i,s)] - sum(l,REG, CASHRS(i,l,s) * n(i,l)) EQUATION (r : REG) : calculate regional utility u(r) INCOME(r) * u(r) = sum(i,TRAD_COMM,VCA(i,r) * qc(i,r)) + SAVE(r)*qsave(r) EQUATION (i,r,s : NIMC_TRAD, REG,REG) : calculate NIMC_COMM qxs(i,r,s) qxs(i,r,s) = sum(k,PROD_COMM, [VFMS(i,k,r,s)NIMS(i,r,s)] * qfsp(i,k,r,s)) + [VCMS(i,r,s)NIMS(i,r,s)]* qcsp(i,r,s) EQUATION (i,r,s : IMC_COMM, REG, REG) : calculate IMC_COMM qxs(i,r,s) qxs(i,r,s) = sum(k,PROD_COMM, [VFMS(i,k,r,s)NIMS(i,r,s)] * qfsm(i,k,r,s) + [VCMS(i,r,s)/VIMS(i,r,s)]* qcsm(i,r,s) EQUATION (i,r : ENDW_COMM, REG) : calculate ENDW_COMM qc(i,r) VOM(i,r) * qc(i,r) = sum(j,PCM_COMM, VFM(i,j,r) * qfep(i,j,r)) + sum(j,IMC_COMM, VFM(i,j,r) * qfem(i,j,r) EQUATION : calculate pcgds VCGDS * pcgds = sum(r,REG, surn(k,CGDS_COMM, VOA(k,r) * ps(k,r))) 131 EQUATION (r : REG) : calculate globalcgds globalcgds = surn(k,CGDS_COMM, qo(k,r)) EQUATION (i,r,s : TRAD_COMM, REG, REG) : calculate pms(i,r,s) pms(i,r,s) = tms(i,r,s) + pw(i,r,s) EQUATION (i,r,s : TRAD_COMM, REG, REG) : calculate pw(i,r,s) pw(i,r,s) = ps(i,r) - txs(i,r,s) EQUATION (i,r : ENDW_COMM, REG) : calculate pme(i,r) pme(i,r) = ps(i,r) - to(i,r) EQUATION (i,j,r : ENDW_COMM, PROD_COMM, REG) : calculate pfe(i,j,r) pfe(ij,r) = tf(i,j,r) + pme(i,r) EQUATION (i,r,s : TRAD_COMM, REG, REG) : calculate pcs(i,r,s) pcs(i,r,s) = tcs(i,r,s) + pms(i,r,s) EQUATION (i,j,r,s : TRAD_COMM, PROD_COMM, REG, REG) : calculate pfs(i,j,r,s) pfs(i,j,r,s) = tfs(i,j,r,s) + pms(i,r,s) EQUATION (i,r : TRAD_COMM, REG) : calculate qc(i,r) VOA(i,r) * qc(i,r) = sum(s,REG, VXA(i,r,s) * qxs(i,r,s)) EQUATION (i,r : IMC_COMM, REG) : calculate mkr(i,r) mkr(i,r) = {1 - TELA(i,r) / [TELA(i,r) -1]} * e1a(i,r) EQUATION : (i,r,s : IMC_COMM, REG, REG) : calculate elas(i,r,s) elas(i,r,s) = - ALPHA(i,r,s) * [pm(i,s) - pms(i,r,s)] EQUATION (i,r : IMC_COMM, REG) : calculate e1a(i,r) e1a(i,r) = - qc(i,r) + sum(s,REG, SHRPELAS(i,r,s) * [elas(i,r,s) + qxs(i,r,s)]) 132 EQUATION (r : REG) : calculate EV(i,r) EV(r) - [INC(r)/ 100] * u(r) = o EQUATION (r : REG) : calculate gp(i,r) [sum(i,TRAD_COMM, VCA(i,r)) + SAVE(r)] * gp(r) = sum(i,TRAD_COMM, VCA(i,r) * pc(i,r)) + SAVE(r) * pcgds EQUATION (r : REG) : calculate regional income y(r) y(r) = {1/INCOME(r)} * {sum(m,ENDW_COMM, VOA(m,r) * (PS(m,r) + c10(mJ)))+ surn(mJMCLCOMM, VOA(m,r) "' lPS(m,r) + q0(m,r)] - sum(h,TRAD_COMM, VFA(h,m,r) * [pf(h,m,r) + qf(h,m,r)])- VA(m,r) * [Pvammfi + {[VVA(mJ)/VA(mJ)] * qua(m,r)} + {[FVA(m,r)/VA(m,r)l * fqva(marfll) + suin(m,PCM_C0MM, VOA(mJ) * [PS(m,r) + q0(m,r)] - sum(h,TRAD_COMM, VFA(h,m,r) * [pf(h,m,r) + qf(h,m,r)])- VA(m,r) * [pvap(m,r) + qo(m,r)]) + sum(m,ENDW_COMM, (VOM(m,r) * (pme(m,r) + q0(m,r))) - (VOA(mJ) * (PS(m,r) + q0(m,r)))) + surn(h,ENDW_COMM, sum(m,PCM_COMM, (VFA(h,m,r) * (pfe(h,m,r) + qfep(h,m,r))) -(VFM(h,m,r) * (pme(h,r) + qfep(h,m,r)))) + sum(m,IMC_COMM, (VFA(h,m,r) * (pfe(h,m,r) + qfem(hamJD) - (VFM(h,m,r) * (Pm€(h,r) + qfem(hamJD)» + sum(m,PROD_COMM, sum(h,NIMC_TRAD, sum(s,REG, (VFAS(h,m,s,r) * (PfS(h,m,SJ) + quP(h,m,SJ))) - (V FMSGLHLSJ) * (Pm5(h,5,r) + quPmamasar)))))) + sum(m,PROD_COMM, sum(h,IMC_COMM, sum(s,REG, (VFAS(h,m,s,r) * (pfs(h,m,s,r) 1+ n(h,s)! + qfsm(h,m,s,r))) — (VFMS(h,m,s,r) * (pms(h,s,r) + qfsm(h,m,s,r)))))) + sum(h,NIMC_TRAD, sum(s,REG, 133 (VCAS(h,s,r) * (pcs(h,s,r) + qcsp(h,s,r))) - (VCMS(h,s,r) * (pms(h,s,r) + qcsp(h,s,r))))) + sum(h,IMC_COMM, sum(s,REG, (V CAS(h,s,r) * (pcs(h,s,r) + qcsm(h,s,r))) - (V CMS(h,s,r) * (pms(h,s,r) ! + n(h,s)! + qcsm(h,s,r))))) + sum(h,TRAD_COMM, sum(s,REG, (VIWS(h,r,s) * (pw(h,r,s) + qxs(h,r,s))) - (V XA(h,r,s) * (ps(h,r) + qxs(h,r,s))))) + sum(h,TRAD_COMM, sum(O,REG, (VlMS(h,o,r) "‘ (pms(h,o,r) + qxs(h,o,r))) - (VIWS(h,o,r) * (pw(h,o,r) + ClXS(h,0,r)))))} EQUATION : check Walras law VSAVE * walras_dem = sum(r,REG, SAVE(r) * qsave(r)) EQUATION : check Walras law walras_sup = globalcgds EQUATION : Walras law walras_sup = walras_dem + walraslack 134 LIST OF REFERENCES : Armington, Paul (1969), “A Theory of Demand for Products Distinguished by Place of Production,” W 16, pp. 159-178. Ballard, Charles L. (1990), “Marginal Welfare Cost Calculations. Differential Analysis vs. Balanced Budget Analysis,” WWW 41(2), pp. 263- 276. Brown, Drusilla K. (1991), “Tariffs and Capacity Utilization of Monopolistically Competitive Firms,’ ’lgmmamflmemmmnaLEggngmms 30(3-4), pp. 371 -.81 Brown, Drusilla K. (1992), “Properties of Computable General Equilibrium Trade Models with Monopolistic Competition and Foreign Direct Investment,” Working Paper, Tufts University. Brown, Drusilla K., Alan V. Deardorff, and Robert M. Stern (1992), “A North American Free Trade Agreement: Analytical Issues and a Computational Assessment,” Wm, London, Basil Blackwell. Brown, Drusilla K. and Robert M. Stern (1989a), “Computable General Equilibrium Estimates of the Gains from US-Canadian trade Liberalization,” in David Greenaway, Thomas Hyclack, and Robert .1. Thorton (eds. ) W W, New York, New York University Press. Brown, Drusilla K. and Robert M. Stern (1989b), “U. S.- Canada Bilateral Tariff Elimination. The Role of Product Differentiation and Market Structure,” in Robert C Feenstra(ed)Irade_Belicies_fQLIntemamnal_QQmpetlti1/eness Chicago, University of Chicago Press. Choc, Myung-Gun (1992), “Provisional PARF TA and Its Economic Effect,” 103mm Intematienalfimnemielmegmtien 7 (1)91). 102112 Christelow, Dorothy (1990), “Japan’s Intangible Barriers to Trade In Manufactures,” Philip Klng, ed ,IntszmatienaLE952n9mis;s_andIntematienaLEs‘aQanis‘aEQliclt,m New York. McGrawHill. 135 Cox, David and Richard G. Harris (1985), “Trade Liberalization and Industrial Organization: Some Estimates for Canada,” MW 93, pp. 115-45. Cox, David and Richard G. Harris (1986), “A Quantitative Assessment of the Economic Impact on Canada of Sectoral Free Trade with the United States,” CanadianlmzmalefiEchomics 19, pp 371-94 Cox, David and Richard G. Harris (1992), “North American Free Trade and Its Implication for Canada: Results from a CGE Model of North American Trade,” The NAFTA Economywide Modeling Conference, U.S.I.T.C., Washington, DC, February 24-25. Dauphin, Roma (1978), WWW, Economic Council of Canada. de Melo, Jaime and Sherman Robinson (1989), “Product Differentiation and the Treatment of Foreign Trade In Computable General Equilibrium Models of Small Economies ,”.Io_umal_Q£IntematiQnaLE9_Qanics 27 pp 47-67 Dixit, Avinash and Victor Norman (1988), “Scale Economics and Imperfect Competition,” in W, Cambridge University Press, pp. 265- 295. Dixit, Avinash, and Joseph E. Stiglitz (1977), “Monopolistic Competition and Optimum Product Diversity,” WW 67(3), pp. 297-307. Drobnick, Richard L. (1992), “Economic Integration in the Pacific Region,” Paris : Organization for Economic Co-Operation and Development, Technical Paper No. 65. Eastman, H.C. and S. Stykolt (1960), “A Model for the Study of Protected Oligopolies,” Economielmunal 70, pp. 336-47. Eastman, HQ and 3- Stykolt (1967), IheIanffandCempetmoninCanada, Toronto: Macmillan. English, H. Edward (1989), "International Economic Integration and National Policy Diversity," in K L Brownsey (ed ), Canadazlapanaflewmnfllchahzatien. Halifax, Nova Scotia, Canada: The Institute for Research on Public Policy. Flam, Harry, and Elhanan Helpman (1987), “Industrial Policy Under Monopolistic Competition,” Ionmalpflntematicnalficenomics 22, pp 79-102 136 Gehlhar, Mark J. (1994), “Historical Analysis of Growth and Trade Patters in the Pacific Rim: A Validation Exercise,” Fifth International CGE Modeling Conference, University of Waterloo, Canada, October 27-29. Harris, Richard (1984), “Applied General Equilibrium Analysis of Small Open Economies with Scale Economics and Imperfect Competition,” Amerim EggngmicReyim December, 74(5), pp. 1016-1032. Harrison, Glenn W., Thomas F. Rutherford and David G. Tarr (1995), “Quantifying the Uruguay Round,” World Bank Conference on “The Uruguay Round and the Developing Economies,” Washington DC, January 26 - 27. Harrison, Jill, and Ken Pearson (1993), Wan, The Impact Project and KPSOFT, Monash University, Clayton, Vic., Australia. Helpman, Elhanan (1981), “International Trade in the Presence of Product Differentiation, Economies of Scale and Monopolistic Competition: A Chamberlin-Heckscher—Ohlin Approach, ” W 11, pp. 305- 340. Helpman, Elhanan and Assaf Razin (1983), “Increasing Returns, Monopolistic Competition, and Factor Movements: A Welfare Analysis,” Jenmalnflntemhonalfimnomlcs 14(3-4), pp. 263- 76. Hertel, Thomas W. (1992), “The ‘Procompetitive’ Effects of Trade Policy Reform 1n a Small Open Economy,” Ioumamflmmationalficonomics, May Hertel, Thomas W., J. Mark Horridge, and K. R. Pearson (1993), “Mending the Family Tree : A Reconciliation of the Linearization and Levels Schools of CGE Modeling”, EmngmigMQdeljng 3(2), pp. 71-140. Hertel, Thomas W., and M. E. Tsigas (1994), WWW, London : Cambridge University Press. Hunter, Linda, James R. Markusen, and Thomas F. Rutherford (1992), “Trade Liberalization in a Multinational-Dominated Industry: A Theoretical and Applied General Equilibrium Analysis,” The NAFTA Economywide Modeling Conference, U.S.I.T.C., Washington, DC, February 24-25. Johansen, Leif. (1960), WWW Amsterdam: North-Holland Publishing Company, (2nd edition, 1974). Jomini, P., J.F. Zeitsch, R. McDougall, A. Welsh, S. Brown, J. Hambley, and J. Kelly (1991), “Model Structure, Database and Parameters,” in W 137 EgmhhnumModeLefthefloddEcenomy, Vol. 1, Industry Commission : Canberra, Australia. Karikari, John A. (1988), “International Competitiveness and Industry Pricing in Canadian Manufacturing,” Canadianlnumalnffienanmien 21(2), pp. 410-26. Koo, Bon Ho (1990), “Korea’s Perspective on Asia-Pacific Economic Cooperation,” in John P Hardt and Young C Kim eds ,EmmmmgperationintheAsiazflacific Regina, San Francisco. Westview Press, pp. 89-108. Kreinin, Mordechai E. ,and Micheal G. Plummer (1994), "Economic Effects of the North American Free-Trade Area on Australia and New Zealand, " lmnaainfiEenanmin 1111221811911 9(1), pp. 1-28. Kreinin, Mordechai E. and Micheal G. Plummer (1992), “Effects of Economic Integration in Industrial Countries on ASEAN and the Asian NIEs,” WnrlgL Dexelopmem 21 (9), pp- 1345-66- Krugman, Paul (1979), “Increasing Returns, Monopolistic Competition, and International Trade,” WW 9, pp 469-479. Krugman, Paul (1980), “Scale Economics, Production Differentiation, and the Pattern of Trade,” AmericanEeonomieReym 70, pp- 950-959. Krugman, Paul (1991), “Increasing Returns and Economic Geography,” Jnumalnfl BolitiealEconomy 99, pp. 483-499. Krugman, Paul (1981), “Intraindustry Specialization and the Gains from Trade,” ImmalnffloliticaLEconomy 89 PP 959 973 Lee, Hiro, and David Roland-Holst (1994), “Trade Liberalization and Employment Linkages in the Pacific Basin,” OECD Technical Paper No. 94, produced as part of the research programme on Sustainable Development : Environment, Resource Use, Technology and Trade. Magun, Sunder, Someshwar Rao, and Bimal Lodh (1987), “Impact of Canada-US Free Trade on the Canadian Economy,” Paper no. 331, Economic Council of Canada. Markusen, James R., and Lars E.O. Svensson (1986), “Factor Endowments and Trade with Increasing Returns : Generalization and Extensions,” Working Paper, Stockholm and London : University of Stockholm and University of Western Ontario. 138 Markusen, James R. and Randall M. Wigle (1989), “Nash Equilibrium Tariffs for the United States and Canada: The Roles of Country Size, Scale Economics, and Capital Mobility,”,[namal_n££_nmieaLEnnanmy 97, pp. 368-386. Mercenier, Jean (1995), “Can‘ 1992’ Reduce Unemployment In Europe? On Welfare and Employment Effects of Europe’s Move to a Single Market,” Lnumalnfl’niiey Modelinglfll), pp. 138 Mercenier, Jean, and Nicolas Schmitt (1992), "Sunk Costs, F rec-Entry Equilibrium, and Trade Liberalization in Applied General Equilibrium : Implications for 'Europe 1992' ," mimeo, University of Montreal and Simon Fraser University, Canada. Nguyen, Trien T., and Randall M. Wigle (1992), “Trade Liberalization with Imperfect Competition : The Large and Small of It,” W 36, pp. 17-35. Norman, Victor D. (1990), “Assessing Trade and Welfare effects of Trade Liberalization,” EarnneaafinnanminRuiew 34, pp. 725-51. Pearson, Charles, S. (1994), “The Asian Export Ladder,” in Shu-Chin Yang, ed., U. 1.9 .'0 .HI 0 .-. .1211..- '0. .i° OHII‘-'o .- Regienalmnmtian, pp. 35 52 Prattem, Cliff (1988), “A survey of the Economics of Scale,” in Smdiesnnjhe Ennanminsnflmegmina, Vol. 11, Commission of the European Communities, Document. Sakong, Il (1993), Knreajajheflnfldfienamay, Institute for International Economics, Washington, DC. Scarf, Herbert E- (1973), W New Haven : Yale University Press. Shibusawa, Masahide, Zakaria Haji Ahmade, and Brian Bridges (1992), Eaeifie Asia in 11111229, London : Routledge. Sobarzo, Horacio E. (1992), “A General Equilibrium Analysis of the Gains From Trade for the Mexican Economy of a North American Free Trade Agreement,” Worm: 15(1), pp- 33-100- United States International Trade Commission (1989), “The Pros and Cons of Entering into Negotiations on Free Trade Area Agreements with Taiwan, The Republic of Korea, and ASEAN, or The Pacific Rim Region in General,” Report to the Senate Committee on Finance on Investigation, U.S.I.T.C. Publication # 2166, Washington DC: U.S.I.T.C. 139 Whalley, John (1985), IradeLiberalizatimlAmonRMajoLWoflflradingAreas, Cambridge, MA: MIT Press. Whalley, John (1986), “Impacts of a 50% Tariff Reduction In an Eight Regional Global Trade Model,” in T. N. Srinivansan and J. Whalley, eds. ,Wfim ImdeMndeliag, Cambridge, MA. MIT Press. Yang, Shu-Chin (1994), “Open Industrialization in East Asia and the Quest for Regional Cooperation: An Overview,” in Shu-Chin Yang, ed. ,MaanfaenneiExannenf .‘t -..I 10.. lo. AI' OIHI'- '0 0‘ {”H H"... H,pp.-.334 ll MICHIGAN STATE UNIV. LIBRARIES 11111111111111111111111111111111111 31293014109098