IFPRI Discussion Paper 01102 July 2011 The Price and Trade Effects of Strict Information Requirements for Genetically Modified Commodities under the Cartagena Protocol on Biosafety Antoine Bouët Guillaume Gruère Laetitia Leroy Environment and Production Technology Division Markets, Trade and Institutions Division INTERNATIONAL FOOD POLICY RESEARCH INSTITUTE The International Food Policy Research Institute (IFPRI) was established in 1975. IFPRI is one of 15 agricultural research centers that receive principal funding from governments, private foundations, and international and regional organizations, most of which are members of the Consultative Group on International Agricultural Research (CGIAR). PARTNERS AND CONTRIBUTORS IFPRI gratefully acknowledges the generous unrestricted funding from Australia, Canada, China, Denmark, Finland, France, Germany, India, Ireland, Italy, Japan, the Netherlands, Norway, the Philippines, South Africa, Sweden, Switzerland, the United Kingdom, the United States, and the World Bank. AUTHORS Antoine Bouët, International Food Policy Reasearch Institute Senior Research Fellow, Markets, Trade and Institutions Division Guillaume Gruère, International Food Policy Reasearch Institute Research Fellow, Environment and Production Technology Division Laetitia Leroy, International Food Policy Reasearch Institute Senior Research Assistant, Environment and Production Technology Division Notices IFPRI Discussion Papers contain preliminary material and research results. They have been peer reviewed, but have not been subject to a formal external review via IFPRI’s Publications Review Committee. They are circulated in order to stimulate discussion and critical comment; any opinions expressed are those of the author(s) and do not necessarily reflect the policies or opinions of IFPRI. Copyright 2011 International Food Policy Research Institute. All rights reserved. Sections of this material may be reproduced for personal and not-for-profit use without the express written permission of but with acknowledgment to IFPRI. To reproduce the material contained herein for profit or commercial use requires express written permission. To obtain permission, contact the Communications Division at ifpri-copyright@cgiar.org. iii Contents Acknowledgments vi Abstract vii 1. Introduction 1 2. Conceptual Framework 3 3. Model and Scenarios 6 4. Simulation Results 9 5. Conclusions 22 Appendix: Supplementary Tables 23 References 33 iv List of Tables 2.1—Characterization of the four groups by GM production status and CPB membership 3 3.1—List of exporting and importing countries included in the model by group 7 3.2—Data sources for key parameters 7 3.3—Definitions of scenarios 8 4.1—Relative changes (%) in world market variables compared to the Base in Set A 9 4.2—Relative changes (%) in key variables compared to the base in each group of countries 9 4.3—Changes in maize and soybeans export volume (metric tons) relative to the base under the scenarios in Set A 13 4.4—Relative changes (%) in world market variables under scenarios in Set B compared with the base 14 4.5—Relative changes (%) in key variables in each group of countries under scenarios in Set B compared with the base 14 4.6—Changes in maize and soybean export volume (metric tons) relative to the base under the scenarios in Set B 16 4.7—Change in welfare effects in Sub-Saharan African countries under Scenario B3 compared with the base scenario, in U.S. dollars per year 21 A.1—Change in export volume (metric tons) relative to the base under the six scenarios for maize 23 A.2—Change in export volume (metric tons) relative to the base under the six scenarios for soybeans 23 A.3—Change in trade volume (metric tons) relative to the base under the six scenarios for maize 24 A.4—Change in trade volume (metric tons) relative to the base under the six scenarios for soybeans 25 A.5—Change in export volume (metric tons) relative to the base under Scenario A1: Top 10 and bottom 10 trade flows 26 A.6—Change in export volume (metric tons) relative to the base under Scenario A2: Top 10 and bottom 10 trade flows 26 A.7—Change in export volume (metric tons) relative to the base under Scenario A3: Top 10 and bottom 10 trade flows 27 A.8—Change in export volume (metric tons) relative to the base under Scenario B1: Top 10 and bottom 10 trade flows 27 A.9—Change in export volume (metric tons) relative to the base under Scenario B2: Top 10 and bottom 10 trade flows 28 A.10—Change in export volume (metric tons) relative to the base under Scenario B3: Top 10 and bottom 10 trade flows 28 A.11—Welfare effects for maize (Scenario B3 compared with base) by country in decreasing order of total surplus, in U.S. dollars per year 29 A.12—Welfare effects for soybeans (Scenario B3 compared with base) by country in decreasing order of total surplus, in U.S. dollars per year 31 v List of Figures 4.1—Changes in maize trade volume (metric tons) relative to the base under the three scenarios in Set A 11 4.2—Changes in soybean trade volume (metric tons) relative to the base under the three scenarios in Set A 12 4.3—Changes in maize trade volume (metric tons) relative to the base under the three scenarios in Set B 15 4.4—Changes in soybean trade volume (metric tons) relative to the base under the three scenarios in Set B 16 4.5—Change in consumer surplus (U.S. dollars per year) for each group under each scenario 18 4.6—Change in producer surplus (U.S. dollars per year) for each group under each scenario 19 4.7—Change in total surplus (U.S. dollars per year) for each group under each scenario 19 vi ACKNOWLEDGMENTS This research was supported by the Program for Biosafety Systems, a project managed by the International Food Policy Research Institute and funded by the United States Agency for International Development. Views expressed in this paper are the authors’ alone. vii ABSTRACT This paper assesses the global economic implications of the proposed strict documentation requirements on traded shipments of potentially genetically modified (GM) commodities under the Cartagena Protocol on Biosafety. More specifically, we evaluate the trade diversion, price, and welfare effects of requiring all shipments to bear a list of specific GM events (the does contain rule) in the maize and soybean sectors. Using a spatial equilibrium model with 80 maize- and 53 soybean-trading countries, we show that information requirements would have a significant effect on the world market for maize and soybeans. But they would have even greater effects on trade, creating significant trade distortion that diverts exports from their original destination. The measure would also lead to significant negative welfare effects for all members of the Protocol and nonmembers that produce GM maize, soybeans, or both. While non-GM producers in Protocol member countries would benefit from this regulation, consumers and producers in many developing countries would have to pay a proportionally much heftier price for such a measure. Keywords: genetically modified food, international trade, Cartagena protocol on biosafety 1 1. INTRODUCTION The Cartagena Protocol on Biosafety (CPB), a supplementary agreement to the United Nations Convention on Biological Diversity introduced in 2000 (Secretariat of the COB 2000), entered into force in September 2003 with the goal of setting up a harmonized framework of risk assessment, risk management, and information sharing on the transboundary movements of living modified organisms (LMOs).1 Within the Protocol, a number of rules focus on LMOs intended for direct uses as food or feed, or in food or feed processing (noted as LMO-FFPs), which are essentially unprocessed genetically modified (GM) agricultural commodities.2 Article 18.2(a) of the Protocol requires that each traded shipment of LMO-FFPs be labeled as may contain LMO-FFPs not intended for release in the environment, though the Convention also noted that a more specific rule on information requirements should be determined at a later date (Secretariat of the COB 2000). At a March 2006 meeting in Brazil, after a contentious debate, Protocol members agreed to adopt a two-option rule consisting of a more stringent option and the less stringent one that had previously been in effect (ICTSD 2006). Under the stringent option, shipments containing LMO-FFPs identified through means such as identity-preservation (IP) systems would be labeled as does contain LMO-FFPs and would include a list of all GM events present in each shipment. Shipments containing LMO-FFPs that are not well identified would follow previous practice and would be labeled as may contain LMO-FFPs. At the same time, a complete list of GM events commercialized in the exporting country would be available to importers via the Biosafety Clearing House (BCH), an internet database. At the same meeting, Protocol members also agreed that the two-option rule would be reconsidered, with the possibility of making the stringent does contain option mandatory for all countries (ICTSD 2006). At the latest meeting of parties in October 2010 in Nagoya, Japan, discussion on the issue was postponed to leave time for data collection, but it is planned to get the topic back to the table in 2014. While the benefits of this proposed change are highly debatable, its implementation would generate significant new costs (see, for example, Gruère and Rosegrant 2008; Kalaitzandonakes 2004; and Redick 2007). More specifically, under the does contain rule, countries that produce and export only non-GM products would be exempt from verifications and tests, while countries that export GM products would have to test each shipment to verify the accuracy of GM event identification. Importers that are ratifying parties of the CPB would also need to pay for the IP system or to conduct tests to confirm the validity of shipment statements in order to ensure enforcement of mandatory information requirements. Previous studies have analyzed the economic implications of adopting the does contain rule in different countries, such as Argentina (Direccion Nacional de Mercados Agroalimentarios 2004), the United States (Kalaitzandonakes 2004), and Australia (Foster and Galeano 2006), reporting that the costs of such a change would be potentially significant. More recently, Huang et al. (2008) showed that the cost of implementation would be high globally but not really significant for China (their focus country). Gruère and Rosegrant (2008) assessed the potential implementation costs of article 18.2(a) on all member countries of Asia–Pacific Economic Cooperation (APEC)3 and provided a range of cost estimates for exporters and importers, noting the disproportionate cost for developing countries that have been supportive of this measure. They also showed that it would effectively constitute a new entry cost for GM adoption and for Protocol membership in the APEC region. Yet most of these studies provide short-run, partial-cost estimates of the strict rule in particular regions, leaving aside potential price and trade diversion effects. Huang et al. (2008) do did use a multiregion computable general equilibrium (CGE) model to assess the potential trade effects of this new measure, showing that it would affect the prices of maize and soybeans. But their approach focused only 1 Also called genetically modified organisms. 2 These products represent more than half the total import value of the four main GM commodities. Approximately 51 percent of the import value of soybeans and 88 percent of that of maize comes from unprocessed commodities (Gruère 2006). 3 APEC is a regional trade body covering 21 countries located around the Pacific Ocean, from Chile to New Zealand, including large traders like Mexico, the United States, Canada, China, Japan, South Korea, Indonesia, and Australia. 2 on a few regions (China, the Americas, and the world), used the 2001 Global Trade Analysis Project (GTAP) database, and did not provide a detailed assessment of potential trade diversions. Furthermore, the CGE modeling approach prevents the appearance of new trade flows, leaving aside significant possibilities in trade diversion. While their results showed that the cost of implementation would be high for all but not really significant for China, they noted that other developing countries would likely pay a higher price. The objective of this paper is to complement previous studies by providing a comprehensive global trade assessment of strict documentation requirements in all member countries of the Protocol. In particular, our analysis intends to evaluate the market effect it would have on developing countries that are members of the Protocol. To do so, we develop a spatial trade model and simulate scenarios to evaluate the trade diversion, price, and welfare effects of implementing the does contain rule on the maize and soybean sectors in all significant trading countries, using data from multiple sources in the reference period 1995–2005. The model incorporates transportation costs, uses a lower level of product analysis (four-digit codes of the Harmonized Commodity Description and Coding System [HS]), includes more countries than GTAP-based models include, and accounts for trade diversion and the creation of new trade flows. The results of our policy simulation intend to provide an overview of the medium- to long-run effects of mandating the does contain rule to all members of the Protocol, ahead of future negotiations on the issue. Developing countries that are members of the CPB have been vocal supporters of using precautionary measures for trade of GM commodities, such as Article 18.2(a), but they may have underestimated the cost of such measures on their economies. Beyond the cost estimates and their geographic and product differentiation, our findings aim at giving an outlook of a possible future trade scenario for GM commodities in the presence of increasingly stricter trade regulations in specific trade blocs. In the following section we provide a conceptual framework for analysis. We then present the simulation model, data, and policy scenarios. The fourth section presents and discusses the first results of our simulations, and we close the paper with some policy conclusions. 3 2. CONCEPTUAL FRAMEWORK While the may contain and does contain rules may share usefulness for regulatory purposes, their costs of implementation widely differ. Under the does contain rule, countries that export GM products would have to test each shipment to verify the accuracy of the list of GM events, whereas the may contain rule would not require additional tests beyond those to reject unapproved events in the importing countries. Even if all GM events were approved in all importing nations, the exporter would be required to provide precise information on each shipment. This could also include additional insurance costs for shippers against the rejection of shipments. On the importing side, CPB member countries would need to pay for the IP system or pay to conduct tests to confirm the validity of shipment statements in order to ensure enforcement of the requirements. Naturally, importers would also have to pay the price for the information, given the additional testing and insurance applied to shipments. Given these considerations, we propose an analytical framework based on the characterization of Gruère and Rosegrant (2008) that categorizes countries to assess the cost of information requirements. More specifically, we divide countries into four groups according to their membership in the CPB and whether they produce GM maize or GM soybeans, as presented in Table 2.1. Table 2.1—Characterization of the four groups by GM production status and CPB membership Group Produces GM?* Member of the CPB Example of country 1 Yes No Argentina, U.S.A. 2 No Yes Japan, Mexico 3 Yes Yes Brazil, South Africa 4 No No Russia Source: Authors, based on Gruère and Rosegrant (2008). Note: * In our case the question will be “Produces GM maize?” or “Produces GM soybeans?” depending on the commodity. This categorization4 is used to impose the effect of strict information on specific trade flows, that is, those flows that link groups of GM maize– or GM soybeans–producing countries to groups of CPB members. Two types of trade relationships are bound to be affected, those that will request testing at the import and export sides, linking GM producers (Groups 1 and 3) to CPB members (Groups 2 and 3), and those that would affect only exporters, linking CPB member GM-producing countries (Group 3) to non- CPB member countries (Groups 1 and 4). We use this framework to set up a simplified partial equilibrium model of trade with four countries (A, B, C, and D) representing the four groups (1 to 4), to illustrate the potential price effect of such regulation. A and C produce GM, B and D do not, and B and C are members of the CPB while A and D are not. Each country I faces a linear supply SI defined by the inverse relationship 𝑝𝐼 = 𝑐𝑘𝐼𝑄𝐼, whose slope coefficient depends on whether the country adopts GM (k = g) or not (k = n). We assume that the slope coefficients are ranked as follows: 0 < 𝑐𝑔𝐴 < 𝑐𝑔𝐶 < 𝑐𝑛𝐵 < 𝑐𝑛𝐷, and that A and C are net exporters, while the two others are net importers. The demand in each country is linear and defined by the inverse demand equation 𝑝𝐼 = 𝑎𝐼𝑄𝐼 + 𝑏𝐼 where (𝑎𝐼 < 0). The equilibrium price is reached when all excess supply equals excess demand. The original world price (𝑝0𝑊) is 𝑝0𝑊 = − 𝑏 𝐴 𝑎𝐴 − 𝑏 𝐵 𝑎𝐵 − 𝑏 𝐶 𝑎𝐶 − 𝑏𝐷 𝑎𝐷 1 𝑐𝑔𝐴 + 1 𝑐𝑔 𝐶+ 1 𝑐𝑛 𝐵+ 1 𝑐𝑛 𝐷− 1 𝑎𝐴 − 1 𝑎𝐵 − 1 𝑎𝐶 − 1 𝑎𝐷 . (1) 4 Naturally the groups will differ by commodity, depending on whether a country produces GM corn, GM soybeans, or both. 4 The proposed regulation is modeled as an additional transport cost for GM and non-GM products from A to B, from A to C,5 and from C to B, for simplification.6 Let us assume a per-unit cost of τ, applied as a relative tariff on the affected trade flows. At the equilibrium, there are two prices for commingled commodities: one with affected flows and the other with non affected. The affected equilibrium is going to be defined by the relationship between A, B, and C, while the non affected one will be defined by the relationship of A and D. Naturally A and C will only export to B and D under price arbitrage conditions. The main equations are the following. The world price is 𝑝𝑊 = 𝑐𝑔𝐴𝑄𝐴 = 𝑐𝑔𝐴�𝑄𝐶𝑃𝐵𝐴 + 𝑄𝑂𝑈𝑇𝐴 � (2) The trade volume of non affected (subscript OUT) and affected (subscript CPB) commodities are 𝑄𝑂𝑈𝑇𝐴 = 𝑝𝑂𝑈𝑇𝑊 − 𝑏𝐴 𝑎𝐴 + 𝑝𝑂𝑈𝑇𝑊 − 𝑏𝐷 𝑎𝐷 − 𝑝𝑂𝑈𝑇𝑊 𝑐𝑛𝐷 (3) 𝑄𝐶𝑃𝐵𝐴 = 𝑝𝐶𝑃𝐵𝑊 − 𝑏𝐶 𝑎𝐶 + 𝑝𝐶𝑃𝐵𝑊 − 𝑏𝐶 𝑎𝐶 − 𝑝𝐶𝑃𝐵𝑊 𝑐𝑛𝐵 − 𝑝𝐶𝑃𝐵𝑊 − 𝜏 𝑐𝑔𝐶 (4) The arbitrage condition for A to export in both market is: �𝜋𝐶𝑃𝐵𝐴 = 𝜋𝑂𝑈𝑇𝐴 � ⇒ � �𝑝𝑂𝑈𝑇𝑊 �2 2𝑐𝑔𝐴 = �𝑝𝐶𝑃𝐵𝑊 − 𝜏�2 2𝑐𝑔𝐴 � (5) Using these equations, we find that at the equilibrium, the price of the non affected area (subscript OUT), affected area (subscript CPB), and overall world prices are, respectively, 𝑝𝑂𝑈𝑇𝑊 = 𝑝𝐶𝑃𝐵𝑊 − 𝜏 (6) 𝑝𝐶𝑃𝐵𝑊 = − 𝑏 𝐵 𝑎𝐵− 𝑏 𝐶 𝑎𝐶 + 𝜏 � 1 𝑐𝑔𝐴 + 1 𝑐𝑔𝐶 � 1 𝑐𝑔𝐴 + 1 𝑐𝑔𝐶 + 1 𝑐𝑛𝐵 − 1 𝑎𝐵 − 1 𝑎𝐶 (7) 𝑝𝑤 = 𝑎𝐴�𝑝𝐶𝑃𝐵𝑊 � + 𝑝𝑂𝑈𝑇𝑊 � 1 𝑎𝐷 − 1 𝑐𝐷 � + 𝑏𝐵 𝑎𝐵 + 𝑏𝐴 𝑎𝐴 (8) In this simplified case, the cost of the regulation acts as a wedge between the two prices—the higher the cost, the larger the difference between the two. The international price may or may not differ, but the local consumer price will increase in B and C, and may decrease in A and D. Therefore consumers in A and D may experience welfare gain, but because of the tariff-like effect, producers in A, C, and potentially D will lose, while producers in B will gain. In the long term, countries may decide to produce 5 The basic transport costs are not included explicitly here, because we focus on the new costs associated with the regulation, but they are treated with care in the empirical application. 6 We exclude trade flows from C to C and from C to A and D. 5 or abandon GM, while others may decide to join or abandon the CPB. If the effect on price is significant, A producers may try to avoid planting new GM crops, to lower additional losses. Naturally the use of this aggregate trade model can only provide a crude, medium-term, and inaccurate appreciation of what information requirements will do. Not all GM producers are large exporters; not all importers are the same; and transport costs, tariffs, and the structure of supply and demand vary widely from one country to another, even within the same group. We will now turn to our simulation model to explore the observable effects of the strict option under specific scenarios in the case of GM maize and GM soybeans. 6 3. MODEL AND SCENARIOS We built a spatial trade equilibrium model (Samuelson 1952, Takayama and Judge 1971) of the international market for maize and soybeans, which includes N countries (80 and 53 for maize and soybeans, respectively), that produce, export, or import these two commodities. All countries are maximizing their welfare function subject to a set of spatial trade arbitration equations. The structure of the model is based on the application by Devadoss et al. (2005) in the case of trade of timber.7 The objective function is a quasi-welfare function (QW) that Devadoss et al (2005) called a net social monetary gain function, defined as 𝑄𝑊 = �(𝛼𝑖 − 𝛽𝑖𝑦𝑖)𝑦𝑖 𝑁 𝑖=1 −�(𝛾𝑖 − 𝛿𝑖𝑥𝑖)𝑥𝑖 𝑁 𝑖=1 −��𝑥𝑖𝑗𝑡𝑖𝑗 𝑁 𝑗=1 𝑁 𝑖=1 −���𝜌𝑗𝐷 − 𝜌𝑖𝑆�𝑥𝑖𝑗 𝑁 𝑗=1 𝑁 𝑖=1 + ��� 𝜌𝑗𝐷 1 + 𝜀𝑖𝑗 − 𝜌𝑖𝑆� 𝑥𝑖𝑗 𝑁 𝑗=1 𝑁 𝑖=1 (9) where αi, βi, γi, and δi are the positive demand and supply coefficients, respectively; yi is the quantity demanded and xi the quantity produced in country i; tij is the transportation cost from i to j and xij the volume exported from i to j; 𝜌𝑗𝐷 and 𝜌𝑖𝑆 are the market supply and demand prices for maize (which accounts for constraints in and access to the international market); and 𝜀𝑖𝑗 is the ad valorem tariff equivalent for an import of maize from i to j. The market prices should not be confused with the country prices, �𝑝𝑖𝐷,𝑝𝑖𝑆�, which are defined by the inverse demand and supply equations 𝑝𝑖𝐷 = 𝛼𝑖 − 𝛽𝑖𝑦𝑖 and 𝑝𝑖𝑆 = 𝛾𝑖 + 𝛿𝑖𝑥𝑖. This objective function is maximized subject to the following set of feasibility constraints, capacity constraints, and arbitrage conditions: ∀𝑖 ∈ ⟦1,𝑁⟧ ∑ 𝑥𝑖𝑗𝑁 𝑗=1 ≤ 𝑥𝑖 (10) ∀𝑗 ∈ ⟦1,𝑁⟧ ∑ 𝑥𝑖𝑗𝑁 𝑖=1 ≥ 𝑦𝑗 (11) ∀𝑖 ∈ ⟦1,𝑁⟧ 𝛼𝑖 − 𝛽𝑖𝑥𝑖 ≤ 𝜌𝑖𝐷 (12) ∀𝑖 ∈ ⟦1,𝑁⟧ 𝛾𝑖 + 𝛿𝑖𝑥𝑖 ≥ 𝜌𝑖𝑆 (13) ∀(𝑖, 𝑗) ∈ ⟦1,𝑁⟧ 2 �1 + 𝜀𝑖𝑗��𝜌𝑖𝑆 + 𝑡𝑖𝑗� ≥ 𝜌𝑗𝐷 (14) ∀(𝑖, 𝑗) ∈ ⟦1,𝑁⟧ 2 �𝑥𝑖 ≥ 0,𝑦𝑖 ≥ 0, 𝑥𝑖𝑗 ≥ 0 � (15) Equations (10) and (11) imply that the total quantity exported by i does not exceed I’s production and that the total quantity imported by j is greater than or equal to j’s demand. Equations (12) and (13) state that the market demand price should not exceed the country demand price and that the market supply price should be greater than or equal to the country supply price. When these inequalities are binding, in the case of an interior solution, market and country prices are equal, and the country produces or consumes a nonzero quantity of maize. The fifth constraint, given by equation (14), relates the market supply price (accounting for transport costs and tariffs) to the market demand price, and the last condition, in equation (15), is that demand, supply, and trade are nonnegative. 7 Sobolevsky, Moschini, and Lapan (2005) also used a spatial equilibrium model in the case of the GM soybeans, focusing on different actors in the chain in four regions, and used it to simulate different scenarios on segregation and GM import bans. 7 Table 3.1—List of exporting and importing countries included in the model by group Group Maize Soybeans Net exporters 1 Argentina, U.S.A. Argentina, Canada, U.S.A. 2 Austria, Bulgaria, China, France, Hungary, India, Namibia, Paraguay, Romania, Swaziland, Thailand, Uganda, Ukraine Austria, Ecuador, India, Russia, Slovakia, Tanzania, Ukraine, Uganda, Vietnam 3 Brazil, Czech Republic, South Africa Brazil, Paraguay 4 Moldova Malawi, Moldova Net importers 1 Canada, Uruguay Uruguay 2 Algeria, Bangladesh, Belgium–Luxembourg, Bolivia, Colombia, Costa Rica, Croatia, Cuba, Cyprus, Ecuador, Egypt, El Salvador, Greece, Guatemala, Honduras, Indonesia, Iran, Italy, Japan, Jordan, Kenya, Lebanon, Libya, Malaysia, Mauritius, Mexico, Mozambique, Nigeria, Netherlands, North Korea, Panama, Peru, Saudi Arabia, Slovenia, South Korea, Sri Lanka, Sudan, Syria, Tanzania, Turkey, Venezuela, Vietnam, Yemen, Zambia, Zimbabwe Bolivia, Bulgaria, China, Colombia, Czech Republic, Croatia, Egypt, El Salvador, France, Germany, Guatemala, Greece, Honduras, Indonesia, Italy, Japan, Kenya, North Korea, Peru, Philippines, Poland, Romania, Slovenia, Spain, Sri Lanka, Thailand, Turkey, United Kingdom, Venezuela, Yugoslavia, Zambia, Zimbabwe 3 Germany, Philippines, Spain Mexico, South Africa 4 Angola, Chile, Israel, Jamaica, Kuwait, Malawi, Morocco, Pakistan, Russia Bosnia and Herzegovina Source: Authors. Note: The groups are based on the year 2009 for protocol membership and 2008 for GM maize production (James 2008). Table 3.1 shows the list of countries retained for the simulation; this includes all countries with maize production, export, or import volume during the period 1995–2005 exceeding 0.1 percent of total volume, and for which key data were available. Because spatial trade models allow for only unidirectional bilateral trade flows, we distinguish net exporters from net importers based on United Nations Comtrade data at the HS four-digit level (HS-4 1005 for maize and 1201 for soybeans) from 1995 to 2005. Table 3.2—Data sources for key parameters Parameter Years Sources of original data Production 1995–2005 FAOSTAT, UN Food and Agricultural Organization Domestic prices 1995–2005 FAOSTAT, UN Food and Agricultural Organization Consumer prices 1995–2005 FAOSTAT, UN Food and Agricultural Organization Elasticities of supply 2001–2005 IMPACT model, International Food Policy Research Institute Elasticities of demand 2001–2005 IMPACT model, International Food Policy Research Institute Net trade flows 1995–2005 UN Comtrade 1005 and 1201 (HS-4) bilateral trade data Transportation costs 2004–2006 Ocean freight rates from the International Grains Council Ad valorem tariffs 2005 MAcMap database Source: CEPII (2010), FAOSTAT (2010), International Grains Council (2010), Rosegrant et al (2008), UN COMTRADE (2010). Table 3.2 summarizes the major sources of data used for key parameters. As noted above, we assume linear supply and demand in each country, with initial coefficients based on production data from the Food and Agricultural Organization (FAO) FAOSTAT database, and supply and demand elasticities obtained from the International Model for Policy Analysis of Agricultural Commodities and Trade (IMPACT) of the International Food Policy Research Institute (IFPRI). Transportation costs for each bilateral trade flow are estimated using reported ocean freight rates from the International Grains Council 8 as references, and distances between ports are computed with data from the Centre d’Etudes Prospectives et d’Informations Internationales (CEPII). Tariff rates are based on the MAcMap HS-6 database8 of ad valorem–equivalent aggregate tariffs developed by the CEPII and the International Trade Centre (ITC). Producer and consumer prices are derived from above-listed data and consumer and producer support equivalents from IMPACT and the Organization for Economic Cooperation and Development (OECD).9 Because of the inconsistency across data sources and incomplete datasets for some of the parameters, we use cross-entropy methods to calibrate the data, following a procedure used by Robinson, Cattaneo, and El-Said (2001) and by You and Wood (2006). More specifically, the parameterization is completed in two stages. In the first stage, bilateral trade data are entered and rebalanced in order to respect two constraints: There is no bilateral trade flow between two countries and a country cannot simultaneously import from one partner and export to another one. In the second stage, transport costs are adjusted to fit with the rest of the data, in particular the trade data. In these two stages, the prior distributions of probabilities for the parameters of interest (bilateral trade flows and transport costs) are based on distributions of frequencies of trade volume per exporter and of transport costs directly derived from available data. The support used for these cross-entropy stages is therefore a uniform distribution. The third stage runs the model of quasi-welfare maximization in a standard fashion using a nonlinear solver in the General Algebraic Modeling System (GAMS). Table 3.3—Definitions of scenarios Affected trade flows by group of countries Additional cost imposed on trade flows (US$) Base Scenario 1 Scenario 2 Scenario 3 Set A 1→2, 1→3, 2→2, 2→3 $0 $1.50/ton $6.50/ton $13/ton Set B 1→2, 1→3, 2→2, 2→3 $0 $1.50/ton $6.50/ton $13/ton 3→1, 3→4 $0 $1/ton $5/ton $9/ton Source: Authors, based on Gruère and Rosegrant (2008). We run six scenarios of simulations by implementing marginal increases in transport costs of affected trade flows of potentially GM maize and soybeans. Table 3.3 presents each of the scenarios. Set A imposes additional transport costs only on flows between GM producers (Groups 1 and 3) and CPB members (Groups 2 and 3). Set B includes the same shocks but also includes additional costs for exports from GM-producing countries that are CPB members (Group 3) toward nonmembers (Groups 1 and 4). In other words, Set A provides a minimum (or pragmatic) implementation of the requirements by CPB members, and Set B shows the situation if CPB members implement the requirements to all their exports as long as they produce GM crops. Under each set, the base scenario represents the initial situation, which can be interpreted as the may contain option. Scenario 1 introduces a small per-volume cost on affected trade flows, based on a sum of the export and import costs assumed by Gruère and Rosegrant (2008) but also consistent with the costs estimated by Huang et al. (2008) in the case of China. Scenarios 2 and 3 impose higher additional costs, following Gruère and Rosegrant (2008), who cite JRG Consulting Group (2004) and Kalaitzandonakes (2004), that represent less efficient testing systems10 and that may be more representative of the costs for less developed trading countries. In each case, we focus on three key variables: the relative changes in trade volume, prices, and quantities in major countries. For notification simplicity, the scenarios in Set A will be written A1, A2, and A3, and the scenarios in Set B will be designated B1, B2, and B3. 8 See Bouët et al. 2008. 9 The complete procedures, though not presented here, are available from the authors. 10 JRG Consulting Group (2004) and Kalaitzandonakes (2004) studied the cases of major exporters of GM products, with very advanced infrastructure, and therefore their proposed costs may still be low compared with the actual transport costs for smaller, developing countries. But because they are much higher than those of Huang et al. (2008), which appear to be more precise, we take them as benchmark values for a possible high-end cost of implementation. 9 4. SIMULATION RESULTS Scenarios in Set A Changes in Main Market Variables At the global level (Table 4.1), the additional cost imposed on the main affected trade flows decreases the total production of maize by 190,771 metric tons11 (Scenario A1) to 1.6 million tons (A3) and the production of soybeans by 98,689 tons (Scenario A1) to 855,276 tons (A3). As expected, an increase in the cost of information requirements amplifies the effects the requirements have on the world market. The average country’s supply price increases by up to 3.06 percent in Scenario A3 for maize and 3.1 percent for soybeans. The average country’s demand price also increases, signifying a drop in demand, by up to 4.72 percent for soybeans and 5.76 percent for maize in Scenario A3. However, these results do not provide a good overview of the changes experienced at a lower level of aggregation. Since the shocks are implemented by group, it is useful to first analyze differences in groups, as shown in Table 4.2. Table 4.1—Relative changes (%) in world market variables compared to the Base in Set A Maize Soybeans Scenario A1 A2 A3 A1 A2 A3 Aggregate quantity -0.04% -0.16% -0.33% -0.05% -0.22% -0.45% Average pS +0.37% +1.41% +3.06% +0.37% +1.57% +3.10% Average pD +0.69% +2.70% +5.76% +1.05% +2.66% +4.72% Source: Derived from simulation results. Table 4.2—Relative changes (%) in key variables compared to the base in each group of countries Group 1 Group 2 Group 3 Group 4 Scenario A1 A2 A3 A1 A2 A3 A1 A2 A3 A1 A2 A3 Maize Supply -0.14 -0.69 -1.38 +0.09 +0.53 +1.0 -0.13 -0.63 -1.16 -0.05 -0.38 -0.73 Demand +0.24 +1.22 +2.42 -0.19 -0.95 -1.86 -0.34 -1.37 -2.96 +0.09 +0.63 +1.22 Average pS -0.57 -2.85 -6.34 +0.63 +2.7 +5.7 -0.78 -4.16 -7.87 -0.06 -0.7 -1.39 Average pD -0.57 -2.85 -6.34 +0.89 +3.73 +7.9 +1.43 +5.44 +11.3 -0.27 -2.19 -4.25 Soybeans Supply -0.79 -0.35 -0.69 +0.04 +0.18 +0.35 -0.06 -0.27 -0.55 +0.01 +0.05 +0.09 Demand +0.05 +0.21 +0.43 -0.15 -0.63 -1.27 -0.16 -0.72 -1.43 -0.04 -0.16 -0.31 Average pS -0.81 -3.51 -7.02 +0.5 +2.22 +4.43 -0.71 -3.08 -6.17 +0.8 +2.67 +4.34 Average pD -0.81 -3.51 -7.02 +0.6 +2.55 +5.09 +0.67 +2.9 +5.81 +0.63 +1.43 +1.24 Source: Derived from simulation results. Group 1, including GM-producing countries whose exports are affected (see tables A1 and A2 in the appendix), does experience a decrease in supply and supply prices. Between Scenarios A1 and A3, the production of maize and soybeans declines by 285,345 tons to 2.8 million tons and 84,235 tons to 729,976 tons respectively, which are non-negligible amounts. In the case of maize, most of this decline is experienced by the United States (-271,375 to -2.65 million tons), but all Group 1 countries do reduce 11 All weights are given in metric tons. 10 their production. With the average demand price decreasing by 0.57 to 6.34 percent for maize and by 0.8 to 7 percent for soybeans, internal demand does increase by about 433,993 to 4.3 million tons for maize and by 47,244 to 409,418 tons for soybeans, which have to come from a reduction of exports or an increase in imports. Results for Group 2, which includes non-GM producers who are members of the CPB, are opposite in direction and amplitude. Countries in this group slightly increase their production of maize and soybeans due to the effect of the new tariff-like measures, but they decrease their demand because of an increase in consumer prices. This region is the largest consumer of maize, and a decline in demand of 1.86 percent for maize and 1.28 percent for soybeans in Scenario A3 translates into a reduction of consumption by about 4.3 million tons for maize and 747,615 tons for soybeans. This suggests that the region does increase its imports of maize and soybeans more than it increases its exports overall. However, the group includes a large number of countries that do not all share the same trend. India, Italy, France, and Nigeria experience large decrease in demand for maize (exceeding 230,000 tons), but China is the only country with a drop exceeding 1 million tons. On the supply side, China, France, Italy, and India lead the group in production increase of maize (ranging from 133,889 tons in Italy to 806,902 tons in China in Scenario A3). These increases suggest significant domestic changes, resulting in increased maize exports to countries of other groups. China and India also experience the largest decrease in demand for soybeans between Scenario A1 and A3 (exceeding 300,000 tons and 130,000 tons, respectively). On the supply side, China increases its production of soybeans from 6,953 tons to 60,271 tons; India and Bolivia also experience a large increase in their production of soybeans between A1 and A3 (from 3,967 tons to 34,392 tons and from 1,311 tons to 11,364 tons respectively). Group 3 countries, which include GM producers that are CPB members, represent an intermediate situation for maize, with decreased supply and demand, a higher demand price, and a reduction in the supply price. But the drop in demand (from -180,775 tons in Scenario A1 to -1,568,318 tons in Scenario A3) largely exceeds the decrease in supply (from -83,438 tons to -732,749 tons), signifying a growing maize surplus. Most of the decline in demand is borne by Brazil (-946,456 tons in A3), South Africa (-291,435 tons), and Spain (-214227 tons). The decrease in supply is also experienced greatly by Brazil (-416,133 tons in A3) and, to a lesser extent, South Africa (-142,217 tons in A3), but much less by other countries in the group. Similar results can be highlighted for soybeans in terms of variation of supply, demand, and prices. Soybeans also experience a growing surplus as evidenced by the drop in demand (from -59,311 tons in A1 to -514,072 tons in A3), which greatly exceeds the decrease in supply (from -29,310 tons to -254,000 tons between A1 and A3). Once again, Brazil experiences the larger decline in demand (from -57,214 tons in A1 to -495,896 tons in A3) and in supply (from -25,586 tons to -221,731 tons). To a lesser extent, Paraguay follows with a decrease in demand (-14,202 tons in A3) smaller than the decrease in supply (-30,176 tons in A3). Lastly, Group 4, which includes non-GM producers that are not CPB members, has a distinct pattern with, on average, negligible changes in supply and demand, following the same pattern as Group 2. However, the minor observed decrease in demand is associated with a minor decrease in demand prices, suggesting the presence of heterogeneous effects in countries with differentiated demand elasticities. Indeed, unlike other groups, there is a significant variation of demand effects within Group 4. Moldova experiences lower maize demand (-1,636 tons in A1 to -14,858 tons in A3) while larger countries, like Chile (+41,126 tons in A3), Russia (+21,639 tons), and Pakistan (+16,305 tons), increase their demand for maize. Israel (+21,691 tons in A3) and, to a greater extent, Malawi (+55,225 tons) also experience a significant growth in demand. These variations are consistent with demand price fluctuations across countries. Significant variations are also observable on the supply side, with Moldova producing more maize (up to +13,847 tons) to take advantage of higher supply prices, while other countries decrease their production, especially larger countries such as Russia, Chile, and Pakistan. To a lesser extent, these results also apply to soybeans, with a slight decrease in demand (-3,065 tons in A3) and a growing supply (+3,839 tons) in Russia, while Bosnia and Herzegovina records the opposite trend. 11 Trade Effects The simulation results on trade are generally consistent with our expectations; trade flows that bear additional costs are affected. But the magnitude of trade diversion is perhaps more significant than expected and varies across regions and scenarios. Figures 4.1 and 4.2 show a decomposition of aggregate bilateral trade flows by group under the three scenarios for the two crops, based on Tables A.3 (maize) and A.4 (soybeans) in the Appendix. The top 10 and bottom 10 variations of disaggregated bilateral trade flows are reported by scenario in Tables A.5, A.6, and A.7 in the Appendix. Figure 4.1—Changes in maize trade volume (metric tons) relative to the base under the three scenarios in Set A . Source: Created by authors based on simulation results. -9,000 -7,000 -5,000 -3,000 -1,000 1,000 3,000 5,000 To 1 To 2 To 3 To 4 Th ou sa nd s Scenario A1 From 1 From 2 From 3 From 4 -9,000 -7,000 -5,000 -3,000 -1,000 1,000 3,000 5,000 To 1 To 2 To 3 To 4 Th ou sa nd s Scenario A2 From 1 From 2 From 3 From 4 -9,000 -7,000 -5,000 -3,000 -1,000 1,000 3,000 5,000 To 1 To 2 To 3 To 4 Th ou sa nd s Scenario A3 From 1 From 2 From 3 From 4 12 Figure 4.2—Changes in soybean trade volume (metric tons) relative to the base under the three scenarios in Set A . Source: Created by authors based on simulation results. In Scenario A1, an additional cost is imposed on trade flows going from Group 1 and Group 3 to Group 2 and 3 (see Table 3.3). These trade flows are greatly reduced because of the additional transport cost. In particular, Groups 1 and 3 export around 3.7 million (124,000) tons of maize (soybeans) to Group 2, and 203,353 fewer tons of maize and 64,000 fewer tons of soybeans to Group 3 than in the base scenario. But these deficits are partially compensated for by exports from other groups; Groups 2 and 4 export 3.3 million tons of maize and 37,700 tons of soybeans to Group 2. A domino effect follows, with countries in affected groups (1 and 3) diverting their exports toward non-affected groups (1 and 4), and countries in compensating groups (2 and 4) reducing their exports to affected exporters (1 and 3). Still, in aggregate, the total net trade volume is reduced by 190,772 tons of maize and 98,690 tons of soybeans, and all groups import less maize and soybean than before. But only Group 1 reduces its total export of maize and soybeans because of the additional cost, as shown in Table 4.3. -1,800 -1,400 -1,000 -600 -200 200 600 To 1 To 2 To 3 To 4 Th ou sa nd s Scenario A2 From 1 From 2 From 3 From 4 -1,800 -1,400 -1,000 -600 -200 200 600 To 1 To 2 To 3 To 4 Th ou sa nd s Scenario A3 From 1 From 2 From 3 From 4 -1,800 -1,400 -1,000 -600 -200 200 600 To 1 To 2 To 3 To 4 Th ou sa nd s Scenario A1 From 1 From 2 From 3 From 4 13 Table 4.3—Changes in maize and soybeans export volume (metric tons) relative to the base under the scenarios in Set A Group 1 Group 2 Group 3 Group 4 Total Maize A1 -718,557 162,866 84,097 3,162 -468,433 A2 -3,611,490 685,926 285,293 12,889 -2,627,381 A3 -7,017,858 1,497,366 734,081 28,705 -4,757,705 Soybeans A1 -131,280 16,892 29,784 838 -83,765 A2 -568,851 73,204 129,084 3,632 -362,931 A3 -1,137,667 146,412 258,192 7,264 -725,799 Source: Created by authors based on simulation results. The same general effects are observed on a larger scale under Scenarios A2 and A3 (Figures 4.1 and 4.2 and Table 4.3). Overall, compared with the base, the total trade volume of maize and soybeans decreases by 771,366 and 427,643 tons and 1.6 million (855,276) tons in scenarios A2 and A3 compared to the basis, respectively. Results from Scenario A2 are fully consistent with those from A1 but on a larger scale. Results from Scenario A3, however, do deviate minimally; instead of diminishing, exports of maize from Group 2 to Group 3 increase slightly by a non significant amount (+17,228 tons as shown in Table A.3 in the Appendix). This increase is more significant for soybeans (+87,644 tons) than for maize. This may be because exports to Group 2 are so much diminished (-9.5 million tons of maize from Groups 1 and 3, and -1.6 million tons of soybeans from Group 1) that compensating Group 2 sends an even larger volume to this group than to Group 3, creating an excess demand in Group 3 that may be met by minimal additional amounts from exporters in Group 2 (see Table A.4 in the Appendix). At the country level, the largest changes are experienced by major trading countries in Groups 1 and 3, as expected. For instance, in Scenario A1, the United States decreases its trade balance of maize by about 694,465 tons overall, but it decreases mainly its exports to Japan (Group 2) and Spain (Group 3), by 2 million tons and 22,000 tons, respectively. Its exports of maize increase toward Group 4: Morocco (+509,259 tons) and Chile (+783,185 tons). In the same scenario, Brazil (Group 3), India (Group 2), and France (Group 2) export more maize overall (with the trade balance increasing by 53,544, 67,218, and 46,500 tons, respectively), but Brazil exports 44,000 and 11,000 fewer tons to Venezuela and Colombia, respectively (Group 2), which is compensated for by an increase in exports of 110,912 tons to Morocco (Group 4). South Africa (Group 3) also decreases its exports of maize to various Group 2 countries by 33,819 tons but compensates by exporting 75,960 tons more to Russia (Group 4). Results for soybeans are similar but with lower volumes, perhaps because of a lower flexibility in the market.12 The United States experiences the largest decrease in net exports (-65,520 tons overall in Scenario A1), especially toward Group 2: China (-46,087 tons) and Japan (-17,392 tons). Brazil and Paraguay (Group 3) increase their exports (+69,219 tons overall in Scenario A1) toward countries of Group 2 (Japan, Zimbabwe, and Germany) and Group 1 (Uruguay). 12 Soybean production and exports are concentrated in a few North and South American countries. 14 Scenarios in Set B Change in Main Market Variables Table 4.4 shows the relative changes in prices and quantities at the global level. These results are almost identical to the ones under Set A when comparing the three scenarios (Table 4.4). In particular, the volume of production and the average of supply and demand prices experience identical relative changes. This may indicate that the additional changes have only minor effects on the market, given that they do not represent major trade flows. Table 4.4—Relative changes (%) in world market variables under scenarios in Set B compared with the base Maize Soybeans Scenario B1 B2 B3 B1 B2 B3 Aggregate quantity -0.04% -0.16% -0.33% -0.05% -0.22% -0.45% Average pS +0.4% +1.4% +3.1% +0.38% +1.58% +3.10% Average pD +0.7% +2.7% +5.8% +0.50% +2.13% +4.20% Source: Created by authors based on simulation results. Table 4.5 presents the same relative changes by group. Once again, the results are similar to the ones obtained under Set A, in terms of both signs and quantitative relative changes from the base. A few changes appear for selected scenarios and variables, but never exceeding +/-0.2 percent. The only visible difference concerns Group 4 in the case of soybeans. This group, a relatively lower trader of soybeans than others, experiences additional costs for its imports from CPB members (countries of Group 3) compared with Set A. This results in slightly different effects on the demand side under scenarios B2 and B3, compared with A2 and A3. While Group 1 also witnesses the same changes for imports from CPB members, the effects of additional transport costs are negligible, because the group is made up of mostly net exporting regions, or regions that may compensate for their losses. Table 4.5—Relative changes (%) in key variables in each group of countries under scenarios in Set B compared with the base Group 1 Group 2 Group 3 Group 4 Scenario B1 B2 B3 B1 B2 B3 B1 B2 B3 B1 B2 B3 Maize Supply -0.13 -0.7 -1.38 +0.07 +0.53 +0.99 -0.13 -0.63 -1.17 -0.01 -0.32 -0.69 Demand +0.23 +1.2 +2.4 -0.17 -0.95 -1.86 -0.35 -1.37 -2.9 +0.06 +0.57 +1.2 Average pS -0.53 -2.86 -6.34 +0.65 +2.68 +5.7 -0.85 -4.36 -8.1 +0.02 -0.58 -1.3 Average pD -0.53 -2.86 -6.34 +0.9 +3.7 +7.9 +1.36 +5.24 +11 -0.1 -1.9 -4.08 Soybeans Supply -0.08 -0.35 -0.7 +0.04 +0.18 +0.35 -0.06 -0.27 -0.5 +0.01 +0.05 +0.09 Demand +0.05 +0.21 +0.43 -0.15 -0.63 -1.27 -0.16 -0.7 -1.4 -0.05 -0.22 -0.45 Average pS -0.8 -3.5 -7 +0.5 +2.2 +4.4 -0.7 -3.1 -6.2 +0.8 +2.67 +4.3 Average pD -0.8 -3.5 -7 +0.58 +2.54 +5.1 +0.7 +2.9 +5.8 +0.6 +1.4 +1.24 Source: Created by authors based on simulation results. 15 Trade Effects The trade effects of the shocks implemented under scenarios B1, B2, and B3 are presented in Figures 4.3 and 4.4 and Table 4.6 (total exports by group). In addition, detailed bilateral group flows are shown in Tables A.3 and A.4 in the Appendix, as was done in the case of Set A. At first view, the aggregate results of Table 4.6 appear quite similar to those observed in Table 4.3 (Set A). Total trade volume of maize is reduced by 468,433 tons and soybeans by 83,765 tons under B1; the reductions for the two commodities are 2.6 million tons and 362,931 tons under B2, and 4.8 million under B3 (725,799); slightly less than under A3. All groups reduce their imports, and only Group 1 reduces its exports. But in the detail, the amplitude and direction of intra- and intergroup trade change greatly, as visible in Figures 4.3 and 4.4. Figure 4.3—Changes in maize trade volume (metric tons) relative to the base under the three scenarios in Set B . Source: Created by authors based on simulation results. -12,000 -10,000 -8,000 -6,000 -4,000 -2,000 0 2,000 4,000 6,000 To 1 To 2 To 3 To 4 Th ou sa nd s Scenario B1 From 1 From 2 From 3 From 4 -12,000 -10,000 -8,000 -6,000 -4,000 -2,000 0 2,000 4,000 6,000 To 1 To 2 To 3 To 4Th ou sa nd s Scenario B2 From 1 From 2 From 3 From 4 -12,000 -10,000 -8,000 -6,000 -4,000 -2,000 0 2,000 4,000 6,000 To 1 To 2 To 3 To 4 Th ou sa nd s Scenario B3 From 1 From 2 From 3 From 4 16 Figure 4.4—Changes in soybean trade volume (metric tons) relative to the base under the three scenarios in Set B . Source: Created by authors based on simulation results. Table 4.6—Changes in maize and soybean export volume (metric tons) relative to the base under the scenarios in Set B Group 1 Group 2 Group 3 Group 4 Total Maize B1 -718,557 162,866 84,097 3,162 -468,433 B2 -3,611,490 685,926 285,293 12,889 -2,627,381 B3 -7,017,858 1,497,366 734,081 28,705 -4,757,705 Soybeans B1 -131,280 16,892 29,784 838 -83,765 B2 -568,851 73,204 129,084 3,632 -362,931 B3 -1,137,667 146,412 258,192 7,264 -725,799 Source: Created by authors based on simulation results. -2,500 -2,000 -1,500 -1,000 -500 0 500 1,000 1,500 To 1 To 2 To 3 To 4Th ou sa nd s Scenario B1 From 1 From 2 From 3 From 4 -2,500 -2,000 -1,500 -1,000 -500 0 500 1,000 1,500 To 1 To 2 To 3 To 4Th ou sa nd s Scenario B2 From 1 From 2 From 3 From 4 -2,500 -2,000 -1,500 -1,000 -500 0 500 1,000 1,500 To 1 To 2 To 3 To 4Th ou sa nd s Scenario B3 From 1 From 2 From 3 From 4 17 For maize and soybeans, Group 1 countries follow the same pattern as under Set A: They export much less to Groups 2 and 3, and compensate by exporting more to Groups 1 and 4. But the magnitude of these diversions is much greater than under Set A. In particular, Group 1 reduces its maize (respectively soybean) exports to Group 2 by 3.2 million (0.13 million tons) to 10.8 million tons (2.53 million tons), depending on the scenario, instead of 2 to 8 million tons (maize) and 0.12 million to 1.6 million tons (soybeans) under Set A. Interestingly, these changes happen despite the fact that Group 1 is not directly affected by the new transport costs. Group 2 also follows the same diversion pattern as under Set A, diverting its exports to Groups 2 and 3 to compensate for the losses due to Group 1’s trade reductions. Group 2’s overall imports and exports are very similar to those under Set A. On the import side, however, Group 2 experiences a much larger shift in maize suppliers, notably because of the much larger drop in exports from Group 1. But instead of making up the volume from itself and from Group 4 (as under Set A), it receives a large amount of maize (-188,151 tons to 1.2 million tons) and soybeans (2,300 tons to 1.2 million of tons) from Group 3. Group 3 is in fact the most affected by these additional changes, as expected. It faces additional costs for all its maize and soybean exports, regardless of their destination, but with more costs imposed on trade to Group 2 and itself than to Group 1 and 4. Interestingly, however, these relatively smaller changes on exports to Groups 1 and 4 lead to a complete switch in export diversion for Group 3. Group 3 reduces its exports to Groups 1 and 4, and significantly increases its exports to Group 2 under Scenarios B2 and B3. This may be due to different market considerations, but likely mostly to trade preference factors, such as regular tariffs and transport costs, as well as Group 3 exporters’ own competitiveness compared with that of other countries. The drastic drop in exports from Group 1 to Group 2 may also be a driver of this preference for exporting to Group 2, which has the largest set of importers. Despite these significant changes, the aggregate exports and imports under Set B scenarios are virtually identical to those under Set A scenarios (they are actually identical for soybeans). The effect is a simple and pure trade diversion. Lastly, trade from and to Group 4 is relatively unaffected by the new measure compared with the results for Scenario A. Its total imports do decrease more than under Set A but by relatively small volumes. At the country level (Tables A.8, A.9, and A.10 in the Appendix), the largest changes can be seen in major trading nations of Group 1 and especially Group 3. For instance, in the case of Scenario B1, the United States (Group 1) reduces its maize exports by 4.16 million tons to Japan (Group 2). It also reduces its maize exports to Vietnam (-14,954 tons) and Turkey (-10,205 tons), both from Group 2. These reductions are compensated for by increased maize exports to Group 4: Morocco (+932,173 tons), Chile (+782,839 tons), Malawi (+590,932 tons), Kuwait (+403,115 tons), and to a smaller extent, Angola (+77,944 tons) and Bosnia and Herzegovina (+12,750 tons). The United States also experiences increased maize exports toward Group 2 countries by 1.17 million tons to Malaysia and by more than 600,000 tons each to Ecuador, Belgium, Bangladesh, Greece, Honduras, Libya, Mauritius, Italy, and Slovenia. Its exports to Germany (Group 3) also increase significantly, by 485,300 tons.13 Brazil (Group 3) significantly decreases its maize exports to Colombia and Cuba (Group 2) by 1.2 million tons and 853,414 tons, respectively, and to a smaller extent, toward Vietnam (Group 2) by about 87,600 tons. These decreases are compensated for by increased exports to other countries within Group 2: Tanzania (+717,000 tons), Sudan (+573,214 tons), Mozambique (+394,000 tons), Kenya (+ 325,912 tons), and Yemen (+190,876 tons). South Africa reduces its exports to Germany (Group 3) by about 459,000 tons, and to Group 2 countries Ecuador (-116,806 tons); Belgium (-87,596 tons); Greece and Honduras (down more than 70,000 tons each); and to a lesser extent, Slovenia, Israel, Egypt, and Syria (down less than 4,000 tons each). These reductions are compensated for by exporting an additional 590,729 tons and 300,971 tons toward closer Japan and Bangladesh (Group 2), respectively. As expected, each of these changes is consistent with a cost-minimizing effort on behalf of the exporting country; substitutions are made only to countries at similar distance or closer, or those that have similar or not significantly different trade policies. 13 Overall the United States does decrease its maize exports by about 650,200 tons in this scenario. 18 As with maize, the largest changes for soybeans can be seen in major trading nations of Group 1 and Group 3. Under Scenario B1, the United States (Group 1) decreases its soybean exports by more than 50,000 tons to China and Poland (Group 2), while Paraguay (Group 3) reduces its exports by 33,135 tons to Zimbabwe (Group 2) and by about 6,600 tons to Spain (Group 2). Brazil (Group 3) also decreases its exports to Group 2 countries Bolivia, Italy, and Spain by more than 4,000 tons each and compensates by additional exports of 24,480 tons and 19,654 tons of soybean exports to Uruguay (Group 1) and Japan (Group 2), respectively. Paraguay (Group 3) increases its exports to Germany (Group 2) by about 38,000 tons. Discussion: From Markets to Welfare Effects The results from the simulations have shown that implementing strict information requirements with the does contain option on maize and soybeans could have significant market and especially trade effects. However, although there is less trade and smaller volume of maize or soybeans, which constitute clear market losses, not all countries will experience similar welfare outcomes. In this section we look further by analyzing economic welfare for countries in different regions. We use the slope and intercept coefficients and the supply and demand variables to compute Marshallian consumer and producer surpluses for each country and group in each scenario. Figures 4.5, 4.6, and 4.7 show the absolute changes in consumer surplus, producer surplus, and total surplus for each group compared with the base. Tables A.11 and A.12 in the Appendix provide the results by country for Scenario B3. Figure 4.5—Change in consumer surplus (U.S. dollars per year) for each group under each scenario . Source: Created by authors based on simulation results. -1,500 -1,000 -500 0 500 1,000 1,500 A1 A2 A3 B1 B2 B3 M ill io ns Maize 1 2 3 4 -1,500 -1,000 -500 0 500 1,000 A1 A2 A3 B1 B2 B3M ill io ns Soybeans 1 2 3 4 19 Figure 4.6—Change in producer surplus (U.S. dollars per year) for each group under each scenario . Source: Created by authors based on simulation results. Figure 4.7—Change in total surplus (U.S. dollars per year) for each group under each scenario Source: Created by authors based on simulation results. -1,500 -1,000 -500 0 500 1,000 A1 A2 A3 B1 B2 B3M ill io ns Maize 1 2 3 4 -1,500 -1,000 -500 0 500 1,000 A1 A2 A3 B1 B2 B3M ill io ns Soybeans 1 2 3 4 -800 -600 -400 -200 0 200 A1 A2 A3 B1 B2 B3 M ill io ns Maize 1 2 3 4 -800 -600 -400 -200 0 200 A1 A2 A3 B1 B2 B3 M ill io ns Soybeans 1 2 3 4 20 The results show that the distribution of welfare effects for both maize and soybeans is indeed quite heterogeneous. Generally, the effects observed for maize are replicated in the case of soybeans at a smaller scale (except for Group 4); the main differences are observed across groups. On the consumer side, Group 1 is bound to gain from lower prices, while Groups 2 and 3 lose. These effects are amplified when moving toward more costly scenarios. On the producer side, Group 2 gains from increased trade costs imposed on competitors (which act like protectionist rents), and Groups 1 and 3 lose from loss of market access and price restrictions. The amplitude of these gains and losses also increases with more costly scenarios. On both sides, Group 4 experiences small changes in welfare, positive or negative, that increase with more drastic scenarios. Still, when considering both producer and consumer surplus (Figure 4.7), Group 4 is the only one that derives net welfare gains, which grow from A1 to A3 and B2 to B3, with minimal losses under B1. In contrast, Group 3 does experience non-negligible net losses for maize from US$89 million14 under Scenarios A1 and B1 to $773 million under Scenarios A3 and B3. To a smaller extent, the total welfare losses (which include lost tax revenues) are also quite significant for soybeans, from $62 million under Scenarios A1 and B1 to more than $535 million under Scenarios A3 and B3. Interestingly, Group 2 countries overall, representing CPB members, lose more than Group 1 countries, which are not members of the CPB, but that result is mostly because Group 2 countries together are net importers of maize and soybeans, and because consumers bear more of the surplus than do producers. Overall, these results show that most countries (54 out of 80 countries for maize and 41 out of 53 for soybeans, as shown in tables A.11 and A.12 in Appendix) are bound to lose with strict information requirements, which confirms the conclusions of other studies. But they also shed light on some of the key supports for such requirements as the Cartagena Protocol. Nonmembers have only an indirect role to play in negotiation, so even if the large trading countries in Group 1 (like Argentina, Canada, or the United States) continue to push against it, they may not advance much. Group 4 countries are absent from discussions, as smaller traders and non-members. The core of the support obviously needs to come from member countries in Groups 2 and 3, groups that are both bound to lose overall, especially Group 3 countries for maize (Brazil and Romania are the biggest losers, as shown in Table A.11 in the Appendix) and Groups 2 and 3 countries for soybeans (Brazil and Japan are the biggest losers, as shown in Table A.12 in the Appendix). Yet member countries of these groups (especially Brazil, European countries, and African countries) have generally been very supportive of the strict requirements in meetings of the Protocol. So why do they support a measure that could be economically detrimental for them? As in other political forums, a well-known result from the literature (Olson 1965) is that the best- organized groups are bound to be the most influential. In developed countries, the most influential or well-organized parties tend to be on the production side. Results presented in Figures 4.6 and 4.7 suggest that producers, especially in countries of Group 2, are bound to gain from this measure significantly. In other countries of Group 2, notably those in Africa, producers and consumers are typically not well represented, and the support for such measure has been seen from anti-GM organizations, which are pushing for any restriction in the marketing of GM food. Yet these countries are bound to be directly affected by the measure, with potentially high losses at stake. Table 4.7 shows the welfare results for countries in Sub-Saharan Africa (SSA) in our study in the case of Scenario B3. Of the twelve CPB member SSA countries included in the study, only Uganda (maize and soybeans), Tanzania (soybeans), Namibia (maize), and Swaziland (maize) would experience welfare gains overall, due to production gains and the small number of consumers in the first three countries, and due to the gains of consumers in a small producing country in the third country. Consumers in Group 4 countries (Angola and Malawi) would gain. But at the same time, eight countries, and especially Nigeria and South Africa, would bear nontrivial welfare losses. 14 All dollar amounts are in U.S. dollars. 21 Table 4.7—Change in welfare effects in Sub-Saharan African countries under Scenario B3 compared with the base scenario, in U.S. dollars per year Group Country Maize Soybeans Consumer surplus (USD$) Producer surplus (USD$) Total surplus (USD$) Consumer surplus (USD$) Producer surplus (USD$) Total surplus (USD$) 2 Kenya -16,863,690 14,953,707 -1,909,983 0 0 0 2 Mozambique -11,424,342 8,616,742 -2,807,600 n.a. n.a. n.a. Mauritius -4,337,626 0 -4,337,626 n.a. n.a. n.a. 2 Namibia 0 12,780,857 12,780,857 n.a. n.a. n.a. 2 Nigeria -31,825,904 0 -31,825,904 n.a. n.a. n.a. 2 Senegal -5,052,061 862,223 -4,189,838 n.a. n.a. n.a. 2 Swaziland -453,620 12,915,522 12,461,902 n.a. n.a. n.a. 2 Tanzania -19,693,684 14,974,529 -4,719,155 0 9,332,821 9,332,821 2 Uganda -5,153,379 17,194,193 12,040,814 -770,629 9,551,946 8,781,316 2 Zambia -8,312,453 3,308,317 -5,004,136 -3,864,878 88,967 -3,775,911 2 Zimbabwe -12,215,160 7,929,417 -4,285,743 -4,462,323 502,835 -3,959,488 3 South Africa -50,317,745 -43,092,970 -93,410,716 -4,878,721 -954,005 -9,515,636 4 Angola 4,864,907 -1,813,459 3,033,795 n.a. n.a. n.a. 4 Malawi 10,282,904 -6,995,551 3,287,353 0 9,306,972 9,306,972 Source: Created by authors based on simulation results. Note: n.a. = Not available—countries not included in the soybean model. While small producers in SSA countries (mostly in Group 2) do not always connect to the market (an implicit assumption here), urban consumers are immediately affected by price increases, as observed during the food price increase of 2008. This means that the presumed gains for producers may be overestimated here while consumers’ losses may be underestimated. Producers in Groups 3 and 4 that are connected to the market will also lose. Even South Africa will experience large losses for both producers and consumers. All these groups will probably pay a much larger proportional price than consumers in developed nations of Groups 2 and 3, given their resources, or even producers in some of the most productive countries of Group 3. Naturally, these welfare effects would change if countries were to change their group. Countries like Kenya, Tanzania, and Uganda are part of a large public–private partnership to develop drought- tolerant GM maize, and if they adopted this promising crop, they would join Group 3 and would thus bear welfare losses for both producers and consumers. This penalty effect for GM adoption provides a rationale for why the measure is so strongly supported by anti-GM groups. If in place, it could further discourage developing countries in Africa that are already subject to external influence (see, for example, Paarlberg 2008) from moving toward GM crop adoption. 22 5. CONCLUSIONS In this paper we investigate the economic effects of implementing a strict information requirement (does contain LMO-FFPs with a list of specific GM events) under Article 18.2(a) of the Cartagena Protocol on Biosafety. More specifically, our analysis focuses on evaluating the effect on prices, trade, and welfare of implementing this regulation at the global level. Using a simple analytical model, we first show that such regulation would create price tension with losers and winners. We then use an empirical model to validate our hypothesis in the case of maize and soybeans. We find that under relatively conservative cost assumptions, information requirements would have a significant effect on the world market for both maize and soybeans. But they would have even greater effects on trade, creating significant trade distortion that diverts exports from their original destination. In particular, nonmember countries that produce GM products would reduce their exports to Protocol members, and GM-producing countries that are part of the Protocol would also divert their exports to new destinations, depending on the scenario. The measure would reduce world trade and production in maize and soybeans, with significant welfare effects. At the global level, under the more costly scenario, total welfare effects (consumer and producer surplus plus tax revenue) would decline by up to $1.036 billion annually for maize and by up to $716 million annually for soybeans, with significant heterogeneity across countries and agents. While non-GM producers in Protocol member countries would benefit from increased protection, consumers and producers in selected countries of SSA would proportionately pay a much heftier price for the regulation. Even those that derive gains from new protectionist rents would lose if they decided to adopt potentially beneficial GM crops currently under development, like drought-resistant maize. This situation calls for governments in African and other affected countries to reconsider their support for this new regulation, which does not present any clear benefit for regulators but, if implemented, would be associated with significant costs for generations to come. 23 APPENDIX: SUPPLEMENTARY TABLES Table A.1—Change in export volume (metric tons) relative to the base under the six scenarios for maize A1 A2 A3 B1 B2 B3 Argentina -24,093 -121,090 -235,302 -22,557 -121,304 -235,450 USA -694,465 -3,490,400 -6,782,556 -650,195 -3,496,566 -6,786,838 Austria 7,437 30,317 67,518 7,762 30,272 67,486 Bulgaria 1,678 14,052 31,294 1,827 14,031 31,280 France 46,502 189,577 422,199 48,538 189,296 422,004 Hungary 14,882 60,669 135,114 15,533 60,579 135,052 India 67,218 265,288 541,286 69,609 264,955 541,054 Namibia 75 304 677 78 304 677 Paraguay 3,617 14,748 32,844 3,776 14,726 32,828 Swaziland 944 3,849 8,571 985 3,843 8,567 Thailand 3,110 36,174 99,858 551 36,097 99,804 Uganda 6,565 26,765 59,607 6,853 26,725 59,580 Ukraine 10,838 44,184 98,399 11,312 44,118 98,354 Brazil 53,544 190,792 530,323 60,579 189,821 529,649 Czech Republic 851 1,521 3,073 189 377 1,664 Romania 8,073 23,333 51,467 9,145 23,183 51,363 South Africa 21,629 69,647 149,218 23,845 69,339 149,004 Moldova 3,162 12,889 28,705 3,300 12,870 28,692 Source: Created by authors based on simulation results. Table A.2—Change in export volume (metric tons) relative to the base under the six scenarios for soybeans A1 A2 A3 B1 B2 B3 Argentina -58,748 -254,564 -509,113 -58,748 -254,564 -509,113 Austria 87 378 756 87 378 756 Brazil 31,628 137,072 274,166 31,628 137,072 274,166 Canada -7,011 -30,381 -60,760 -7,011 -30,381 -60,760 Ecuador 146 633 1,266 146 633 1,266 India 15,872 68,781 137,566 15,872 68,781 137,566 Moldavia 37 159 318 37 159 318 Malawi 5 20 41 5 20 41 Paraguay -1,844 -7,988 -15,974 -1,844 -7,988 -15,974 Russia 797 3,452 6,905 797 3,452 6,905 Slovakia 9 39 79 9 39 79 Tanzania 0 2 4 0 2 4 Uganda 39 170 340 39 170 340 Ukraine 492 2,131 4,262 492 2,131 4,262 U.S.A. -65,520 -283,906 -567,794 -65,520 -283,906 -567,794 Venezuela 3 12 25 3 12 25 Vietnam 238 1,031 2,062 238 1,031 2,062 Zambia 6 26 53 6 26 53 Source: Created by authors based on simulation results. 24 Table A.3—Change in trade volume (metric tons) relative to the base under the six scenarios for maize To 1 To 2 To 3 To 4 Total A1 From 1 433,993 -2,066,237 -22,724 1,369,623 -285,345 From 2 0 3,312,822 -4,026 -3,126,336 182,460 From 3 0 -1,716,922 -180,629 1,814,113 -83,438 From 4 0 39,281 0 -43,730 -4,449 Total 433,993 -431,056 -207,379 13,670 -190,772 To 1 To 2 To 3 To 4 Total A2 From 1 2,181,258 -4,466,579 -85,498 936,668 -1,434,151 From 2 0 4,239,306 -14,689 -3,126,336 1,098,281 From 3 5 -2,033,984 -731,929 2,367,559 -398,349 From 4 0 49,008 0 -86,158 -37,150 Total 2,181,263 -2,212,248 -832,116 91,734 -771,367 To 1 To 2 To 3 To 4 Total A3 From 1 4,335,510 -8,426,293 -216,570 1,470,409 -2,836,944 From 2 0 5,155,959 17,228 -3,126,336 2,046,851 From 3 -526 -1,120,165 -1,583,204 1,971,145 -732,750 From 4 0 64,824 0 -136,463 -71,639 Total 4,334,984 -4,325,675 -1,782,546 178,756 -1,594,481 To 1 To 2 To 3 To 4 Total B1 From 1 406,854 -3,238,347 461,436 2,102,902 -267,155 From 2 0 2,990,845 30,006 -2,870,562 150,289 From 3 -526 -188,151 -706,715 816,163 -79,229 From 4 0 39,419 0 -40,274 -855 Total 406,328 -396,234 -215,273 8,228 -196,951 To 1 To 2 To 3 To 4 Total From 1 2,185,643 -7,178,650 -99,770 3,656,093 -1,436,684 From 2 0 4,208,457 14,253 -3,126,336 1,096,374 B2 From 3 -526 712,501 -744,823 -366,637 -399,485 From 4 0 48,989 0 -79,915 -30,926 Total 2,185,117 -2,208,702 -830,340 83,205 -770,720 B3 To 1 To 2 To 3 To 4 Total From 1 4,338,328 -10,799,805 -176,226 3,798,926 -2,838,777 From 2 0 5,211,903 -39,272 -3,126,336 2,046,295 From 3 -526 1,198,826 -1,565,476 -366,637 -733,813 From 4 0 64,811 0 -132,545 -67,734 Total 4,337,802 -4,324,265 -1,780,974 173,409 -1,594,028 Source: Created by authors based on simulation results. Note: Shaded cells represent affected trade flows. 25 Table A.4—Change in trade volume (metric tons) relative to the base under the six scenarios for soybeans To 1 To 2 To 3 To 4 Total From 1 47,045 -126,280 -5,000 0 -84,235 From 2 -27,281 36,882 4,783 -17 14,367 A1 From 3 27,480 2,304 -59,095 0 -29,311 From 4 0 838 0 -349 489 Total 47,244 -86,256 -59,311 -366 -98,690 To 1 To 2 To 3 To 4 Total A2 From 1 203,852 -557,040 -11,811 0 -364,999 From 2 -27,281 695,544 10,872 -616,875 62,260 From 3 28,144 -515,931 -256,089 616,871 -127,005 From 4 0 3,632 0 -1,532 2,100 Total 204,715 -373,796 -257,028 -1,535 -427,644 To 1 To 2 To 3 To 4 Total A3 From 1 407,692 -1,665,099 -89,522 616,954 -729,976 From 2 -27,281 681,036 87,644 -616,875 124,525 From 3 29,008 229,184 -512,194 0 -254,002 From 4 0 7,264 0 -3,087 4,177 Total 409,418 -747,615 -514,072 -3,008 -855,276 To 1 To 2 To 3 To 4 Total B1 From 1 47,045 -130,918 -361 0 -84,234 From 2 -27,281 41,520 145 -17 14,384 From 3 27,480 2,304 -59,095 0 -29,311 From 4 0 838 0 -349 489 Total 47,244 -86,256 -59,311 -366 -98,672 To 1 To 2 To 3 To 4 Total From 1 1,216,747 -1,835,457 -363,161 616,871 -365,000 From 2 -27,281 368,224 338,192 -616,875 62,260 B2 From 3 -984,751 1,105,826 -248,079 0 -127,004 From 4 0 -12,388 16,020 -1,532 2,100 Total 204,715 -373,796 -257,028 -1,535 -427,644 To 1 To 2 To 3 To 4 Total From 1 1,421,451 -2,528,565 -239,816 616,954 -729,976 From 2 -27,281 530,743 237,938 -616,875 124,525 B3 From 3 -984,751 1,242,943 -512,194 0 -254,002 From 4 0 7,264 0 -3,087 4,177 Total 409,418 -747,615 -514,072 -3,008 -855,276 Source: Created by authors based on simulation results. Note: Shaded cells represent affected trade flows. 26 Table A.5—Change in export volume (metric tons) relative to the base under Scenario A1: Top 10 and bottom 10 trade flows Maize Soybeans Exporter Gp. Importer Gp. A1 Exporter Gp. Importer Gp. A1 Bottom 10 U.S.A. 1 Japan 2 -2,043,532 U.S.A. 1 China 2 -46,087 Uganda 2 Morocco 4 -931,942 Argentina 1 Poland 2 -27,622 Czech Rep. 3 Nigeria 2 -669,621 India 2 Uruguay 1 -27,281 France 2 Chile 4 -603,303 Ecuador 2 Germany 2 -25,808 France 2 Pakistan 4 -568,343 U.S.A. 1 Japan 2 -17,392 Czech Rep. 3 Sri Lanka 2 -484,336 Argentina 1 UK 2 -12,848 Hungary 2 Iran 2 -460,190 Paraguay 3 Zimbabwe 2 -12,806 France 2 Russia 4 -452,677 Brazil 3 Zimbabwe 2 -9,720 Swaziland 2 Iran 2 -451,490 Argentina 1 Colombia 2 -7,394 Austria 2 Bosnia-Herz. 4 -282,808 Canada 1 Japan 2 -7,011 Top 10 Czech Rep. 3 Bosnia-Herz. 4 283,154 India 2 Hungary 2 5,469 U.S.A. 1 Morocco 4 509,259 India 2 Philippines 2 5,469 France 2 Sri Lanka 2 531,194 India 2 Poland 2 5,469 Czech Rep. 3 Russia 4 546,623 Ecuador 2 UK 2 5,469 Swaziland 2 Nigeria 2 582,529 India 2 UK 2 5,469 Czech Rep. 3 Pakistan 4 594,509 Russia 4 Colombia 2 6,882 U.S.A. 1 Chile 4 783,185 Paraguay 3 Germany 2 17,447 Hungary 2 Japan 2 896,275 Ecuador 2 Poland 2 21,876 France 2 Iran 2 950,620 India 2 Zimbabwe 2 22,426 Uganda 2 Japan 2 1,098,893 Brazil 3 Japan 2 24,292 Source: Created by authors based on simulation results. Table A.6—Change in export volume (metric tons) relative to the base under Scenario A2: Top 10 and bottom 10 trade flows Maize Soybeans Exporter Gp. Importer Gp. A2 Exporter Gp. Importer Gp. A2 Bottom 10 U.S.A. 1 Japan 2 -4,159,330 Brazil 3 Bolivia 2 -501,060 Thailand 2 China 2 -1,199,270 Uganda 2 Bosnia-Herz. 4 -432,243 Uganda 2 Morocco 4 -931,942 India 2 Turkey 2 -234,542 Argentina 1 Malaysia 2 -841,513 India 2 Bosnia-Herz. 4 -184,632 Czech Rep. 3 Nigeria 2 -727,999 India 2 Egypt 2 -182,532 Argentina 1 Libya 2 -653,104 Argentina 1 Poland 2 -162,159 France 2 Chile 4 -603,303 U.S.A. 1 Thailand 2 -130,349 Hungary 2 Philippines 3 -574,536 Ukraine 2 China 2 -123,903 France 2 Pakistan 4 -568,343 Ecuador 2 Zimbabwe 2 -104,966 U.S.A. 1 Egypt 2 -564,026 U.S.A. 1 China 2 -80,452 Top 10 Czech Rep. 3 Pakistan 4 607,664 Russia 4 Colombia 2 20,346 Romania 3 Libya 2 632,932 Brazil 3 Uruguay 1 28,144 Argentina 1 Venezuela 2 704,129 Brazil 3 Japan 2 95,385 Argentina 1 Japan 2 719,280 Ukraine 2 Thailand 2 117,054 Brazil 3 Morocco 4 789,684 Ecuador 2 Poland 2 146,735 U.S.A. 1 Chile 4 805,020 India 2 Zimbabwe 2 158,979 U.S.A. 1 Malaysia 2 838,654 Uganda 2 Egypt 2 179,617 France 2 Iran 2 932,177 Uganda 2 Turkey 2 230,674 Uganda 2 Japan 2 1,119,093 India 2 Bolivia 2 488,289 Hungary 2 Japan 2 1,312,283 Brazil 3 Bosnia-Herz. 4 616,871 Source: Created by authors based on simulation results. 27 Table A.7—Change in export volume (metric tons) relative to the base under Scenario A3: Top 10 and bottom 10 trade flows Maize Soybeans Exporter Gp. Importer Gp. A3 Exporter Gp. Importer Gp. A3 Bottom 10 U.S.A. 1 Japan 2 -4,159,330 Argentina 1 UK 2 -595,491 Thailand 2 China 2 -1,857,582 Uganda 2 Bosnia-Herz. 4 -432,243 Uganda 2 Morocco 4 -931,942 Argentina 1 Poland 2 -295,261 Argentina 1 Malaysia 2 -820,895 U.S.A. 1 China 2 -284,820 U.S.A. 1 Greece 2 -780,970 India 2 Turkey 2 -234,542 Czech Rep. 3 Nigeria 2 -727,999 U.S.A. 1 Japan 2 -196,367 U.S.A. 1 Croatia 2 -692,480 India 2 Bosnia-Herz. 4 -184,632 France 2 Chile 4 -603,303 India 2 Egypt 2 -182,532 Hungary 2 Philippines 3 -574,536 Ecuador 2 Zimbabwe 2 -147,764 France 2 Pakistan 4 -568,343 Ukraine 2 China 2 -123,903 Top Argentina 1 Angola 4 562,121 Brazil 3 Bolivia 2 82,944 10 Czech Rep. 3 Russia 4 582,496 Ukraine 2 South Africa 3 87,644 India 2 Croatia 2 605,554 Russia 4 Colombia 2 89,544 South Africa 3 Greece 2 645,414 India 2 Zimbabwe 2 115,147 U.S.A. 1 Chile 4 716,828 Uganda 2 Egypt 2 176,701 France 2 Iran 2 790,355 Uganda 2 Turkey 2 226,806 France 2 Japan 2 797,160 Ecuador 2 Poland 2 244,635 Swaziland 2 Nigeria 2 823,567 Brazil 3 Japan 2 256,635 Uganda 2 Japan 2 1,151,935 India 2 UK 2 583,537 Hungary 2 Japan 2 1,386,728 Argentina 1 Bosnia-Herz. 4 616,954 Source: Created by authors based on simulation results. Table A.8—Change in export volume (metric tons) relative to the base under Scenario B1: Top 10 and bottom 10 trade flows Maize Soybeans Exporter Gp. Importer Gp. B1 Exporter Gp. Importer Gp. B1 Bottom 10 U.S.A. 1 Japan 2 -4,159,330 U.S.A. 1 China 2 -50,798 U.S.A. 1 Egypt 2 -2,011,723 Argentina 1 Poland 2 -50,567 Namibia 2 Nigeria 2 -1,782,003 Ecuador 2 Germany 2 -46,359 Argentina 1 Malaysia 2 -1,276,783 Paraguay 3 Zimbabwe 2 -33,135 Brazil 3 Colombia 2 -1,206,428 India 2 Uruguay 1 -27,281 Uganda 2 Morocco 4 -931,942 U.S.A. 1 Japan 2 -12,752 Brazil 3 Cuba 2 -853,414 Canada 1 Japan 2 -7,011 Austria 2 Tanzania 2 -734,536 Paraguay 3 Spain 2 -6,595 Argentina 1 Libya 2 -653,104 Brazil 3 Italy 2 -4,733 France 2 Chile 4 -603,303 Brazil 3 Bolivia 2 -4,292 Top 10 Argentina 1 Cuba 2 851,836 India 2 Bolivia 2 1,574 Argentina 1 Peru 2 917,556 Uganda 2 Bosnia-Herz. 4 1,604 U.S.A. 1 Morocco 4 932,173 Russia 4 Colombia 2 2,334 Uganda 2 Japan 2 1,053,263 Vietnam 2 Thailand 2 2,764 U.S.A. 1 Malaysia 2 1,170,383 Ukraine 2 China 2 3,642 Argentina 1 Colombia 2 1,194,559 Brazil 3 Japan 2 19,654 Austria 2 Iran 2 1,378,181 Brazil 3 Uruguay 1 27,480 Ukraine 2 Nigeria 2 1,386,112 Paraguay 3 Germany 2 37,887 Namibia 2 Japan 2 1,782,081 India 2 Zimbabwe 2 38,918 France 2 Egypt 2 1,907,269 Ecuador 2 Poland 2 49,369 Source: Created by authors based on simulation results. 28 Table A.9—Change in export volume (metric tons) relative to the base under Scenario B2: Top 10 and bottom 10 trade flows Maize Soybeans Exporter Gp. Importer Gp. B2 Exporter Gp. Importer Gp. B2 Bottom 10 U.S.A. 1 Japan 2 -4,159,330 Brazil 3 Uruguay 1 -984,751 Thailand 2 China 2 -1,196,722 Canada 1 Japan 1 -684,717 Uganda 2 Morocco 4 -931,942 Argentina 1 UK 1 -595,491 Argentina 1 Libya 2 -653,104 Uganda 2 Bosnia-Herz. 1 -432,243 Namibia 2 Nigeria 2 -649,368 Argentina 1 South Africa 1 -371,394 U.S.A. 1 Greece 2 -628,664 Argentina 1 Poland 1 -241,376 U.S.A. 1 Egypt 2 -627,508 India 2 Turkey 1 -234,542 France 2 Chile 4 -603,303 U.S.A. 1 Japan 1 -196,367 France 2 Pakistan 4 -568,343 India 2 Bosnia-Herz. 1 -184,632 Austria 2 Peru 2 -541,620 India 2 Egypt 1 -182,532 Top 10 Czech Rep. 3 Nigeria 2 601,630 Brazil 3 Bolivia 1 94,726.1 U.S.A. 1 Pakistan 4 604,281 Ecuador 2 Zimbabwe 1 135,266.7 Namibia 2 Japan 2 649,672 Uganda 2 Egypt 1 179,616.7 Romania 3 Libya 2 651,622 Uganda 2 Turkey 1 230,674.1 Swaziland 2 Peru 2 744,549 India 2 Poland 1 294,988.5 France 2 Iran 2 932,213 Ukraine 2 South Africa 1 322,172.2 U.S.A. 1 Morocco 4 937,208 India 2 UK 1 606,368.7 Hungary 2 Japan 2 958,186 Canada 1 Bosnia-Herz. 1 616,871.4 Uganda 2 Japan 2 1,119,053 Argentina 1 Uruguay 1 1,012,896 Austria 2 Japan 2 1,347,846 Brazil 3 Japan 1 1,039,526 Source: Created by authors based on simulation results. Table A.10—Change in export volume (metric tons) relative to the base under Scenario B3: Top 10 and bottom 10 trade flows Maize Soybeans Exporter Gp. Importer Gp. B3 Exporter Gp. Importer Gp. B3 Bottom 10 U.S.A. 1 Japan 2 -4,159,330 Brazil 3 Uruguay 1 -984,751 U.S.A. 1 Egypt 2 -2,025,745 Argentina 1 UK 2 -595,491 Thailand 2 China 2 -1,857,582 Argentina 1 South Africa 3 -583,634 Uganda 2 Morocco 4 -931,942 Argentina 1 Poland 2 -507,527 U.S.A. 1 Greece 2 -780,970 Uganda 2 Bosnia-Herz. 4 -432,243 U.S.A. 1 Croatia 2 -692,480 Vietnam 2 Indonesia 2 -426,121 Czech Rep. 3 Nigeria 2 -654,687 Austria 2 Japan 2 -403,471 Romania 3 Italy 2 -652,705 Argentina 1 Colombia 2 -370,125 Argentina 1 Libya 2 -610,167 U.S.A. 1 China 2 -368,575 France 2 Chile 4 -603,303 Russia 4 Hungary 2 -354,721 Top 10 Argentina 1 Morocco 4 567,779 India 2 Hungary 2 314,520.2 U.S.A. 1 Bosnia-Herz. 4 574,425 U.S.A. 1 South Africa 3 343,371.8 India 2 Croatia 2 614,018 Russia 4 Colombia 2 365,690.8 South Africa 3 Greece 2 683,471 Austria 2 Indonesia 2 404,226.8 U.S.A. 1 Chile 4 830,764 Vietnam 2 Thailand 2 428,183.8 Swaziland 2 Egypt 2 849,990 Ecuador 2 Poland 2 479,653.3 Czech Rep. 3 Egypt 2 921,275 India 2 UK 2 598,088.5 Uganda 2 Japan 2 1,151,908 Argentina 1 Bosnia-Herz. 4 616,954 Hungary 2 Japan 2 1,386,666 Brazil 3 Japan 2 666,545.5 France 2 Japan 2 1,432,215 Argentina 1 Uruguay 1 1,013,759 Source: Created by authors based on simulation results. 29 Table A.11—Welfare effects for maize (Scenario B3 compared with base) by country in decreasing order of total surplus, in U.S. dollars per year Group Consumer Surplus (USD$) Producer Surplus (USD$) Total Surplus (USD$) France 2 -45,070,877 70,382,039 25,311,162 India 2 -77,003,315 95,668,532 18,665,217 Hungary 2 -34,203,209 50,622,027 16,418,819 Ukraine 2 -25,388,278 40,027,374 14,639,096 Austria 2 -8,931,600 23,321,346 14,389,746 Bulgaria 2 -8,214,527 21,333,467 13,118,941 Moldova 4 -8,061,054 21,031,769 12,970,715 Namibia 2 0 12,780,857 12,780,857 Swaziland 2 -453,620 12,915,522 12,461,902 Uganda 2 -5,153,379 17,194,193 12,040,814 Thailand 2 -22,255,120 34,071,475 11,816,355 Paraguay 2 -3,321,440 13,590,844 10,269,404 Morocco 4 5,214,748 -554,835 4,659,913 Israel 4 4,488,106 0 4,488,106 Chile 4 8,597,715 -4,374,424 4,223,291 Malawi 4 10,282,904 -6,995,551 3,287,353 Jamaica 4 3,272,035 0 3,272,035 Kuwait 4 3,139,964 0 3,139,964 Angola 4 4,864,907 -1,813,459 3,051,448 Uruguay 1 3,818,429 -887,761 2,930,669 Pakistan 4 9,459,700 -6,959,766 2,499,934 Bosnia & Herzegovina 4 5,219,822 -2,878,746 2,341,076 Russia 4 9,598,192 -7,279,777 2,318,415 Canada 1 17,631,074 -17,465,876 165,198 Indonesia 2 0 0 0 Mexico 2 0 0 0 South Korea 2 -770,675 320,610 -450,065 Vietnam 2 -17,994,403 16,635,835 -1,358,568 Kenya 2 -16,863,690 14,953,707 -1,909,983 Mozambique 2 -11,424,342 8,616,742 -2,807,600 China 2 -316,930,047 313,699,946 -3,230,101 Italy 2 -68,848,536 65,568,770 -3,279,767 Croatia 2 -17,133,267 13,776,914 -3,356,353 Bolivia 2 -7,863,384 3,912,010 -3,951,374 Sudan 2 -4,211,087 108,605 -4,102,483 Sri Lanka 2 -4,301,331 186,266 -4,115,065 Belgium-Luxembourg 2 -5,365,103 1,179,088 -4,186,015 Senegal 2 -5,052,061 862,223 -4,189,838 Zimbabwe 2 -12,215,160 7,929,417 -4,285,743 Mauritius 2 -4,337,626 0 -4,337,626 30 Table A.11—Continued Group Consumer Surplus (USD$) Producer Surplus (USD$) Total Surplus (USD$) Yemen 2 -4,693,129 173,332 -4,519,797 Bangladesh 2 -5,246,789 616,462 -4,630,327 Lebanon 2 -4,655,887 15,818 -4,640,069 Cyprus 2 -4,666,032 0 -4,666,032 Tanzania 2 -19,693,684 14,974,529 -4,719,155 Honduras 2 -8,028,624 3,220,785 -4,807,840 Slovenia 2 -7,230,180 2,261,006 -4,969,174 Panama 2 -5,402,121 430,733 -4,971,388 Jordan 2 -5,075,638 93,294 -4,982,344 Zambia 2 -8,312,453 3,308,317 -5,004,136 Libya 2 -5,066,729 8,511 -5,058,218 Costa Rica 2 -5,421,085 119,126 -5,301,960 El Salvador 2 -9,059,598 3,607,092 -5,452,507 Ecuador 2 -8,310,395 2,386,726 -5,923,669 Greece 2 -20,496,971 14,139,394 -6,357,577 Cuba 2 -6,562,586 0 -6,562,586 Guatemala 2 -11,682,662 4,967,934 -6,714,728 Peru 2 -13,113,143 6,258,229 -6,854,914 Algeria 2 -7,671,730 3,406 -7,668,324 Netherlands 2 -8,669,929 977,389 -7,692,539 Syria 2 -7,911,981 0 -7,911,981 Venezuela 2 -15,698,768 7,711,697 -7,987,071 Turkey 2 -24,084,676 16,034,538 -8,050,138 Saudi Arabia 2 -8,359,529 151,557 -8,207,972 Colombia 2 -16,650,701 7,662,577 -8,988,125 Malaysia 2 -10,363,537 349,630 -10,013,907 Iran 2 -15,293,149 4,883,617 -10,409,531 Egypt 2 -52,555,962 38,927,681 -13,628,281 Czech Republic 3 -2,194,608 -12,101,598 -14,296,206 Argentina 1 22,709,989 -47,099,824 -24,389,835 Nigeria 2 -31,825,904 0 -31,825,904 Germany 3 -23,256,453 -13,052,688 -36,309,141 North Korea 2 -39,569,164 0 -39,569,164 Japan 2 -40,103,626 0 -40,103,626 Spain 3 -37,066,976 -18,195,357 -55,262,332 Philippines 3 -34,583,943 -24,878,740 -59,462,683 South Africa 3 -50,317,745 -43,092,970 -93,410,716 USA 1 884,791,079 -986,061,468 -101,270,389 Romania 3 -69,400,487 -53,422,320 -122,822,807 Brazil 3 -205,837,242 -185,974,222 -391,811,464 Source: Created by authors based on simulation results. 31 Table A.12—Welfare effects for soybeans (Scenario B3 compared with base) by country in decreasing order of total surplus, in U.S. dollars per year Group Consumer Surplus (USD$) Producer Surplus (USD$) Total Surplus (USD$) India 2 -31,333,891 43,529,664 12,195,773 Ukraine 2 -712,144 10,248,947 9,536,803 Moldova 4 0 9,440,711 9,440,711 Austria 2 -217,046 9,583,355 9,366,309 Tanzania 2 0 9,332,821 9,332,821 Slovakia 2 -51,195 9,371,420 9,320,226 Vietnam 2 -986,436 10,237,045 9,250,609 Russia 4 -2,151,588 11,131,251 8,979,663 Ecuador 2 -425,826 9,404,555 8,978,729 Uganda 2 -770,629 9,551,946 8,781,316 Uruguay 1 7,583,946 -727,801 6,856,145 Bosnia & Herzegovina 4 2,115,105 -19,665 2,095,439 Kenya 2 0 0 0 Venezuela 4 0 24,517 0 Malawi 2 0 9,306,972 0 Romania 2 -4,177,058 1,001,095 -3,175,964 Sri Lanka 2 -3,317,201 8,441 -3,308,759 El Salvador 2 -3,535,429 12,577 -3,522,853 Poland 2 -3,667,056 527 -3,666,529 Bolivia 2 -11,016,874 7,304,446 -3,712,428 Zambia 2 -3864878 88967 -3775911 Germany 2 -3809443 4395 -3805048 Bulgaria 2 -3864116 25280 -3838836 Czech Republic 2 -3870776 30931 -3839845 Hungary 2 -4103575 262728 -3840847 Slovenia 2 -3843806 566 -3843240 Yugoslavia 2 -5088723 1214110 -3874613 Croatia 2 -4315811 436330 -3879481 United Kingdom 2 -3948360 0 -3948360 Zimbabwe 2 -4462323 502835 -3959488 Peru 2 -3974962 14977 -3959985 Colombia 2 -4370028 366736 -4003293 Honduras 2 -4050602 21288 -4029314 Guatemala 2 -4247123 204975 -4042148 France 2 -5448555 1237356 -4211200 Egypt 2 -4421879 187015 -4234863 Greece 2 -4320172 22261 -4297912 Philippines 2 -4351975 7373 -4344602 Turkey 2 -4717451 308842 -4408609 32 Table A.12—Continued Group Consumer Surplus (USD$) Producer Surplus (USD$) Total Surplus (USD$) Italy 2 -8979589 4310612 -4668977 Indonesia 2 -11481954 6590847 -4891108 South Africa 3 -4878721 -954005 -5832726 Spain 2 -6309439 31406 -6278033 South Korea 2 -7525700 760536 -6765164 Thailand 2 -9155722 1839736 -7315986 Canada 1 13545563 -24910303 -11364740 Mexico 3 -10563844 -842515 -11406359 China 2 -102190929 87415271 -14775658 USA 1 475362089 -496772896 -21410807 Argentina 1 129875487 -165597097 -35721609 Paraguay 3 -18537941 -31789362 -50327303 Japan 2 -89364565 1097388 -88267177 Brazil 3 -189163459 -278298108 -467461567 Source: Derived from simulation results. 33 REFERENCES Bouët, A., Y. Decreux, L. Fontagné, S. Jean, and D. Laborde. 2008. “Assessing Applied Protection across the World.” Review of International Economics 16 (5): 850–863. Centre d’Etudes Prospectives et d’Information Internationales (CEPII). 2009. MacMaps Database consulted December 2009. Available online. www.cepii.fr/anglaisgraph/bdd/macmap.htm Devadoss, S., A. H. Aguiar, S. Shook, and J. Araji. 2005. “A Spatial Equilibrium of U.S.–Canadian Disputes on the World Softwood Lumber Market.” Canadian Journal of Agricultural Economics 53 (2 & 3): 177–192. Direccion Nacional de Mercados Agroalimentarios. 2004. Contexto y Opciones para la Exportación Segregada de Maíz y Soya OVM y No OVM en Condiciones de Bioseguridad, Conforme al Protocolo de Cartagena. Proyecto FAO/SAGPYA TCP/ARG 2903. Rome: UN Food and Agriculture Organization. FAOSTAT .2010. Database. Available Online: http://faostat.fao.org/faostat/collections?subset=agriculture. Foster, M., and D. Galeano. 2006. Biosafety Protocol-Implications of the Documentation Regime. ABARE eReport 06.2. Canberra, Australia: Australian Bureau of Agricultural and Resource Economics. Gruère, G. P. 2006. An Analysis of Trade Related Regulations of Genetically Modified Food and their Effects on Developing Countries. EPT Discussion Paper 147, Washington DC: International Food Policy Research Institute. Gruère, G. P., and M. W. Rosegrant. 2008. “Assessing the Implementation Effects of the Biosafety Protocol’s Proposed Stringent Information Requirements for Genetically Modified Commodities in Countries of the Asia Pacific Economic Cooperation.” Review of Agricultural Economics 30 (2): 214–232. Huang, J., D. Zhang, J. Yang, S. Rozelle, and N. Kalaitzandonakes. 2008. Will the Biosafety Protocol Hinder or Protect the Developing World? Learning from China’s Experience. Food Policy 33 (1): 1–12. ICTSD (International Centre for Trade and Sustainable Development). 2006. Constructive Ambiguity Saves LMO Labelling Discussions at MOP-3. 2006. Bridges Trade BioRes 6 (5): 1–3. International Grains Council. 2010. Ocean Freight Rates. Available online: www.igc.int/en/grainsupdate/igcfreight.aspx James, C. 2008. Global Status of Commercialized Biotech/GM Crops: 2008. ISAAA Brief No. 39. Ithaca, NY: International Service for the Acquisition of Agri-Biotech Applications. JRG Consulting Group. 2004. Economic Assessment of the Ratification of the Cartagena Protocol, Based on the Outcome of the 1st Meeting of Parties. Final Report Prepared for Agriculture and Agri-Food Canada, Ottawa. Kalaitzandonakes, N. 2004. The Potential Impacts of the Biosafety Protocol on Agricultural Commodity Trade. IPC Technology Issue Brief. Washington, DC: International Food and Agricultural Trade Policy Council. Olson, M. 1965. The Logic of Collective Action. Cambridge, MA, US: MIT University Press. Paarlberg, R. 2008, Starved for Science. How Biotechnology is Being Kept Out of Africa. Harvard University Press, Cambridge, MA. Redick T. P. 2007. “The Cartagena Protocol on Biosafety: Precautionary Priority in Biotech Crop Approvals and Containment of Commodities Shipments.” Colorado Journal of International Environmental Law and Policy 18(Winter 2007):51–116. Rosegrant, M.W., S. Msangi, C. Ringler, T.B. Sulser, T. Zhu and S.A. Cline 2008. International Model for Policy Analysis of Agricultural Commodities and Trade (Impact): Model Description. Washington, D.C.: International Food Policy Research Institute. Robinson, S., A. Cattaneo, and M. El-Said. 2001. “Updating and Estimating a Social Accounting Matrix Using Cross Entropy Methods.” Economic Systems Research 13 (1): 47–64. 34 Samuelson, P. 1952. “Spatial Price Equilibrium and Linear Programming.” American Economic Review 21: 283– 303. Secretariat of the COB (Convention on Biological Diversity). 2000. Cartagena Protocol on Biosafety to the Convention on Biological Diversity: Text and Annexes. Montreal, Canada: Secretariat of the Convention on Biological Diversity. Sobolevsky, A., G. Moschini, and H. Lapan. 2005. “Genetically Modified Crops and Product Differentiation: Trade and Welfare Effects in the Soybean Complex.” American Journal of Agricultural Economics 87 (3): 621– 644. Takayama, T., and G. Judge. 1971. Spatial and Temporal Price Allocation Models. Amsterdam: North Holland. UN Comtrade . 2010. New York: United Nations Available at: http://comtrade.un.org/db/dqBasicQuery.aspx You, L. and S. Wood. 2006. “An Entropy Approach to Spatial Disaggregation of Agricultural Production.” Agricultural Systems 90 (1–3): 329–347. RECENT IFPRI DISCUSSION PAPERS For earlier discussion papers, please go to http://www.ifpri.org/publications/results/taxonomy%3A468. All discussion papers can be downloaded free of charge. 1101. Beyond fatalism: An empirical exploration of self-efficacy and aspirations failure in Ethiopia. Tanguy Bernard, Stefan Dercon, and Alemayehu Seyoum Taffesse, 2011. 1100. Potential collusion and trust: Evidence from a field experiment in Vietnam. Maximo Torero and Angelino Viceisza, 2011. 1099. Trading in turbulent times: Smallholder maize marketing in the Southern Highlands, Tanzania. Bjorn Van Campenhout, Els Lecoutere, and Ben D’Exelle, 2011. 1098. Agricultural management for climate change adaptation, greenhouse gas mitigation, and agricultural productivity: Insights from Kenya. Elizabeth Bryan, Claudia Ringler, Barrack Okoba, Jawoo Koo, Mario Herrero, and Silvia Silvestri, 2011. 1097. Estimating yield of food crops grown by smallholder farmers: A review in the Uganda context. Anneke Fermont and Todd Benson, 2011. 1096. Do men and women accumulate assets in different ways?: Evidence from rural Bangladesh. Agnes R. Quisumbing, 2011. 1095. Simulating the impact of climate change and adaptation strategies on farm productivity and income: A bioeconomic analysis. Ismaël Fofana, 2011. 1094. Agricultural extension services and gender equality: An institutional analysis of four districts in Ethiopia. Marc J. Cohen and Mamusha Lemma, 2011. 1093. Gendered impacts of the 2007–08 food price crisis: Evidence using panel data from rural Ethiopia. Neha Kumar and Agnes R. Quisumbing, 2011 1092. Flexible insurance for heterogeneous farmers: Results from a small-scale pilot in Ethiopia. Ruth Vargas Hill and Miguel Robles, 2011. 1091. Global and local economic impacts of climate change in Syria and options for adaptation. Clemens Breisinger, Tingju Zhu, Perrihan Al Riffai, Gerald Nelson, Richard Robertson, Jose Funes, and Dorte Verner, 2011. 1090. Insurance motives to remit evidence from a matched sample of Ethiopian internal migrants. Alan de Brauw, Valerie Mueller, and Tassew Woldehanna, 2011. 1089. Heterogeneous treatment effects of integrated soil fertility management on crop productivity: Evidence from Nigeria. Edward Kato, Ephraim Nkonya, and Frank M. Place, 2011. 1088. Adoption of weather index insurance: Learning from willingness to pay among a panel of households in rural Ethiopia. Ruth Vargas Hill, John Hoddinott, and Neha Kumar, 2011. 1087. Was the global food crisis really a crisis?: Simulations versus self-reporting. Derek Headey, 2011. 1086. The economics of desertification, land degradation, and drought: toward an integrated global assessment. Ephraim Nkonya, Nicolas Gerber, Philipp Baumgartner, Joachim von Braun, Alex De Pinto, Valerie Graw, Edward Kato, Julia Kloos, and Teresa Walter, 2011. 1085. Agriculture’s role in the Indian enigma: help or hindrance to the undernutrition crisis?. Derek Headey, Alice Chiu, and Suneetha Kadiyala, 2011. 1084. Policy options for improving regional fertilizer markets in West Africa. Balu L. Bumb, Michael E. Johnson, and Porfirio A. Fuentes, 2011. 1083. Loss prevention for hog farmers: Insurance, on-farm biosecurity practices, and vaccination. Yue-hua Zhang, Chu-Shiu Li, Chwen-Chi Liu, and Kevin Z. Chen, 2011. 1082. Institutions, geography, trade, and income per capita: A spatial-simultaneous equation approach. Guyslain Ngeleza, 2011. 1081. Assessing the livelihood impacts of a livestock disease outbreak: An alternative approach. Ekin Birol, Lydia Ndirangu, Devesh Roy, and Yorbol Yakhshilikov, 2011. 1080. Can conditional cash transfers improve maternal health and birth outcomes?: Evidence from El Salvador’s Comunidades Solidarias Rurales. Alan de Brauw and Amber Peterman, 2011. http://www.ifpri.org/publications/results/taxonomy%3A468 INTERNATIONAL FOOD POLICY RESEARCH INSTITUTE www.ifpri.org IFPRI HEADQUARTERS 2033 K Street, NW Washington, DC 20006-1002 USA Tel.: +1-202-862-5600 Fax: +1-202-467-4439 Email: ifpri@cgiar.org ACKNOWLEDGMENTS Abstract 1. Introduction 2. Conceptual Framework 3. Model and Scenarios 4. Simulation Results Scenarios in Set A Changes in Main Market Variables Trade Effects Scenarios in Set B Change in Main Market Variables Trade Effects Discussion: From Markets to Welfare Effects 5. Conclusions APPENDIX: Supplementary Tables References Recent IFPRI Discussion Papers