II.4. Trade and Climate Blocs. Claudia Kemfert and Barbara Buchner *

II.4. Trade and Climate Blocs Claudia Kemfert and Barbara Buchner * Abstract Substantial reduction of GHG emissions requires co-operation between countries. However, greenhouse gas emissions reduction is an international public good necessitating long term and global economic efforts. The formulation of the Kyoto Protocol and its subsequent specifications represent an initial outcome of co-operative international climate control policy actions. Nevertheless, a variety of incentives exist to free ride on climate control. As shown in the previous chapter, these incentives to free ride can be addressed by using trade sanctions. In this chapter we focus on a different link between trade and climate policy, Indeed, we assume that the existing trade cooperation can provide incentives to undertake an effective cooperative climate control. In this scenario, we analyse whether climate blocs based on existing trade blocs could constitute an effective approach to climate policy. Therefore, in this chapter, the economic and environmental impacts of different trade coalitions cooperating on climate control are assessed. 1. Introduction In contrast to environmental coalitions, agreements on other economic issues (e.g. trade) are characterised by a limited amount of free-riding incentives. Typically, economic co-operations are proposed for the provision of private club goods. These goods induce benefits which are excludable, i.e. cannot be reaped by free-riders. Therefore, cooperation is more likely to emerge than in the case of environmental issues, as shown by the number of implemented agreements on trade or R&D cooperation. Nonetheless, even in the case of trade, global co-operation is difficult to achieve and has not yet been reached. Although global free trade has been encouraged by the WTO, a resurgence of regionalism has been observed. This movement is demonstrated by the recent formation of various strong trading blocs, a trend initiated by the European Union (EU) and continued by the North American Free Trade Agreement (NAFTA), Mercosur (consisting of Argentina, Brazil, Paraguay, Uruguay) and other trade blocs. A strand of literature has focused on the formation of trade blocs and their characteristics, most importantly regarding size, stability and welfare effects (e.g., Viner, 1950; Baldwin, 1993; Casella, 1995; Bloch and Ferrer, 1999; Bond and Syropoulos, 1996; Krugman, 1991; Yi, 1996a, 1996b and 1998). In international trade, the resurgence of regionalism has thus become a crucial subject, underlined by the formation of competing customs unions and the debate about free trade areas. Substantial attention has been given to the efficiency and implications of regional co-operations. It has been shown that incentives to develop regional trade cooperations are much stronger than those behind global trade cooperation (see, for example, Yi 1996a, b). Therefore, one may wonder whether the existence of trade blocs can facilitate the emergence of climate blocs. Several reasons explain why this is often the case. For example, countries which are used to cooperate on one issue can cooperate more easily on other issues. Or, countries which cooperate on one issue can link cooperation on that issue to cooperation on other issues (as discussed in chapter I.4). In practice, there are examples of climate University of Oldenburg and FEEM. Fondazione Eni Enrico Mattei, Venice.

KEMFERT & BUCHNER cooperations superposed to the existing trade cooperation. The EU bubble represents such a case, but the recent talks about climate cooperation within NAFTA or among ASEAN countries are also interesting examples. In this chapter we fist analyse the economic theory of trade blocs and we discuss the incentives for the formation of these blocs. Then, we assume that the incentives to the formation of trade blocs, the links between different economic issues inside a trade blocs, the culture of cooperation which develops inside a trade bloc, are such as to induce countries to cooperate also on climate control. Given this assumption, we explore the consequences of this type of cooperation, which is not the one aimed to by the UNFCCC, but corresponds to a more realistic approach in which trade blocs act as catalysers of climate cooperation and may achieve better results than other forms of partial cooperation. 2. A Game Theoretic Analysis of Trade Blocs Before providing some empirical evidence which could help answer the question whether trade blocs committed to climate control could effectively reduce GHG emissions, this section addresses the trade blocs issue from a theoretical viewpoint. The methodological framework used in the analysis is the game-theoretic approach to the endogenous formation of economic coalitions. To overcome some of the basic difficulties of traditional cooperative game theory, and to appropriately handle the issue of competition among coalitions, a new strand of literature has embedded the coalition formation process in a non-co-operative framework. This more general approach enables the analysis of co-operations which have emerged in various economic fields such as international trade, public finance or environmental economics. The games which are used to analyse the formation of coalitions can usually be divided into two categories: (i) games with positive spillovers, if the formation of a coalition leads to positive effects on non-members, and (ii) games with negative spillovers, when the formation of a co-operation induces adverse effects on the non-participating countries. In contrast to agreements on the provision of public goods, which can be represented as games of the first category, trade co-operations can usually be described by games with negative spillovers. Indeed, a co-operation in which regional customs unions form induces adverse effects on the non-member countries, e.g. by levying an external tariff on their imports. In the discussion about trade blocs, the distinction between customs unions and free trade areas is particularly important. A customs union is a coalition which abolishes any tariff on imports among the members and instead sets a common external tariff on non-member countries. A free trade area is characterised by a set of countries which eliminate their internal tariffs but let the individual members set their welfare-maximising external tariffs. Both these economic coalitions in international trade raise a number of important issues for economists and policy makers. Two factors have received considerable interest (Yi, 2001): First, it would be important to identify the private and social benefits of the formation, expansion, and merger of trading blocs. How does the formation of a customs union or a free trade area affect the volume and terms of trade and welfare of the participating countries and the outsiders? Second, the outcome of the endogenous coalition formation process needs a more detailed specification. In international trade, when economic agents can form coalitions freely, does a stable coalition structure of trading blocs exist as an equilibrium outcome? How is it characterised? Is the grand coalition and therefore global free trade a stable outcome? And, one of the most often posed questions: Are regional trading blocs stepping stones or stumbling blocks towards global free trade? From a theoretical viewpoint, the existing literature on coalition formation provides some reliable indications on the main features of the final equilibrium outcome. However, a difficulty exists in 2

TRADE AND CLIMATE BLOCS finding general results due to the differences between customs unions and free trade areas. Therefore, this section will discuss both cases. 1 Customs Unions With the emergence of three major trading blocs North America, Europe and an Asian bloc the formation of customs unions in international trade has received a lot of attention. According to Krugman (1991) a customs union is a coalition which abolishes any tariff on imports among the members and instead sets a common external tariff on non-member countries. Within a trade bloc, and in the sense of a customs union, members trade freely within the group and have a common trade policy for non-members, thereby maximising the joint welfare of the coalition. Following the strong concern over the impacts of an increased number of regional trade blocs on the effectiveness of multilateral trading systems, the implications of the expansion of trading blocs on their external tariffs, and thus on the level of world welfare, became an important question. As indicated above, the main doubt is whether regional trade blocs represent stumbling blocks against free trade, by e.g. increasing the group s external barriers to trade in order to constrain the other countries market access, 2 or whether they can be seen as stepping stones towards global free trade, enabling better negotiations with non-member countries in terms of tariff reductions. To better characterise games with negative spillovers, Yi (1997) introduces three conditions. 3 The first condition defines these games and states that when coalitions merge to form a larger coalition, the non-participating players are worse off. Condition 2 says that a member of a coalition becomes better off if its coalition merges with larger or equal-sized coalitions whereas the last condition says that a member of a coalition becomes better off if it defects from its present coalition to join another coalition which is larger or of equal size. Starting from this framework, Yi (1996a) demonstrates that at least under the open regionalism rule where non-members can join a coalition without the consensus of the coalition members 4 regional trade unions can indeed be stepping stones towards global free trade, since the expansion of trade blocs improves the aggregate welfare of the existing and new member countries (even without side payments). 5 By characterising the functioning of customs unions as a two-stage game, Bloch (2001) agrees that if in the first stage players decide to participate in a coalition, then in the second stage they can increase profits. While the aggregate welfare of member countries is thus increased, the welfare-maximising trade coalition always makes non-members worse off. 6 In particular, the welfare of non-members is reduced after the trade coalition has imposed welfare-maximising tariffs on their imports, because the non-members suffer a deterioration of their terms of trade effects. In general, the formation or expansion of trade unions induces an ambiguous effect on world welfare: Krugman (1991), considering only symmetric cases, associates an expansion of trade blocs with an increase in market power and thus with a reduction in global welfare. However, extending Krugman s analysis, Bond and Syropoulos (1996) find that depending on the tariffs the expansions may either decrease or increase global welfare. Yi (1996a) confirms the ambiguity of the effects on global welfare 1 For excellent overviews on games with positive and negative spillovers, also focusing on the formation of trade cooperation, see Yi (2001) and Bloch (2001). 2 For example, Krugman (1991) argues that the expansion of trade blocs reduces world welfare by increasing their market power. However, he analyzes only symmetrical coalition structures. 3 For a formal analysis and a more detailed discussion of the axioms see Yi (1997) and Yi (2001). 4 In accordance with the membership rules used in coalition formation theory, Yi (1996a) uses the open regionalism rule as a synonym for the open membership rule in the context of trade agreements and the unanimous regionalism rule as a synonym for the unanimous membership rule. 5 However, when the membership to trade blocs becomes exclusive, i.e. requiring the consent of the actual coalition members, then regional trade co-operations can become stumbling blocks against global free trade. 6 Apparently, an important factor is the size of the trade blocs. It is generally recognised that a cooperating country of a large bloc has a higher level of welfare than a country belonging to a smaller trade bloc. 3

KEMFERT & BUCHNER but emphasises an important exception: When membership in a trade coalition is open to nonmembers, the grand coalition maximises global welfare (since all the countries are involved). Extensive research has focused on the stability of trade blocs. Baldwin s domino theory of regionalism shows that the expansion of a regional trade bloc triggers membership requests from countries which were not yet participating in the coalition, which may ultimately result in global free trade (Baldwin, 1993). These insights are also supported by Bond and Syropoulos (1996) and by Kennan and Riezman (1990) who demonstrate that two countries can enhance their welfare with respect to the free trade situation if they decide to form a trading bloc. Bond and Syropoulos (1996) show that there is always a bloc size at which the expanding bloc can increase its welfare with respect to the level attainable without trade coalitions. This finding implies that there is an incentive for trade blocs to induce other countries to enter and for non-members to join the large trade blocs. Nonetheless, Macho-Stadler et al. (1998), using a co-operative approach, find that although multilateral trade negotiations may yield gains to all participants, they can be unstable. Looking at the final equilibrium outcome, the grand coalition, i.e. global free trade, is found to be the unique equilibrium outcome if membership to the customs unions is open to all countries (Yi, 1996a and 1996b). 7 However, under more restrictive conditions, the equilibrium outcome is typically formed by several trade blocs. In particular, Yi (1996a) finds that under an exclusive membership rule, which means that membership to the trade blocs is dependent on the approval by all existing coalition members, two asymmetric trade blocs typically emerge at the equilibrium. Free Trade Areas The main part of the theoretical literature on regional trade blocs focused on customs unions. Recently, more attention has been given to the analysis of free trade areas (Yi, 1998; Bond et al., 1999). The reason for this interest is that a number of emerging trade co-operations takes the form of free trade areas, as is demonstrated by the examples of NAFTA and Mercosur. As in the customs unions, members of a free trade area eliminate tariffs among themselves. However, the distinguishing characteristic of a free trade area is that each coalition member sets its external tariff in order to maximise its own individual welfare. As a consequence, free trade areas represent a special case of games with spillovers since this different way of determining the external tariffs has important implications on the welfare of non-participating countries (Yi, 1998). Similar to the previous type of trade co-operation, the formation or expansion of free trade areas induces an ambiguous effect on global welfare. However, due to the positive externalities caused by free trade areas, the motivation for this result is different from the one previously described. In strong contrast to customs unions, Yi (1998) shows that free trade areas may induce a sufficiently large reduction of external tariffs to make also non-members better off. 8 Indeed, countries in the free trade agreement may decide to impose lower tariffs on the imports of some countries to help restore a balanced consumption portfolio (Yi, 1998). This latter result is supported by a recent study by Bond et al. (1999) who confirm Yi s findings by demonstrating that the formation of a free trade area indeed improves the terms of trade and welfare of non-members by creating an incentive for the co-operating countries to reduce their external tariffs. In general, the aggregate welfare of member countries in free trade areas is improved because the formation of free trade areas is Pareto-improving. As shown by Yi (1998), only privately beneficially 7 This result is in strong contrast with the theoretical findings in the case of provision of public goods, e.g. environmental agreements. In particular, self-enforcing environmental agreements may emerge as an equilibrium outcome, but their size remains limited for any functional specification of countries welfare functions. For a detailed discussion see chapter I.3. 8 The underlying reason for this result is the supermodularity of the welfare function in tariffs which exists when countries have love of variety preferences (Yi, 1998). 4

TRADE AND CLIMATE BLOCS agreements are signed voluntarily. However, Bond et al. (1999) identify two opposing effects with respect to the welfare of member countries. The worsening of the terms of trade at the equilibrium reduces their welfare whereas the expansion of international trade results to be welfare increasing. As long as the size of the member countries is sufficiently large, the positive effect dominates and the formation of this trade co-operation indeed represents a Pareto improvement. However, in contrast to the situation identified with respect to customs unions, the global free trade agreement may not be an equilibrium outcome because of the existence of free-riding incentives. In particular, Yi (1998) explains that a country can become better off by leaving the global free trade agreement because the remaining members raise their welfare-maximising tariffs on the deviator only by a small amount since they impose no tariffs on the other members. Moreover, Bond et al. (1999) find that the formation of a free trade agreement may increase the probability that a country prefers this trading bloc to free trade. This implies that the long-run objective of a multilateral trade liberalisation may be hurt by the presence of a regional free trade agreement. Summing up, the welfare effects of trade blocs are ambiguous. This ambiguity in the context of a preferential reduction in tariffs has indeed been emphasised since Viner (1950). In particular, we can conclude that the effect on non-members welfare is different in the two cases (custom unions and free trade areas). As Yi (2001) points out, the monopoly power of the coalition members is responsible for this result. If monopoly power is exerted, as in the case of customs unions, non-members may suffer, whereas they may benefit when the joint monopoly power of the trade bloc is not used, as in the context of free trade areas. As a consequence of the differences in the two types of trade blocs, the effects on welfare and on the stability of the equilibrium outcome induced by the formation of trade blocs are difficult to generalise. In the case of customs unions, global free trade is the stable outcome when non-members can join the coalition without any constraint. Regional trading blocs in the sense of customs unions can thus be stepping stones towards a global co-operation. By contrast, the formation of free trade areas rarely leads to global cooperation in international trade. 3. Impacts of Trade Coalitions Co-operating on Climate Change Control Let us now analyse what would be the environmental impact of a scenario in which one or more trade coalitions decide to cooperate on climate control as well. However, before going into details, we would like to stress that the results of this chapter are not intended to represent the actual situation in climate policy, but are rather to be seen as counterfactual experiments which could give useful insights into the future design of climate policy. First, we discuss general results obtained using the applied general equilibrium model WIAGEM and explain the modelling framework. Then, we will focus on a number of existing trade blocs. Finally, the last subsection will analyse possible evolutions of single co-operations. 3.1 The Modelling Framework and General Results The analysis presented in this chapter has been carried out using the multi-regional, multi-sectoral trade model WIAGEM, an applied general equilibrium model described in detail at the end of Chapter II.2. The basic framework of the model is an intertemporal computable general equilibrium and multiregional trade model covering 25 world regions that have been aggregated to achieve 11 trading regions which are linked through bilateral sectoral trade flows. The model is based on GTAP 4.0 data 9. The game-theoretic framework is the one proposed in Carraro and Siniscalco (1997) and Buchner, Carraro et al. (2002) where countries play a two-stage game. Negotiating countries first decide non-cooperatively whether or not to join a coalition. Then, in the second stage, if they join the coalition, they 9 See McDougall, R.A. (1995). 5

KEMFERT & BUCHNER play a co-operative, open-loop Nash game to determine their welfare maximising policy variables. If they do not join the coalition, they play as singletons against the coalition in a non-cooperative way. ASIA CHN CNA EU15 JPN LSA MIDE REC ROW SSA USA Regions India and other Asian Countries (Republic of Korea, Indonesia, Malaysia, Philippines, Singapore, Thailand, China, Hong Kong, Taiwan) China Canada, New Zealand and Australia European Union Japan Latin America (Mexico, Argentina, Brazil, Chile, Rest of Latin America) Middle East and North Africa Russia, Eastern and Central European Countries Other Countries (Rest of the World) Sub-Saharan Africa United States of America Table 1: World Regions in WIAGEM We assume that coalitions cooperating on international trade also cooperate on climate control, imposing thus on the coalition a climate policy which is modelled as a 10 percent reduction of emissions. 10 In the analysis, we distinguish between climate control (CC) scenarios where emissions permit trade is allowed (ET for emission trading) or not. Before focusing on selected coalition scenarios, this subsection discusses some general insights that have been identified by this study. First, we found high terms of trade losses for those regions acting unilaterally. However, trade spillover effects lead to negative impacts on trade also in those regions not reducing emissions. For example, if the USA reduces its emissions, negative terms of trade spillover effects occur in nearly all other world regions (see Figure 1). This result can be explained by the fact that the USA is a large economic player influencing the international trade market considerably. If on the other hand Japan reduces emissions unilaterally, negative terms of trade effects to other world regions are much smaller. Second, full cooperation between all players leads to a worsening of terms of trade. The reason behind this finding is that all regions have to decrease emissions by 10 percent implying thereby production losses, especially in the USA, Europe and Japan. Russia basically suffers from terms of trade spillover losses. 10 Since we impose a 10% reduction target, the commitments that countries face are stronger than the ones embedded in the Kyoto Protocol. 6

TRADE AND CLIMATE BLOCS Terms of Trade Changes 0,0100 0,0050 0,0000 in % to BAU -0,0050-0,0100-0,0150-0,0200 JPN CHN USA SSA ROW CNA EU15 REC LSA ASIA MIDE Euuni Japuni USuni Figure 1: Terms of Trade Changes in to BAU We compare unilateral action of the diverse regions with different climate control and trade coalitions. Unilateral emissions reductions are only profitable if no emissions trading is authorised. By contrast, a cooperative behaviour makes all players better off if emissions permit trading is allowed. Small coalitions benefit cooperating nations with binding emissions reductions targets. Russia as a main seller of permits wants to cooperate with as many potential purchasing nations as possible. These insights are confirmed by terms of trade effects in comparison to a situation allowing no emissions trading. The USA and EU always prefer joining a coalition with Russia if permit trading is allowed. These results provide support to those presented in part I of this study where a different modelling framework was used. Taking the impacts of climate into account, we generally find stronger impacts on the terms of trade. The explanation for this finding is related to the welfare and production losses in the individual regions induced by climatic impacts. Countries must spend more money in climate protection as well as climate mitigation. This reaction leads to stronger terms of trade effects. In general, all developing regions prefer to cooperate on climate policy in climate blocs if emissions trading is allowed compared to when permit trading is not an option. However, a climate control policy implies that developing regions must reduce their emissions with respect to their baseline emissions, which results in welfare losses. These negative welfare implications induce terms of trade losses. 3.2 The Impact of Selected Trade Blocs that also Cooperate on Climate Change Control This subsection analyses a variety of different trade coalitions which are also committed to GHG emission control. We compare their implications, focusing in particular on the terms of trade effects. 7

KEMFERT & BUCHNER EUREC The first scenario is characterised by a coalition among Europe and Russia (EUREC). From a trade point of view, this case makes a lot of sense because of the strong trade interlinkages between these two countries. The cooperation between Europe and Russia now includes both international trade and climate control. In this case, the negative terms of trade s in Europe do not reach the extent they have in the case of EU unilateral action. However, with respect to the business as usual case, Europe would suffer strong negative terms of trade effects. In addition, the revenues of permit trading are much lower for Russia than in other coalitions where large emission permit demand increases permit prices. Thus, positive terms of trade s are relatively low. Furthermore, as we have also seen in the unilateral action scenario, the USA benefit if Europe or Japan reduce emissions. This advantage explains why the USA achieve positive economic effects in this scenario. The same is true for the other large economic player, Japan, although benefits for Japan are smaller. In addition, some developing countries like China achieve positive economic benefits, a result basically due to positive terms of trade spill over effects through production increases in the USA. The positive spillover economic effects are even higher if permit trading between the cooperating nations is implemented. Permit trading indeed implies that the negative economic impacts of Europe are lower while the positive economic effects of Russia are higher. As before, the introduction of permit trading thus leads to positive spillover effects. Terms of Trade Changes 0,0150 0,0100 0,0050 in % to BAU 0,0000-0,0050-0,0100-0,0150 JPN USA EU15 REC CHN LSA ASIA MIDE CNA SSA ROW EUREC, no ET EUREC, with ET Figure 2: Regional Terms of Trade Changes in % to BAU of EUREC Co-operation Scenario in 2010 However, the positive economic effects lead to a boost in the countries economic activities, resulting into increased emissions. In particular, this effect can be seen in the USA and in China. Nonetheless, although regional emissions in the USA and China increase in comparison to the BAU scenario, global GHG emissions can be reduced by around 1.6 % due to the large emissions reductions in Europe and in Russia. 8

TRADE AND CLIMATE BLOCS NO ET EUREC ET 2010 11.6926-1.57% 11.8873 +0.07% 2050 15.7034-1.43% 15.9444 +0.08% NO ET MERCOSUR ET 2010 11.8200-0.49% 11.8428-0.30% 2050 15.8661-0.41% 15.8972-0.22% NO ET NAFTA ET 2010 11.5499-2.77% 11.6464-1.96% 2050 15.5254-2.55% 15.6429-1.81% NO ET ASIA BLOC ET 2010 11.5974-2.37% 11.6273-2.10% 2050 15.5286-2.51% 15.5600-2.33% Table 1: Environmental Effectiveness as Emissions Changes of Different Scenarios If emissions permit trading is allowed, additional positive economic and spillover effects induce a further increase in regional and global emissions. Indeed, although emissions in Europe and Russia increase only marginally in comparison to the no emissions permit trading scenario, global emissions increase by around 0.07 % in the year 2010 (see Table 1). NAFTA If we assume that NAFTA countries (USA, Mexico and Canada) form a coalition on free trade, terms of trade improve, as described earlier. However, if free trade is associated with climate control, terms of trade effects worsen, as Figure 3 illustrates. This situation is induced by huge welfare and terms of trade losses which the USA has to accept if their domestic emissions are reduced by 10 percent. The same effects are also true for the other GHG reducing nations (Mexico and Canada) while the strong negative terms of trade effects for the USA induce additional negative spillover effects to nearly all world regions. 9

KEMFERT & BUCHNER Terms of Trade Changes 0,0050 0,0000-0,0050 in % to BAU -0,0100-0,0150-0,0200-0,0250-0,0300 JPN CHN USA SSA ROW CNA EU15 REC LSA ASIA MIDE MEX NAFTA, NO ET NAFTA, ET Figure 3: Terms of Trade Changes in the NAFTA Coalition in Comparison to the BAU Scenario As for emissions, if the NAFTA regions cooperate on climate control and emission trading is implemented, total emissions of these regions are reduced from a total 4.72 to 4.61 billion. From a global point of view, the NAFTA coalition could lead to a reduction of about 2% of global GHG emissions in 2010 (without emissions trading emissions are reduced by almost 3%). This increase in environmental effectiveness can be explained mainly by the terms of trade losses which lead to production losses and therefore emissions reductions in other world regions. Mercosur In the WIAGEM model, the Mercosur scenario corresponds to a unilateral action by Latin South America (LSA). In this scenario, Latin South America reduces regional emissions by 10 percent. The emission reduction requirements lead to terms of trade and welfare losses in this region. However, due to the fact that Latin South America does not represent a major global economic player, no relevant negative spillover effects to other world regions can be detected. In addition, the negative economic impacts suffered by the Mercosur regions are even smaller if emissions trading is allowed. In particular, when emission trading is introduced, we assume that LSA trades GHG emission permits with Mexico and Asia. This relation explains why both Mexico and Asia improve terms of trade when permit trading is allowed. 10

TRADE AND CLIMATE BLOCS Terms of Trade Changes 0,0040 0,0020 0,0000-0,0020 in % to BAU -0,0040-0,0060-0,0080-0,0100-0,0120-0,0140-0,0160 JPN CHN USA SSA ROW CNA EU15 REC LSA ASIA MIDE MEX MERCOSUR MERCOSUR with ET Figure 4: Regional Terms of Trade Changes in % to BAU in MERCOSUR Scenario, with and without Emissions Trading The scenario in which the Mercosur free trade area reduces GHG emissions by 10% leads to a small reduction of global GHG emissions in the year 2010. Emission s range from 0,3% to 0,5% (if emissions trading is not adopted). Because no large economic spillover effect was identified, emissions of other regions are almost invariant in comparison to the BAU scenario. In the case of emissions trading, the increase of emissions through the main trading partners Asia and Mexico induce a lower environmental effectiveness. Asian Bloc (CHNASIAJPN) The Asian bloc scenario refers to a climate control and trade coalition formed by Japan, China and Asia. The climate control coalition achieves, as in the other scenarios, an emissions reduction of 10 percent in comparison to the business as usual scenario. Again, the emission reduction target induces production and welfare losses for coalition members (Japan, Asia and China). The results demonstrate that emission trading is not of great help in lightening these negative terms of trade and welfare effects. Indeed, the strong economic growth in China and Asia implies that only Asia can sell a small amount of permits in the time period 2000-2010. Therefore, the price of permit is high and benefits from trading are low. 11

KEMFERT & BUCHNER Terms of Trade Changes 0,03 0,02 0,01 in % to BAU 0-0,01-0,02-0,03-0,04 JPN CHN USA SSA ROW CNA EU15 REC LSA ASIA MIDE without ET with ET Figure 5: Terms of Trade Changes in Comparison to the BAU of the Asian Bloc Scenario, with and without Trade Co-operation Interestingly, Latin South America benefits by free riding in this scenario because this region acts as a trade substitute. As a consequence of the trade declines in trade-intensive regions like China and Asia and Japan, parts of international trade are shifted from these regions to Latin South America. Apart from LSA, also the USA takes over some of the international trade flows. Instead, Europe as a net exporter suffers from trade losses with the Asian regions. This scenario leads to a greater environmental effectiveness than in the previous cases. The reason for this positive result lies in the high economic growth expected for both China and Asia. As a consequence, not only Japan but also China need to buy emissions permits, whereas the high growth assumptions in the Asian region imply a very low supply of emissions permits. Accordingly, emissions reductions are harder to be achieved, inducing more drastic domestic emissions reduction. In addition, also Europe as another strong emitter decreases its emissions because of trade losses. Only the USA experience an increase in emissions which is however not strong enough to offset the other countries emissions decline. In total, this scenario leads to the strongest emissions reductions (more than 2% if emissions trading is allowed) and the greatest environmental improvement in comparison to the previous scenarios (see Table 1). 12

TRADE AND CLIMATE BLOCS 3.3 An Analysis of the Possible Evolution of Climate and Trade Coalitions EUREC + (EUJPNREC-EUUSREC-EUJPNRECUSA) To see whether the coalition EUREC (analysed in the previous section) could further evolve, we compare its outcome to a number of other cooperation and coalition scenarios. In particular, Figures 6 and 7 illustrate the terms of trade s of different scenarios for the original Annex B countries, both when emissions trading is absent and when it is allowed. Without the possibility of permit trading, full cooperation between all players leads to a worsening in terms of trade. This result is induced by the emission reduction requirements which lead to production losses, especially in the USA, Europe and Japan. The introduction of emissions trading alleviates the situation, benefiting above all Russia and countries which are engaged in permit trading with this country. Indeed, Russia s terms of trade strongly improve and simultaneously also all countries with binding emissions targets profit from emissions trading with Russia. Terms of Trade Changes, no ET 0,0100 0,0050 0,0000 in % to BAU -0,0050-0,0100-0,0150-0,0200 Full cooperation Euuni Japuni USuni RECuni EUjap EUUS JapUS EUJPNREC JPNREC EUREC USAJPNREC USAREC EUUSREC JPNUSAEU JPN USA EU15 REC Figure 6: Terms of Trade Changes in % to BAU No Emission Trading Allowed 13

KEMFERT & BUCHNER Terms of Trade Changes 0,0150 0,0100 0,0050 in % to BAU 0,0000-0,0050-0,0100-0,0150 Full cooperation Euuni Japuni USuni RECuni EUjap EUUS JapUS EUJPNREC JPNREC EUREC USAJPNREC USAREC EUUSREC JPNUSAEU JPN USA EU15 REC Figure 7: Terms of Trade Changes in % to BAU- Emissions Trading is Allowed As Figure 7 illustrates, in the case of emissions trading Russia could increase exports and improve its trade balance if it joins a small or grand coalition with regions that demand emissions permits such as the USA, Japan, or Europe. Therefore, following the actual policy development in climate control, a probable extension of the initial coalition EUREC could be the cooperation between Europe, Russia and Japan, EUJPNREC. However, when analysing the trade effectiveness of this new coalition, we see that both Japan and Europe suffer from worse terms of trade. The high marginal abatement costs faced by Europe and Japan are responsible for this negative effect. Although emissions trading with Russia reduces the compliance costs of these two countries, the required emissions reductions still lead to production, welfare and terms of trade losses. Both Europe and Japan have therefore to accept economic losses because of domestic emissions abatement. At the same time, being outside of this coalition provides high economic benefits to the USA. As a consequence, emissions increase and a lower degree of environmental effectiveness is achieved. One possible extension of the previous coalition could be that the USA instead of Japan joins the coalition (EUUSREC). Because of high domestic emissions abatement costs, Europe and the USA suffer from welfare, production and terms of trade losses in this scenario. These effects are lower if emissions trading with Russia is allowed. However, Europe and the USA compete as permit demanders with each other, inducing in this way a high permit price and increased compliance costs. These higher costs combined with the binding emissions reductions lead to an improved environmental effectiveness, strengthened also by the fact that both Europe and the USA are important emitters. In addition, emissions decrease also in the countries outside the coalition, since this cooperation causes negative terms of trade spillover effects on these regions. 14

TRADE AND CLIMATE BLOCS NO ET EUUSREC ET 2010 11.4953-3.22% 11.6857-1.62% 2050 15.4553-2.91% 15.6899-1.51% NO ET EUJPNREC ET 2010 11.6424-1.99 % 11.8507 +0.02% 2050 15.6369-1.85% 15.8920 +0.02% EUUSJPNREC (Full Cooperation) NO ET ET 2010 11.1180-5.81% 11.5972-2.36% 2050 15.0352-5.62% 15.5791-2.21% EUUSJPNREC +NAFTA NO ET ET 2010 11.076-6.75% 11.696-1.53% 2050 14.876-6.62% 15.701-1.44% EUUSJPNREC +NAFTA +ASIA BLOC NO ET ET 2010 10.699-9.92% 11.286-4.98% 2050 14.355-9.89% 15.127-5.05% EUUSJPNREC +NAFTA+ ASIA BLOC +MERCOSUR NO ET ET 2010 10.658-10.27% 11.251-5.26% 2050 14.311-10.16% 15.092-5.27% Table 2: Environmental Effectiveness of Different Scenarios EUREC could also be broadened to achieve the coalition among the original Kyoto countries (EUUSJPNREC). As it has already been indicated above, full cooperation between EU, JPN, USA and REC would lead to a strong gain in terms of trade for Russia when emissions trading is allowed. However, all the remaining countries would worsen their situation with respect to terms of trade. The US would particularly deteriorate its terms of trade s, followed by Japan. Nevertheless, the US prefers to join a coalition with Russia as a partner. This strategy is clearly motivated by the possibility of achieving lower compliance costs due to Russia s sufficient supply of emissions permits. Obviously, a coalition among the original Kyoto countries would also enhance climate control (see Table 3). In particular, global GHG emissions could be reduced by about six percent if there is no cooperation on emissions trading and about three percent if trading is allowed. The difference among these two results stems from the Russian hot air effect: Russia sells emissions permits to all other 15

KEMFERT & BUCHNER cooperating nations. As in previous cases, without permit trading regions suffer welfare and terms of trade losses leading to decreased emissions in all other nations, including Russia. Full Co-operation (EUJPNRECUSA+ NAFTA+ MERCOSUR+ Asian Bloc) In this subsection, we compare the trade and environmental effectiveness of different enlargements of trade regions. Let us begin with the enlargement of the coalition between Europe, Japan, Russia and the USA (EUJPNRECUSA, called Cooperation 1) and add the NAFTA coalition, i.e. Canada and Mexico (EUJPNRECUSA + NAFTA, called Cooperation 2 in Figure 8). As we have seen above, if trade is restricted, welfare losses occur not only to cooperating nations but also to other nations due to negative terms of trade spillover effects. Our results demonstrate that negative economic effects vary in proportion with the size of the coalition. In particular, the larger the coalition on climate control, the greater the negative economic effects. The only exemption is Russia who can benefit from permit trading if more nations cooperate on climate control. The reason for this result is the increase in the demand for permits which is induced by the entry of both Canada and Mexico, as Canada is net buyer and Mexico cannot offer a large amount of emissions permits. Terms of Trade Changes 0,020 0,010 0,000 in % to BAU -0,010-0,020-0,030-0,040 JPN CHN USA SSA ROW CNA EU15 REC LSA ASIA MIDE MEX COOP1, NO ET COOP1, with ET COOP2, no ET COOP2, withet Figure 8: Terms of Trade Changes in % to BAU of Different Cooperation Scenarios If Latin South America joins this coalition (EUJPNRECUSA + NAFTA+ MERCOSUR, called Cooperation 3), economic losses increase not only in Latin South America, but also in other nations like China and Asia because of negative economic spillover effects. Without emissions trading, negative economic effects, expressed in terms of trade losses, are larger than if permit trading is allowed. However, even the introduction of emissions trading does not notably improve the situation. As seen above, Latin South America and Mexico cannot offer a large amount of permits; therefore compliance costs of demanding nations cannot be reduced significantly via emission trading. Finally, if China and Asia join the coalition (EUJPNRECUSA + NAFTA+ MERCOSUR+ Asia Bloc, called Cooperation 4), negative economic effects are very large. A large participation and the 16

TRADE AND CLIMATE BLOCS high growth expected in the developing countries increase the competition for emission permits and render the achievement of emission reduction requirements difficult and expensive. Emissions targets become thus binding in all cooperating nations, inducing in this way production, welfare and terms of trade losses. Because of the strong development in China, welfare and terms of trade losses occur even in the presence of emission trading. As expected, this scenario is beneficial for environmental effectiveness: the more nations cooperate on climate control, the stronger are the positive environmental effects (for more details see Table 2). If emissions trading is introduced, emissions increase compared to the no-trade case, but the overall environmental effectiveness is still strong. Nonetheless, the positive environmental effects are associated with economic costs that are high even if emissions trading is allowed. Terms of Trade Changes 0,03 0,02 0,01 in % to BAU 0-0,01-0,02-0,03-0,04 JPN CHN USA SSA ROW CNA EU15 REC LSA ASIA MIDE MEX COOP3, no ET COOP3, ET COOP4, no ET COOP4, ET Figure 9: Terms of Trade Changes in % to BAU of Different Cooperating Scenarios 4. Conclusions Using a multi-regional, multi-sectoral trade model embedded in a game theoretic framework, this chapter analysed the implications of inducing trade coalitions to cooperate also on climate control, taking into account both developed and developing countries. As climate control, and especially unilateral emissions reduction measures, induce negative production s in all regions committed to binding emission reduction targets, negative terms of trade effects occur. This development can be explained by the fact that production and welfare losses induce export reductions which in turn deteriorate the trade balance. Because of strong trade relations between regions, the allowance of emissions permit trading improves the terms of trade considerably. As a consequence, regions with relevant emission reduction targets want to cooperate with regions that can offer emissions permits. 17

KEMFERT & BUCHNER Developing countries are shown to play an important role in the coalitions aimed at trade and mitigation cooperation. In the case of binding domestic emission reduction targets, developing countries need emission permits to comply with their commitments. As a consequence, compliance costs increase considerably for coalition members but at the same time the intense efforts on climate control induce significant environmental benefits. However, if we assume that developing nations do not have binding emissions reduction targets but follow the business as usual emissions path, free trade leads to large economic benefits and consequently a lower environmental effectiveness. In general, the larger the cooperation among countries and the higher the marginal abatement costs of the participants, the stronger are the negative economic effects of climate control, expressed in terms of trade losses. As expected, also the degree of environmental effectiveness is related to the size of the coalition: the larger the cooperation, the higher the positive environmental effects. However, our results confirm that a strong environmental effectiveness comes at the price of high economic costs. 18

TRADE AND CLIMATE BLOCS References Baldwin, R. 1993. A Domino Theory of Regionalism. NBER Working Paper No. 4465. Barrett, S (1994) Self-Enforcing International Environmental Agreements. Oxford Economic Papers 46: 878-894. Barrett, S (1998) The Credibility of Trade Sanctions in International Environmental Agreements. in Friedriksson, P (ed) Trade, Global Policy and the Environment, World Bank Discussion Paper, 1999. Bloch, F. 2001. Non-cooperative Models of Coalition Formation in Games with Spillovers. Forthcoming in C. Carraro, ed., 2003, The Endogenous Formation of Economic Coalitions, Edward Elgar. Bloch, F. and H. Ferrer. 1999. Trade Fragmentation and Coordination in Bilateral Oligopolies. CORE Discussion Paper No. 9908. Bond E. and C. Syropoulos. 1996. The Size of Trading Blocs. Market power and world welfare effects. Journal of International Economics, Vol. 40 (1996), 417-37. Bond E., R. Riezman and C. Syropoulos. 1999. The Theory of Free Trade Areas. Florida International University. Buchner, B., Carraro, C., Cersosimo, I. (2001) On the Consequences of the U.S. Withdrawal from the Kyoto/Bonn Protocol, FEEM, Nota di Lavoro 102. 2001. Carraro, C. (1997) The Structure of International Agreements on Climate Change. in: International Environmental Agreements on Climate Change, Kluwer Academic Publications, Dordrecht. Carraro, C. (1999) The Structure of International Agreements on Climate Change in Carraro, C. (ed) International Environmental Agreements on Climate Change. Dordrecht, Kluwer Academic Publisher. Carraro, C. and Hourcade, C. J. (1998) Climate Modeling and Policy Strategies. The Role of Technical Change and Uncertainty. Energy Economics 20: 463-471. Carraro, C. and M. Galeotti. 2002. The Future Evolution of the Kyoto Protocol. Costs, Benefits, Incentives to Ratification and New International Regimes, in C. Carraro and C. Egenhofer, eds., Firms, Governments and Climate Policy: Incentive-Based Policies for Long-Term Climate Change, Edward Elgar, Cheltenham, 2002. Carraro, C. and Siniscalco, D (1993) Strategies for the International Protection of the Environment. Journal of Public Economics 52: 309-328. Carraro, C. and Siniscalco, D. (1997) R&D Cooperation and the Stability of International Environmental Agreements. International Environmental Agreements: Strategic Policy Issues. C. Carraro. Cheltenham, Edward Elgar. Carraro, C. and Siniscalco, D (1998) International Environmental Agreements: Incentives and Political Economy. Amsterdam. Carraro, C. and Siniscalco, D. (2002) Global Governance and Policy Coordination, forthcoming in C. Carraro and D. Siniscalco, eds., Advances in Environmental Economics. Theory and Applications, Cambridge University Press, Cambridge. Casella, A. 1995. Large Countries, Small Countries and the Enlargement of Trade Blocs. NBER Working Paper No. 5365. Cesar, H. (1994) Control and Game Models of the Greenhouse Effect. Berlin, Springer-Verlag. Kemfert, C., Lise, W., Tol, R. S. J. (2002) Games of Climate Change with International Trade. in: Environmental and Resource Economics, forthcoming. 19

KEMFERT & BUCHNER Kennan, J. and R. Riezman, 1990. Optimal Tariff Equilibria with Customs Unions, Canadian Journal of Economics, 70-83 Vol. 23 (1) pp. 70-83. Canadian Economics Association Krugman, O. 1991. Is Bilaterism Bad? In E. Helpman and A Razin (eds.), International Trade and Trade Policy. MIT Press: Cambridge. Lise, W., Tol, R. S. J. and van der Zwaan, B. C. C. (2001) Negotiating Climate Change as a Social Situation. Nota di Lavoro 44(01), Fondazione Eni Enrico Mattei, Milan. Macho-Stadler, I., D. Perez-Castrillo and C. Ponsati. 1998. Stable Multilateral trade Agreements. Economica (1998) 65, 161-77. McDougall, R. A. (1995). "The GTAP 3 Data Base, Center for Global Trade Analysis." Purdue University. Tjornhom, J. 2000. Dynamic Trade Bloc Formation: Building Blocs or Stumbling Blocs? University of Minnesota, Department of Economics. Viner, J. 1950. The Customs Union Issue. Carnegie Endowment for International Peace: New York. Yi, S.-S. 1996a. Endogenous Formation of Customs Unions under Imperfect Competition: Open Regionalism is Good. Journal of International Economics, Vol. 41 (1996), 153-177. Yi, S.-S. 1996b. Open Regionalism and World Welfare. Eastern Economic Journal, Vol. 22, No. 4, Fall 1996, 467-475. Yi, S.-S. 1997. Stable Coalition Structures with Externalities. Games and Economic Behaviour, 20, 201-237. Yi, S.-S. 1998. Free trade Areas and Welfare: An Equilibrium Analysis. Dartmouth College, Department of Economics. Yi, S.-S. 2001. Endogenous Formation of Economic Coalitions: A Survey on the Partition Function Approach. Forthcoming in C. Carraro, ed., 2003, The Endogenous Formation of Economic Coalitions, Edward Elgar. 20