INTERNATIONAL PRICES AND FOOD SECURITY: AN ANALYSIS OF FOOD AND FERTILIZER PRICE TRANSMISSION IN CENTRAL AMERICA Viviana M.E. Perego, Melissa Brown, Francisco Ceballos, Manuel Hernandez, María Lucía Berrospi, Luis Dias Pereira, Salomon Salcedo, McDonald P. Benjamin, Luis Flores, Elena Mora 1 © 2024 The World Bank   1818 H Street NW, Washington, DC 20433   Telephone: 202-473-1000; Internet: www.worldbank.org  Some rights reserved. This work is a product of the staff o f the World B ank with external contributions. The fi ndings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of the Executive Directors of the World Bank or the governments they represent. Although the World Bank makes reasonable efforts to ensure all the information presented in this document is correct, its accuracy and integrity cannot be guaranteed. Use of any data or information from this document is at the user’s own risk and under no circumstances shall the World Bank or any of its partners be liable for any loss, damage, liability or expense incurred or suffered which is claimed to result from reliance on the data contained in this document. The boundaries, colors, denomination, and other information shown in any map in this work do not imply any judgment on the part of the World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions  The material in this work is subject to copyright. Because the World Bank encourages the disse- mination of its knowledge, this work may be reproduced, in whole or in part, for non-commercial purposes as long as full attribution to this work is given.  Photo credits Cover photos: Kraiwut, Serghei V - Stock Adobe Attribution Please cite this work as follows: World Bank. 2024. International prices and food security: An analysis of food and fertilizer price transmission in Central America. Perego, V.M.E., Brown, M., Ceballos, F., Hernandez, M., Berrospi, M.L., Pereira, L.D., Salcedo, S., Benjamin, M.P., Flores, L., Mora, E. World Bank: Washington, DC Any queries on rights and licenses, including subsidiary rights, should be addressed to: World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202- 522-2625; e-mail: pubrights@worldbank.org 1 Table of content ACKNOWLEDGEMENTS .................................................................................................................................................................... 5 ACRONYMS AND ABBREVIATIONS..............................................................................................................................................6 EXECUTIVE SUMMARY........................................................................................................................................................................9 I – INTRODUCTION..............................................................................................................................................................................20 II – ARE INTERNATIONAL FOOD PRICE INCREASES AND VOLATILITY TRANSMITTED TO DOMESTIC MARKETS IN CENTRAL AMERICA?...................................................................................................................................... 32 III – THE WELFARE EFFECTS OF GLOBAL FOOD AND FERTILIZER PRICE INFLATION IN CENTRAL AMERICA.............................................................................................................................................................................................49 IV – WHAT POLICY RESPONSES HAVE ACCOMPANIED RECENT FOOD PRICE CRISES?....................... 59 V – CONCLUSIONS AND RECOMMENDATIONS................................................................................................................ 76 REFERENCES.........................................................................................................................................................................................84 ANNEX 1 - ECONOMETRIC MODELLING AND RESULTS..............................................................................................96 ANNEX 2 - FINDINGS FROM AGRICULTURAL IMPACT EVALUATIONS............................................................... 126 ANNEX 3 – SETTING FOOD SECURITY WITHIN THE BROADER CONTEXT OF SUSTAINABLE DEVELOPMENT..............................................................................................................................................................................137 ANNEX 4 – HOW A CHANGING CLIMATE AND NATURAL DISASTERS ARE EXACERBATING FOOD INSECURITY IN CENTRAL AMERICA................................................................................................................................. 139 ANNEX 5 – DETAILED POLICIES ADOPTED IN CENTRAL AMERICA IN RESPONSE TO FOOD PRICE CRISES................................................................................................................................................................................................. 141 ANNEX 6 - GLOBAL TRENDS IN POLICY SUPPORTS TO ADDRESS FOOD PRICE CRISES....................147 LIST OF FIGURES Figure ES.1: People with moderate and severe food insecurity in Central American countries: 2020-2022 compared to 2014-2016...................................................................................................................................................................9 Figure ES.2: Food price inflation vs. general inflation in Central America, January-June 2022......................10 Figure ES.3: Evolution of real global food prices 1961-2023.............................................................................................. 11 Figure ES.4: Global prices of fertilizers (Urea And DAP) and energy (Natural Gas), 2019-2022..................... 11 Figure ES.5: Tariff and Non-tariff trade barriers...................................................................................................................... 16 Figure 1.1: Households Reporting Running Out Of Food In The Last 30 Days (2021)......................................... 22 Figure 1.2: Acute Food Insecurity (Panel A) And Chronic Food Security (Panel B) In Honduras (2022)... 23 Figure 1.3: Food Price Inflation In Central America (2005-2022).................................................................................. 24 Figure 1.4: Food Price Inflation in Guatemala and Honduras (February 2023)....................................................... 25 Figure 1.5: Trends In Fertilizer (Urea And DAP) And Energy (Natural Gas) Prices (January 2019 – August 2022)................................................................................................................................................... 27 Figure 1.6: Urea And DAP Farmer Prices In The Dominican Republic, Guatemala, and Honduras (January 2019 – August 2022)................................................................................................................................................... 27 2 Figure 1.7: Latin America And The Caribbean: Change In Net Fertilizer Trade, By Type Of Nutrient 2018- 2022 (Tons Per 1,000 Hectares Of Cultivated Land)..................................................................................................... 28 Figure 1.8: Local Food Prices In Tegucigalpa (2021, 2022 And Most Recent Five-Year Averages By Month).................................................................................................................................................................................................... 28 Figure 2.1: FAO Food Price Index, 1961-2023 (2014-16 = 100)........................................................................................ 33 Figure 2.2: Elasticities of Price Return Transmissions........................................................................................................ 42 Figure 2.3: Elasticities of Price Volatility Transmission.......................................................................................................44 Figure 2.4: Time-Varying Conditional Correlations For Rice In Nicaragua And Maize In Panama.................46 Figure 4.1: Support Estimates for Agriculture in Countries in Central America (US$ Million).......................... 62 Figure 4.2: GSSE Expenditures in Central America and Combined GSSE Investments by Key International Financial Institutions........................................................................................................................................... 65 Figure 4.3: GTA alerts for agricultural products and fertilizers in Central America, 2008-2022 .................. 67 Figure 4.4: Type of “red and amber alert” interventions implemented in Central America, 2008-2022 ...68 Figure 4.5: Count of NTMs on agriculture products and fertilizers in Central America, 2008-2021 ..........68 Figure 4.6: GTA red and amber alerts and NTMs.................................................................................................................. 69 Figure 5.1: Central America: Domestic Food CPI versus International Commodity Prices, 2019-2023............................................................................................................................................................................................77 Figure A1.1: Evolution Of Domestic And International Prices By Commodity And By Country....................... 112 Figure A1.1: Evolution Of Domestic And International Prices By Commodity And By Country (Cont.)....... 113 Figure A1.1: Evolution Of Domestic And International Prices By Commodity And By Country (Cont.)....... 114 Figure A1.1: Evolution Of Domestic And International Prices By Commodity And By Country (Cont.)....... 115 Figure A1.2: Evolution Of Domestic And International Price Returns By Commodity And By Country..... 116 Figure A1.2: Evolution Of Domestic And International Price Returns By Commodity And By Country (Cont.)....................................................................................................................................................................................................117 Figure A1.2: Evolution Of Domestic And International Price Returns By Commodity And By Country (Cont.)................................................................................................................................................................................................... 118 Figure A1.2: Evolution Of Domestic And International Price Returns By Commodity And By Country (Cont.)................................................................................................................................................................................................... 119 Figure A1.3: Conditional Correlations Between International And Domestic Markets...................................... 120 Figure A1.3: Conditional Correlations Between International And Domestic Markets (Cont.)...................... 122 Figure A1.3: Conditional Correlations Between International And Domestic Markets (Cont.)...................... 123 Figure A1.4: Three-Month Elasticities of Price Return Transmissions....................................................................... 124 Figure A1.5: Three-Month Elasticities of Price Volatility Transmissions................................................................... 125 Figure A2.1: Results Per Intervention Category As Classified By IEG........................................................................ 128 Figure A4.1: Changes In Reported Household Incomes, Mid-2021 Vs. Pre-Pandemic (Percent Of Households)............................................................................................................................................................140 Figure A4.2: Agricultural Losses And Damages In Guatemala Due To Tropical Storms Eta And Iota......140 Figure A6.1: Policy Actions Adopted Around The Globe To Address The 2007-08 Global Food Crisis... 148 Figure A6.2: Share Of Global Food And Feed Exports Affected By Restrictions In 2022 (Percentage By Product)............................................................................................................................................................ 148 3 Figure A6.3: Latin America And The Caribbean: Number Of Countries That Implemented Measures In Response To Rising Food Prices, By Subregion (During February To May, 2022)........................................ 149 Figure A6.4: Latin America and the Caribbean: Number of Countries that Implemented Measures in Response to Rising Fertilizer Prices, by Subregion (During February to May, 2022).................................. 150 LIST OF BOXES Box 1.1: Food Security And Food Crises...................................................................................................................................... 21 Box 1.2: Food insecurity In Honduras........................................................................................................................................... 23 Box 4.1: How restrictive has trade been in Central America during and between crises?................................ 67 LIST OF TABLES Table 1.1: Prevalence Of Food Insecurity In Central America (Percent Of Population)........................................ 22 Table 1.2: Overall Inflation Versus Food Price Inflation In Central America (January-June 2022)................. 25 Table 3.1: Compensating Variation As A Percentage Of Household Expenditure After A 10 Percent Increase In International Food Prices............................................................................................ 54 Table 3.2: Simulated Impact On Fertilizer Use, Crop Production, And Rural Incomes Of A Given Percentage Increase In International Fertilizer Prices...................................................................................... 55 Table 4.1: OECD Indicators of Producer, Consumer and General Services Supports for Agriculture.........60 Table 4.2: General Support Services Estimates in Central America, 2017................................................................ 62 Table 4.3: Assessment of the Effects of Policies Measures Implemented in Central America...................... 69 Table A1.1: Summary Statistics And Selected Normality, Autocorrelation And Stationary Tests.................100 Table A1.2: Selected Domestic And International Price Time Series And Sources of Data............................ 101 Table A1.2: Selected Domestic And International Price Time Series And Sources of Data (Cont.)........... 103 Table A1.3: Ranking Of Daily Calorie Contributions To Per Capita Diets By Country........................................104 Table A1.3: Ranking Of Daily Calorie Contributions To Per Capita Diets By Country (Cont.)........................ 105 Table A1.4: Selected Model Results And Residuals Tests...............................................................................................106 Table A1.4: Selected Model Results And Residuals Tests (Cont.)............................................................................... 107 Table A1.4: Selected Model Results And Residuals Tests (Cont.)...............................................................................108 Table A1.4: Selected Model Results And Residuals Tests (Cont.)...............................................................................109 Table A1.5: Net Food Imports As A Share Of Domestic Availability.............................................................................. 111 Table A2.1: OECD Producer And General Service Categories ..................................................................................... 133 Table A3.1: Poverty Rates In Central America.........................................................................................................................137 Table A5.1: Policy Measures Adopted In Central America Since 2008 ..................................................................... 141 4 ACKNOWLEDGEMENTS This report was prepared by a World Bank team including Viviana Maria Eugenia Perego (Task Team Leader, Senior Agriculture Economist, World Bank), Melissa Brown (Task Team Leader, Senior Agriculture Specialist, World Bank), McDonald Benjamin (Consultant, World Bank), Elena Mora (Agriculture Analyst, World Bank), and Luis Flores (Consultant, World Bank). Original analysis and inputs were provided by a team of the International Food Policy Research Institute (IFPRI) including Francisco Ceballos, Manuel Hernandez, and María Lucía Berrospi; as well as a team of the Food and Agriculture Organisation of the United Nations (FAO) including Luis Dias Pereira and Salomon Salcedo. The team is grateful for the guidance and support of Diego Arias (Practice Manager, World Bank) and David Tréguer (Program Leader, World Bank). Leah Germer (Agriculture Specialist, World Bank), Jan Nijhoff (Senior Agriculture Economist, World Bank), and Alberto Portugal (Senior Economist, World Bank) kindly served as Peer Reviewers for this report. The team acknowledges advice and brainstorming from Edward Bresnyan (Lead Agriculture Economist, World Bank), Hans Jansen (Senior Agriculture Economist, World Bank), Francisco Bueso (Senior Agriculture Specialist, World Bank), Eva Hasiner (Agriculture Economist, World Bank), Hira Channa (Agriculture Economist, World Bank), Hector Peña (Consultant, World Bank), Emilio Urteaga (Consultant, World Bank), Joaquín Arias and Eugenio Díaz-Bonilla (Inter-American Institute for Cooperation on Agriculture), Valeria Piñeiro and Nicholas Minot (IFPRI), Ricardo Rapallo, Iván León, Rodrigo Castañeda, Adoniram Sanches, Maya Takagi, Cristian Morales Opazo, Yerania Sánchez (FAO), Dalila Cervantes-Godoy (Organisation for Economic Co-operation and Development), Carolina Trivelli (Instituto de Estudios Perunanos), Paolo de Salvo and Gustavo Rondinone (Inter-American Development Bank), Ángel Murillo (Central American Bank for Economic Integration), Edith Flores (Secretariat for Central American Economic Integration), Lucrecia Rodríguez Peñalba (Secretariat of the Central American Agricultural Council), Manuel González Tejera (Ministry of Agriculture, Dominican Republic). Domestic fertilizer data for Honduras were kindly shared by the Honduras Foundation for Agricultural Research (FHIA) and the Information System for Markets of Agricultural Produce in Honduras (SIMPAH). The team recognizes with gratitude the support from the Inter-American Institute for Cooperation on Agriculture (IICA) in co-hosting a validation workshop of the preliminary results of this report at the IICA headquarters in San José, Costa Rica, in February 2024. The contributions made by participants in the workshop were essential to validate, refine, and finalize the contents of this report, and the team wishes to acknowledge all attendees for their active participation and insights. The team expresses its gratitude for logistical and administrative support from Sofia Neiva (Team Assistant, World Bank), as well as Eugenia Salazar and Viviana Chacón (IICA). Graphic design services were provided by Jaime Sosa. 5 ACRONYMS AND ABBREVIATIONS ADF Augmented Dickey-Fuller test AMIS Agricultural Market Information System CAFTA Central America Free Trade Agreement BCIE Banco Centroamericano de Integración Económica (Central American Bank for Economic Integration) BSO Buffer stock operation CC Climate Change CEPAL/ Comisión Económica para América Latina y el Caribe (Economic Commission ECLAC for Latin America and the Caribbean) CERC Contingency Emergency Response Component CIAT Centro Internacional para la Agricultural Tropical (International Center for Tropical Agriculture) CIDES Centro Internacional para el Desarrollo Sostenible (International Center for Sustainable Development) COP Conference of the Parties (United Nations Climate Change Conference) COVID-19 Coronavirus Pandemic 2019 CPI Consumer Price Index DAP Diammonium phosphate DCC Dynamic Conditional Correlation model DR-CAFTA Dominican Republic - Central America Free Trade Agreement ENAHO Encuesta Nacional de Hogares (National Household Survey, of Costa Rica) ENCOVI Encuesta Nacional sobre Condiciones de Vida (National Survey of Living Conditions, in Guatemala) EHPM Encuesta de Hogares de Propósitos Múltiples (Multipurpose Household Survey, of El Salvador) ENHOGAR Encuesta Nacional de Hogares de Propósitos Múltiples (National Multipurpose Household Survey, of the Dominican Republic) EPHPM Encuesta Permanente de Hogares de Propósitos Múltiples (Multipurpose Permanent Household Survey, of Honduras) EU European Union FAO Food and Agriculture Organization of the United Nations FAOSTAT FAO Food and Agriculture Database FEWS NET Famine Early Warning Systems Network FIDA/IFAD Fondo Internacional de Desarrollo Agrícola (International Fund for Agricultural Development) FPMA Food Price Monitoring and Analysis tool GAFSP Global Agriculture and Food Security Program GAPs Good Agricultural Practices GDP Gross Domestic Product GHG Greenhouse Gases 6 GMO Genetically Modified Organism GSSE General Services Support Estimate GV Gross Value GWP Global Warming Potential IBRD International Bank for Reconstruction and Development ICO International Coffee Organization ICT Information and Communications Technology IDB Inter-American Development Bank IE Impact Evaluation IFPRI International Food Policy Research Institute IICA Instituto Interamericano de Cooperación para la Agricultura (Inter-American Institute for Cooperation on Agriculture) IPA Innovations for Poverty Action ISFP Initiative on Soaring Food Prices LAC Latin America and the Caribbean Region LB Lyung-Box Statistic LM Engle ARCH Lagrange Multiplier Statistic MGARCH Multivariate Generalized Auto-Regressive Conditional Heteroskedasticity Model MSMEs Micro-, Small- and Medium-scale Enterprises NDC Nationally Determined Contribution NRM Natural Resources Management OEC Observatory of Economic Complexity OECD Organisation for Economic Co-operation and Development OPS/PAHO Organización Panamericana de la Salud (Pan American Health Organization) P4P Purchase for Progress Program (of the World Food Programme) PPP Public-Private Partnership PRONAF Programa Nacional de Fortalecimento da Agricultura Familiar (National Program for the Strengthening of Family Farming, in Brazil) PROSAP Programa de Servicios Agrícolas Provinciales (Provincial Agricultural Services Program, in Argentina) PSE Producer Support Estimate R&D Research and Development ROW Rest of the World SAG Secretaría de Agricultura (Ministry of Agriculture, of Honduras) SBIC Schwarz Bayesian information criterion SDGs Sustainable Development Goals SESAN Secretariat of Food and Nutritional Security SFP School Feeding Program SIECA Secretaría de Integración Económica Centroamericana (Secretariat for Central American Economic Integration) SIMMAGRO Sistema Regional de Inteligencia y Monitoreo de Mercados Agrícolas (Regional System for Intelligence on and Monitoring of Agricultural Markets) 7 SIMPAH Sistema de Información de Mercados de Productos Agrícolas de Honduras (Information System on Markets for Agricultural Products of Honduras) SIRSS Sistema Informática de Reconocimiento Mutuo de Registros Sanitarios (System for Mutual Recognition of Sanitary Records) SMS Short Message Service SPFS Special Programme for Food Security T-BEKK Bivariate T-Student Baba, Engle, Kraft, and Kroner Model tCO2e Tons of Carbon Dioxide Equivalent UN United Nations UNCTAD United Nations Conference on Trade and Development UNICEF United Nations Children’s Fund USA United States of America USAID United States Agency for International Development USDA United States Department of Agriculture VAM Vulnerability Analysis and Mapping VAR Vector Autoregressive Process VAT Value-Added Tax VEC Vector-Error Correction process PMA/WFP Programa Mundial de Alimentos (World Food Programme) WHO World Health Organization 8 EXECUTIVE SUMMARY 1. The region composed of Costa Rica, El Salvador, Guatemala, Honduras Nicaragua, Panama, and the Dominican Republic, (referred to as “Central America” in this report) is currently facing an emerging food security crisis. Central America had long experienced chronic food insecurity, but in recent years food insecurity has increased sharply, with more than 40 percent of the population of El Salvador, Guatemala and Honduras (19.4 million people) classified as moderately or severe food insecure during the COVID-19 pandemic (2020-22) compared to the previous period 2014-16 (see Figure ES.1). Between a quarter and a third of populations surveyed in mid-2021 in the seven countries analyzed in this study reported having run out of food in the previous 30 days, a large increase from the pre-pandemic situation. Figure ES.1: People with moderate and severe food insecurity in Central American countries: 2020-2022 compared to 2014-2016 70 60 59.8 56.1 54.2 52.1 50 48.4 42.7 41.6 42.2 40 39 30 27.6 23.5 24.3 21.1 22 20 16.2 16.2 16.1 13.8 14.2 13 12.2 10 7.9 1.8 2.9 0 Guatemala Honduras Dominican El Salvador LAC average Costa Rica Republic Prevalence of severe food insecurity Prevalence of moderate and severe food 2014-2016 insecurity 2014-2016 average Prevalence of severe food insecurity Prevalence of moderate and severe food 2020-2022 insecurity 2020-2022 average 2. This report explores the dynamics between domestic food security in Central America and global price inflation. The report analyzes the extent to which international food and fertilizer prices have been passed through to Central American countries, their impacts in terms of household expenditure and income, and the effectiveness of government’s domestic policy responses in the face of high food prices. The report also explores the historic evolution of agriculture and food public policies in Central America during previous instances of international food price hikes, so as to derive lessons for Central American policymakers on the adequacy of the policy ecosystem to prevent the emergence of, and respond to, food security crises. 3. This work is part of an effort to better understand the impact of the recent food price spikes on developing countries. The study draws on primary data from key informant interviews with actors in the agriculture sector; secondary data 9 sources, including international databases and national databases and surveys; and policy trackers developed by regional and international organizations. This report is intended for a broad audience of policymakers, program managers, development professionals and academics in Central America and the broader development community. A. CONTEXT 4. Countries in the Central America region that have long experienced chronic food insecurity experienced significant increases in domestic food prices in 2022, when global food prices reached historic highs in real terms. Overall domestic price inflation was fairly high in Central American countries in 2022, but food price inflation substantially exceeded overall inflation, and in various countries the gap between the two widened during the course of the year (see Figure ES.2). This mirrored major spikes in global food prices, which started to increase in 2020 and by 2022 reached their highest levels in sixty years (see Figure ES.3). The increase in food prices in Central America has also been accompanied by a doubling of fertilizer prices since 2020 (see Figure ES.4). Figure ES.2: Food price inflation vs. general inflation in Central America, January-June 2022 20 % YoY inflation 10 0 Costa Rica Dominican El Salvador Guatemala Honduras Nicaragua Panama Republic Consumer Price Index (CPI), All items Jan-22 Food and non-alcoholic beverages CPI Jan-22 Consumer Price Index (CPI), All items Jun-22 Food and non-alcoholic beverages CPI Jun-22 Source: IMF 10 Figure ES.3: Evolution of real global food prices 1961-2023 Weather, export bans, low stocks, high oil prices, high demand Bad weather, for biofuels, speculation, groing incomes in emerging 160,00 export bans economies, exchange rate fluctuations 140,00 120,00 100,00 80,00 COVID, Ukraine war 60,00 40,00 20,00 0,00 1 3 5 7 9 1 3 61 963 965 967 969 971 973 975 977 979 981 983 985 987 989 991 993 995 997 999 00 00 00 00 00 011 013 015 017 019 02 02 19 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 2 Nominal Real Note: FAO Food Price Index, 2014-16 = 100. Source: FAO, https://www.fao.org/worldfoodsituation/foodpricesin- dex/en/ Figure ES.4: Global prices of fertilizers (Urea And DAP) and energy (Natural Gas), 2019-2022 1200.00 400.00 350.00 1000.00 900.50 925.00 890.00 300.00 872.50 846.38 800.00 744.17 250.00 707.50 695.00 690.00 601.00 600.00 200.00 591.25 441.54 150.00 446.88 418.75 400.00 393.25 352.88 335.00 331.63 328.10 100.00 260.00 263.50 265.00 262.50 250.00 250.63 247.50 247.50 247.50 235.00 245.00 0 247.50 247.50 237.50 237.00 224.50 231.12 235.00 249.50 250.50 245.50 245.4 217.50 215.40214.38 214.40 200.00 201.90 202.00 50.00 0.00 0.00 01 02 03 04 05 06 07 08 09 10 M11 12 01 02 03 04 05 06 07 08 09 M10 M1 1 12 01 02 03 04 05 06 07 08 20 01 20 02 20 03 20 04 20 05 06 20 07 20 08 20 09 20 10 20 11 12 M M M M M M M M M 0M M 1M M M M M M M M 1M 1 M M M M M M M M M M 20 21 02 21 M 21 M M 20 20 02 21 021 021 021 021 21 22 M 22 022 022 022 022 022 022 M M M M M M M 20 020 020 020 020 020 020 020 02 19 2 19 19 19 20 19 20 20 19 19 19 19 19 19 19 20 20 20 2 20 2 2 2 20 20 20 2 20 20 20 2 2 2 20 2 2 2 2 2 2 2 2 2 2 2 2 DAP (S/mt) Urea (S/mt) Natural gas in dex (2010-100) Source: Own elaboration, based on data from World Bank (2022) 5. Why are international prices important to food security? If international prices are transmitted to local markets, rising food and fertilizer prices can have an impact on food security by affecting consumers’ ability to purchase food and by affecting farmers’ production decisions (including whether, how, and what to produce), although they can also potentially increase net producers’ incomes. 6. Why does volatility transmission matter for food security? Food price volatility has important implications for food security. First, a lack of stability in food prices can have detrimental effects on consumers’ ability to meet their food needs, especially low-income households who spend a large share of their income on food. Second, food price volatility affects poor, small-scale farmers who rely on food sales for a significant part of their income and possess limited capacity to time their sales. Third, especially in the absence of efficient risk- sharing mechanisms, price volatility is likely to distort input allocation, inhibit 11 agricultural investment, and reduce agricultural productivity growth, with long-run implications for farmers, the aggregate domestic food supply, and consumers. 7. The high global food price inflation since 2020 was accompanied by broader volatility (i.e., frequent and considerable fluctuations in food prices) around mean prices in international markets. Food price volatility has important implications for food security. First, a lack of stability in food prices can have detrimental effects on consumers’ ability to meet their food needs, especially low-income households who spend a large share of their income on food. Second, food price volatility affects poor, small-scale farmers who rely on food sales for a significant part of their income and possess limited capacity to time their sales. Third, especially in the absence of efficient risk-sharing mechanisms, price volatility is likely to distort input allocation, inhibit agricultural investment, and reduce agricultural productivity growth, with long-run implications for farmers, the aggregate domestic food supply, and consumers. B. ESTIMATING INTERNATIONAL PRICE PASS- THROUGHS TO CENTRAL AMERICAN MARKETS 8. Econometric analysis was used to test the extent to which international food prices and volatility were transmitted to domestic markets in Central America. The analysis assessed how increases in international prices and volatility translated into changes in domestic prices and volatility in Central America. The degree of transmission of international to local food prices varied significantly across countries and across crops, but overall it was found to be relatively low. The lowest degrees of transmission of mean prices were found to occur in the case of maize, beans, wheat, and bananas, while moderate elasticities of price transmission were found for rice, coffee, and fertilizers. Honduras exhibited some of the highest levels of price transmission across most products (including fertilizers). For example, in the case of rice in Honduras, when the international price changed by 1 percent, the local price changed by 0.67 percent after one month. Guatemala and Nicaragua had more moderate transmission for rice (0.5 per cent and 0.3 per cent, respectively), and little to no transmission was found in the cases of Costa Rica, the Dominican Republic, El Salvador, and Panama. 9. The degree of volatility transmission from international to domestic markets was found to be considerably stronger than that for price levels. Domestic price fluctuations for maize, black beans, and bananas correlated strongly with international price fluctuations in several cases, despite their low degree of mean price transmission. Costa Rica, Honduras, and Panama were characterized by some of the highest degrees of volatility transmission, with El Salvador, Guatemala, and Nicaragua showing more mixed results, depending on the commodity under consideration. In terms of the degree of co-movement between international and domestic food and fertilizer price variations over time, 12 the analysis found moderate changes in the extent of co-movement following the 2007-2008 food price crisis, the COVID-19 pandemic, and the Russia-Ukraine war, depending on the market and commodity under consideration. 10. The price elasticities estimated in this analysis could generally be regarded as low, and are in line with existing evidence for developing countries. In principle, an elasticity greater than one is generally considered relatively high, and values lower than one are deemed relatively low. The results in this analysis are thus indicative of moderate levels of international-to- domestic price transmission in Central America. These findings are consistent with similar studies for Central America, and in line with existing evidence for other developing countries globally. Price pass-through is typically stronger in countries with more modern and dynamic agrifood sectors. In the United States, for example, domestic producer prices for wheat and maize have been found more consistently associated with international prices. Within Latin America, the degree of pass-through for maize and rice prices has been shown to be larger in Brazil, Chile, and Uruguay than in other countries in the region. C. THE WELFARE EFFECTS OF GLOBAL FOOD AND FERTILIZER PRICE INFLATION IN CENTRAL AMERICA 11. To assess how international price shocks and volatility translate into welfare changes in Central America, the findings on the degree of pass- through from international to domestic food prices are used to extrapolate the likely impact of international food inflation on livelihoods and food and nutrition security. The overall impact on household welfare is estimated considering both income and substitution effects, and the balance between net food producers and consumers in the countries under study. The analysis also assesses how the ongoing fertilizer price crisis is affecting input access and incomes for farmers using granular fertilizers, addressing impacts in terms of costs increase and resulting production choices. 12. Consistent with a moderate pass-through, international food and fertilizer price inflation was found to have modest but non-negligible welfare effects on producers and consumers in Central American countries, which however do not match the magnitude of the crisis currently observed in the region. Simulation exercises were conducted to assess the impact of increases in international food prices on household welfare in Central American countries, as well as the impact of international fertilizer price increases on rural producers in Costa Rica and Honduras. Simulations of a hypothetical increase of 10 percent in the international price of different food commodities revealed small yet non- negligible effects on both urban and rural household welfare that varied across commodities. An increase in international fertilizer prices was found to have non-negligible negative effects on rural incomes, especially in Honduras. For certain crops, such as rice in Guatemala, Honduras, and Nicaragua or beans in 13 Costa Rica, the positive effect of a price increase on local households that are net sellers and/or have access to substitute food groups seems to more than compensate, on average, the negative effect on net food consumers stemming from an increase in its international price. The opposite appears to occur, although to a lower extent, with maize in Costa Rica, Guatemala, Honduras, and Nicaragua, where an increase in the international price is found to reduce household welfare. 13. The results of the fertilizer analysis suggest that increases in international prices of fertilizers result in reduced rural incomes in Costa Rica and Honduras. Crop production was found to decline by 1 to 5 percent in Honduras and Costa Rica in response to simulated international fertilizer price increases in the range of 23 to 38 percent, resulting in depressed rural incomes by 0.2 to 1.8 percent. The differences between the two countries are explained by the larger price elasticity and the larger share of crop sales to total rural incomes in Honduras, particularly among poorer rural households. 14. It is important to note that the results of the simulation stem from using the highest observed month-to-month increase in international fertilizer prices, because of the focus of the analysis on short-term adjustments. However, compounded international fertilizer price increases have been significantly larger over longer time spans: in the case of urea, the international price increased by 125 percent in 2022, while for phosphorous and potassium fertilizers, prices increased by 67 percent and 100 percent, respectively. Thus, the impact on crop production and rural incomes of the recent inflation in international fertilizer prices is likely to have been substantially larger than the simulations presented in this analysis. D. WHAT POLICY RESPONSES HAVE ACCOMPANIED RECENT FOOD PRICES? 15. The results on imperfect international to domestic price pass-through suggest that the policy ecosystem in Central American countries may be dampening the local transmission of undistorted market signals, for better or worse. The most recent era of global food price spikes that began with the 2008 global crisis has elicited a range of policy responses in Central American countries, typically an array of short-term measures. A common policy response has been to lower import tariffs on food and fertilizers, which served to dampen the pass-through effects from international to domestic food prices. Budgetary measures adopted by Central American countries in the 2008-11 period included the implementation or expansion of school feeding programs in El Salvador and Honduras, support for community seed banks in El Salvador, and reduced production taxes on grains in Honduras. Guatemala invested in policy responses 14 to reduce longer-term risks, including by providing payments for soil recovery and fertility improvements, implementing a program to boost strategic food reserves, piloting increased agricultural insurance, and encouraging the participation of family farmers as suppliers in the national school feeding program. 16. In response to the latest food and fertilizer price increases in 2022, Central American countries have adopted additional policy measures. For example, Costa Rica joined other Central American countries in reducing import tariffs (in this case for red beans). Guatemala introduced a temporary propane gas subsidy, as well as cash transfers to improve health and nutrition. El Salvador also introduced a price control on fuel in April 2022 that remains active, eliminated customs duties for food and fertilizers for one year (March 2022-March 2023), and increased minimum wages by 20 percent towards the end of 2021. Nicaragua introduced price controls on fuel and liquified propane gas. Moreover, various countries implemented food and energy price controls, food and energy subsidies, reduced tariffs on fertilizers and fertilizer subsidies for farmers, as well as support for the use of (local) organic fertilizers and for more efficient use of inputs. 17. Overall, it appears that short-term measures have largely prevailed over strategic long-term policy interventions that could increase food production and resilience, thereby enhancing food security. On the whole, the policies enacted by Central American countries in this period have been in line with practices applied in other contexts around the world, and in most cases have avoided the worst international practices, such as increased trade restrictions in response to food price spikes. Looking at the overall ecosystem of public support to agriculture, however, a number of potentially distortionary trends can be identified in Central America as governments adopted a set of interventions with mixed expected effectiveness. The policies that are currently being enacted by Central American countries can be expected to produce positive food- security effects in the short term, but would benefit from a more long-term vision. For example, despite their immediate benefits to producers, input subsidies (for example, for fuel) have proven to be inefficient and costly for governments with tight budgets. Furthermore, while trade barriers may have helped smooth price changes for producers when international prices have fallen, in the long run they are likely to keep prices higher for domestic consumers and make national (protected) production systems non-competitive at international prices (Figure ES5). Based on global evidence and research, it can be argued that most of the measures that have been adopted in Central America have been sound in terms of their short-term potential to impact positively on food security. However, there is no evidence on how lasting the effects of these policies could have been, nor on whether they have increased countries’ resilience to future crises. 15 Figure ES.5: Tariff and Non-tariff trade barriers Number of GTA alerts for agriculture products Type of red/amber alerts implemented since 2008 by and fertilizers in Central America Central American countries 16 25 14 1 20 1 2 4 2 12 1 2 1 1 10 15 3 8 10 1 6 1 15 16 3 5 12 1 4 7 5 2 0 2008-2010 2011-2013 2014-2016 2017-2019 2020-2022 0 08 09 10 1 2 3 4 5 6 7 8 9 20 1 22 20 01 01 01 01 01 01 01 01 01 02 Import tari Import tari quota Anti-dumping 20 20 2 2 2 2 2 2 2 2 2 20 2 20 Import ban Safeguard Price stabilisation Export ban Tax or social insurance relief Green Red/Amber Quantity of Non-tari barriers applied by GTM, NIC, HON, SLV 70 60 50 Guatemala 40 Honduras 30 Nicaragua 20 El Salvador 10 0 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 Source: Own elaboration, based on data from Global Trade Alert Database and UNCTAD, TRAINS NTMs database (2024). E. CONCLUSIONS AND RECOMMENDATIONS 18. This report concludes that there has been little transmission of international to domestic food prices in the Central America region, although there has been greater transmission of international price volatility to domestic markets. Possible factors which may be responsible for the low extent of price pass-through include domestic market structures and policy ecosystems, which future research should address rigorously. Consistent with low levels of pass-through of mean international price movements, the welfare losses produced by the current spikes in international food prices only account for a fraction of the food security crisis currently experienced in Central America. Domestic considerations, which may include climate-change-induced disruptions and natural disasters, high cost of inputs (like petroleum) which drive up agrologistics costs and underinvestment in modern technologies, and the toll of widespread violence and fragility, among others, are factors that can explain the recent situation in Central American domestic markets. 19. Central American countries are left with the imperative to improve the efficiency of their domestic markets, while at the same time introducing countercyclical and non-distortive public policies and programs for both food consumers and producers. Options for strengthening policy responses in efficient ways that can also prevent and mitigate future food crisis include measures involving consumer support, producer support and general services support. These measures include: 16 Consumer support » Re-assessing trade measures, including options to lower import tariffs, which can be an effective tool to decrease price distortions and reduce consumer prices. » Retaining expanded safety nets, which proved to be valuable instruments to ensure food access during the 2008, 2011, and ongoing food crises. Producer support » Reviewing direct transfers to producers via the national budget through input distribution interventions or measures with similar objectives, such as conditional cash transfers and vouchers tied to technology adoption with the aim of moving beyond short-term goals to also achieve broader medium to long term welfare impacts and larger positive externalities. » Promoting family farmers’ access to markets, finance, and services (for example via the productive alliances approach) can increase incomes and resilience for producers. General Services Support » Improving the information, research and development systems that enable the sector to innovate and build competitiveness, and to identify and address effectiveness of policies. » Comprehensive long-term investment in infrastructure, services strategy in order to generate increased productivity, higher incomes for farmers, and enhanced climate resilience. 20. Looking ahead and beyond the current food crisis, redirecting public support towards the provision of public goods and services will contribute to constructing a solid basis to attenuate the severity of impacts of future crises, or to reduce their likelihood of occurring. Moreover, while trade and recurrent expenditure measures (subsidies) have been favored in the course of the recent crises, it is essential to direct more resources to public investment in areas that can increase agricultural productivity and resilience, reduce food losses, and improve domestic markets. These investments could, for example, include improved water resources management, soil improvement, feeder roads, (cold) storage and warehousing facilities, as well as improved ports handling. Such investments can serve to mitigate not only against acute risks arising from shocks in international markets, but also the more chronic risks associated with climate change. 21. Beyond national boundaries, coordinating and promoting a regional agenda for food security can also play a major role in strengthening agricultural production and resilience, as well as food security in Central America. The Central American Integration System (SICA) offers a valuable potential platform for coordinating regional policy initiatives and maintaining 17 consistent, best practice approaches, and achieving synergies across the various countries’ initiatives, notably in areas involving general services support. Moreover, regional data coordination via SICA would allow for the improved collection of agrometeorological, climate, production, prices, and food security data to enable both policymakers and producers to improve planning and risk management. 22. Finally, intra- and inter-regional collaboration can be strengthened in order to increase transparency and efficiency in food and fertilizer trade. In particular, it will be important to continue to systems such as the Agricultural Market Information System (AMIS) to assess global food supplies and coordinate policy action in times of market uncertainty, while alternative mechanisms for financing imports, such as import facilities, can be set up for highly indebted countries.1 1  See ECLAC, FAO and WFP (2022). 18 1 INTRODUCTION Наталья Устинова - Stock Adobe 19 I – INTRODUCTION A. AN EMERGING FOOD SECURITY CRISIS IN CENTRAL AMERICA 1.1 The region composed by Costa Rica, El Salvador, Guatemala, Honduras Nicaragua, Panama, and the Dominican Republic, (referred to as “Central America” in this report) is currently facing an emerging food security crisis (Box 1.1). While Central America had long experienced chronic food insecurity, with more than 40 percent of the population of El Salvador, Guatemala and Honduras classifying as moderately or severely food insecure prior to the COVID-19 pandemic, in recent years all countries in the region, with the exception of the Dominican Republic,2 have experienced a further worsening of food security indicators (Table 1.1). An estimated 10.6 million Guatemalans (59.8 percent of the population) are food insecure, up from 6.9 million (43 percent of the population) during 2014-16.3 The prevalence of severe food insecurity in Guatemala has also increased in the country, from 16.1 percent of the population in the period 2014-2016 (2.6 million people), to 21.1 percent (3.7 million people) in the period 2020-2022.4 In Honduras, 56.1 percent of the population faced moderate or severe food insecurity in the period 2020-2022, compared to 50 percent of the population in the period 2019-2021 and 42 percent of the population in the period 2014-2016. The worsening food insecurity is aggravating the nutritional health of the Honduran population: in 2021, 44.8 percent of Hondurans (around 4.6 million people) were unable to afford a healthy diet,5 while chronic child malnutrition rates were around 29 percent in rural areas and 14.6 percent in urban areas, and reached as much as 58 percent of the population in the Dry Corridor (Box 1.2).6 In El Salvador, heavy rains in 2022 depressed grains production, leading to low grains stocks and continued upward pressure on prices until the harvests at the end of August 2023. At the same time, populations in El Salvador’s Dry Corridor are experiencing lPC Phase 3 (Crisis) levels of food insecurity. More broadly, one-quarter to one-third of the populations surveyed in mid-2021 in the seven countries analyzed in this study reported running out of food in the previous 30 days, up sharply from pre-pandemic situations (Figure 1.1). The Famine Early Warning Systems Network (FEWSNET) reported that Guatemala, Honduras, Nicaragua, and El Salvador would require external food assistance in August 2023.7 2  The decline in food insecurity in the Dominican Republic has been attributed to enhanced food production and availability. It is important to notice, however, that 52.1 percent of the population still experienced moderate or severe food insecurity from 2020 to 2022, which calls for continued efforts to ensure adequate food access for more than half of the population (https://presidencia.gob.do/noticias/fao-afirma-republica- dominicana-logra-bajar-el-indice-de-hambre-de-83-67). 3  See FAO (2022b), and World Bank (2023d). 4  The levels of mild, moderate, and severe food insecurity more than doubled in rural areas across the western region of the country, which have large indigenous populations (see Ceballos, Hernandez and Paz, 2021; Ceballos, Hernandez and Paz, 2022). Indeed Guatemala has the highest rate of malnutrition among children under five in LAC: nearly 50 percent, according to the World Food Program. For indigenous children the malnutrition rate is even higher: close to 70 percent. During the first half of 2022, Guatemala registered 16 children under the age of five who had died due to lack of food, while another 11,437 were reported as suffering from acute malnutrition, according to statistics from the Secretariat of Food and Nutritional Security, SESAN (https://elperiodico.com.gt/cultura/salud/2022/07/21/guatemala-desatiende-problemas-de-crecimiento-por- desnutricion/). 5  See FAO, IFAD, UNICEF, WFP and WHO (2023). 6  See Global Nutrition Report (2023). 7  See FEWSNET (2023). 20 Box 1.1: Food Security And Food Crises The United Nations’ Committee on World Food Security (CFS) defines food security as follows:8 “Food security exists when all people, at all times, have physical, social, and economic access to sufficient, safe, and nutritious food that meets their food preferences and dietary needs for an active and healthy life.”9 This standard definition divides food security into four further components: availability; access; utilization; and stability of the food supply: » Availability refers to the physical presence of food that could be acquired by a population, either through production or by importation, but does not imply that the population in fact has access to it. » Access refers to having the means to obtain that food, whether by purchase, or granted by an institution such as a school-feeding program or food stamps, or by local production for sale, barter, or direct consumption. » Utilization represents the extent to which a household or individual receiving the food can extract nutritional benefit from it to lead healthy lives. The obstacles to physiological utilization include low nutritional quality of the food, food safety issues such as spoilage and biological or chemical contamination, potable water quality, and compromised health, which can limit the full absorption of the food’s caloric, vitamin, and mineral content. » Stability of the food supply refers to the absence of discontinuities and disruptions in the supply chain occasioned by such things as pandemics, military hostilities, natural disasters, embargos, port strikes, terrorist attacks, shipping interruptions, or shortages elsewhere in the world caused by economic upheavals such as those triggered by warfare or cataclysmic weather events. More recently, the concept of food security has been expanded to include the notion of nutrition security. The World Bank defines nutrition security as “the ongoing access to the basic elements of good nutrition, i.e., a balanced diet, safe environment, clean water, and adequate health care (preventive and curative) for all people, and the knowledge needed to care for and ensure a healthy and active life for all household members”10. Situations of food insecurity can arise that can be chronic or acute, or both. Chronic food insecurity arises when there is persistent food insecurity in a given area, even in absence of exceptional circumstances, in terms of a persistent shortfall in the quantity (i.e. energy/caloric intake) and quality (i.e. micro-nutrients) relative to a minimum daily intake required for the population to lead a health life. This shortfall can arise because of a chronic lack of available food supplies (whether produced domestically or imported) in a given country of an adequate quality, or where there is an inability for a large share of the population to access such food supplies or to process such foods nutritionally (inadequate utilization). Acute food insecurity arises when exceptional circumstances prompt a reduction in food availability, access, utilization or stability (e.g., due to armed conflict, natural disasters, or sudden sharp increases in global food prices), resulting in an acute spike in food insecurity as populations cannot access or afford the minimum quantity and quality of daily intake required for a healthy life. 8  See Annex 3 for the relationship between food security and the Sustainable Development Goals (SDGs). 9  This consensus definition emerged from the 1996 World Food Summit. See FAO (2006). 10  This definition was included in the report Improving Nutrition Through Multisectoral Approaches. See World Bank (2013). 21 Table 1.1: Prevalence Of Food Insecurity In Central America (Percent Of Population) Prevalence of severe food Prevalence of moderate and insecurity severe food insecurity 2014-2016 2020-2022 2014-2016 2020-2022 average average average average Costa Rica 1.8 2.9 12.2 16.2 Dominican Republic 24.3 22.0 54.2 52.1 El Salvador 13.8 16.2 42.2 48.4 Guatemala 16.1 21.1 42.7 59.8 Honduras 14.2 23.5 41.6 56.1 Nicaragua N/A N/A N/A N/A Panama N/A N/A N/A N/A LAC average 7.9 13.0 27.6 39.0 Source: FAO, IFAD, UNICEF, WFP and WHO (2023) Figure 1.1: Households Reporting Running Out Of Food In The Last 30 Days (2021) 45 40 38.6 35 34.8 34.8 34.0 32.4 30.1 30 27.9 26.5 27.0 24.0 24.7 25 23.0 22.1 21.1 20.5 20.5 20 18.8 15 10 5 0 El Salvador Guatemala Costa Rica Nicaragua Panama LAC Honduras Dominican Republic Mid 2021 End 2021 Pre-pandemic (feb 2020) Source: World Bank and UNDP (2021) 1.2 Food insecurity has also affected the relatively wealthier countries in Central America. In Costa Rica, one in six persons (around 800,000 people) faced moderate or severe food insecurity during 2019-21. According to an IPC analysis, 15 percent of the Dominican population faced emergency (Phase 3 - 14 percent) or crisis (Phase 4 - 1 percent) levels of food insecurity during the period October 2022 - February 2023, while another 34 percent of the population faced Phase 2 (Stressed) conditions.11 In the case of Panama, according to IDB data, substantial progress was made in reducing undernourishment of the population over the period 2002-19, however child malnutrition has remained high, at 15.8 percent, above the LAC average of 11.8 percent, and indeed is reported as the fourth highest percentage in the region.12 11  See IPC (2023). 12  See Deza and Ruiz-Arranz (2022). 22 Box 1.2: Food insecurity In Honduras An analysis of food insecurity in 2022 in Honduras revealed a high prevalence of both chronic and acute food insecurity in the country. Chronic food insecurity at Levels 3 and 4 affected more than 1 million Hondurans in 2022, especially in the Dry Corridor.13 An additional 1.6 million Hondurans faced Level 2 (Stressed) food insecurity conditions (see Panel B of Figure 1.2). However, the global food crisis resulted in critical acute food insecurity, with almost 2 million Hondurans facing Phase 3 (Crisis) food insecurity and an additional 310,000 Hondurans facing Phase 4 (Emergency) food insecurity.14 Moreover, 3.3 million Hondurans were facing Phase 2 (Stressed) food insecurity conditions (see IPC (2023b), and Panel A of Figure 1.2). Figure 1.2: Acute Food Insecurity (Panel A) And Chronic Food Security (Panel B) In Honduras (2022) Total # (pp) Phase 1 Phase 2 Level 1 Level 2 Level 3 Level 4 # % # % Total # 9,745,149 4,133,979 42% 3,310,243 34% (pp) # % # % # % # % Phase 3 Phase 4 Phase 5 # % # % # % 4,601,030 1,930,340 42% 1,621,941 35% 804,844 17% 266,337 6% 1,990,219 20% 310,707 3% 0 0% Sources: IPC (2023b) and IPC (2023c) B. THE PRICE INCREASES BEHIND THE CRISIS 1.3 Among other factors, such as compounded climate-related events and natural disasters and the COVID-19 impacts that affected agricultural supply chains and food trade,15 the current emerging food security crisis in Central America has been attributed to sustained international food price inflation. Beginning in 2020, global food prices began to rise sharply, and by 2022 real global food prices had reached the highest levels on record in more than 60 years (3.2 percentage points higher in real terms than the previous record reached in 1974). Food supply chains around the world were disrupted significantly as a result of the COVID-19 pandemic. This was followed by the Russian invasion of Ukraine in February 2022, which contributed further to the most recent spikes in global food prices. In particular, the Ukraine war disrupted the grains, oilseeds, and vegetable oil exports of two of the world’s leading agricultural exporters, namely Ukraine and the Russian Federation. 13  See IPC (2023c). 14  See IPC (2023b). 15  See Annex 2 for a summary of the effect of climate-induced hazards on recent food insecurity in the region. 23 1.4 The sharp increases in global food prices have been felt in the Latin America and Caribbean (LAC) region in general and in Central America in particular (Figure 1.3). In September 2022, the 12-month average increase in the food price index for LAC reached 11.7 percent, compared to 7.1 percent for headline inflation.16 In February 2023, the price of white maize rose for the third consecutive month in Guatemala and Nicaragua, and for the second consecutive month in El Salvador and Honduras, to levels in all four countries that were at least 20 percent above their levels one year earlier. In February 2023, the year-on-year price difference was largest (45 percent) in El Salvador.17 Figure 1.3: Food Price Inflation In Central America (2005-2022) 30 25 20 15 10 5 0 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 -5 Costa Rica Dominican Republic El Salvador Guatemala Honduras Nicaragua Panama Source: Own elaboration, based on data from FAO’s FAOSTAT database 1.5 Overall domestic price inflation was fairly high in Central American countries in 2022, but food price inflation18 substantially exceeded overall inflation,19 and in various countries the gap between the two widened during the course of the year (Table 1.2). For example, in January 2022, El Salvador registered an overall inflation of 6.5 percent and a food inflation of 7.8 percent year on year. By June 2022, El Salvador registered an overall inflation rate of 8.9 percent and a food inflation index of 14.4 percent year on year.20 Similarly, in the Dominican Republic, food price inflation of 9.5 percent in January 2022 exceeded overall consumer price inflation of 8.7 percent by 0.8 percentage points, but by June 2022 the gap between food price inflation and overall consumer price inflation had widened to 3.7 percentage points, with food price inflation at 13 percent. In mid-June 2022, Honduras maize was added to FAO’s update of food price warnings, as the price of white maize continued to rise and was nearly 80 percent higher year-on-year.21 16  See ECLAC, FAO and WFP (2022). 17  See FAO (2023). 18  IMF data. Measured as year-on-year change in the food and non-alcoholic beverage component of a country’s Consumer Price Index (CPI). 19  IMF data. Measured as year-on-year change in the overall CPI. 20  IMF data. Accessed September 11, 2022. 21 https://www.fao.org/giews/food-prices/pricew-arnings/en/?page=1&ipp=10&no_cache=1&tx_dynalist_ pi1%5bpar%5d=YToxOntzOjE6IkwiO3M6MToiMCI7fQ== 24 Table 1.2: Overall Inflation Versus Food Price Inflation In Central America (January-June 2022) Food and non-alcoholic Consumer Price Index, All items beverages CPI January 2022 June 2022 January 2022 June 2022 Costa Rica 3.5 10.6 - - Dominican Republic 8.7 9.5 9.3 13.0 El Salvador 6.5 7.8 8.9 14.4 Guatemala 2.9 7.5 3.1 10.6 Honduras 6.2 10.2 7.5 14.7 Nicaragua 7.7 10.4 10.3 15.5 Panama 2.6 5.2 2.1 4.2 Source: IMF data, available at .imf.org/en/Data 1.6 The food price inflation has continued into 2023 and indeed worsened relative to 2022 in various countries in Central America (Figure 1.4). For example, in Costa Rica, food price inflation reached 14.5 percent in February 2023, compared to an overall inflation rate of 5.5 percent. Similarly, by February 2023, food price inflation in Guatemala had surged to 15.4 percent, compared to 10.6 percent in June 2022, and to headline inflation in February 2023 of 10 percent. In Honduras, food price inflation had surged to 17 percent as of February 2023, compared to 14.7 percent in June 2022, and to headline inflation in February 2023 of 10 percent. Figure 1.4: Food Price Inflation in Guatemala and Honduras (February 2023) Guatemala Annual Inflation, by category, February 2023 Honduras Annual Inflation, by category, February 2023 15 15 15 15 10 10 10 10 5 5 5 5 0 0 0 0 Housing & Core Transportation Headline Food Housing & Core Transportation Headline Food Utilities Utilities Target Band Target Band Source: National Sources/Haver Analytics 1.7 These escalating food prices exacerbate food and nutrition insecurity in the region, particularly amongst the most vulnerable populations. Surges in food price inflation have affected the most vulnerable households disproportionally, because they spend a larger share of their incomes on food. For example, in Honduras, moderately and extremely poor households spend 36.1 and 44.5 percent of their income on food, respectively22. Recent evidence on low and middle income countries 23 also highlights the severe impacts of food prices on child malnutrition and dietary quality, especially in rural, asset-poor households, and those lacking access to farmland. 22  See World Bank (2023d). 23  See Headey, D., & Ruel, M. (2023). The study, examining spanning 44 low- and middle-income countries, reveals that a 5 percent increase in the real price of food significantly increases the risk of wasting by 9 percent and severe wasting by 14 percent among children under five. The heightened risk of wasting, suggesting prenatal harm from declining maternal nutrition, is especially prevalent among younger children, particularly those from rural, asset-poor households, and those lacking access to farmland. The study also finds that a 5 percent real food price increase in the past 12 months predicts a 3 percent decrease in the likelihood of having an adequately diverse diet as consumption shifts to cheaper, starchy staples. 25 1.8 The increase in food prices in Central America has been accompanied by a doubling of fertilizer prices since 2020, which has held back a production response to higher food prices. Fertilizer prices have been on the increase since late 2020, as a consequence of logistics issues due to COVID-19-induced disruptions to global supply chains, high prices of energy and natural gas, and the occurrence of several tropical storms destroying natural gas factories in the Gulf of Mexico. With 67 percent of the global fertilizer output being nitrogen fertilizers,24 and as the key raw material inputs for nitrogen-based fertilizer production are air and natural gas (respectively a source of nitrogen and energy), fertilizer prices are strongly linked to this latter energy input: for instance, natural gas represents 86 percent of the production cost of ammonia and 81 percent of the production cost of urea at the manufacturing plant. Since September 2021, as the earliest signs of COVID-19 recovery led to higher inflation, higher fuel prices, and higher natural gas prices, countries in LAC have experienced price increases of up to 300 percent in urea and up to 200 percent in other popular fertilizer formulas such as diammonium phosphate (DAP), ammonium nitrate and potassium chloride compared to prices in 2019 and 2020 (Figure 1.5). Countries that introduced outright bans on (or imposed additional licensing processes for) exports of fertilizers also accentuated the increase in global fertilizer prices. Indeed, the International Food Policy Research Institute (IFPRI) estimates that such export restrictions affected approximately 20 percent of global fertilizer trade25. 1.9 The Russia-Ukraine conflict is putting further pressure on the international price of fertilizers and fuel. Russia and Belarus are respectively the largest and fourth-largest exporters of fertilizer worldwide, and most Central American countries are net fertilizer importers mostly from Russia, Belarus, and Ukraine. Starting in February 2022, the Russia-Ukraine conflict triggered a more aggressive hike in prices from one week to another. Moreover, sanctions imposed on the Russian Federation have affected oil, gas and fertilizer trade and prices, resulting in higher agricultural input prices and transportation costs.26 In the case of urea, the price increased by 125 percent in one year, while for phosphorous and potassium fertilizers, the price increased by 67 percent and 100 percent, respectively.27 The international price of oil has followed a similar same trend: in one year, the price of a barrel of West Texas Intermediate oil has increased by 64.5 percent.28 24  Source: FAOStat, Global production 2019-2020. 25  See World Bank (2023b). 26  In 2021, the Russian Federation was the world’s largest exporter of nitrogen fertilizers, the second-largest supplier of potassium and the third-largest exporter of phosphate fertilizers. Although fertilizers, as well as food, have been excluded from the sanctions imposed on the Russian Federation by the international community, logistical and financial restrictions operate as barriers to trade in these inputs, driving up prices. Added to this are the high prices of natural gas, a fundamental input in the production of nitrogen fertilizers. See ECLAC, FAO and WFP (2022). 27  https://www.dtnpf.com/agriculture/web/ag/crops/article/2022/01/19/fertilizer-prices-continue-mostly 28  https://datosmacro.expansion.com/materias-primas/petroleo-wti 26 Figure 1.5: Trends In Fertilizer (Urea And DAP) And Energy (Natural Gas) Prices (January 2019 – August 2022) 1200 400 350 1000 900.5 925 890 300 872.5 846.38 800 744.17 250 695 707.5 690 600 601 200 591.25 446.88 441.5 150 400 393.25 418.75 335 331.63 260 328.1 250.63 265 352.88 100 247.5 247.5 263.5 237.75 249.5 245 231.13 202 245 247.5 247.5 262.5 224.5 215.4 235 200 214.4 250.5 245 237 217.5 214.38 201.9 50 0 0 DAP ($/mt) Urea ($/mt) Natural gas index (2010=100) Source: Own elaboration, using data from World Bank (2022) 1.10 Increases in fertilizer prices are affecting the affordability and availability of food in Central America. The inflation in global fertilizer markets has been mirrored by sustained increases in local fertilizer prices (Figure 1.6). Moreover, increased production costs were exacerbated by higher transportation costs.29 Based on interviews conducted at the local level, the cost of granular fertilizers as a share of total production cost per hectare has increased between 45 and 66 percent for basic food staples in the region. These increases are much higher for farms far away from the central warehouses of fertilizer distribution companies. National agriculture research and innovation centers estimate that resource-constrained farmers may apply only 50 percent of the recommended amounts or only apply the recommended doses at planting. In Honduras, one out of five farmers stated they were reducing the planting area for maize, beans, sorghum or rice in 2022.30 Most of the Central American countries analyzed in this report experienced a decline in the use of chemical fertilizers (relative to their cultivated areas) higher than the median reduction for the LAC region, with the largest declines occurring in Guatemala and Costa Rica (Figure 1.7). The combination of higher production costs and reduced planting areas is resulting in a reduction in food available to subsistence producers, as well as in increased food prices in local markets (Figure 1.8). Figure 1.6: Urea And DAP Farmer Prices In The Dominican Republic, Guatemala, and Honduras (January 2019 – August 2022) Urea DAP 1600 1600 1400 1400 1200 1200 1000 1000 800 800 600 400 600 200 400 0 200 0 Urea Price DR ($/mt) Urea Price Gua ($/mt) Urea Price Hon ($/mt) DAP Price DR ($/mt) DAP Price Gua ($/mt) DAP Price Hon ($/mt) Source: Own elaboration, using data from fertilizer retailers in target countries 29  See FAO (2023) and World Bank (2023d). 30  See Unidad Técnica de Seguridad Alimentaria y Nutricional (2022). 27 Figure 1.7: Latin America And The Caribbean: Change In Net Fertilizer Trade, By Type Of Nutrient 2018-2022 (Tons Per 1,000 Hectares Of Cultivated Land) 600 400 200 0 -200 -400 -600 -800 Dominican Rep. Guatemala Argentina Nicaragua Barbados Mexico Dominica Cuba Brazil El Salvador Saint Lucia Costa Rica (Plur. State of) Peru Panama Guyana Surname (Bol. Rep. of) Colombia Antigua and Barbuda Paraguay Ecuador Honduras Bolivia Jamaica Venezuela Belize Chile Bahamas Uruguay Saint Kitts and Nevis Potassium fertilizers Nitrogenous fertilizers Phosphate fertilizers Source: ECLAC, based on the FAOSTAT data base Figure 1.8: Local Food Prices In Tegucigalpa (2021, 2022 And Most Recent Five-Year Averages By Month) White maize Red beans 45 100 40 90 35 80 HNL/5 pounds 70 HNL/5 pounds 30 25 60 20 50 40 15 30 10 20 5 10 0 0 Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul 5 - year avg 2021 2022 Source: FEWS NET Honduras Price Bulletin, Aug. 2022. C. SCOPE AND STRUCTURE OF THE REPORT 1.11 This report takes a critical look at past international food price crises to unveil the complex dynamics of the current food security crisis in Central America. The narrative presented so far depicts a clear link between international to domestic markets, whereby major events occurring on the global arena reverberate onto the local scene, with direct consequences for domestic food security. However, the degree to which Central American markets of food products are integrated into international ones is still a subject of debate. In fact, a previous World Bank study31 examined the pass-through of international to domestic prices of key agriculture commodities in Honduras and Nicaragua from 2000 to 2009, and concluded that the degree of price transmission in those two countries in that period had been limited at best. Clearly, policy recommendations on how to respond to a situation of mounting inflation vary substantially depending on whether international and domestic markets co-move, or whether instead the home market is insulated from international market signals and domestic food price movements depend more on domestic factors. If the former is true, then efforts to reduce price spikes and volatility can potentially rely more on concerted 31  See Arias and De Franco, 2011. 28 regional and international actions (e.g. through the World Trade Organization and/or other multilateral bodies), while on the domestic side strengthened safety nets can shield consumers while farmers can receive undistorted market signals and take advantage of higher international prices. Alternatively, if food price spikes and volatility are mostly attributed to domestic factors, then the most effective policy remedies would likely include domestic policy actions to stabilize or expand food production, reduce food storage and transport costs, and improve market efficiency and competition vis-à-vis existing distortions. The report thus analyzes the historic co-evolution of international and domestic food and fertilizer prices in Central American countries, to assess the expected impact of changes in the international price of given food and fertilizer products on rural livelihoods and food and nutrition security in the region. The analysis also simulates the expected welfare impacts of the current international food and fertilizer inflation, to benchmark it against the real-life crisis observed in Central American countries. 1.12 To complement the retrospective analysis on the transmission from international to domestic prices, the report also explores the historic evolution of agriculture and food public policies in Central America. The study takes a historical look at the policies that were deployed in Central America in support of agriculture and during previous instances of international food price hikes, to derive lessons on the adequacy of the policy ecosystem to respond to and prevent the emergence of food security crises. The exercise looks at the broad categories of public support to agriculture (producer support, consumer support, general support services) in Central America, and assesses policies that were enacted during recent international food price increases (2007/8, 2011/12, and during the 2020/21 COVID-19 pandemic) as well as the current one. The analysis documents both successful experiences and those that had a more limited impact, with an eye to highlighting resilient approaches focusing on productive inclusion, climate-smart agriculture, and nutrition-smart agriculture, as well as the expansion of the support provided to public goods and services. 1.13 This report is intended to reach a broad audience of policy makers, program administrators, development professionals, and academics in Central America and in the broader development community. The contents of this report are based on information derived from a wide range of primary, secondary, and private data sources. These include: » primary data from key informant interviews, including with smallholders, farmer organizations, agro-input stores, women-led farming businesses, fertilizer and agribusiness companies, logistics service providers, government agencies, and development partners; » secondary data sources, including the FAOStat database of the Food and Agriculture Organization of the United Nations (FAO), the World Bank’s “Pink Sheet” on commodity prices, global trade databases (e.g., the United Nation’s COMTRADE database, and Panjiva Inc.’s database of commercial import and export shipments), national databases, such as Honduras’ Information System on Markets for Agricultural Products (SIMPAH), national surveys (e.g., Guatemala’s National Survey of Living Conditions (ENCOVI), Honduras’ Multipurpose Permanent Household Survey (EPHPM), El Salvador’s Multipurpose Household Survey (EHPM), and the Dominican Republic’s National Multipurpose Household Survey (ENHOGAR), among others; » private sources, such as information from major fertilizer companies and retailers. In order to track policy measures, the report relied on key policy trackers, including the Inter-American Institute for Cooperation on Agriculture’s (IICA’s) Observatory of Public Policies for Agrifood 29 Systems, as well as ad hoc reviews prepared by FAO, the World Bank, the International Food Policy Research Institute (IFPRI), and the International Monetary Fund (IMF).32 1.14 In addition to this introductory chapter, the report is organized as follows. Chapter II uses time-series techniques to analyze the extent to which global increases and volatility of food and fertilizer prices have historically been transmitted from international to local markets in Central America. Chapter III estimates the  impact of the current global food and fertilizer inflation  on livelihoods and food and nutrition security in Central America region, based on the findings on the degree of pass-through from international to domestic food prices. Chapter IV reports on the nature and effectiveness of responses to past food crises by Central American governments, and comments on the expected soundness of policies and programs that are currently being enacted in the region. Finally, Chapter V concludes and proposes a set of broad policy recommendations, that can be considered and potentially operationalized at the level of individual countries. 32  See FAO, IFAD, PAHO, UNICEF and WFP (2023); IICA (2022); IFPRI (2023), IMF (2022), and World Bank (2023). 30 2 ARE INTERNATIONAL FOOD PRICE INCREASES AND VOLATILITY TRANSMITTED TO DOMESTIC MARKETS IN CENTRAL AMERICA? Paul - Stock Adobe 31 II – ARE INTERNATIONAL FOOD PRICE INCREASES AND VOLATILITY TRANSMITTED TO DOMESTIC MARKETS IN CENTRAL AMERICA? A. A RECAP OF RECENT FOOD PRICE TRENDS 2.1 This chapter examines the degree of transmission (pass-through) of average prices and of price volatility from international to domestic food and fertilizer markets in the Central America region. It uses an econometric analysis to assess the degree of price transmission especially in the short run, focusing on the domestic price response after one period (one month) to a shock (the impulse) in the international price. The analysis aims to identify the elasticity of average price transmissions from international to domestic markets, and the elasticity of price volatility transmissions from international to domestic markets, so as to assess how increases in international prices and volatility, respectively, translate into changes in domestic prices and volatility. The chapter also calculates conditional correlations (co-movements) over time between international and domestic price variations, so as to assess whether the observed trends may be linked to specific international and/or domestic events, shocks, and/or policies in place. As the analysis uses historical data spanning over the last two decades, this chapter starts with a quick overview of the major events that shaped the evolution of international prices of food in the same period. 2.2 Following a general downward trend since the mid-1970s, global food prices started increasing in real terms in 2001, with three major peaks in 2008, 2011, and 2022. International food prices had been declining in real terms for more than 25 years from 1974 to 2001, but began to increase slowly after 2001 (Figure 2.1). In 2006, the global Food Price Index measured by FAO rose by 7 percent, and in 2007 it surged by 27 percent, with a further sharp acceleration in the first half of 2008 to reach the highest levels in three decades. Overall, between January 2006 and April 2008, international prices for rice soared 257 percent, while prices for yellow corn spiked 140 percent, and those of soybean and wheat more than doubled by 126 and 117 percent, respectively.33 Although food prices decreased in 2009-10, they did not return to their pre-peak levels, and the Food Price Index surged again in real terms in 2011, reaching a level that was 4.5 percentage points above the spike in 2008.34 After almost a decade of declining prices, beginning in 2020, global food prices began to rise sharply again, and by 2022 real global food prices had reached the highest levels on record in more than 60 years (3.2 percentage points higher than the previous record reached in 1974). While the current context of international food inflation has been described in Chapter I, the following paragraphs provide a short historical account of the food 33  See World Food Programme (2008). 34  See FAO (2011). 32 price crises of 2008 and 2011, to set the stage for the price and volatility transmission analysis that follows. Figure 2.1: FAO Food Price Index, 1961-2023 (2014-16 = 100) Weather, export bans, low stocks, high oil prices, high demand Bad weather, for biofuels, speculation, groing incomes in emerging 160,00 export bans economies, exchange rate fluctuations 140,00 120,00 100,00 80,00 COVID, Ukraine war 60,00 40,00 20,00 0,00 1 3 5 7 9 1 3 61 963 965 967 969 971 973 975 977 979 981 983 985 987 989 991 993 995 997 999 00 00 00 00 00 011 013 015 017 019 02 02 19 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 2 Nominal Real Source: FAO, https://www.fao.org/worldfoodsituation/foodpricesindex/en/ A.1. The 2008 Global Food Crisis 2.3 A combination of factors drove the increases in global food prices in 2007-08. On one hand, demand for agricultural products sharply increased: this was the case for example for grains, whose demand increased both for conversion into biofuel production and as an input into livestock production (notably in China and India but also in other emerging markets), in response to growing demand for meat consumption in light of rising incomes. On another hand, persistent droughts in key producing countries resulted in severe harvests losses, which in turn contributed to the lowest levels of cereal stocks in more than 30 years. A simultaneous increase in global fuel prices aggravated the situation by driving up costs of transportation for inputs and food, as well as fertilizer production costs, while the 2008 global financial crisis that followed the collapse of Lehman Brothers sparked a flight to safety and therefore large withdrawals of international finance from emerging markets.35 2.4 Domestic food prices increased sharply in most countries during the global food crisis, the exceptions being a few large countries that were more able to insulate themselves from world markets. However, this trade insulation by large producers itself contributed to higher food prices and volatility in international markets, resulting in larger domestic price increases in smaller, import-dependent countries. For example, estimates of the impacts of food export restrictions during the 2007–08 food price crisis suggest that such policies were responsible for 40 percent of the increase in food prices.36 Thus, in many instances, unpredictable government policies were a 35  Aggregate net capital flows to selected emerging markets fell from a net inflow of US$175 billion at the start of 2008 to a net outflow of US$230 billion by the end of 2008. See the United Nations Conference on Trade and Development (UNCTAD) 2015. At the same time, there was an inflow of speculative funds into agricultural commodity futures markets as the global financial downturn weakened, which may also have contributed to the crisis. See FAO (2009). 36  See ECLAC, FAO and WFP (2022). 33 more important contributor to domestic price volatility than world market price fluctuation.37 2.5 While the high prices encouraged an expansion in cereals production by major producers, such as Brazil, China and India, cereals production actually fell across most developing countries between 2007-08.38 An important reason was that smallholder farmers find it hard to invest in expanding production, especially when price changes are unpredictable or when the farmers do not benefit completely from the transmission of higher food prices. In addition, the poorest households usually buy more food than they sell.39 In net importing countries whose local currencies were pegged to the dollar (e.g., El Salvador and Panama in Central America), or whose local currencies were relatively weak due to macro-fiscal factors, the depreciation in the dollar following the 2008 financial crisis further increased the cost of procuring food and resulted in an unfavorable environment for smaller producers to increase food production.40 The rise in fertilizer prices following fuel price increases also held back expansion of food production. 2.6 Central American countries were badly affected by the 2008 global food crisis in light of their high dependence on grain imports, ranging from 70 percent for rice to almost 100 percent for wheat.41 However, even domestically produced grains experienced sharp price increases due to delayed rains followed by storms (notably Hurricane Felix) in Central America in 2007. Thus, white maize prices increased 20 percent while prices for red beans rose 80 percent between January 2007 and April 2008. 42 These products are integral parts of the basic diet in Central America, accounting for around 47 percent of the basic food in Honduras, 50 percent in Nicaragua, 54 percent in Guatemala and as much as 75 percent in rural El Salvador.43 Since one-fifth of the populations of El Salvador, Guatemala, Honduras and Nicaragua were undernourished, i.e., had insufficient caloric intakes prior to the 2008 food crisis,44 the chronic food situation in these countries was greatly aggravated by the crisis. The World Food Programme (WFP) estimated that one in eight households in El Salvador reduced their food consumption as a result of the 2008 crisis, while more than one-quarter of household did so in Nicaragua.45 Although many households may have been able maintain food consumption levels by substituting cheaper for more expensive foods, nutritional quality would have declined. Thus, the WFP estimates that in El Salvador, 87 percent of poor households reduced either the quantity or quality or both of the foods consumed as a result of the price increases.46 The particularly large adjustment in food consumption among poorer households (who spend a larger share of their incomes on food than do rich households) reflects the differential impact of the food crisis on poorer versus wealthier households, especially in the Central America region, which the Economic Commission for Latin America and the Caribbean (ECLAC) has characterized as the most unequal in the world.47 2.7 The World Bank estimates that the number of poor in Honduras rose by 150,000 in 2008, while in Guatemala the combination of the 2008 global food price rises and a drought in 2009 (which led to famines and harvest losses of certain crops of up to 90 percent) resulted in the 37  See FAO (2011). 38  See FAO (2009). 39  See FAO (2011). 40  See FAO (2009). 41  See World Food Programme (2008). 42  See World Food Programme (2008). 43  See World Food Programme (2008). 44  According to an FAO assessment of the food conditions in these countries for the years 2002-04. See FAO (2007). 45  See World Food Programme (2008). 46  See World Food Programme (2008). 47  See Krozer (2010). 34 declaration of a state of national emergency, as over 4,000 communities were at risk of acute malnutrition.48 Moreover, according to World Bank data, 8.2 million persons were undernourished in 2008 in the seven countries analyzed in this report.49 A.2. Global Food Prices Spiked Again in 2011 2.8 Demand for grain continued to be a driving factor for international food inflation in 2011. A doubling of demand for grains (from an average of 21 million tons per year during 1995-2005 to 41 million tons during 2005-10) and the related depletion of global grains stocks was a key factor behind the renewed spike in global food prices. For example, the stocks-to-use ratio for maize fell to around 13 percent in 2011, the lowest level since the 1970s.50 Other factors include continuing structural changes, notably climate change-related heat waves in key North American and Eastern European producing countries, as well as the plateauing of yields on major crops and the conversion of croplands and water uses to urban uses, due to population pressures, all at a time of soaring demand for grains.51 2.9 The crisis was felt most severely in Eastern Africa, notably in Somalia and Ethiopia, with widespread famine, but Central America also witnessed a further spike in food insecurity. For example, in the Dominican Republic, the price of maize soared by 82 percent between April and June 2011.52 Concerns about worsening food security conditions in the region led to country-level coverages of El Salvador and Guatemala being launched under the multi-partner Integrated Food Security Phase Classification (IPC) system in 2012 – coverage of Honduras had already begun in 2010.53 The prevalence of undernourishment in Guatemala, which had declined from 18.3 percent in 2009 to 17.1 percent in 2010, rose again to 18 percent by 2012, i.e. by an additional 170,000 people, aggravating a chronic food insecurity situation reflected in a 50 percent stunting rate in 2011 for Guatemalan children under 5 years of age.54 B. ESTIMATING INTERNATIONAL PRICE PASS- THROUGHS TO CENTRAL AMERICAN MARKETS 2.10 In the context of recurring international food price crises documented above, this chapter formally investigates the extent to which food price increases and volatility have been transmitted from international to local markets in the Central America region.55 Rising food and fertilizer prices are major factors that can have an impact on food security, by increasing the cost of food and affecting farmers’ production decisions (including whether, how, and what to produce), but at the same time potentially increasing net producers’ incomes. In addition to food price 48  See World Bank (2010) and World Bank (2010b). 49  According to data from the World Bank’s World Development Indicators, see: https://databank.worldbank.org/source/health-nutrition-and- population-statistics# 50  See World Bank (2011). 51  See Brown (2011). 52  See World Bank (2011). 53  The Integrated Food Security Phase Classification (IPC) system was developed by FAO to provide a rigorous standard for classifying food insecurity contexts in five stages, to inform policymaking and responses: 1: Minimal; 2: Stressed; 3: Crisis; 4: Emergency; and 5: Famine. See https:// www.ipcinfo.org/ipcinfo-website/ipc-overview-and-classification-system/en/. Partners include UN agencies such as the WFP and UNICEF, bilateral donors such as the European Union and the United States Agency for International Development (USAID), and non-governmental organizations (NGOs) such as Oxfam, Action Against Hunger, Care, and the Famine Early Warning System Network (FEWS NET). 54  See the Global Nutrition Report (2023b). 55  The analysis in this chapter is drawn from a background paper prepared for this report by the International Food Policy Research Institute (IFPRI). See IFPRI (2023). The analysis in this chapter updates and expands on the results of Arias and De Franco (2011), who applied time series analysis to evaluate the likely results of the Central America Free Trade Agreement (CAFTA) on food markets in Honduras and Nicaragua in the early 2000s. 35 inflation, food price volatility (i.e. frequent and sizeable fluctuations in food prices) can have severe impacts on food security, because it makes food prices the unpredictable. On one hand, vulnerable households tend to react to unexpected food price shocks through a reduction in caloric intake and in dietary quality (as well as other coping strategies such as increased work by women and children, asset sale, and interruption of schooling and/or medical treatment).56 On the other hand, price uncertainty affects poor small-scale farmers who rely on food sales for a significant part of their income and possess limited capacity for timing their sales.57 In the absence of efficient risk-sharing mechanisms, volatility also inhibits physical and human investment and distorts input allocation, reducing agricultural productivity growth and creating a poverty trap for poor farmers.58 2.11 It is important to note that the transmission of average food price increases from international to local markets does not necessarily go hand in hand with the transmission of food price volatility. Although it seems reasonable to assume that markets with high transmission of prices would also be characterized by high transmission of volatility, prices from highly volatile world markets may be transmitted to local markets only with a certain delay, thus insulating local markets from international price fluctuations and resulting in local prices that exhibit much less volatility. Alternatively, even if there were no direct price transmission, it is possible for local market volatility to be determined by the degree of uncertainty among local traders, which could be influenced by frequent and sizeable fluctuations of food prices (i.e., high volatility) in world markets. B.1. Methodological Approach 2.12 The empirical strategy for the analysis relies on a multivariate generalized auto-regressive conditional heteroskedasticity (MGARCH) model.59 Full details on the study’s methodology, as well as supplementary tables and figures, are presented in Annex 1 to this report. 2.13 The starting point is the collection of country-commodity pairs of domestic and international prices. A single country-commodity pair entails two price series that span the same time period and capture the domestic and the international price of a fixed unit of weight for a given commodity, both internationally and in a specific country in Central America (for example, the international price of 1 kilogram of red beans and the domestic price of 1 kilogram of red beans in Honduras). While the series reflect actual prices collected at markets, it is standard to work with ‘price returns’, or the percentage change in each price from one period to the next to assess how changes in prices internationally relate to changes in prices domestically.60 2.14 Based on these pairs of domestic and international price returns, a two-stage approach is then followed. First, in line with the economic time-series literature, each pair of domestic- international price returns is modelled by assuming that the domestic (international) return for a given period oscillates around an average value (i.e., the constant), and depends on both domestic (international) returns and international (domestic) returns from previous months (the number of previous months depending on the number of ‘lags’ in the model), in addition to an unknown shock (the error term). The dependence of current returns on past returns is usual in time series analysis 56  Jacquet (2012). 57  According to recent work by Magrini, Morales Opazo, and Baile (2015), estimates of households’ willingness to pay to eliminate cereal price volatility in five developing countries range from only 0.06% of household income where price volatility is lo) to 1% where price volatility is higher. 58  Cf. Fafchamps (1992) and Poulton et al. (2006). 59  The analysis uses a two-stage procedure that involves modeling both a conditional mean equation and a conditional variance-covariance equation for the modelled international and domestic percentage changes in prices (price return series). See Ceballos, Hernandez, Minot and Robles (2017). 60  While the information contained in a price series is the same as that in a return series, the latter has certain statistical properties that make it appropriate to work with in the models under consideration, namely stationarity. 36 of prices, because price downturns and upturns that reflect underlying market events typically occur over periods of time, sometimes spanning several months. Thus, observing the price rise in the previous month(s) is indicative of a potential upward trend in prices. Of course, such a trend can be counteracted by an unknown shock (the error term) in either market (domestic or international). A mathematical model of the joint dynamic of domestic and international prices enables one to estimate parameters, using statistical techniques, and derive certain stylized facts about the relationship between domestic and international prices, and to formally test hypotheses.61 2.15 This process obtains a mathematical model for the so-called ‘conditional mean’ behavior of a given pair of domestic and international prices (a country-commodity pair), i.e. the relationship between mean international and domestic prices for a given commodity and country. At this stage, a time-series of residuals or shocks can also be estimated, i.e., any movements in the domestic and international prices not being explained by the mathematical model. Second, using the residuals or unexplained shocks from the conditional mean models, the corresponding domestic and international conditional variance-covariance equations are estimated.62 The conditional variance-covariance equations do not measure the relationship between mean international and domestic prices, but rather the transmission of volatility from international to domestic markets, be they immediate spillovers or persistent effects across markets. Thus, this approach enables one to characterize the magnitude and persistence of volatility transmission from international to domestic markets. It also enables the derivation of time-varying conditional correlations to assess the degree of co-movement between international and domestic price variations.63 2.16 As the main interest is in the degree of price transmission in the short run, the focus of the analysis in this chapter is on the domestic price return response after one period (one month) to a simulated shock in the international price return.64 Since the size of the shock that can be introduced into this mathematical system can be manipulated, the reaction observed in the domestic price is standardized as a fraction of the shock in the international price. This enables a comparison of the size of price shocks from international to domestic markets across different commodities, and the resulting measure is equivalent to the elasticity of price transmission from the international market to the domestic market for given commodities. The same process is also undertaken for price volatility, i.e., for variations in the domestic conditional variance after a 61  In more technical terms, the conditional mean equation for each country-commodity pair is modelled as either a vector autoregressive (VAR) or vector-error correction (VEC) process, two alternatives for jointly modelling a pair of time series that are appropriate under different underlying characteristics of the country-commodity pair under consideration. For each country-commodity pair, the selection between models and the number of lags used for each is decided based on standard tests. If a Johansen trace test finds evidence for cointegration between the domestic and international returns, a VEC model is used for that pair, and a VAR model is used otherwise. The number of lags (k) for either model is selected based on the Schwarz Bayesian information criterion (SBIC). This criterion favors more parsimonious models over more complex ones, adding a penalty based on the number of parameters estimated in the model (Schwarz, 1978; Bauldry, 2015). Table A1.4 shows the optimal number of lags, as determined by the SBIC criterion, which are either one or two in every country-commodity pair. More details on the econometric modeling are provided in Annex 1. 62  This is done using a bivariate T-BEKK model, as proposed by Engle and Kroner (1995). 63  In addition, the selection of the above models and their goodness of fit is determined following a number of statistical tests that determine whether the statistical properties of the price return series under consideration align with the features of the mathematical models chosen to represent their behavior. Thereafter, both models, i.e., conditional mean and conditional variance, are calibrated based on the price data for a given country-commodity pair. 64  It is important to distinguish between the number of lags used in the underlying econometric model depicting the international-domestic price and volatility relationships and the number of periods considered for the impulse-response function calculations. The optimal number of lags used to model the bivariate (international-domestic) conditional mean process for each country-commodity pair is determined empirically, based on the SBIC criterion. Once the above models are estimated, a hypothetical shock to the international price or volatility is introduced in the model, and the effect after one period (month) on the domestic market is calculated (the impulse-response function), providing the basis for the international-to-domestic price and price volatility elasticities. The choice of calculating elasticities after one single period allows for a direct (immediate) estimate of the international-to-domestic price and price volatility transmission that is not dependent on additional shocks to either the domestic or international market and follows what is commonly found in the price transmission literature. However, Annex 1 presents, as a robustness check, elasticities that account for the effect of an international shock on the domestic market after three months, assuming that no other shocks are introduced in the system. 37 shock in the international market to derive the elasticity of domestic price volatility with respect to international price volatility. The resulting volatility transmission indicator shows the reaction (after one period) of domestic price volatility to a shock to (i.e., a change in) the volatility of the international market. An elasticity equal to one implies that the domestic price volatility increases in the same proportion as the international price volatility in the presence of a shock in the international market. 2.17 Caveat: what this exercise does not do. The analysis in this chapter is mainly motivated by the recent important price spikes and high volatility in global agricultural markets.65 In this context, the empirical strategy focuses on measuring how increases in price and price volatility in international markets transmit to local markets in Central America. Admittedly, the underlying price transmission process may vary across different periods of time, and it might be asymmetric to price rises and declines (e.g., local prices may be sticky and not decrease after international prices start to fall). Since the exercise was not intended to uncover varying short-term price and volatility transmission across different periods, however, the analysis presented in this chapter is not formally testing for structural breaks, nor is it attempting to identify asymmetric transmission (i.e., whether positive and negative shocks in international markets affect domestic markets differently).66 The potential presence of structural breaks in the modeled series would have been of concern had the analysis been focused on price forecasting, but this was not the objective of this study. Another potential risk of not accounting for structural breaks would be spurious volatility transmission (over-rejecting the null hypothesis that there is no transmission),67 but this does not seem to be the case here as most elasticity estimates presented in the rest of this chapter are not statistically significant. The inclusion of three-month price elasticities in Annex 1 (see Figures A1.4 and A1.5) could provide some preliminary hints or suggestive evidence around the issue of price stickiness. Addressing these issues formally will be a promising avenue for future research. B.2. Data Sources 2.18 The analysis is based on a large set of monthly prices for different agricultural commodities and fertilizers. This set includes 26 domestic prices of food staples, cash crops, and agricultural inputs in Costa Rica, the Dominican Republic, El Salvador, Guatemala, Honduras, Nicaragua and Panama.68 The analysis focuses on domestic prices of key food staples (notably rice, beans, maize and wheat flour) with at least 10 years of available data that are considered important for local diets, as determined by the proportion of the daily caloric intake per person that they represent in each country.69 Bananas and coffee are also included, given their importance as cash crops and because of their employment-generating potential in the region. Finally, the analysis includes prices of ammonium-based fertilizers, with a focus on ammonia, di-ammonium 65  Cf. Glauber, Hernandez, Laborde, Martin, Rice, and Vos (2022); Headey and Ruel (2022). 66  Considering that the period of analysis encompasses the COVID-19 period, it would be informative to assess whether the pandemic induced any structural change in the degree of price and/or volatility transmission. However, given the short period after the end of the peak of COVID-19 to date, the available data is not sufficient to estimate post-COVID parameters and test whether they are different to pre-COVID ones. Identifying potential asymmetric effects would have also required estimating an alternative MGARCH model imposing additional restrictions to estimate all necessary parameters, which would have been hampered by the relatively small sample size for several of the domestic price series included in the analysis. As more post-pandemic data become available, future research could further explore price and volatility transmission in the presence of sudden price changes (breaks), using models that allow for time-varying and asymmetric price transmission (see e.g. Koop, Leon-Gonzalez and Strachan, 2011; Saikkonen and Choi, 2004; Götz, Qiu, Gervais, and Glauben, 2016). 67  See Caporin and Malik (2020). 68  See Annex 1 Table A.1. 69  Data on dietary importance are derived from FAO (2020) - see Annex 1 Table A.2. By way of example, in the case of Costa Rica the analysis considers rice, which is second in importance in terms of per capita daily caloric intake; wheat (third in importance); maize (seventh), and beans (ninth). Other foods representing an important fraction of the caloric intake in Costa Rica, such as sugar (first), milk (fourth), or soyabean and palm oil (fifth and sixth) are not included in the analysis for Costa Rica, due to a lack of domestic price data series fulfilling the requisite criteria. The minimum length for data availability was set according to the data requirements of the mathematical models used in this analysis. 38 phosphate (DAP), and urea, given their importance for smallholder farming in the region. Overall, the crops and agricultural inputs that are analyzed include: bananas, black beans, coffee, maize (white), plantains, red beans, rice, and wheat, as well as ammonia, DAP, and urea. Annex 1, Figure A1.1 provides line graphs of domestic price trends for the various commodities in Central American countries, together with the corresponding trends in international prices over time. Annex 1, Figure A1.2 does the same for returns (i.e. month-to-month changes in prices, rather than the prices themselves), which serve as the inputs for the statistical models described above. 2.19 The data for this analysis come from a range of different sources. The two main portals from which price data for food staples were retrieved are the Food Price Monitoring and Analysis (FPMA) Tool of the Food and Agriculture Organization (FAO) and the Vulnerability Analysis and Mapping (VAM) platform of the World Food Program (WFP). Additionally, some price data on other agricultural commodities were obtained from SIMMAGRO (the Regional System for Intelligence on and Monitoring of Agricultural Markets, of the Secretariat for Central American Economic Integration, SIECA); the Honduran Institute of Coffee, and the Dominican Ministry of Agriculture. Data regarding fertilizers were obtained from the Honduran Information System of Markets of Agricultural Products (SIMPAH) and the Costa Rican National Council of Production. The choice of price series for each country was ultimately determined by data availability. For each food staple under consideration, the most important market was selected if more than one market price series was available, with priority given to retail prices if more than one price type was available (e.g., wholesale and retail), given the central interest of the study on the relationship between food price fluctuations and the livelihoods of lower income households. Series with more than 20 percent of missing or repeated values were excluded. 2.20 International monthly prices of food staples were obtained from the FAO International Commodity Prices Database (FAOSTAT). In particular, the international price series for rice correspond to Long Grain 2.4 percent rice (US). This time series is relevant since Central America is one of the top five export markets for U.S. long-variety rice.70 Domestic prices for maize are paired with the price for No. 2 Yellow Maize (US), since the United States is the largest producer and exporter of corn in the world.71 In the case of bananas, the United States is the main destination for banana exports by all Central American countries, entering the country through US East Coast ports; therefore, the domestic price series are paired with the U.S. East Coast (Main Brands) price. For wheat, the price of wheat flour in Costa Rica was paired with that of Canadian wheat, since the country’s main source of wheat imports in recent years has been Canada.72 The international price for coffee used in the study is the International Coffee Organization (ICO) Composite Index, retrieved from ICO’s portal. This indicator provides an overall benchmark for the price of green coffee worldwide, since it weights prices according to the major coffee markets, exports, and varieties.73 For beans, different prices were considered, based on the type of bean that is most relevant to each country’s consumption (i.e., red or black beans), with international prices derived from each country’s main regional supplier. The international price of red beans corresponds to the unit value of Nicaraguan red bean exports from the Central Bank’s website, since this country is a surplus producer of red beans and supplies the region’s deficit.74 Domestic price series for beans in Honduras and Panama were paired with that of Nicaragua for this reason. In the case of black 70  See USA RICE (2022), and USA RICE (2018). 71  See USDA Economic Research Service (2023). 72  See OEC. (2022). 73  See International Coffee Organization (2022). 74  See FEWS NET (2022). 39 beans, the price in the United States was considered as the international price – specifically the unit value of black bean exports from the United States Department of Agriculture Foreign Agricultural Service portal – since the United States has played a key role in recent years in covering the deficit of production versus consumption in Central America.75 For fertilizers, the international price series come from Bloomberg indices for DAP (US Gulf NOLA, i.e. US Gulf New Orleans, Louisiana) used for domestic DAP prices; for urea (US Gulf NOLA Urea) for domestic urea prices, and for ammonia (US Gulf NOLA) for other ammonium-based fertilizer prices (ammonium nitrate in this analysis). The United States is one of the largest producers and exporters of fertilizers in the world, and it supplies a significant share of the fertilizers imported by Central America.76 C. HOW ARE DOMESTIC PRICES RESPONDING TO INTERNATIONAL STIMULI? 2.21 The results of the analysis described above in Section B are presented next. This section first presents the degree of transmission of prices in levels from international commodity markets to domestic food markets across countries and commodities. Thereafter, it presents the estimates for transmission of volatility from international to domestic markets. Finally, it presents the estimated conditional correlations which indicate the degree of co-movement over time between international and domestic price variations. The full model results for all series considered in the analysis are presented in Annex 1, Table A1.4. C.1. Transmission of Mean Price Movements from International to Local Markets 2.22 The elasticity of price transmission from international to domestic markets across the full array of available commodities for the seven countries ranges from slightly negative to 0.67 (Figure 2.3).77 An elasticity of 0.67 indicates that, when the international price varies by 1 percent, the domestic price varies by 0.67 percent after one month.78 Honduras is the country in the region with the highest degrees of international-domestic price transmissions across most products (particularly for coffee and rice), while Guatemala and Nicaragua evince moderate levels of pass-through for rice but not for maize, (nor for beans in the case of Guatemala). Costa Rica, El Salvador, Panama, and the Dominican Republic, in turn, are characterized by little to no response of domestic prices to variations in international prices. 2.23 Domestic maize, beans, and wheat prices show the lowest degree of pass-through from international to local prices, with maize only revealing elasticities that are not statistically significant (i.e., that are statistically indistinguishable from zero). This is arguably related in part to the low degree of dependence on imports of these foods among certain Central American countries, as shown in Annex 1 Table A1.5.79 For example, only 6 percent of beans consumed in 75  See FEWS NET (2021). 76  See IICA (2023). 77  Since the dataset for this analysis does not have price series covering comparable products across all markets, the findings in Figure 2.3 may be partly driven by data availability and should be interpreted with the corresponding caution. 78  The 1-month period considered to calculate the elasticities is standard in the literature, and was chosen because it allows for a more direct estimate of price and price volatility transmission from international to domestic markets. Considering longer-period elasticities would imply imposing the additional unrealistic assumption that no additional shocks materialize in the domestic or international market during that longer period. Still, Annex 1 presents a robustness test deriving 3-month elasticities of price and price volatility transmission (Figures A1.4 and A1.5). Broadly, the exercise delivers an overall higher degree of transmission than the 1-month estimates (with only a few exceptions). This is to be expected, since in a closed model of international-domestic price transmission, the initial shock continues to be propagated until the system reaches a new steady state. Nevertheless, in most cases the elasticities of price and price volatility transmission over one versus three months are not substantially different, especially in terms of statistical significance. 79  The degree of dependence on food imports is calculated as (M - X)/A, where M is the volume of imports, X is the volume of exports, and A is 40 Guatemala and 4 percent consumed in Honduras are imported. The low degree of pass-through for these three staples may also be due to the importance of these crops in Central American diets, so that these staples are sometimes subject to direct price controls, trade restrictions, and buffer stock mechanisms that can limit the pass-through of international shocks.80 On the other hand, the domestic price of rice is moderately associated with the variations in the international price of long grain rice in the U.S., particularly in Guatemala, Honduras, and Nicaragua. Indeed, Guatemala and Honduras import around three-quarters of the rice they consume (see Annex 1, Table A1.5). 2.24 The results regarding international to local price transmissions for cash crops are mixed. The international-domestic pass-through in the price for bananas in both Panama and the Dominican Republic is close to null, although the price transmission for coffee in Honduras is relatively high, with an elasticity of 0.63. Both bananas and coffee are export crops, so the lack of price transmission in bananas is somewhat puzzling considering that these countries export approximately one percent to two percent of the world’s total banana exports, and would be expected to be price takers.81 It could be the case that local banana markets operate differently, possibly due to quality differences in local and international bananas (with lower-quality bananas supplied to the local market). 2.25 Finally, fertilizer prices exhibit a moderate degree of price transmission. The elasticities of price transmission were found to be close to 0.3 for DAP, ammonium nitrate, and urea in Honduras, while lower elasticities (around 0.1) were found to apply in the case of ammonium nitrate and urea in Costa Rica. Even though these products are almost exclusively imported, the moderate transmission suggests that local markets in these countries have to a certain extent been insulated from international price spikes, although with the doubling of global fertilizer prices in 2022, prices have also generally increased in the region.82 For example, in Honduras, prices increased by around 50 percent in spite of the expansion of the country’s national fertilizer subsidy program during the same year. total domestic availability, defined as production plus net imports plus change in stocks. In other words, the measure roughly captures, for each food type, what fraction of the total volume ultimately consumed in-country comes from international imports. 80  See García-Jiménez and Gandlgruber (2014), and IICA (2013). 81  See International Trade Center (2021). 82  https://blogs.worldbank.org/es/latinamerica/de-la-planta-al-tenedor-una-evaluacion-rapida-de-la-crisis-de-fertilizantes-y 41 Figure 2.2: Elasticities of Price Return Transmissions83 Panel. A Costa Rica Panel B. Dominican Republic 1.0 1.0 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0.0 BLACK BEANS^ 0.0 WHEAT FLOUR^ RICE 80-20 AMMONIUN UREA 46%^** FIRST QUALITY NITRATE^** BLACK BEANS^ BANANA^ -0.2 -0.2 MAIZE RICE Panel C. El Salvador Panel D. Guatemala 1.0 1.0 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.0 BLACK BEANS^ 0.2 MAIZE^ FIRST QUALITY RICE^** 0.0 RICE^** MAIZE Panel E. Honduras Panel F. Nicaragua 1.0 1.0 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0.0 QUALITY RICE^** RED BEANS^** 0.0 DAP^** UREA 46%^ COFFEE^** AMMONIUN NITRATE^** -0.2 FIRST QUALITY MAIZE^ SECOND -0.2 MAIZE^ RICE^** Panel G. Panama 1.0 0.8 0.6 0.4 0.2 0.0 PLANTAINS RED BEANS MAIZE 83  The elasticity of price transmission is defined as the cumulative one-period domestic impulse response arising from a one-time shock in the international market (after orthogonalizing the residuals using the Cholesky decomposition) and standardizing it by the size of the shock introduced (equal to one standard deviation of the international price returns). 42 D.2 Transmission of Volatility in Prices from International to Local Markets 2.26 Figure 2.4 shows the elasticity of volatility transmission for the same countries and commodities and follows the same structure as in Figure 2.3.84 The elasticity can be interpreted as the reaction of domestic price volatility to a shock in international volatility after one month.85 An elasticity equal to one implies that, in the face of an exogenous shock that increases the volatility of a given international price by 1 percent, domestic volatility for that commodity increases by 1 percent. Annex 1, Figure A1.2 shows local and international price volatility for all countries/ commodities. 2.27 In contrast to price transmission in levels, the results show an overall higher degree of volatility transmission from international to domestic markets. In other words, recurrent price fluctuations in international markets have a considerable effect on domestic price fluctuations in Central American markets, in spite of no strong evidence of direct pass-through of price levels (as seen above). There are some outlier elasticity values – which is a usual occurrence in this type of models – particularly the 20.6 elasticity for maize in Panama.86 Such values should be taken with caution, since they could be indicative of the presence of noise in the price series or structural breaks in the international-domestic volatility relationship during the time period analyzed.87 2.28 The analysis points to a relatively high transmission of price volatility from international markets into Honduras, Guatemala, and Panama across most analyzed products, except for maize in Guatemala and red beans in Panama. The case of Honduras is notable in that it shows a high degree of transmission in terms of both price levels and price volatility. El Salvador and Nicaragua evince mixed results, with high volatility transmission in the case of rice but very little in the case of maize. Costa Rica and the Dominican Republic show very low levels of price volatility transmission overall, the only exception being black beans in the case of Costa Rica. In the case of maize, a low degree of volatility transmission is observed, except for the outlier that was mentioned for Panama. Recall that maize also had low price level transmissions across the region. In the case of beans, the results are mixed, with low volatility transmission in most cases, but moderate volatility transmission in Costa Rica and Guatemala. A higher level of price volatility transmission is observed for rice, following the same patterns observed for the transmission of rice prices in levels. 2.29 While the pass-through in banana price levels was found to be almost null for Panama, the pass-through for volatility transmission is relatively high, close to 0.6, so that shocks in international price volatility are largely transmitted to domestic volatility. This is not the case for the Dominican Republic, with an almost null international-domestic transmission of either price levels or price volatility for bananas. Finally, consistent with the findings regarding price level transmission, the volatility of domestic prices for coffee in Honduras seems to move almost 1-to-1 with that of the international price. 2.30 With regard to fertilizer prices, price volatility in Honduras seems to be highly sensitive to international turmoil in fertilizer markets (the exception being DAP, for which the volatility transmission is more moderate). Hence, local fertilizer prices in Honduras are more responsive to 84  As described in more detail in Annex 1, these elasticities are derived from the estimated T-BEKK models for each country-commodity pair. 85  See Annex 1 for three-month elasticities. 86  This issue does not seem to be related to issues with the model’s convergence. 87  Testing for potential structural breaks in the modelled series would require the implementation of additional estimations approaches that are beyond the scope of the study. 43 price fluctuations in international markets (with volatility elasticities above 1) than to price variations (with price elasticities around 0.3). By contrast, in the case of Costa Rica, there appears to be little to no transmission of international to domestic price volatility for either ammonium nitrate or urea. Figure 2.3: Elasticities of Price Volatility Transmission88 Panel. A Costa Rica Panel B. Dominican Republic 1.0 1.0 0.8 0.8 0.6 0.6 1.0 0.4 0.4 0.2 0.2 0.0 0.0 ** Y ^* N A ZE E IT ^ T E IU * AN UR % 0* * S S C AL RA N AI AN 46 AN N -2 O IT O RI U M BA FL 80 N MM BE BE Q EA T T E UR A K K EA RS C AC C RI A H FI BL BL W Panel C. El Salvador Panel D. Guatemala 1.0 1.0 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0.0 0.0 RICE^** MAIZE** BLACK FIRST QUALITY MAIZE** BEANS** RICE^** Panel E. Honduras Panel F. Nicaragua 1.0 1.0 0.8 0.8 0.6 0.6 0.4 0.4 0.2 1.2 1.0 1.9 0.2 0.0 0.0 N EE ** S ** ** Y AN TE IU AP % ZE * LIT FF RA N 46 BE AI D IT O E* A O FIRST QUALITY MAIZE** M C U EA N MM C D RI Q RE UR RICE D A N O C SE Panel G. Panama 1.0 0.8 0.6 0.4 0.2 20.6 0.0 MAIZE** PLANTAINS RED BEANS** Note: The figure is truncated to preserve scale, with outlier values indicated in bold, and ** denotes a statistically significant estimate at the 5% level. Source: Authors’ elaboration. 88  The elasticity of price volatility is defined as the percentage change in the standard deviation of the domestic price return (relative to its steady-state value), versus that of the standard deviation of the international price return. Statistical significance is approximated by a Wald test for the joint significance of α21 and g21 in the conditional variance equation specified in equation (3), where α21 shows the short-term effect of an international price shock on domestic volatility (direct spillover effect), and g21 captures the short-term effect of changes in international price volatility on domestic volatility (direct persistence effect). 44 D.3 The Level of Interdependence of International and Local Markets 2.31 In addition to examining price and volatility transmission from international to domestic markets, the level of interdependence between these markets can be assessed. These may vary over time due to specific events or shocks. This is done by deriving time-varying conditional correlations from the estimated models, which measure the degree of co-movement between month-to-month variations in international and domestic prices.89 Correlation values can range in an interval between -1 and +1, with values closer to +(-)1 indicating a strong positive(negative) co-movement between two variables, and values closer to 0 reflecting a weaker relationship. By way of example, Figure 2.5 below plots estimated time-varying conditional correlations for two country-commodity pairs, namely rice in Nicaragua and maize in Panama. The full set of conditional correlations between international and domestic markets for all country-commodity pairs is presented in Annex 1, Figure A1.3. The horizontal red lines in the figures depict the average correlation over the whole period analyzed, and the dotted lines are 95 percent confidence intervals for that average. Thus, we can conclude that rice markets in Nicaragua exhibit a modest positive conditional correlation (0.2) with international rice markets, with a high degree of confidence (95 percent) that this correlation is very close to 0.2, while maize markets in Panama exhibit a modest negative conditional correlation to international maize markets (around -0.1), again with a high degree of confidence (95 percent) that this correlation is indeed very close to -0.1. The three vertical lines in the figures mark three major events: the 2008 food price crisis (January 2008), the COVID-19 pandemic (March 2020), and the Russia-Ukraine war (February 2022), respectively. 2.32 Overall, the average correlations between international and domestic price variations align closely to the estimated elasticities of price volatility transmission discussed above.90 The magnitudes of the correlations for certain products can be associated in some cases with their significant dependence on international markets (see also Annex 1 Table A1.5), while in other cases there might be additional factors at play that limit the co-movement of international and domestic prices, such as local policies and regulations and local market structures (i.e., the degree and nature of competition between local suppliers). The correlations generally show a high month- to-month fluctuation with no clear trend over time, though with a relatively small amplitude (range) in most cases. Still, several interesting patterns emerge for specific countries and commodities. 2.33 For Costa Rica and El Salvador, the co-movement of international and domestic maize prices shows an increase after the 2007-2008 food price crisis. This is particularly the case during 2010 and 2011, when global maize prices experienced important peaks and increased variability.91 For example, while the conditional correlation of both Costa Rican and Salvadoran white maize with international white maize market prices was very modest for the overall period (under 0.1 in both cases, with a high degree of confidence), the range of values around this 0.1 average expanded in both cases to a range of -0.3 to +0.4 in 2010. Similarly, the correlation between domestic wheat prices in Costa Rica and international wheat prices increased after the 2008 food price crisis, although it decreased later, following Russia’s (a major global producer) export ban in 2010-2011 and the crisis in Crimea in 2013-2014, when wheat futures prices spiked.92 89  See Annex 1 for more details on the derivation of the conditional correlations that result from the T-BEKK model. The correlations could alternatively be derived using a Dynamic Conditional Correlation (DCC) model proposed by Engle (2002). However, when compared to the T-BEKK model, the DCC model implies less flexibility in the structure of the conditional variance-covariance matrix. See Engle (2002). 90  The close alignment is explained by the fact that the price correlation between two markets is defined as the price covariance divided by the multiplication of the variances in each market. 91  See Gardebroek, Hernandez and Robles (2015). 92  See Svanidze, Götz and Serebrennikov (2022), and Rosenberg (2014). 45 Figure 2.4: Time-Varying Conditional Correlations For Rice In Nicaragua And Maize In Panama 1 Rice first quality wholesale conditional correlations in Nicaragua Maize wholesale conditional correlations in Panama 1 5 5 0 0 -5 -5 -1 -1 2000m1 2005m1 2010m1 2015m1 2020m1 2006m1 2008m1 2010m1 2012m1 2014m1 2016m1 2018m1 2020m1 2022m1 Conditional correlations Conditional correlations Average Average Lowe 95% confidence interval/Upper 95% confidence interval Lowe 95% confidence interval/Upper 95% confidence interval Source: Authors’ elaboration. 2.34 In the case of Honduras, the 2008 food price crisis and the Russia-Ukraine war appear to have exacerbated the fluctuations in the conditional correlations of fertilizers (albeit not for most food commodities). The average conditional correlation of domestic wholesale prices for DAP in Honduras with international prices was positive at 0.2, but spiked briefly to 0.7 following the outbreak of the Russia-Ukraine war February 2022. During the COVID-19 pandemic, fluctuations were intensified for maize and fertilizers, while co-movements were reduced for red bean prices. The food price crisis and the COVID-19 pandemic appear to have exacerbated the fluctuations in the correlations between international and domestic markets in Nicaragua, while in Panama the COVID-19 pandemic was associated with a negative shift in the co-movement of maize prices. Lastly, in the Dominican Republic, the range of variations in the correlations increased over time, especially for bananas (and to a lower extent rice) following the 2008 food price crisis, although the average values of the dynamic correlations are close to zero. For Guatemala, the dynamic correlations fluctuate around zero for both black beans and maize, which are to a great extent produced locally, but for rice the average correlation is close to 0.2, indicating a certain degree of interdependence with international markets, in line with the higher volatility transmission from international to Guatemalan rice markets discussed in the section above on volatility. E. SUMMARY OF RESULTS 2.35 This chapter analyzed the degree of transmission of mean prices and of price volatility from international to domestic food and fertilizer markets in Central America. It applied advanced time-series methodologies that allowed for the estimation of price and volatility transmission from international to domestic markets while accounting for both volatility spillovers and persistence across markets. Furthermore, it calculated conditional correlations that depict the co-movement between international and domestic price variations over time. The chapter reached the following conclusions: » The degree of transmission of international to local food prices varied significantly across countries and across crops, although overall they were found to be relatively low. » The lowest degrees of transmission of mean prices were found to occur in the case of maize, beans, wheat, and bananas, while moderate elasticities of price transmission were found for rice, coffee, and fertilizers. 46 » In terms of differences across countries, Honduras exhibited some of the highest levels of price transmission across most products (including fertilizers), while Guatemala and Nicaragua had more moderate transmission for rice but not for maize, and little to no transmission was found in the cases of Costa Rica, the Dominican Republic, El Salvador, and Panama. » The degree of volatility transmission from international to domestic markets was found to be considerably stronger than that for price levels. Several domestic markets for maize, black beans, and bananas correlated strongly with international price fluctuations, despite their low degree of price transmission in levels. » Costa Rica, Honduras, and Panama were characterized by some of the highest degrees of volatility transmission among the analyzed markets, with El Salvador, Guatemala, and Nicaragua showing more mixed results, depending on the commodity under consideration. » In terms of the degree of co-movement between international and domestic food and fertilizer price variations over time, the analysis finds moderate changes in the extent of co-movement following the 2007-2008 food price crisis, the COVID-19 pandemic, and the Russia-Ukraine war, depending on the market and commodity under consideration. » The price elasticities estimated in this chapter could generally be regarded as low, and are in line with existing evidence for developing countries. In principle, an elasticity greater than one is generally considered relatively high, and values lower than one are deemed relatively low. The results in this chapter are thus indicative of moderate levels of international-to-domestic price transmission in Central America. These findings are consistent with similar studies for Central America, and in line with existing evidence for other developing countries globally.93 » Price pass-through is typically stronger in countries with more modern and dynamic agrifood sectors. In the United States, for example, domestic producer prices for wheat and maize have been found more consistently associated with international prices.94 Within Latin America, the degree of pass-through for maize and rice prices has been shown to be larger in Brazil, Chile, and Uruguay than in other countries in the region.95 93  Previous studies had found low and statistically insignificant pass-through for maize prices in El Salvador, Guatemala, Honduras, and Nicaragua between 2000 and 2013 (Arias and De Franco, 2011; Ceballos, Hernandez, Minot, and Robles, 2017). Another study on maize and rice prices in Costa Rica, El Salvador, Honduras, Nicaragua, and Panama had highlighted low transmission from the mid-1980s to the late 2000s, with the exception of strong transmission of maize in Costa Rica (Dutoit, Hernández, and Urrutia, 2010). Price elasticities generally lower than one and not always statistically significant have also been estimated over time for several countries in Latin America (Argentina, Colombia, Ecuador, Mexico, Peru) and worldwide (Egypt, Ghana, India, Indonesia, Kenya, Madagascar, Malawi, the Philippines, Thailand) for maize, wheat, and rice (cf. Baffes and Gardner, 2003; Dutoit, Hernández, and Urrutia, 2010; Ceballos, Hernandez, Minot, and Robles, 2017; Mittal, Hariharan and Subash, 2018; Nzuma and Kurui, 2021). On the other hand, while evidence on volatility transmission is much scarcer, and absent for Central American countries, one study for a sample of 12 developing countries globally finds larger elasticities in maize and wheat than the ones estimated in this report (Ceballos, Hernandez, Minot, and Robles, 2017). 94  Guo and Tanaka (2020). 95  Dutoit, Hernández, and Urrutia (2010). 47 3 THE WELFARE EFFECTS OF GLOBAL FOOD AND FERTILIZER PRICE INFLATION IN CENTRAL AMERICA chris - Stock Adobe 48 III – THE WELFARE EFFECTS OF GLOBAL FOOD AND FERTILIZER PRICE INFLATION IN CENTRAL AMERICA A. HOW DO INTERNATIONAL PRICE MOVEMENTS TRANSLATE INTO WELFARE CHANGES IN CENTRAL AMERICA? 3.1 Given the international-domestic price elasticities calculated in Chapter II, a key question is how international price shocks and volatility translate into welfare changes in Central America.96 In a scenario of mounting international prices, the question arises whether domestic producers can reap the benefits of favorable conditions on international markets, and the extent to which domestic food consumers can be shielded from food price inflation. In this chapter, the findings on the degree of pass-through from international to domestic food prices are used to extrapolate the likely impact of international food inflation on livelihoods and food and nutrition security, considering the balance between net food producers and consumers in the countries under study. Furthermore, the analysis assesses how the ongoing fertilizer price crisis is affecting input access and incomes for farmers using granular fertilizers, addressing impacts in terms of costs increase and resulting production choices. The analysis involves simulations and calculations that draw on the international-domestic price elasticities derived in Chapter II for each country, as well as on household-level data from nationally representative surveys and results other related studies available in the literature. The following sections describe the separate simulations performed to assess the impact of an increase in international food prices on household welfare and of an increase in international fertilizer prices on farmers’ income. B. THE IMPACT OF INTERNATIONAL FOOD AND FERTILIZER PRICE INCREASES ON WELFARE IN CENTRAL AMERICA 3.2 When the international price of any food commodity changes, the domestic price faced by consumers also changes, as reflected in the estimated price elasticities. Once the local price increases, this leads to two simultaneous effects on the household’s budget constraint. First, the household’s purchasing power decreases, as more income is required to maintain the same level of consumption. However, if a household also produces the food group, the price increase induces a positive impact by increasing the value of production. This combination is known as the income effect, and its direction depends on whether a household is a net food buyer or seller. A net buyer is a household whose demand for a given food product exceeds its production of that product, whereas a net seller is a household that produces more than it consumes of the given product. In 96  The analysis in this chapter is drawn from a background paper prepared for this report by the International Food Policy Research Institute (IFPRI). See IFPRI (2023). 49 addition to the income effect, households can also react by buying less of the food product that has become more expensive and substituting it with other foods that are now relatively cheaper. This is known as the substitution effect and occurs at the same time as the income effect. 3.3 The overall impact on household welfare, which accounts for both the income and the substitution effect, can be estimated with the method of compensating variations. The approach consists in calculating a hypothetical income amount that a household would need to earn (or give up) to compensate for the price increase, in order to maintain a constant level of welfare, as proxied by monthly household expenditure. This hypothetical income adjustment is known as the compensating variation.97 The procedure used to estimate compensating variations relies on the previously calculated international-domestic price elasticities, local price-demand and cross-elasticities, consumer and producer shares by food staples, and average household expenditures within each country. 98 Annex 1 provides further details on the empirical approach for the estimation. 3.4 The simulation exercise to assess the impact of increasing international prices of food on household welfare mimics the price increases observed on global markets in the year 2022. While the increase through the monthly peak of the FAO Global Food Price Index in March 2022 was 13.8 percent,99 an increase of 10 percent is used in the exercise for ease of exposition and for comparison purposes with other studies in the literature.100 The analysis further distinguishes between rural and urban areas in each country, acknowledging the differing consumption patterns and income sources and levels of rural vs. urban populations. To control for the varying balance between net food producers and consumers in the countries under study, country-area-product consumer and producer shares (i.e., the fraction of households that are net food buyers or net food sellers in each food group and rural/urban area in each country) are included in the analysis.101 Finally, the inclusion of cross price elasticities enables the analysis to account for likely substitution effects across food products after the price of a given food changes. 3.5 In addition to estimating the household welfare effects of food price increases, the analysis in this chapter also estimates the impact of an increase in international fertilizer prices on agricultural production decisions and rural incomes. The exercise aims to measure changes in fertilizer use and crop production, as well as the ultimate effect on rural incomes.102 The estimation relies on the estimated international-domestic price elasticities for ammonium nitrate, Urea N46% (a commonly applied urea fertilizer with a 46 percent content of nitrogen), and DAP, as well as elasticities of crop production to fertilizer use obtained from other related studies103 and the relative weight of crop sales in rural incomes. 97  According to Hicks (1942), the compensating variation is defined as the amount a person should receive as income to make the person as well off after a change of prices as he or she was in the initial situation. If the compensating variation is positive, the individual is worse off under the new situation, whereas if it is negative, the person is better off under it. See Hicks (1942). 98  These were obtained from Robles and Torero (2009), Robles and Torero (2010), and Robles and Keefe (2011). 99  See https://www.fao.org/worldfoodsituation/foodpricesindex/en/. 100  See, for example, Robles and Torero (2009), Robles and Torero (2010), and Robles and Keefe (2011). 101  The specific methodology followed is similar to Robles and Torero (2009), who divide household expenditures into seven food groups. Four of these groups were used for this simulation exercise: (i) rice; (ii) corn; (iii) bread and dried (where “dried” refers to pasta, cookies, cakes, wheat, oats, quinoa, and the flour of wheat, oats or quinoa); and (iv) beans, roots, vegetables, and fruits. The other three food groups – not included in this analysis – are: meat, fish, and dairy; other food; and nonfood. 102  The approach followed is similar to that applied by Hernandez and Torero (2013) to assess the impact of increased concentration in the global fertilizer industry on local markets in Africa. 103  See Robles and Torero (2009), Robles and Torero (2010), and Robles and Keefe (2011). 50 3.6 The fertilizer impact simulation considers the peak of the monthly evolution of international prices for each fertilizer during 2022. For ammonia and Urea N46%, the peak occurred in March 2022 with increases of 38.4 percent and 22.6 percent, respectively, while for DAP the peak occurred in February 2022, with an increase of 28.2 percent. The analysis considers a price elasticity of fertilizer use of -1.745 and an elasticity of crop production to fertilizer use of 0.235.104 Finally, the share of crop sales over rural income is calculated by dividing self-generated income by rural gross income, using data from national household surveys for Costa Rica and Honduras, the only two countries for which time series of domestic fertilizer prices are available.105 B.1. Caveats and Restrictive Assumptions 3.7 A first empirical limitation of the household welfare approach described above is its partial- equilibrium nature, which assumes that international prices of the various food products under study increase one at a time, while the prices of the remaining ones remain unvaried. This is a substantial simplification that departs from what was observed in practice in 2022, where most food prices were found to increase simultaneously. To the extent that simultaneous increases in multiple food prices reduce consumers’ incomes and choice set (as it is less easy to substitute an expensive food with a cheaper alternative), the results of the analysis in this chapter are best considered as a lower bound for the magnitude of the welfare effect of a generalized food price crisis on net food consumers. On net food producers, the effect of several products increasing simultaneously depends on the specific production set of the producer (whether they are producing one or more products whose prices are increasing), but again it is likely that the analysis in this chapter finds a lower bound, especially as it would be less straightforward for consumers to substitute away from producers’ food offer in the absence of cheaper products. Even though this chapter does not attempt to build a full general-equilibrium model of food price crises calibrated to Central America, the analysis in the following sections presents some simple, partial-equilibrium calculations to try and estimate the aggregate impacts of different price spikes. 3.8 Similarly, the fertilizer simulations assume that the only affected margin is the intensity of fertilizer application. That is, in the analysis performed in this chapter the only explored effect is decreased crop production due to lower fertilizer use because of higher prices, while producers cannot substitute across fertilizer products nor change or diversify the basket of crops produced. While this undoubtedly represents an over-simplification, the approach nonetheless depicts a situation not too distant from the reality of how smallholder farmers respond to shocks in the short-term. A related issue is that, presumably, increasing fertilizer prices would translate into higher food prices for (especially urban) consumers, either because producers pass on the higher input prices to consumers, or because they reduce their production and scarcer food becomes more expensive on the market. While a general equilibrium model would be more suited to provide insights to such a question, the analysis in this report is based on partial-equilibrium simulations together with additional strong assumptions. Therefore, the estimates presented in the following sections should be interpreted bearing these limitations in mind. 3.9 A second limitation of the empirical strategy in this chapter is its reliance on aggregate data and its resulting focus on the average household, which abstracts from fundamental 104  The price elasticity of fertilizer use of -1.745 is derived from the average price elasticities estimated by Bumb, Johnson, and Fuentes (2011) and by Tenkorang and Lowenberg-DeBoer (2008), while the assumed elasticity of crop production to fertilizer use of 0.235 reflects the average elasticity estimated by Gruhn, Goletti, and Roy (1995), and by Williamson (2011). 105  See https://inec.cr/estadisticas-fuentes/encuestas/encuesta-nacional-hogares for Costa Rica and https://ine.gob.hn/V3/ephpm/ for Honduras. 51 differences across households of different socio-economic statuses. Since the compensating variation is expressed as a percentage of the average household’s expenditures, having higher or lower expenditures would not affect results directly. However, poorer households statistically devote a higher share of their income on food than richer ones (a phenomenon known as “Engel’s law”), and the very composition of the food basket differs substantially across income levels (i.e., consumption of given food items represent different shares of total consumption for poorer and for richer households). A parallel caveat applies to the analysis of fertilizer impacts, which focuses on the average producer and does not distinguish between subsistence and non-subsistence producers, who would nevertheless respond quite differently to increased input costs, in light of their differing budget constraints, but also differential access to finance and information. As such, the results presented in this chapter are likely to mask important heterogeneity at the tails of the income distribution. In the absence of country-specific micro-data, which would enable an in-depth analysis of these patterns on a country-by-country basis by scale of production, the following sections present simpler estimations of the heterogeneity of fertilizer impacts to shed some preliminary light on this question. C. DOES AVERAGE WELFARE IN CENTRAL AMERICAN COUNTRIES DECREASE IN RESPONSE TO INCREASING INTERNATIONAL FOOD PRICES? 3.10 Compensating variations are computed as a percentage of average household expenditure after a 10 percent increase in the international price, for each food group, country, and rural/ urban area. In terms of interpretation, a positive compensating variation implies that the average household would be adversely impacted by an increase in international prices, because they would require a higher income to achieve the same level of welfare as enjoyed before the food price increase. Conversely, a negative compensating variation implies that households can achieve the same level of welfare with less income, indicating that the international price increase benefitted them. This could be the case, for instance, if local prices decreased and net consumer households required less income to maintain the same consumption level, or if local prices increased so that net seller households experienced an increase in the value of their food production. 3.11 The compensating variations estimated in this analysis range from -1.45 to 0.16 percent, with significant heterogeneity across countries and commodities (Table 3.1). Where the compensating variations are positive, a 10 percent increase in the international price of a given food commodity is associated with a required increase in average household income of up to 0.16 percent to maintain the same level of expenditures prior to the price increase. Where the compensating variations are negative, it means that average household income can be up to 1.45 percent lower after a 10 percent increase in the international price of a given food commodity and the household can still maintain the same level of expenditures prior to the price increase. 3.12 Costa Rica, for example, is estimated to benefit from an increase in the international price of beans, with a compensating variation for beans of approximately -0.5 percent in both urban and rural areas. In other words, an increase in international prices would induce a greater increment in the value of domestic bean production (in rural areas) or greater substitution for other food products (in urban areas). In contrast, a hypothetical increase in the international price of maize is estimated to have a slight adverse effect on the average rural household’s welfare, equivalent to a 0.16 percent reduction of the average household expenditure. 52 3.13 In the case of the Dominican Republic and Panama, the estimated average effect of an international price increase on households is adverse but very small in all cases. In Guatemala, Honduras, and Nicaragua, an increase in the international price of rice has an average positive effect on household welfare, and the estimated effects are relatively high compared to the other commodities and countries (up to a 1.45 percent estimated improvement in welfare in Nicaragua after a hypothetical 10 percent increase in the international price of rice). Since the estimated international-domestic price elasticities are all positive in these cases, the results suggest that households in rural areas in these countries may benefit overall from the higher value of their domestic rice production, while in urban areas households could be substituting rice for cheaper products that result in lower expenditures. 3.14 Intuitively, an increase in the international price of maize has an adverse effect on households in Guatemala, Honduras, and Nicaragua. The effects are generally higher in Honduras and Nicaragua, pointing to a relatively important adverse effect on local producers, combined with a limited adjustment in maize consumption: this result is however to be interpreted with caution, given that the underlying estimated international-domestic price elasticities were not statistically significant (as reported in Chapter II). An increase in the price of beans in international markets also has a negative impact on households in Honduras, while in the case of Guatemala the effect is roughly zero. 3.15 These results should be considered as a lower-bound for the real welfare effect of simultaneous price increases across food commodities. As mentioned, the analysis performed in this chapter only considers a partial-equilibrium situation in which food prices change one at a time, thereby allowing for substituting consumption between food groups, and attenuating the effect on consumers’ budget constraints. The limitation of this approach is evident. By comparison, an earlier analysis estimated a 1.7 to 7.1 percent welfare reduction (depending on the country under consideration) for Guatemala, Honduras, Nicaragua, and Peru, following a 10 percent increase in the price of all food commodities.106 106  See Robles and Torero (2010). In another study, Dong, Stewart, Dong and Hahn (2022) estimate a welfare loss of 0.25 percent among U.S. consumers faced with higher local beef and pork prices (of, respectively, 25 percent and 12 percent year-over-year) during March to December 2020, at the peak of the COVID-19 pandemic. See Dong, Stewart, Dong and Hahn (2022), as well as Robles and Torero (2010). See Robles and Torero (2009), Robles and Torero (2010), and Robles and Keefe (2011). 53 Table 3.1: Compensating Variation As A Percentage Of Household Expenditure After A 10 Percent Increase In International Food Prices Panel A. Costa Rica Panel B. Dominican Republic Panel C. Guatemala Product Urban Rural Product Urban Rural Product Urban Rural Beans -0.51% -0.57% Banana 0.05% 0.06% Beans -0.01% 0.00% Maize 0.04% 0.16% Beans 0.02% 0.03% Maize 0.01% 0.04% Rice 0.01% -0.03% Rice 0.00% 0.00% Rice -0.29% -0.21% Wheat 0.02% 0.03% Panel D. Honduras Panel E. Nicaragua Panel F. Panama Product Urban Rural Product Urban Rural Product Urban Rural Beans 0.09% 0.01% Maize 0.13% 0.01% Beans 0.04% 0.05% Maize 0.07% 0.04% Rice -1.45% -0.40% Maize 0.00% 0.00% Rice -0.51% -0.30% Plantains 0.04% 0.05% This table shows the compensating variation as a percentage of household expenditure by food staple and by urban and rural area within a country. Each row shows the effect of a 10 percent increase in the international price of a given food on the average household welfare. Positive compensating variations indicate that the ave- rage household has been negatively affected and needs an increase in their income to achieve the same (prior) level of expenditures. Negative compensating variations are interpreted as households requiring less money to reach the same level of expenditures, indicating that the international price increase has benefitted them. D. WHAT HAPPENS TO FARMERS’ INCOMES IN CENTRAL AMERICA WHEN INTERNATIONAL FERTILIZER PRICES INCREASE? 3.16 The results of the fertilizer analysis suggest that increases in international prices of fertilizers result in reduced rural incomes in Costa Rica and Honduras (Table 3.2). In Costa Rica, the simulated effects reveal that a 38.4 percent increase in the international price of ammonium nitrate results in a 3.7 percent increase in the domestic price, resulting in a 6.4 percent reduction in the use of ammonium nitrate. This in turn results in a decrease in crop production by 1.5 percent, leading to an overall decrease in rural income of 0.3 percent. In the case of Urea N46%, a 22.6 percent increase in the international price leads to a 2.9 percent increase in the domestic price, which in turn results in a 5.1 percent decrease in the use of the fertilizer, a 1.2 percent reduction in crop production, and a 0.2 percent decline in rural income (which averaged below US$400 per month per capita in July 2022). Considering that the share of crop sales to income is 25% for poor households and 17% for non-poor households in Costa Rica, this translates into a slightly higher income decrease for poor households than for non-poor households (respectively 0.4 and 0.3 percent in the case of ammonium nitrate and 0.3 and 0.2 percent in the case of Urea N46%).107 3.17 Adverse impacts on rural incomes are even higher in Honduras. For ammonium nitrate, a 38.4 percent simulated increase in the international price results in a much larger hike in the domestic price in Honduras (12.4 percent) than Costa Rica (3.7 percent). Consequently, fertilizer use is found to decrease by a larger proportion, 21.5 percent, causing a decline in crop production by 5 percent and leading to a 1.8 percent drop in rural income. In the case of DAP, the domestic price increases by almost 9 percent after a 28.2 percent increase in the international price, decreasing fertilizer use by 15.3 percent and, as a result, stifling crop production by 3.6 percent and rural income by 1.3 percent. Finally, local Urea N46% domestic prices increase by 7 percent in 107  See Trejos Solorozano (2012). 54 response to a 22.6 percent increase in the international price, leading to a decrease in fertilizer use of 12 percent, a decrease in crop production of 3 percent and a fall in rural income of 1.1 percent. Thus, rural incomes decrease by 1 to 1.8 percent across the three cases, which is a non-negligible and much larger decrease than the estimated decrease in rural incomes for Costa Rica. Moreover, rural households in the bottom half of the income distribution in Honduras have a crop sales to income ratio of 0.57, whereas the ratio is 0.3 among those in the top 10 percent.108 So this effects translate into decreases of 3 percent for ammonium nitrate and 2 percent for DAP and Urea N46% among poor households, and decreases of 2 percent for ammonium nitrate and 1 percent for DAP and Urea N46% among wealthier households. Overall, the differences between the two countries are explained by the larger price elasticity and the larger share of crop sales to total rural incomes in Honduras, particularly among poorer rural households. 3.18 Increases in fertilizer prices also have the potential to affect consumer prices. The cost of fertilizers represents, for example, about 16 percent of the total costs of producing maize (sweet corn) in Honduras.109 Assuming that such a cost structure is similar for other crops and that producers pass on the full cost increase to consumers, the above increases in domestic prices of DAP, ammonium nitrate, and Urea N46% would increase consumer food prices in, respectively, 1, 2, and 1 percent. Since small producers are usually price takers, they would not be able to pass on the full increase in their costs, but an increase in fertilizer prices would normally also result in an increase in the total production cost of food crops, increasing their price. Table 3.2: Simulated Impact On Fertilizer Use, Crop Production, And Rural Incomes Of A Given Percentage Increase In International Fertilizer Prices Panel A. Costa Rica Ammonium Urea Nitrate N46% International price increase 38.4% 22.6% Prices Elasticity of domestic to international prices 0.10 0.13 Change in prices 3.68% 2.94% Fertilizer use Elasticity of fertilizer use to prices -1.75 -1.75 Change in fertilizer use -6.43% -5.12% Crop production Elasticity of production to fertilizer use 0.24 0.24 Change in crop production -1.51% -1.20% Rural income Share of crop sales to rural income 19.00% 19.00% Change in rural income -0.29% -0.23% 108  See Srinivasan and Rodriguez (2016). 109  See RED-Programa de Diversificacion Economica Rural (2005). 55 Panel B. Honduras Ammonium DAP Urea 46% Nitrate International price increase 38.4% 28.2% 22.6% Prices Elasticity of domestic to international 0.32 0.31 0.32 prices Change in prices 12.37% 8.80% 7.15% Fertilizer use Elasticity of fertilizer use to prices -1.75 -1.75 -1.62 Change in fertilizer use -21.58% -15.36% -12.47% Crop production Elasticity of production to fertilizer use 0.24 0.25 0.25 Change in crop production -5.07% -3.61% -2.93% Rural income Share of crop sales to rural income 36.20% 36.20% 36.20% Change in rural income -1.84% -1.31% -1.06% Note: The elasticities of international to domestic prices correspond to those estimated in Chapter III. The elasticities of fertilizer use to prices and crop production to fertilizer use are based on the average across the studies of Gruhn et al. (1995), Tenkorang and Lowenberg-DeBoer (2008), Williamson (2011), and Bumb et al. (2011). The share of crop sales to rural income is obtained from national household surveys for Honduras and Costa Rica. E. SUMMARY OF RESULTS 3.19 This chapter presented simulation exercises conducted to assess the impact of increases in international food prices on household welfare in Central American countries, as well as the impact of international fertilizer price increases on rural producers in Costa Rica and Honduras. Simulations of a hypothetical increase of 10 percent in the international price of different food commodities revealed non-negligible effects on both urban and rural household welfare that varied across commodities. An increase in international fertilizer prices was similarly found to have non-negligible negative effects on rural incomes, especially in Honduras. » For certain crops, such as rice in Guatemala, Honduras, and Nicaragua or beans in Costa Rica, the positive effect of a price increase on local households that are net sellers and/or have access to substitute food groups seems to more than compensate, on average, the negative effect on consumers stemming from an increase in its international price. » The opposite appears to occur, although to a lower extent, with maize in Costa Rica, Guatemala, Honduras, and Nicaragua, where an increase in the international price is found to reduce household welfare. » Crop production was found to decline by 1 to 5 percent in Honduras and Costa Rica in response to simulated international fertilizer price increases in the range of 23 to 38 percent, resulting in depressed rural incomes by 0.2 to 1.8 percent. » It is important to note that this latter result stems from using the highest observed month-to- month increase in international fertilizer prices, because of the focus of the analysis on short- 56 term adjustments. However, compounded international fertilizer price increases have been significantly larger over longer time spans: in the case of urea, the international price increased by 125 percent in 2022, while for phosphorous and potassium fertilizers, prices increased by 67 percent and 100 percent, respectively. Thus, the impact on crop production and rural incomes of the recent inflation in international fertilizer prices is likely to have been substantially larger than the simulations presented in this chapter. » Even though non-marginal, the effects simulated in this chapter do not capture the full extent of the real food security crisis currently faced by countries in Central America, in line with the limited fraction of domestic inflation explained by international food and fertilizer price increases. 57 4 WHAT POLICY RESPONSES HAVE ACCOMPANIED RECENT FOOD PRICE CRISES? Ingo Bartussek - Stock Adobe 58 IV – WHAT POLICY RESPONSES HAVE ACCOMPANIED RECENT FOOD PRICE CRISES? A. A FRAMEWORK FOR ASSESSING PUBLIC INTERVENTIONS IN THE FACE OF FOOD PRICE CRISES 4.1 The results on imperfect international to domestic price pass-through suggest that the policy ecosystem in Central American countries may be dampening the local transmission of undistorted market signals, for better or worse. While shielding the home economy from the wildest events on international markets may seem like a desirable outcome, such result often comes at the cost of corresponding welfare losses on the part of consumers or producers who are not able to reap the benefits of favorable market conditions when they arise. In general terms, policies that result in a distortion of market prices for products and inputs create a framework in which demand and supply are not guided by market signals: as such, these policies do not encourage competition in the market and can perpetuate inefficient and unsustainable production or consumption practices, to the detriment of general welfare. 4.2 The most widely accepted method for measuring and assessing public support to agriculture is the one developed by the Organisation for Economic Cooperation and Development (OECD).110 The OECD indicators reflect the provision of support, or the level of effort made by governments, implicit in their agricultural policies. In the OECD methodology, “support” is understood as gross transfers to agriculture from consumers and taxpayers, arising from government policies that support agriculture. This includes both budgetary expenditures and other estimated transfers which do not require actual monetary disbursements (e.g. credit concessions). 4.3 The OECD indicators can be distinguished according to the recipient of the transfer, the unit of measurement in which they are expressed, and the type of aggregation. When analyzed together, these indicator provide a comprehensive picture of the level and composition of support. The three broad categories of agriculture support measures identified by the OECD are: (i) Producer Support (PSE); (ii) Consumer Support (CSE); (iii) General Services Support (GSSE); and (iv) Total Support Estimate (TSE), which is the sum of the previous three. Table 4.1 provides a definition and examples of policies for each type of support. In most cases, producer and consumer support comprise short-term measures that, if sustained in the long-term, will keep prices distorted or place a burden on government budgets. General services support normally comprises investments with long-term results that do not entail transfers to specific beneficiaries, and have been broadly demonstrated to represent more efficient and cost-effective public tools to promote agriculture development. 110  The OECD methodology is currently used by 39 OECD members and emerging economies, including Mexico, Brazil and Chile. For other Latin American countries, agriculture support estimates are currently collected by the Inter-American Development Bank (IDB) through its Agrimonitor database. See https://agrimonitor.iadb.org/en. For a recent analysis of Central American countries see Flores Agreda et al. (2020). 59 Table 4.1: OECD Indicators of Producer, Consumer and General Services Supports for Agriculture Indicator Definition SUPPORT TO PRODUCERS Producer Support The annual monetary value of gross transfers from consumers and Estimate (PSE) taxpayers to agricultural producers, measured at the farm gate level, arising from policy measures that support agriculture, regardless of their nature, objectives or impacts on farm production or income (e.g. market price supports or direct budget transfers). Examples of » Reduction of export taxes and increase in import taxes on agricultural Producer Support products » Import bans on agricultural products » Subsidization of production – with required production, or without (area payments). » Subsidization of inputs and equipment Percentage PSE PSE as a share of gross farm receipts (including support). SUPPORT TO CONSUMERS Consumer Support The annual monetary value of gross transfers from (to) consumers of Estimate (CSE) agricultural commodities, measured at the farm gate level, arising from policy measures that support agriculture, regardless of their nature, objectives or impacts on consumption of farm products. Examples of » Increase in export taxes and decrease in import taxes on food Consumer Support products » Export bans » Value-Added Tax reductions » Food vouchers or other support to buy food » Price containment measures Percentage CSE CSE as a share of consumption expenditure (measured at the farm gate) net of taxpayer transfers to consumers. SUPPORT TO GENERAL SERVICES FOR AGRICULTURE General Services The annual monetary value of gross transfers arising from policy Support Estimate measures that create enabling conditions for the primary agricultural (GSSE) sector through development of private or public services, and through institutions and infrastructures regardless of their objectives and impacts on farm production and income, or consumption of farm products. This includes public expenditures on innovation, inspection, infrastructure, marketing and storage service. Moreover, it includes policies in these areas for which primary agriculture is the main beneficiary, but it does not include any payments to individual producers. GSSE transfers do not directly alter producer receipts, costs or consumption expenditures. 60 Indicator Definition Examples of General » Research and innovation support Services Support » Animal health, biosecurity and food safety measures » Infrastructure (roads, railways, ports, information systems) » Education, professional training and extension services » Public warehouses » Subsidies for long-term investment (e.g. matching grants to reduce risk associated with investing in processing units) Percentage GSSE GSSE as a share of Total Support Estimate (TSE). TOTAL SUPPORT TO AGRICULTURE Total Support The annual monetary value of all gross transfers from taxpayers and Estimate (TSE) consumers arising from policy measures that support agriculture, net of associated budgetary receipts, regardless of their objectives and impacts on farm production and income, or consumption of farm products. Percentage TSE TSE as a share of GDP. Sources: OECD (2016) and Agrimonitor: https://agrimonitor.iadb.org/en/results-by-indicator?tab=agriculture 4.4 When reviewing the overall value of producer, consumer and general services support, it appears that most support to agriculture in Central America has been protecting domestic producers at a cost to consumers and long-term efficiency. This is reflected in overall negative CSE time series for six Central American countries that mirror the positive PSE series in the first two charts in Figure 4.1 below. These patterns contrast starkly with the experience of the United States (and other countries with larger fiscal capacity), where CSE and PSE converged until 2007 and remained relatively constant thereafter (see the US chart in Figure 4.1). In all Central American countries, moreover, most of the PSE allocation is overwhelmingly directed to market price support (e.g. through border measures), contrary to what happens in OECD countries where market price support represents roughly a third of PSE (see last panel of Figure 4.1). At the same time, just eyeballing the scale of the vertical axes in the three top panels of Figure 4.1 shows how investment in public goods and services has historically represented but a fraction of policies directly impacting producers and consumers. While there is significant heterogeneity in the portion of TSE accounted for by GSSE across countries, the level of GSSE as a percentage of agricultural gross production value remains low (between 0.76 and 1.94 percent in 2017), compared for instance to the United States where general support accounts for more than 3 percent of agricultural gross production value (see Table 4.2) 4.5 A shift in the structure of agriculture support policies in Central America could enable efficiency gains. The efficiency of agricultural policy instruments depends on the degree to which these respond to economic optimization principles: allocations are efficient if they induce the maximum economic benefit attainable under given market conditions. The current composition of agricultural support in Central American countries as portrayed in this section displays broad efficiency gaps that are bound to impose a significant burden on taxpayers. Recurrent PSE 61 spending on subsidies for private goods, with a heavy reliance on market price support, competes directly with long-term investment allocations for public goods, and is set to crowd out private investment, reduce investment productivity, and lead to inappropriate resource use. In particular, market price support measures de-link supply from market signals, and are as such considered highly distorting, as they hamper market competition and perpetuate inefficient and unsustainable production practices, resulting in negative effects on consumers. Empirical evidence suggests that both the level and composition of rural public spending are important determinants of agricultural performance in terms of productivity, competitiveness, and sustainability, and that a redirection of total support from private subsidies towards public goods and services can promote large efficiency gains.111 Figure 4.1: Support Estimates for Agriculture in Countries in Central America (US$ Million) PSE CSE GSSE 3000 0 350 Panama 2500 2010 2017 2015 2009 2016 2014 2013 2012 2011 -1000 250 Nicaragua 2000 1500 Honduras 150 1000 -2000 Guatemala 500 50 -3000 0 El Salvador -50 2010 2016 2017 2014 2009 2009 2013 2015 2012 2011 2010 2014 2013 2016 2012 2015 2017 2011 -4000 Costa Rica United States of America MPS/PSE OECD 80,000 Panama 60,000 Nicaragua 40,000 Honduras 20,000 Guatemala El Salvador 2009 2019 2005 2003 2015 2007 2013 2017 2011 2001 2021 Dom. Rep. CSE GSSE PSE 0% 50% 100% Costa Rica Note: PSE: Private Support Estimate; CSE: Consumer Support Estimate; GSSE: General Support Services Es- timate; MPS: Market Price Support. Source: Own elaboration, based on Agrimonitor data from: https://agrimo- nitor.iadb.org/es/resultados-por-paises?country=pe&tab=agriculture; MPS/PSE is calculated on last available year (2017 for Honduras and El Salvador, 2018 for Guatemala, 2019 for Dominican Republic and Panama, 2021 for Costa Rica, 2022 for OECD countries). Table 4.2: General Support Services Estimates in Central America, 2017   GSSE/TSE GSSE/Ag. GPV Costa Rica 26.48% 1.26% Dominican Republic 13.63% 0.90% El Salvador 4.37% 1.53% Guatemala 29.68% 0.48% Honduras 25.50% 1.94% Nicaragua 7.38% 0.77% 111  Anríquez et al. (2016), using data on agricultural public spending for nineteen Latin American countries during 1985-2001, show that, although the level of agricultural public spending matters, it is changes in composition, from private to public (and mixed) goods, that explain the largest variation in productivity. A 10 percentage point redistribution of agricultural expenditures from private subsidies to public goods, all else equal, increases agricultural value added per capita by 5 percent. On the other hand, if the structure of spending is maintained, the same increase in value added would require 25 percent or more in total spending. 62   GSSE/TSE GSSE/Ag. GPV Panama 6.30% 1.76% US 11.99% 3.04% Note: GSSE: General Support Services Estimate; TSE: Total Support Estimate; Ag. GPV: Agriculture Gross Pro- duction Value. Source: Own elaboration, based on FAOStat and Agrimonitor data from: https://agrimonitor.iadb. org/es/resultados-por-paises?country=pe&tab=agriculture. B. THE POLICY RESPONSES OF CENTRAL AMERICAN COUNTRIES TO FOOD PRICE CRISES SINCE 2008 4.6 In light of the discussion above on the efficiency of public support to agriculture, this chapter takes stock of the policy responses to food price crises enacted by the governments of Central American112 countries. The stock-taking exercise covers agricultural policies adopted by Central American governments in response to the three major food crises since 2007-08, as described by the literature. The Central American experience is contrasted with global best practices, and, while it is too soon to assess the impacts of the most recent policy measures enacted in response to the 2022 crisis, the chapter draws on results from impact evaluations of similar policies applied around the globe available in the literature, to point to the likely effectiveness of the policies currently being adopted by Central American governments. 4.7 Overall, it can be observed that Central American countries have responded to recent food crises with an array of short-term measures. Table A5.1 in Annex 5 summarizes the policy measures, as evaluated by existing literature, that were enacted by Central American countries in the face of recent food crises, categorizing the interventions as CSE, PSE, GSSE policies.113 Broadly, it can be observed that Central American countries have mostly implemented short-term producer or consumer support policies. Some of these policies were effective and aligned with best practices, such as the introduction or strengthening of safety nets for consumers, or the reduction of trade barriers (rather than restricting trade in food products). However, several more distortionary examples can also be identified, such as price controls or the direct distribution of input subsidies. 4.8 In response to the 2008-2011 global food price crises, Honduras, El Salvador and Guatemala adopted trade measures that eliminated import tariffs on wheat from neighboring countries. Budgetary measures adopted by the countries included the implementation or expansion of school feeding programs in El Salvador and Honduras, support for community seed banks in El Salvador, and reduced production taxes on grains in Honduras, although the country also banned maize exports for a period of time. Guatemala invested in policy responses to reduce longer-term risks, including by providing payments for soil recovery and fertility improvements, implementing a program to boost strategic food reserves, piloting increased agricultural insurance, and encouraging the participation of family farmers as suppliers in the national school feeding program. 4.9 Nicaragua undertook the most wide-ranging set of policy responses among the five countries analyzed in response to the 2008-2011 global food price crises. It supported consumers by providing emergency food aid to households, introduced consumer prices controls for wheat, and invested in its storage and retail network, applying regulated producer and consumer prices, 112  The report is not able to present the same level of detail for each country in the region, in light of the gaps in available information. 113 Information was not available in the literature for the entire period for Dominican Republic and Panama, although the measures they undertook in response to the latest food and fertilizer price increases in 2022 are detailed below.  63 and distributing food in needy areas. Nicaragua also provided direct support to producers via the distribution of inputs and livestock, cash transfers for technology adoption, the provision of certified seeds and support for community seed banks, subsidized credits for farmers, and investments in the storage and distribution network. In addition, Nicaragua provided training on nutrition and farm management, and promoted the consumption of potato bread as an alternative cheaper food.114 4.10 During the pandemic, in light of the closure of schools and the dependence of poor families on school meals to feed their children, Central American countries also invested in strengthening their school feeding programs. Food kits were delivered to homes in Costa Rica and Honduras to replace schoolchildren’s meals. In Guatemala, the Government, with support from the International Fund for Agricultural Development (IFAD), FAO, the World Food Programme (WFP), and the World Bank is strengthening the value chain of its School Feeding Program, which is linked to family farming. In Nicaragua, the government took advantage of the school feeding program (with support from the WFP) to respond to the crises generated by the pandemic and by hurricanes Eta and Iota, providing assistance to students and their families. El Salvador, Guatemala, Honduras and Nicaragua are also implementing a local and regional food procurement policy, together with WFP, to promote fair and inclusive local food value chains and improve the efficiency of smallholder market systems.115 Direct food deliveries through already established mechanisms have proven to be an efficient alternative to cash transfers for countries that have limited fiscal resources but need to ensure food and nutrition security. 4.11 Central American countries have adopted additional policy measures in the wake of the latest food and fertilizer price increases in 2022. For example, Costa Rica joined other Central American countries in reducing import tariffs (in this case for red beans). Guatemala introduced a temporary propane gas subsidy, as well as cash transfers to improve health and nutrition. El Salvador also introduced a price control on fuel in April 2022 that remains active, and eliminated customs duties for food and fertilizers for one year (March 2022-March 2023). Moreover, minimum wages had been increased by 20 percent towards the end of 2021. Nicaragua introduced price controls on fuel and liquified propane gas. And various countries have implemented food and energy price controls, food and energy subsidies, reduced tariffs on fertilizers and fertilizer subsidies for farmers, as well as support for the use of (local) organic fertilizers and for more efficient use of inputs. 4.12 The Dominican Republic took seven significant measures to address the 2022 food (and fuel) price increases. This number is well above the LAC average of 4.5 measures per country. In particular, the Dominican Republic implemented two social assistance measures, by expanding the horizontal and vertical coverage of cash transfers under the Aliméntate and Bono Gas programs (the two programs are large, covering almost 60 percent of the population, and expanded coverages remain in place). In addition, it implemented a price freeze on fuel in March 2022, at the onset of the crisis, which remains in place, as well as price controls (freezes) on the prices of certain staple foods, and direct subsidies for imported foods (although this measure has ended). The Dominican Republic furthermore eliminated custom duties on numerous staple products, especially foods, for six months beginning in April 2022, and restricted flour exports in December 114  See FAO (2009). 115  See ECLAC, FAO and WFP (2022) 64 2022 (the current status of this measure is unclear). These measures are estimated to have cost the government over 35 billion Dominican pesos (US$660 million) in 2022.116 4.13 Panama introduced three policy measures in response to the 2022 food price hikes, all of which involved the provision of subsidies. It fixed the price of gasoline at US$3.25 a gallon through at least April 2023, which was estimated to have cost the government US$200 million. It also implemented price controls on over 70 food products, as well as for some 170 medicines and health-related products, decreasing the prices for these food and health products on average by 30 percent compared to prevailing prices at the time the measure was introduced in July 2022. The food price controls were set to expire in July 2023, while the duration of the health prices controls is unclear. 4.14 Compounding the decisive focus on short-term, reactive measures in the wake of food security crises, governments’ long-term investments in agriculture and food production have remained largely stable (i.e., have not increased in response to food crises) in the last 15 years. The left panel of Figure 4.2 below shows how expenditures on general services support for agriculture in Central America essentially returned to 2009 levels in the last decade. Subdued public investment in strengthening agri-food value chains in the long term is also reflected in sluggish or decreasing borrowing from the World Bank, the Inter-American Development Bank (IDB) and the International Fund for Agricultural Development (IFAD) for projects related to general services support, although the trend has begun to be reversed since 2019 (see the right panel of Figure 4.2). Figure 4.2: GSSE Expenditures in Central America and Combined GSSE Investments by Key International Financial Institutions GSSE in Central America Combined investment IDB, IFAD an WSv (US$ milion 350 350 250 250 150 150 50 50 -50 -50 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 2022 2009 2010 2011 2012 2013 2014 2015 2016 2017 Costa Rica El Salvador Guatemala Costa Rica El Salvador Guatemala Honduras Nicaragua Panama Honduras Nicaragua Panama Note: the chart on the right was constructed by the authors based on projects classified totally or partially as agricultural projects, attributing their total value to the sector. Some projects include PS and CS measures, but their total budget was classified as GSS due to the difficulty to disaggregate budget lines per project. Disbur- sement, in all cases, were considered to be evenly distributed from the date of effectiveness to the date of closure. The vertical bars identify the 2009-2017 period for which Agrimonitor data are available for all Central American countries. Source: authors’ elaboration from Agrimonitor data (https://agrimonitor.iadb.org/es/resulta- dos-por-paises?country=pe&tab=agriculture) and IDB, IFAD and World Bank websites 4.15 The nature of the investment in agriculture and rural development channeled through international financial institutions also reflects a shifting focus towards short-term measures in the periods leading up and immediately following food security crises. The average volume of loans signed per year in the region from 2005 to 2007 was US$241 million and comprised support for the improvement of land administration, rural infrastructure, small rural businesses, forest- 116  See World Bank (2023d). 65 based livelihoods, and small-scale agriculture development. Over the 2008-2011 crisis period, the volume of loans signed per year decreased to US$135 million, with a large share of investment being dedicated to cash transfers in Guatemala and agricultural inputs and disaster recovery in Nicaragua. After 2011, the volume of loans signed in the Central America region nearly halved again reaching a low of US$88 million, and only Nicaragua and Honduras maintained high levels of investment in the sector, with the former investing in rural roads, land administration and cash crops, and the latter emphasizing agribusiness development and food security for vulnerable households. During 2019-2020, the average annual volume of loans to the region for agriculture from the World Bank, IDB and IFAD has almost tripled to US$239 million, and investments in agriculture appear to have focused on export promotion, support for productive livelihoods, and resilience to climate change and extreme weather events. However, interventions in many countries still display a bias towards producer support vis-à-vis key general services support expenditures, such as research and innovation, new technology adoption, irrigation, water resources management, soil fertility, and market infrastructure. A more comprehensive approach to investment in agriculture has been mostly concentrated in Guatemala and Honduras (reflecting these countries’ relatively higher reliance on the sector as a provider of livelihoods and economic growth) and in a few large series of projects, which have become these countries’ main tool to generate rural development and build food system resilience. C. THE EXPECTED EFFECTIVENESS OF CENTRAL AMERICAN COUNTRIES’ POLICY RESPONSES TO THE CURRENT FOOD PRICE CRISIS 4.16 How effective are policy measures in Central America in palliating the effects of food price crises on food and nutrition security? Assessing policy effectiveness relates to the degree to which intended policy results and objectives are achieved, ideally identifying plausible causal impacts between activities and outcomes. While proper impact evaluations of most measures enacted by Central American countries in response to recent food crises are not available, several assessments have been made of policies deployed worldwide that are similar to the policies enacted in Central America. A review of existing studies can, therefore, provide useful insights on the expected results of different kinds of policies adopted in the region, as well as their likely efficacy at improving food and nutrition security while supporting livelihoods and enhancing resilience. To this end, Table 4.3 summarizes existing available evidence on the impacts (or lack thereof) of policies implemented in various regions of the world, categorized according to the OECD classification, and to whether they involve trade measures and price controls, recurrent budget expenditures, or public investment expenditures. Further details on the impact analyses included in the exercise are provided in Annex 2. 4.17 Overall, the policies that are currently being enacted by Central American countries can be expected to produce positive food-security effects in the short term, but lack long-term vision. Based on the literature review summarized in Table 4.3, it can be argued that most of the measures that have been adopted in Central America have been sound in terms of their short-term potential to impact positively on food security. However, there is no evidence on how lasting the effects of these policies could have been, nor on whether they have increased countries’ resilience to future crises. Overall, it appears evident that short-term measures have largely prevailed over strategic long-term policy interventions that could increase food production and resilience, thereby enhancing food security. 66 4.18 One straightforward example in this respect is the fact that countries have been lowering import tariffs with each new crisis, which suggests that trade barriers may tend to be lifted again (arguably for revenue-generation reasons) once food prices decrease (see Box 4.1). This observation clearly resonates with the low level of pass-through from international to domestic food prices documented in Chapter II of this report, insofar as lower tariffs on food imports would be expected to dampen (i.e., partially offset) the impact of higher international food prices. In terms of expected impacts, while raising trade barriers may contribute in the short run to smoothing changes in prices for producers – allowing time for them to adapt to decreasing prices – in the long-run they may: (i) keep prices higher for domestic consumers, and (ii) render (protected) domestic production systems uncompetitive at international prices. Farmers with uncompetitive products cannot expand production levels beyond national consumption levels and may not be able to profit from a next wave of global high food prices. Similarly, despite their immediate benefits for producers, input subsidies (e.g., for fuel) have proven to be inefficient as well as costly for governments with tight budgets. A further consideration, that would need to be analyzed in future, is the extent to which limited competition in domestic wholesale and/or retail markets for food affects the pass-through of international food prices to domestic food prices, or indeed the extent to which limitations in competition may have contributed to domestic food price increases that are not significantly correlated with international food price increases. Box 4.1: How restrictive has trade been in Central America during and between crises? Tariffs and Global Trade Alerts: Central American countries put in place generally healthy trade policies at times of food price crisis. From 2008 to 2022, the Global Trade Alerts (GTA) database records a total of 97 unilateral liberalizing trade interventions (“green alerts” such as tariff reductions, import or export incentives, tariff exemptions or elimination of subsidies) implemented by Central American countries, versus 80 unilateral restrictive policies (“red” or “amber alerts” like import tariffs, import or export bans, import quotas) likely to affect trade in services and goods, including agricultural products (Figure 4.3). As shown in Figure 4.3, around the three major global food price peaks in 2008, 2011, and 2022, Central American countries exhibited a clear commitment to supporting open trade and refraining from outright closures. Soon after the 2011 crisis and in the period leading up to the next, however, a tendency towards a gradual re-closure can be observed: the period between 2014 and 2019 is marked by a prevalence of restrictive measures, mostly including import tariffs and import quotas (Figure 4.4). Figure 4.3: GTA alerts for agricultural products and fertilizers in Central America, 2008- 2022 15 10 5 0 2008 2018 2009 2010 2016 2022 2014 2013 2020 2012 2017 2015 2019 2011 2021 Green Red/Amber Source: Global Trade Alert database, https://www.globaltradealert.org/. Notes: Red and amber alerts signal interventions likely to or almost certainly discriminating against foreign commercial interests; green alerts point to liberalizing interventions on a non-discriminatory basis. 67 Figure 4.4: Type of “red and amber alert” interventions implemented in Central America, 2008-2022 25 Tax or social insurance relief 20 Export ban Price stabilisation 15 Safeguard 10 Import ban 5 Anti-dumping Import tariff quota 0 2008-2010 2011-2013 2014-2016 2017-2019 2020-2022 Import tariff Source: Global Trade Alert database, https://www.globaltradealert.org/ Non-Tariff Measures: Beyond explicit policies on tariffs, trade in Central America has also been affected by non-tariff measures (NTMs) such as sanitary and phytosanitary regulations, standards, testing, and certification (“technical” measures), as well as quantitative restrictions, price measures, and forced logistics or distribution channels (“non-technical” measures), which have an impact on trade flows through information, compliance, and procedural costs (see Figure 4.5). Figure 4.5: Count of NTMs on agriculture products and fertilizers in Central America, 2008-2021 Panel a. Central America Panel b. Central America, excluding PA and CR 700 70 600 60 500 50 400 40 300 30 200 20 100 10 0 0 2008 2020 2010 2009 2019 2017 2018 2014 2012 2016 2013 2015 2011 2021 2008 2020 2010 2019 2009 2017 2018 2014 2012 2016 2013 2015 2011 2021 Costa Rica Guatemala Honduras Guatemala Honduras Nicaragua El Salvador Panama Nicaragua El Salvador Source: UNCTAD, TRAINS NTMs database, 2024. Notes: HS2 codes selected are from 01 to 21 and 31. Partners affected by NTMs include all the world, and both multilateral and bilateral measures. No data available for the Dominican Republic. While Panama appears as a clear outlier, accounting for 77.2 percent of the total number of NTMs deployed in the region between 2008 and 2021 (mostly phytosanitary port import requirements pertaining to trade activities involving the Panama Canal), Panel (b) of Figure 4.5 shows a consistent pattern for El Salvador, Honduras, Guatemala and Nicaragua, whereby the periods between crises feature the maximum intensity of NTM implementation. Remarkably, the peak of NTM implementation lies between 2014 and 2016, which coincides with the period of highest issuance of GTA red and amber alerts for these countries (Figure 4.6). At the same time, in the period 2012-2016 the average ad-valorem equivalent (AVE) of the NTMs enacted by Central American countries for agri-food products amounted to 20.1 for technical measures and 0.5 for non-technical measures, which means that technical and non-technical NTMs had the same impact of a uniform import tariff of 20.1 and 0.5 percent, respectively 68 (for comparison, the average ad-valorem equivalents for the same products and the same period in the United States were 11.2 for technical measures and 4.4 for non-technical measures). Figure 4.6: GTA red and amber alerts and NTMs 12 250 GTA red/amber alert 10 MTMs issued 11 200 GTA 8 226 red/ amber 9 150 6 alerts 138 151 100 4 5 50 Non tariff 2 4 4 Measures 0 56 0 32 0 01 3 01 6 01 9 22 2 01 2 0 08- 11- 2 4- 2 17- 0-2 20 20 201 20 02 2 Table 4.3: Assessment of the Effects of Policies Measures Implemented in Central America Type of measure Expected Impact on food security Consumer support - trade measures and price controls » Import tariff reduction/ These measures benefit consumers, but can be highly elimination distortionary of domestic prices, with long-term consequences for producers. FAO & IPA (2022) showcase a few examples of » Maize exports ban how reduced domestic prices due to trade policies have forced farmers to change to lower value-added products (e.g. from fish » Price controls of staple to rice) or lowered their living standards.* food » Price control of cereals Consumer support - recurrent budget » Food kits WFP (2011) summarized the results of 12 impact evaluations of school feeding programs (SFPs). The review showed clearly » Cash-transfer for the that SFPs where multiple-micronutrients-fortified meals are households most affected provided improve the micronutrient status of the recipients by inflation and sometimes spill over to their families. The evidence on the impact of school feeding as a social safety net is scarcer in the » School meals and related literature, but inferable evidence suggests that school feeding legal framework can be a valuable social assistance tool and that there is a partial redistribution of household resources as household with » Fuel subsidies children benefiting from school feeding do better on selected health and nutrition, and education outcomes. 69 Type of measure Expected Impact on food security FAO et al. (2016) provides evidence on the impact of cash transfer programmes in Ethiopia, Ghana, Kenya, Lesotho, Malawi, South Africa, Zambia, and Zimbabwe. Positives impacts were found in: (i) increased consumption and reduced poverty; (ii) smoothing of consumption within seasons and between years; (iii) statements on happiness and positivity about the future; (iv) improved education and health outcomes; (v) food security and nutrition security (across all countries); (vi) safe transition for adolescents and youths to adulthood (health, mental health, hope, aspirations, and perceptions of the future); (vii) increases in farmed area and production and livestock and agriculture assets ownership, and (viii) more stable employment within agriculture. The FAO et al. (2016) review presents a strong case for unconditional cash transfers in Africa, as compared to conditional transfers (pointing to higher impacts due to greater flexibility for households in managing incomes, while lowering operational costs for the programs). It is possible that, during food crises, the distribution of food kits or food vouchers may be more efficient than cash transfers in ensuring a well-balanced diet, depending on the specifics of each country and situation. Producer support - trade measures and price control » Import tariff reduction/ See considerations on trade measures above. Except for the elimination (negative 2008 maize export ban in Honduras, most trade measures support for outputs, enacted in Central America during the crises did not impact positive support of inputs) directly on the main crops farmed by smallholders. The elimination of import tariffs on staple crops was not enough » Exports bans on food crops to bring prices down to pre-crisis level at the farm level. Food (e.g. maize) price controls in Nicaragua were done through a network of state-run retail shops in targeted poor neighborhoods selling » Reduction of production products about 10-12 percent below the market price. Sales taxes on grains represented less than 4 percent of the national production of staple food. Distortions were not major. Reduction in producer » Food price control support through trade measures and price controls did not seem to have adversely affected production capacity. Producer support - recurrent budget » Distribution of inputs and Eight out of 11 input distribution interventions evaluated in livestock World Bank IEG (2011) and 15 out of 18 further evaluations of these interventions considered for this report showed positive » Conditional cash transfers impacts,* mostly in terms of increased yields. Most of these on technology adoption operations were coupled with extension services and some with investment in infrastructure, which may have contributed » Certified seed and input to, if not determined, their success. One case in Kenya showed distribution that fertilizer application only yields positive results when farmers know the right quantities to be applied.* 70 Type of measure Expected Impact on food security » Producer prices support (public purchases) Output market interventions are evaluated much less often. One evaluation in Ghana, namely World Bank IEG (2011), shows that the measure (buffer stocks) was useful in terms of poverty alleviation during the 10 years it lasted (no information is available on the impact once the policy was discontinued). Krivonos & Dawe’s (2014) case study on Nicaragua suggests that small farmers incomes were affected by high food prices, as some are net buyers, while potentially higher prices for their produce were probably not perfectly transmitted to farmers, thereby affecting their production capacity. FAO & IPA (2022) provide more insights as to why price transmission is incomplete in imperfect markets (with low competitiveness or information asymmetries). Emergency support via the distribution of inputs may have been important to maintain (if not improve) production capacity. General services support - recurrent budget » Conditional transfers to See Consumer support (recurrent budget) improve health and nutrition services. General services support - investment budget Financial services Financial services » Access to credit » There is thin evidence of the impact of improved access to credit or other financial services on food security. IEG (2011) Smallholders’ production reports increased per capita incomes due to access to marketing credit programs in Bangladesh, but mixed results regarding savings groups in Thailand*. However, when credit is available, » Local and regional food farmers tend to take it, and FAO & IPA (2022) find that buyers procurement policy who provide credit are preferred, suggesting it brings benefits to farmers in terms of higher income. Despite the possible » Improvement of the benefits for farmers, in general, improving access to credit is efficiency of smallholder not easy, as banks find transaction costs and risk high when market systems lending to smallholders. For instance, in value chain projects, » Investment in storage and creditors engage more towards the end of the investment retail network – supported by grants – i.e., once income is secured and repayment is more likely. Guarantee funds do not show » Design and financing of conclusive results in improving access to credit.* business plans for small rural enterprises 71 Type of measure Expected Impact on food security Introduction of new Smallholders’ production marketing technologies » There is evidence that, if well designed, interventions to » Production and use of support marketing of produce have positive effects on organic fertilizers farmers’ incomes (and consequently on food access). For instance, IEG (2011) suggests that contract farming has Training and extension largely positive impacts, but it is a complex area, with few examples for LAC: In Peru, a lack of regulatory reforms and of » Trainings on nutrition and investment in infrastructure resulted in potato contracts failing farm management to produce the expected results. On the other hand, in Costa Rica, average farm gate price improved considerably through Cash crop development and participation in both specialty and cooperative marketing export promotion channels. Examples from five countries on contract farming » Support to the development are positive, but they imply strong involvement of private firms of cash crops (coffee) and extension, credit, and input provision and are not risk-free. Other » ComRural, a series of World Bank value chain development projects in Honduras, is investing in the design and » Strengthening of the legal implementation of business plans aimed at product and institutional framework improvement and better access to higher quality markets for on food security several value chains, with positive results in terms of farmers’ income. » Project evaluations in Brazil (undergoing and not published) suggest that public procurement increases sales and incomes of small farmers. » Evaluations of the purchase for progress program (P4P) of WFP (in which small farmers were assisted in meeting the purchasing criteria of WFP and other formal buyers) suggest the program had an impact in terms of increasing production and sales values for farmers. An assessment conducted by FAO (2016) in El Salvador suggests positive results. Only one full-scale impact evaluation of P4P (Democratic Republic of Congo) was found in the literature, namely Oxfam (2016). Introduction of new technologies » The World Bank IEG (2011) review revealed that the nine evaluated natural resource management projects that improved soil structure generated positive and significant increases in yields and family incomes. A project evaluated in El Salvador reported that supporting mulching, minimum tillage, crop rotation, and green manure had a positive effect on income. 72 Type of measure Expected Impact on food security Training and extension » The World Bank IEG (2011) review suggests that (i) farmer schools were effective in decreasing the use of pesticides (e.g., in Peru), but did not reveal significant impact on yields (or food security); (ii) advisory services were more successful, with positive results in Argentina (higher yields and quality of grapes) and Uruguay (increased livestock productivity); and (iii) the provision of market information produced positive results on sales prices*. However, FAO and IPA (2022) reviewed cases in Colombia and India where knowledge of market prices (via SMS messages) did not change farmers bargaining capacity or prices. » An IDB (2010) review conducted also suggest that results from extension interventions are mixed and highly dependent on the context and complementary interventions. The one agricultural research project evaluated in the IDB (2010) review, which provided competitive grants for research projects, was found to have contributed positively to increases in sales by farmers. Cash crop development » The results of the ComRural project in Honduras (see above) suggest that increases in yields and value addition of cash crops for export can contribute to significant increases in farmers’ income. Other areas with potential to increase yields, farmers’ income and access to food: » These include improved irrigation, improved soil management practices, and market information. Interventions in these areas appear to have been relatively less prominent in Central America as a response to the crises of the past 15 years (although there have been interventions in these fields in Central American countries). At the same time, they are highly relevant and can contribute to increasing yields and the resilience of food systems in the region. Notes: *See Annex 2 for additional information – These results may be affected by publication bias (in favor of studies with positive results), yet they are useful to assess if at least in some contexts the evaluated interven- tions produce positive results. 73 D. SUMMARY OF RESULTS 4.19 Central American countries have historically been responding to recent global food price crises with a range of producer support, consumer support, and general services support measures. However, while total government expenditure positively impacts agriculture’s performance, the various types of support do not all have the same level of impact on the agricultural sector, in terms of the efficacy of public action. Evidence from Latin America and other regions of the world suggests that the composition of public expenditure on agriculture affects its effectiveness in promoting agricultural development.117 In particular, increasing the share of expenditures committed to public goods and services is less distortionary and more effective than subsidizing the provision of private goods and services or market price support. » On the whole, the policies enacted by Central American countries have been in line with practices applied in other contexts around the world, and in most cases have avoided the worst international practices, such as increased trade restrictions in response to food price spikes. » While the general thrust in Central America has been to favor direct producer support (in terms of volume of financing), short-term reactions to food price spikes have often involved price controls that supported consumers in the wake of the crises but affected them in the longer run. » Furthermore, expenditures on general services support have lagged, relative to producer and consumer support. » While the short-term responses to protect consumers may have contributed to reducing the impact of international food price spikes on domestic food prices, they have also resulted in price distortions and, potentially in combination with food distribution markets characterized by insufficient competition, have prevented international prices from feeding through to farm gate prices for producers in ways that might have encouraged an increased production response. » Overall, the policy ecosystem displayed by Central American countries can be expected to produce positive food-security effects in the short term, but lack long-term vision. A shift in the structure of agriculture support could enable large efficiency gains that can drive sustainability, equity, and resilience. 117  See, for example, Anríquez et al. (2016), López and Gallinato (2007), and López et al. (2017). 74 5 CONCLUSIONS AND RECOMMENDATIONS Amaiquez - Stock Adobe 75 V – CONCLUSIONS AND RECOMMENDATIONS 5.1 Countries in the Central America region are characterized by chronic food insecurity for large shares of their populations. Even before the 2008 global food crisis, one-fifth of the populations of El Salvador, Guatemala, Honduras and Nicaragua were undernourished, i.e., had insufficient caloric intakes. This chronic food insecurity is associated with high levels of poverty that do not enable significant shares of the population to afford a basic consumption basket, and with worsening climate conditions (increasing frequency of severe droughts, floods and storms) associated with climate change. Overall, one-quarter to one-third of the populations surveyed in mid-2021 in the seven countries analyzed in this report reported running out of food in the previous 30 days, with these numbers rising sharply compared to pre-pandemic situations. 5.2 These chronic conditions are now being aggravated by major spikes in domestic food prices since 2022, which have been increasing more rapidly than average consumer price inflation rates. In each of the countries for which there were data, food price inflation in June 2022 was found to be significantly higher than in January 2022, and for all but Panama, food price inflation in June 2022 was well above average inflation as measured by the national consumer price index. At the same time, the cost of granular fertilizers as a share of total production cost per hectare increased between 45 and 66 percent for basic food staples in Central America in 2022. This situation has led to dire conditions in terms of undernourishment for large segments of the Central American population. For example, an estimated 10 million Guatemalans (56 percent of the population) are food insecure, while 3.7 million (20.7 percent of the population) are severely food insecure. In Honduras, 66 percent of the population faced moderate to severe food insecurity in 2022. Similarly, around one in six Costa Ricans faced moderate to severe food insecurity during the pandemic, while 15 percent of Dominicans faced emergency or crisis levels of food insecurity. 5.3 One explanation that is commonly identified for this emerging food security crisis are major spikes in global food prices since 2022, which reached their highest levels in sixty years. A combination of factors has driven the increased levels and volatility of global food prices, including disruptions in global food supply chains as a result of the COVID-19 pandemic, the Russian invasion of Ukraine in February 2022, and sharp increases in fertilizer prices through 2022. 5.4 The motivation for this report has thus been to assess the extent to which the price dynamics in international food markets can be considered responsible for the sharp domestic price increases documented in Central American countries. While there is clearly some co- movement between the surge of international prices of key food-security items and the recent evolution of domestic food price inflation (Figure 5.1), testing this hypothesis formally is crucial when it comes to identifying effective policy options to respond to the emerging food crisis. 76 Figure 5.1: Central America: Domestic Food CPI versus International Commodity Prices, 2019-2023 Dominican Republic Guatemala Honduras 18.0 18.0 20.0 16.0 18.0 16.0 14.0 16.0 14.0 12.0 14.0 12.0 10.0 12.0 10.0 10.0 8.0 8.0 8.0 6.0 6.0 6.0 4.0 4.0 4.0 2.0 2.0 2.0 0.0 0.0 0.0 2 1 2 11 20 1 1 20 9 20 9 -2.0 2 3 20 5 2 7 2 9 23 3 2 1 2 3 5 2 1 2 5 22 7 19 7 2 1 19 9 19 3 20 M5 21 9 2 3 1 22 1 20 20M7 19 7 2 5 7 20 9 M 1 2 9 2 11 1 5 2 1 1 1 1 3 9 2 1 2 1 23 3 5 20 M1 2 3 20 21 M 3 2 3 2 1 22 1 2 5 2 5 7 20 M1 22 9 23 3 2 1 20 M1 2 3 20 2M 3 21 5 5 2 5 2 1 20 0M 2022M7 20 2M 7 -2.0 2 3 20 3M 20 9 20 2M 2020M5 20 1 9 M 1 20 M 20 3M 1 20 20M7 20 0M 20 22 5 20 0M 20 0M 20 2M 20 M 20 1 9 M M 2 1 20 2M 2 9 20 1 M 2023M 2 11 20 M 20 M 20 9 M 20 M1 1 20 0M 20 9 M 20 1 M 20 9 M 20 2M 20 1 M 20 1 9 M 20 M 20 1 M 20 21 M 20 1 M 22 9 20 2M 2 11 20 9 M 3 20 21 M 1 20 9 M 20 M1 1 20 1 9 M 1 19 9 20 M 1 5 20 3M 20 3M M 20 1 M 7 20 1 M 20 1 M 20 1 M 20 1 M 20 21 M 20 23M 20 0M 20 0M 7 20 M1 M 20 M 20 2M 20 2M 20 22M 2 M 20 M1 2020M 20 0M 20 2M 20 0M 20 1 9 M 20 21 M 20 1 M 20 1 9 M 20 M 20 9 M 20 M 1 1 20 20 2 2 1 20 DAP Urea Coffee Arabica Bananas DAP Urea Coffee Arabica Bananas DAP Urea Coffee Arabica Bananas Maize Rice Wheat Food CPI Maize Rice Wheat Food CPI Maize Rice Wheat Food CPI Nicaragua Panama 18.0 20.0 16.0 18.0 16.0 14.0 14.0 12.0 12.0 10.0 10.0 8.0 8.0 6.0 6.0 4.0 4.0 2.0 2.0 0.0 0.0 2 3 2 7 22 9 22 3 20 9 20 M5 7 2 3 2 5 2 1 19 1 19 9 20 M3 23 3 2 1 2 7 2 9 19 5 5 22 1 1 20 1M5 2 7 1 2 1 2 1 20 19 1 20 9M 3 2 1 2 1 20 0M 20 9M 20 M1 23 3 2 3 2 0 2 2M 7 20 0M 20 0M 20 M1 20 M -2.0 2 9 20 M 2 0 2M 2 0 2M 22 9 2 1 20 1M 20 19M 22 1 20 0M 20 0M 2 1 5 20 1M 20 1M 5 19 9 20 9M 2023M 20 3M M 20 19M 2 0 2M 2 3 20 1M 2 5 7 20 22M1 20 M1 20 20 7 1 5 2 5 7 20 M9 20 11 2 3 1 M 2 1 20 1M 1 2 1 20 3M 1 2 0 2 2M 20 3M 20 M 1 20 21M M 2 0 2M 20 19M 20 21M 20 1M 20 1M 20 1M 20 M1 20 M 20 0M 20 9M 20 0M 20 M 20 0M 20 1M 20 M 2 1 2 1 20 20 -4.0 DAP Urea Coffee Arabica Bananas DAP Urea Coffee Arabica Bananas Maize Rice Wheat Food CPI Maize Rice Wheat Food CPI Source: Own elaboration, based on data from the International Monetary Fund (IMF), accessed June 25th, 2023. Details of data used: Food CPI: Food and non-alcoholic beverages CPI from the IMF; DAP: US Gulf NOLA DAP Export Spot Price (US$ per metric ton); Urea: US Gulf NOLA Urea Granular Spot Price, hundred US$/ST; Coffee Arabica: Coffee, Other Mild Arabicas, International Coffee Organization New York cash price, ex-dock New York, US$ per pound (in hundreds). Bananas: Bananas, Central American and Ecuador, FOB U.S. Ports, US$ per metric ton (in hundreds). Maize: Maize (corn), U.S. No.2 Yellow, FOB Gulf of Mexico, U.S. price, US$ per metric ton (in hundreds); Rice: Rice, 5 percent broken milled white rice, Thailand nominal price quote, US$ per metric ton (in hundreds); Wheat: Wheat, No.1 Hard Red Winter, ordinary protein, Kansas City, US$ per metric ton (in hundreds). 5.5 Time-series analysis performed in this report shows that there is lower-than-expected transmission of international to domestic food prices in the Central America region. The level of pass-through was found to be relatively more pronounced in the case of Honduras, and stronger for certain foods and cash crops (e.g., rice and coffee) than for staples such as maize and beans. Fertilizer prices also seem to be more strongly transmitted to domestic markets, especially in Honduras. But overall, the values retrieved from the analysis are essentially modest: a 1 percent increase in the international price of a certain commodity is translated into a corresponding percent increase in the domestic price ranging from 0.1 to 0.6 percent at best. What seems to be mostly transmitted to domestic markets is international price volatility, so that recurrent price fluctuations in international markets appear to produce recurrent domestic price fluctuations in home markets in Central America. This is particularly the case for maize, black beans, and bananas, and especially so in Costa Rica, Honduras, and Panama. Fertilizer price volatility is similarly strongly passed on to domestic markets, with a 1 percent increase in the international volatility of urea prices producing a corresponding 1.9 increase in the volatility of urea prices in Honduras. 5.6 Consistent with low levels of pass-through, the current spikes in international food prices alone do not seem to be the main responsible for the substantial welfare losses currently documented in Central America. Simulating global food and fertilizer price increases in line with the highest peaks observed in 2022 produces non-negligible but moderate net reductions in average household welfare in Central American countries, accounting for the counteracting impacts on net food buyers and sellers. International fertilizer prices induce slightly more sizeable effects on rural incomes via a reduction in area planted and crop production. While the simulations rely on a range of simplifying assumptions (partial-equilibrium analysis, focus on the average household, unexplored link from fertilizer price increase to reduced food consumption) that 77 jointly dampen the size of the estimated impacts, the overall magnitude of the effect on domestic welfare produced by international food and fertilizer prices still seems too low to account for the widespread crisis-level food-security conditions that are being experienced in Central America. 5.7 The recent domestic price surges and food security crisis observed in Central America seem therefore predominantly driven by domestic considerations.118 One factor that is likely to have a strong impact on food and nutrition security are climate-induced disruptions driven by excess rainfall or droughts as well as natural disasters. Central America is highly vulnerable to climate hazards, and recent years have seen an intensification in the occurrence of highly destabilizing events such as pronounced El Niño/La Niña phenomena and severe tropical storms (see Annex 4). Another issue worth exploring is the influence of increased fuel prices on agrologistics costs along the food value chain, as recent analysis from Northern Central America (World Bank, 2022b) shows that agrologistics accounts for a significant fraction of retail prices of vegetables, coffee, and basic grains in the region. Finally, conflict and violence may be also fueling food insecurity, in a region where several countries exhibit some of the highest rates of violence in the world (UNODC, 2019). Gang violence, political turmoil, threats, extortion, persecution and sexual violence could have multi-faceted reverberations on agricultural production and prices, from preventing farmers from accessing inputs and selling produce, to resulting in the destruction of crops and property, to pushing farmers to abandon production and seek safety elsewhere (the UN Refugee Agency estimates that by the end of 2022 more than 1 million people had either fled to neighboring countries or been internally displaced within the region).119 While these issues were beyond the scope of this report on international to domestic price transmission, future research on food insecurity in Central America should address them rigorously. 5.8 Another important avenue for future research is to understand the determinants of the low levels of price transmission to domestic markets in Central America detected by this report. Although attenuated short-term domestic impacts of international turmoil may appear desirable, there are likely to be domestic factors that bias the international-domestic price relationship and that can prove inefficient and damaging in the long run. One possible explanation relates to the agribusiness market structure in Central American countries: for instance, Arias and De Franco (2011) show high levels of market concentration in Nicaragua, hampering the competitiveness of the sector and its ability to reflect undistorted market prices. Another commonly recognized factor are the policies followed by countries in response to changes in world food prices. A vast literature in food prices and international trade also points to political-economy considerations whereby policy-makers in low- and middle-income countries feel an imperative to stabilize the domestic prices of staple goods (see e.g. Giordani et al., 2016, Timmer, 2010). 5.9 Looking at the overall ecosystem of public support to agriculture, this report has in fact identified a number of potentially distortive trends in the region. Over time, most countries have relied on extensive systems of producer support, usually at the expense of consumers. Furthermore, investment in public goods and services for agriculture, which has been long recognized as the most efficient use of public money, has been limited and stagnating over the years. Moreover, the degree of agriculture support has been volatile over time, especially in Guatemala, Costa Rica, and Panama, which have experienced strong inter-year fluctuations in producer support that can increase risks for producers and affect their investment decisions. 118  The change in local food prices may also reflect relative changes in exchange rates. The relative strength of a country’s currency with respect to its trading partners affects the costs of imported food items in domestic markets, depending on the exchange rate pass-through to domestic prices for tradable staple foods (see e.g. Alper, Hobdari, and Uppal, 2016; Okou, Spray, and Unsal, 2022). Future research could study exchange rate price-through exploiting the differential exchange rate variation of countries in Central America, besides dollarized Panama and El Salvador. 119  https://www.unrefugees.org/emergencies/central-america/ 78 5.10 Even in their response to food price crises (including the current one), Central American governments adopted a set of interventions with mixed expected effectiveness. For example, various countries expanded cash transfer programs, and food kits were delivered to vulnerable households in Costa Rica and Honduras during the pandemic in light of the closure of schools and the dependence of poor families on school meals to feed their children. El Salvador, Guatemala, Honduras, and Nicaragua implemented local and regional food procurement policies, to promote local food value chains. And in most cases, countries in the region have reduced import tariffs and avoided adopting the least appropriate practices applied elsewhere, such as export bans and heavy export taxation (a notable exception being Honduras’ ban on maize exports in response to the 2008 food crisis). Although the overall thrust of these short-term responses has been to support consumers with increased budgetary recurrent expenditures, other measures such as increased fuel, gas and fertilizer subsidies (especially in kind), as well as food and energy price controls, have been proven to be highly distortive and to deliver negative welfare gains to the society in the medium to long run. 5.11 In addition, most interventions still display a bias towards producer support over general services support expenditures to strengthen agrifood value chains in the long term. The fact that the current food security crisis in Central America appears to be driven to a significant extent by more chronic underlying factors, such as climate change and natural disasters (as opposed to international prices movements), underscores the urgency of building a strong resilience framework and expanding efforts to promote climate-smart agricultural technologies and practices at the core of the region’s agri-food system. A. OPTIONS FOR STRENGTHENING POLICY RESPONSES TO PREVENT AND MITIGATE FUTURE FOOD PRICE CRISES 5.12 In light of the above discussion, Central American countries are left with the imperative to improve the efficiency of their domestic markets, while at the same time introducing countercyclical and non-distortive public policies and programs for both food consumers and producers. Options for strengthening policy responses in efficient ways that can also prevent and mitigate future food crisis include measures involving consumer support, producer support and general services support. These measures are described next. A.1. Consumer Support Trade Measures 5.13 Lowering import tariffs is an effective tool to decrease price distortions and reduce consumer prices. However, the inability to keep import tariffs low in light of lost fiscal revenues and of pressures to support local farmers means that these measures are often reversed, thereby harming consumers’ purchasing power. In light of their limited fiscal capacity relative to wealthier countries, Central American countries need to review the efficiency of their broader public expenditures in order to sustain the reduction in tariffs, together with alternative mechanisms to support producers’ competitiveness so as to keep food prices low while ensuring adequate returns to producers. 79 Safety Nets 5.14 The (expanded) safety nets implemented by Central American countries have proved to be valuable instruments to ensure food access during the 2008, 2011, and ongoing food crises. Among short-term measures, cash transfers seem to also display positive effects as policy measures with longer-term impacts. In particular, cash transfers have been found to have clear impacts in alleviating different poverty dimensions in the long run, including food security.120 Rather than staying trapped into safety nets, beneficiaries seem to be able to gradually improve their livelihoods and eventually graduate out of them.121 While low fiscal capacity represents an obstacle to keeping these policies running in the long term, significant coverage can still be achieved for the most vulnerable populations at a relatively low cost in terms of the proportion of GDP (e.g., Dominican safety net programs reach 57 percent of the bottom income quintile and have reduced the poverty headcount by around 10 percent, at a cost of around 0.6 percent of GDP).122 Further work can be done to strengthen the institutional and legal frameworks for social protection systems, and enhance their targeting and coverage, including by improving citizen registry systems, especially for rural populations and undocumented citizens. A.2. Producer Support Trade Measures 5.15 It is likely that current trade measures in Central America distort the prices of important crops for domestic consumption, in particular via significant import tariffs that tend to push domestic prices above international ones.123 Consumers are thus bearing higher food prices than necessary and supporting production systems that are not sufficiently competitive internationally. These trade measures constitute a large share of the producer support documented earlier (Figure 5.5 in Chapter V), and constitute negative support to consumers. During the current and earlier food crises, various countries in the region implemented tariff reductions, and the challenge will be to sustain those reductions while increasing the competitiveness of, and therefore returns to, local producers, accounting for the likely political-economy trade-offs of lifting trade barriers. Direct Transfers to Producers via the National Budget 5.16 Most input distribution interventions in the region (or measures with similar objectives, such as conditional cash transfers and vouchers tied to technology adoption), have been made in response to emergencies. While they have tended to produce the desired immediate effects of maintaining or increasing production capacity, in the medium to long term they are not the most efficient policies in terms of rural development, 124 especially when they come in the form of direct distribution. In particular, the subsidization of private incomes displaces government expenditures that could be dedicated to agro-environmental public goods with broader welfare impacts and larger positive externalities.125 Moreover, the volatile nature of levels of producer support in Central 120  See FAO, UNICEF and Oxford University Press (2016). 121  However, as stated in FAO (2011), inclusive safety net programs that operate for a long period of time imply expenditures that many countries may not be able to afford and other means to graduate people out of poverty and thus make them more resilient to food crises may need to be found. 122  See World Bank (2023e). 123  Available data do not allow for a determination of which particular crops are most supported by trade measures. 124  Moreover, these interventions require adequate targeting to succeed, beginning with the development of family farmer registries, which have yet to be developed fully in the region. Such registries enable the identification of specific measures to address small farmers’ needs and can prove extremely useful for the provision of general services support, such as digital extension, agro-meteorological and marketing support. See, for example, World Bank (2022d) on the Agricultural Information Management System supported in Belize. 125  See Anríquez et.al. (2016); Gautam et al. (2022), and Zavala-Pineda et.al. (2015). 80 American countries increases risks and dampens the intended effect of increasing agricultural investment. 5.17 Promoting family farmers’ access to markets, finance, and services (for example via the productive alliances126 approach) can increase incomes and resilience for producers.127 In particular, linking smallholder producers to formal,128 larger and more diversified markets can not only increase productivity, incomes and resilience for small farmers but also provide greater incentives for the adoption of climate-smart agriculture technologies and practices that enhance climate adaptation and mitigation.129 A.3. General Services Support Information, Research and Development Systems 5.18 It is paramount to improve Central American agri-food information systems. This is particularly the case with regard to the competitiveness of farming systems, logistic hurdles, market imperfections, technical factors and agroclimatic information. More detailed and complete data are also required on public support to agriculture, including: (i) longer and more complete time series; (ii) disaggregation by main product categories (e.g. staple foods vs. cash crops, net imported vs. net exported products, raw vs. processed products, etc.), and (iii) larger breakdowns of policy categories (e.g., trade measures, price control, recurrent budget, investment). 5.19 Likewise, effective policy-making needs to be supported by an increased capacity to monitor and evaluate the results of public programs and policies. There are insufficient rigorous policy and program evaluations in Central America, which makes it challenging to draw conclusions from past interventions in order to improve performance or plan new interventions. 5.20 Central American countries are also among the countries in LAC with the lowest expenditure on agricultural research and development (R&D). In comparison to the widely accepted international benchmark of at least 1 percent of agricultural GDP spent on agricultural R&D, Guatemala only spends 0.1 percent of agricultural GDP on agricultural R&D, while the proportion spent in Honduras is 0.2 percent; in the Dominican Republic, 0.3 percent; and in Nicaragua, 0.4 percent. By contrast, Chile and Brazil, respectively, invest 1.7 percent and 1.9 percent of their agricultural GDPs on agricultural R&D, while in Argentina an increasingly sophisticated ag- tech and ag-biotech innovation ecosystem is emerging, with more than 500 such firms identified,130 in addition to farmer networks such as CREA sharing innovative findings.131 Investment in R&D is not only required to increase farm productivity but to develop clean energy sources for agriculture, organic fertilizers, integrated pest management practices, and technologies for climate change adaptation and mitigation, all of which yield important positive externalities. 126  The approach broadly consists of an agreement between organizations of smallholder producers, buyers, private financial institutions, and technical service providers, which revolves around the implementation of a business plan proposed by producers. Productive investments, technical assistance, and business development are financed through public grants, which are matched by producers’ contributions (which could be in-kind) and, especially in the case of transitioning smallholders and commercial producers, by buyers and participating private financial institutions. 127  See, for example, the positive experience of Honduras with promoting productive alliances, in World Bank (2022e). 128  The informality of economic activities proved to be a risk during the Covid-19 pandemic, as people operating in informal value chains were not covered by social security measures. 129  See World Bank (2022e) for evidence of the experience in this regard under ComRural program in Honduras. 130  See the Argentina Productiva 2030 report prepared by the Ministerio de Economía de la Nación (2023); Bert et al. (2022); Lachman and Lopez (2019); Lachman et al. (2021), Liu et al. (2020), and OECD (2006). 131  Consorcios Regionales de Experimentación Agrícola (CREA), a nonprofit civil association in Argentina founded in 1957, whose members are agricultural entrepreneurs. Currently, CREA comprises almost 2,000 agricultural enterprises engaged in a wide range of agricultural activities. Directed and partially financed by its members, CREA’s main objective is to facilitate the exchange of experiences and knowledge (e.g. on innovations in agronomy) to improve productivity in agri-food value chains. 81 5.21 The 2022 food crisis highlighted the need to diversify sources of fertilizers, including local production of organic fertilizers. Five countries produce 62 percent and export 55 percent of the fertilizers consumed worldwide, according to FAOSTAT data, while five companies account for 45 percent of the global fertilizer market.132 There is ample scope to increase R&D with a view to developing efficient local alternative fertilizers with a higher organic content. This development can draw on locally generated information on local soil and crop characteristics (including by using digital technologies to generate soil maps and estimate nutrient needs), and on techniques to increase soil health.133 Initial investment in R&D in this area can incentivize the private sector in Central America to develop fertilizer solutions that can be deployed at scale. More Comprehensive Long-term Investment 5.22 Resilient food-security strategies need to be built around a comprehensive long-term investment strategy in order to generate increased productivity, higher incomes for farmers, and enhanced climate resilience. High economic returns on public investments can be achieved by focusing public support on securing land rights; improving water resources management and access to irrigation; strengthening soil and forestry management; enhancing access to finance and to markets (including high-value export markets), and investing in research and development – and extension – of locally-adapted climate-smart agriculture technologies. 5.23 In particular, increased investment in critical infrastructure can play a major role in reducing post-farmgate transport costs and food losses, and thereby increase the availability and affordability of food for consumers.134 Post-farmgate value added can be significant relative to farmgate prices: for example, in the case of tomatoes in El Salvador, the farmgate price is only 20 percent of the final exporter price, with post-farm agrologistics accounting for the remaining 80 percent. In the case of Guatemala, product losses account for more than half (54 percent) of the total production of tomatoes, and one-third (34 percent) of the total production of maize. Reducing food loss and waste associated with post-farm food loss and waste, notably during the transport, storage and handling stages of the value chain – especially via improved cold chain management – would also reduce carbon footprints and increase resilient development. Moreover, both travel times and costs are substantially affected by the poor condition of road networks in Central America, especially by the lack of adequate all-weather feeder roads in rural areas, which results in road freight accounting for a high share of food costs, so that investment in improving road infrastructure can greatly enhance returns to producers, reduce food losses and improve prices for consumers. 5.24 Finally, investing in measures to ensure efficient inspection and strengthening of animal health, biosafety and food safety can greatly reduce costs and risks for producers and consumers. Smaller-scale agri-food producers are more likely to make progress in improving food safety levels to the extent that they are part of the formal control system and receive direct specific assistance, training and financial support to meet food safety standards. Simplification and harmonization of regulations and processes is important to ensure that costs to participants in agri-food value chains, in terms of time and money, are maintained as low as possible while ensuring that standards are met. The System for Mutual Recognition of Sanitary Records (SIRSS), 132  The five companies are Nutrien, Yara, Mosaic, EuroChem and Israel Chemicals. See ECLAC, FAO and WFP (2022). 133  See FAO (2022). 134  See the World Bank (2022b), for a thorough analysis and data on agrologistics in Northern Central America. 82 with its harmonized system of sanitary and phyto-zoo-sanitary norms, represents an important step to facilitate regional market integration at lower costs while ensuring safety standards.135 B. CONCLUDING THOUGHTS 5.25 Looking ahead and beyond the current food crisis, it is clear that redirecting public support towards the provision of public goods and services will contribute to constructing a solid basis to attenuate the severity of impacts of future crises – or to reduce their likelihood of occurring. Moreover, while trade and recurrent expenditure measures (subsidies) have been favored in the course of the recent crises, it is essential to direct more resources to public investment in areas that can increase agricultural productivity and resilience, reduce food losses, and improve domestic markets, for example, improved water resources management, soil improvement, feeder roads, (cold) storage and warehousing facilities, as well as improved ports handling. Such investments can serve to mitigate not only against acute risks arising from shocks in international markets, but also the more chronic risks associated with climate change. 5.26 Beyond national boundaries, coordinating and promoting a regional agenda for food security can also play a major role strengthening agricultural production and resilience, as well as food security in Central America. The Central American Integration System (SICA) offers a valuable potential platform for coordinating regional policy initiatives and maintaining consistent, best practice approaches, and achieving synergies across the various countries’ initiatives, notably in areas involving general services support. Moreover, regional data coordination via SICA would allow for the improved collection of agrometeorological, climate, production, prices, and food security data to enable both policymakers and producers to improve planning and risk management. 5.27 Finally, intra- and inter-regional collaboration can be strengthened in order to increase transparency and efficiency in food and fertilizer trade. In particular, systems such as the Agricultural Market Information System (AMIS) to assess global food supplies and coordinate policy action in times of market uncertainty (see Annex 6) should continue to be supported, while alternative mechanisms for financing imports, such as import facilities, could be set up for highly indebted countries.136 135  The Secretariat for Central American Economic Integration (SIECA) announced the initiative in 2016 and with financial support from the United States Agency for International Development (USAID) and the World Bank, and the system is now in the early phases of operation. See World Bank (2022b) and https://www.sirrs.sieca.int/ 136  See ECLAC, FAO and WFP (2022). 83 REFERENCES » Abokyi, E., Strijker, D., Asiedu, K.F., & Daams, M. (2020). The Impact of Output Price Support on Smallholder Farmers’ Income: Evidence from Maize Farmers in Ghana. Elsevier Ltd. / Heliyon 6 (2020) e05013. https:// www.sciencedirect.com/science/article/pii/S2405844020318569 » Alper, E., Hobdari, N.A., & Uppal, A. (2016). Food Inflation in Sub- Saharan Africa: Causes and Policy Implications. Working Paper No. 2016/247. International Monetary Fund: Washington, DC. » Anderson, K., & Valdés, A. (2008). Distortions to Agricultural Incentives in Latin America. The World Bank: Washington, DC » Anríquez, G., Foster, W., Ortega, J., Falconi, C., & De Salvo, C.P. (2016). Public Expenditure and the Performance of Latin American and Caribbean Agriculture. Working Paper 722, Inter-American Development Bank, Washington, DC. » Arias, D., & De Franco, M. (2011). Integrating Central American and international food markets: an analysis of food price transmission in Honduras and Nicaragua. LCSSD Occasional Paper Series on Food Prices, No. 75643. Washington, DC: The World Bank. » Arouna, A., Cichler, J., & Lokossou, J. (2021). Contract Farming and Rural Transformation: Evidence from a Field Experiment in Benin. Journal of Development Economics 151. Elsevier. https://www.sciencedirect.com/science/article/pii/S0304387821000031 » Baffes, J., & Gardner, B. (2003). The transmission of world commodity prices to domestic markets under policy reforms in developing countries. The Journal of Policy Reform, 6(3), 159–180. » Bauldry, S. (2015). Structural Equation Modeling. In J. D. Wright (Ed.), International Encyclopedia of the Social & Behavioral Sciences. 615-620. » Bello, O., & Peralta, L. (coords.). (2021). Evaluación de los efectos e impactos de las depresiones tropicales Eta y Iota en Guatemala. » Benson, T., Minot, N., Pender, J., Robles, M., & von Braun, J. (2008). Global Food Crises - Monitoring and Assessing Impact to Inform Policy Reponses. IFPRI. Washington, DC » Bermeo, S., & Leblanc, D. (2021). Honduras Migration: Climate Change, Violence, & Assistance Policy Brief. Duke Sanford Center for International Development. https://dcid.sanford.duke. edu/wp-content/uploads/sites/2/2021/03/Honduras-Migration-Policy-Brief-Final.pdf » Bert, F., Palmieri Reymond, V., & Ruiz, C. (2022). Digitalización agroalimentaria: Conceptos e ideas para la acción emergentes de diálogos amplios. Instituto Interamericano de Cooperación para la Agricultura (IICA). https://repositorio.iica.int/handle/11324/20756 » Brock, A., & Paasch, A. (2009). International Responses to the Food Crisis. FIAN International. Germany. » Brown, L. (2011). The Great Food Crisis of 2011. Foreign Policy. January 10, 2011. https://foreignpolicy.com/2011/01/10/the-great-food-crisis-of-2011/ 84 » Bumb, B., Johnson, M., & Fuentes, P. (2011). Policy options for improving regional fertilizer markets in West Africa. IFPRI Discussion Paper 1084. International Food Policy Research Institute, Washington, DC. » Caporin, M. & Malik, F. (2020). Do structural breaks in volatility cause spurious volatility transmission? Journal of Empirical Finance, 55, 60-82. » Carter, M.R., Laajaj, R., & Yang, D. (2013). The Impact of Voucher Coupons on the Uptake of Fertilizer and Improved Seeds: Evidence from a Randomized Trial in Mozambique. American Journal of Agricultural Economics - Papers and Proceedings. https://www.povertyactionlab.org/sites/default/files/research- paper/283_384%20voucher%20coupons%20in%20Mozambique%20Apr2013.pdf » Ceballos, F., Hernandez, M.A., Minot, N. & Robles, M. (2017). Grain price and volatility transmission from international to domestic markets in developing countries. World Development, 94, 305-320. » Ceballos, F., Hernandez, M.A., & Paz, C. (2022). COVID-19 and extreme weather: Impacts on food security and migration attitudes in rural Guatemala. IFPRI Discussion Paper #2126, Washington, DC. » Ceballos, F., Hernandez, M.A., & Paz, C. (2021). Short-term impacts of COVID-19 on food security and nutrition in rural Guatemala: Phone-based farm household survey evidence. Agricultural Economics 52(3): 477-494 » CEPAL, FAO, & IICA. (2021). Perspectivas de la Agricultura y el Desarrollo Rural en las Américas – Una mirada hacia América Latina y el Caribe. Santiago, Chile. » CEPAL, FAO, & IICA. (2012). Respuestas de los países de América Latina y el Caribe al alza y volatilidad de los precios de los alimentos y opciones de colaboración. Boletín 1/2012. Santiago, Chile. » Cuevas-Reyes, V. (2018.) Impact Evaluation on Productivity and Profitability of Technology in the Double-purpose Bovine System of the Mexican Tropic. Tropical and Subtropical Agroecosystems 21 (2018). https://www.researchgate.net/publication/327117555_ EVALUATION_OF_IMPACT_ON_PRODUCTIVITY_AND_PROFITABILITY_OF_TECHNOLOGY_ IN_THE_BOVINE_SYSTEM_OF_DOUBLE_PURPOSE_OF_THE_MEXICAN_TROPIC » Davis, K., Nkonya, E., Kato, E., Mekonnen, D., Odendo, M., Miiro, R., & Nkuba, J. (2010). Impact of Farmer Field Schools on Agricultural Productivity and Poverty in East Africa. Discussion Paper 00992. IFPRI. Washington, DC https://ebrary. ifpri.org/utils/getfile/collection/p15738coll2/id/2348/filename/2349.pdf » Deza, M., & Ruiz-Arranz, M. (2022). Food Security in Central America, Panama, the Dominican Republic, Mexico and Haiti. Outlook and policy responses aimed at guaranteeing food security in the region. Washington DC: Inter-American Development Bank. https://publications.iadb.org/es/panorama-y-perspectivas-de- la-seguridad-alimentaria-en-centroamerica-panama-republica-dominicana » Díaz-Bonilla, E. (2016). Volatile volatility: Conceptual and measurement issues related to price trends and volatility. In Food price volatility and its implications for food security and policy, eds. Chapter 2, 35-57. 85 » Díaz-Bonilla, E. (2021). The Role of Institutional, Technological and Financial Innovation in the Transformation of Food Systems in Latin America and the Caribbean, in the Context of the 2030 Agenda and the COVID-19 Pandemic. In Graziano da Silva, et.al. 2021. Food Systems in Latin America and the Caribbean – Challenges in a Post-pandemic World. Panama City, FAO and CIDES. » Dong, D., Stewart, H., Dong, X., & Hahn, W. (2022). Quantifying Consumer Welfare Impacts of Higher Meat Prices During the COVID-19 Pandemic, Economic Research Report 306, Economic Research Service, U.S. Department of Agriculture. » Eckstein, D., Künzel, V., & Schäfer, L. (2021). Global Climate Risk Index 2021: Who Suffers Most from Extreme Weather Events? Weather- Related Loss Events in 2019 and 2000-2019. Germanwatch. » ECLAC, FAO & WFP. (2022). Towards Sustainable Food and Nutrition Security in Latin America and the Caribbean in Response to the Global Food Crisis. Santiago, Chile. » Engle, R. (2002). Dynamic conditional correlation-a simple class of multivariate GARCH models. Journal of Business and Economic Statistics 20, 339–350. » Engle, R., & Kroner, F.K. (1995). Multivariate simultaneous generalized ARCH. Econometric Theory 11, 122–150. » Fafchamps, M. (1992). Cash Crop Production, Food Price Volatility and Rural Market Integration in the Third World. American Journal of Agricultural Economics, 74(1): 90–99. » FAO. (2023). Central America and the Caribbean - Prices of maize and beans rose in most countries and remained above their February 2022 levels. Food price monitoring analysis 10/03/2023. https://www.fao.org/giews/food-prices/regional-roundups/detail/en/c/1633792/ » FAO. (2023b). World Food Situation - FAO Food Price Index. June 2, 2023. https://www.fao.org/worldfoodsituation/foodpricesindex/en/ » FAO. (2023c). FAOSTAT-Food balances (2010-). https://www.fao.org/faostat/en/#data/FBS. » FAO. (2023d). Food Security and Agriculture in Central America - Stock-taking Report of Policy Responses to Food Crises. L. Dias Pereira & S. Salcedo. Mimeo. March 4, 2023. » FAO. (2022). “Alternatives for the sustainable management of soil fertilization and plant nutrition”, Santiago. https://www.fao.org/3/cc0964es/cc0964es.pdf. » FAO. (2022b). The State of Food and Agriculture 2022. Leveraging automation in agriculture for transforming agrifood systems. https://www.fao.org/documents/card/en/c/cb9479en » FAO. (2021). Hacia una Agricultura Sostenible y Resiliente en América Latina y el Caribe: Análisis de siete trayectorias de transformación exitosas. Santiago, Chile. » FAO. (2016). Purchase for progress Country Case Study: El Salvador. https://www.fao.org/3/i5241e/i5241e.pdf » FAO. (2016b). Superación del Hambre y de la Pobreza Rural – Iniciativas Brasileñas. Brasilia, Brazil. https://www.fao.org/publications/card/es/c/ea9eb346-7e24-4059-ac64-e1f8b0ba66ef/ 86 » FAO. (2015). Las Compras Públicas a la Agricultura Familiar y la Seguridad Alimentaria y Nutricional en América Latina y el Caribe – Lecciones Aprendidas y Experiencias. Santiago, Chile. https://www.fao.org/3/i4902s/i4902s.pdf » FAO. (2014). Results-Based Public Management: Tools for the Design and Implementation of Public Rural Development Programs with a Project Cycle Approach. Regional Office for Latin America and the Caribbean. Santiago, Chile. https://www.fao.org/ americas/publicaciones-audio-video/gestionpublicaconbaseenresultados/en/ » FAO. (2011). The State of Food Insecurity in the World 2011 – How does International Price Volatility Affect Domestic Economies and Food Security? Rome. » FAO. (2009). High food prices and the food crisis – experiences and lessons learned. FAO: Rome, Italy. » FAO. (2007). Perspectivas Alimentarias. November 2007. Rome, Italy: FAO. » FAO. (2006). Food Policy Brief. https://www.fao.org/fileadmin/templates/ faoitaly/documents/pdf/pdf_Food_Security_Cocept_Note.pdf. » FAO & Agencia Mexicana de Cooperación Internacional para el Desarrollo. (2021). Mesoamérica sin Hambre: Cooperación y Políticas para la Seguridad Alimentaria. Informe Nicaragua. https://www.fao.org/in-action/mesoamerica-sin-hambre/elprograma/general/es/ » FAO, & ECLAC. (2020). Food systems and COVID-19 in Latin America and the Caribbean: Update on impacts and responses. Bulletin 16. https://www.fao.org/3/cb1433en/CB1433EN.pdf » FAO, FIDA, OPS, PMA, & UNICEF. (2023). Panorama regional de la seguridad alimentaria y nutricional – América Latina y el Caribe 2022: hacia una mejor asequibilidad de las dietas saludables. Sant5iago de Chile. https://doi.org/10.4060/cc3859es » FAO, IFAD, PAHO, UNICEF, & WFP. (2023). Latin America and the Caribbean: Regional Overview of Food Security and Nutrition 2022: Statistics and Trends. https://www. fao.org/3/CC2314EN/online/sofi-statistics-rlc-2022/key-messages-ch-1.html » FAO, IFAD, UNICEF, WFP, & WHO. (2023). The State of Food Security and Nutrition in the World 2023. Repurposing food and agricultural policies to make healthy diets more affordable. Rome, FAO. » FAO, & Innovations for Poverty Action (IPA). (2022). Prices and farmer investment – Evidence from experimental studies. Investment Brief. Rome, FAO. https://doi.org/10.4060/cc3479en » FAO, UNICEF, & Oxford University Press. (2016). From Evidence to Action, The Story of Cash Transfers and Impact Evaluation in Sub-Saharan Africa. Oxford University Press. https://www.fao.org/3/i5157e/i5157e.pdf » FEWS NET. (2023). Global Food Assistance Outlook Brief. February 2023. https://fews.net/global/food-assistance-outlook-brief/february-2023 » FEWS NET. (2022). Central America and Caribbean Price Bulletin July 2022. In FEWS NET Central America. https://fews.net/sites/ default/files/documents/reports/PB_LAC_202207_EN.pdf. 87 » FEWS NET. (2021). Regional supply and market outlook - Central America. November 2021. https://fews.net/sites/default/files/ documents/reports/LAC_Market_Supply_Outlook_20211112.pdf. » Flores, L., Perego, V., & Arias, D. (2022). De la planta al tenedor: una evaluación rápida de la crisis de fertilizantes y alimentos en Centroamérica. https://blogs.worldbank.org/es/ latinamerica/de-la-planta-al-tenedor-una-evaluacion-rapida-de-la-crisis-de-fertilizantes-y » Flores Agreda, R., Muñoz, G., De Salvo, C., & Alcaraz Rivero, A. (2020). Análisis de políticas agropecuarias en Centroamérica y República Dominicana. Informes de política agropecuaria. Washington DC: Inter-American Development Bank. » García-Jiménez, H. & Gandlgruber, B. (2014). Gobernanza y acuerdos institucionales en las cadenas del frijol y del maíz en Centroamérica. CEPAL Documento de Trabajo, México D.F. » Gardebroek, C., Hernandez, M. A., & Robles, M., (2015). Market interdependence and volatility transmission among major crops. Agricultural Economics 47(2), 141–155. » Gautam, M., Laborde, D., Mamun, A., Martin, W., Pineiro, V., & Vos, R. (2022). Repurposing Agricultural Policies and Support: Options to Transform Agriculture and Food Systems to Better Serve the Health of People, Economies, and the Planet. Washington, DC: The World Bank. » Gibbons, Maria Elena, Alessandro Maffioli and Martin A. Rossi. 2016. Money for Wine? Complementarities in the Provision of Private and Public Goods to Wine Producers. IDB Working Paper Series 689. Inter-American Development Bank. Washington, DC https://publications.iadb.org/en/money-wine- complementarities-provision-private-and-public-goods-wine-producers » Giordani, P. E., Rocha, N., & Ruta, M. (2016). Food prices and the multiplier effect of trade policy. Journal of International Economics, 101: 102–122. https://doi.org/10.1016/j.jinteco.2016.04.001. » Global Network against Food Crises, & Food Security Information Network. (2022). 2022 Global Report on Food Crises - Joint Analysis for Better Decisions. https://docs.wfp.org/api/documents/WFP-0000138913/download/?_ ga=2.3960176.1144991785.1687387872-1635469461.1687387872 » Global Nutrition Report. (2023). Honduras Profile. https://globalnutritionreport.org/ resources/nutrition-profiles/latin-america-and-caribbean/central-america/honduras/ » Global Nutrition Report. (2023b). Guatemala Profile. https://globalnutritionreport.org/ resources/nutrition-profiles/latin-america-and-caribbean/central-america/guatemala/ » Götz, L., Qiu, F., Gervais, J.P., & Glauben, T. (2016). Export Restrictions and Smooth Transition Cointegration: Export Quotas for Wheat in Ukraine. Journal of Agricultural Economics, 67(2), 398-419. » Graziano da Silva, J., Jales, M., Rapallo, R., Díaz-Bonilla, E., Girardi, G., del Grossi, M., Luiselli, C., Sotomayor, O., Rodríguez, A., Rodrigues, M., Wander, P., Rodríguez, M., Zuluaga, J., & Pérez, D. (2021). Food Systems in Latin America and the Caribbean – Challenges in a Post-pandemic World. Panama City, FAO and CIDES. https://doi.org/10.4060/cb5441en 88 » Gruhn, P., Goletti, F., & Roy, R. (1995). Proceedings of the IFPRI/FAO Workshop on Plant Nutrient Management, Food Security and Sustainable Agri-culture: The Future through 2020. IFPRI and FAO, Viterbo, Italy. » Guo, J., & Tanaka, T. (2020). Dynamic transmissions and volatility spillovers between global price and U.S. producer price in agricultural markets. Journal of Risk and Financial Management, 13(4), 83. » Headey, D., & Ruel, M. (2022). The global food price crisis threatens to cause a global nutrition crisis: New evidence from 1.27 million young children on the effects of inflation. IFPRI Blog: Research Post. December. https://www.ifpri.org/blog/global- food-price-crisis-threatens-cause-global-nutrition-crisis-new-evidence-127-million. » Headey, D., & Ruel, M. (2023). Food inflation and child undernutrition in low and middle income countries. Nature Communications 14: 5761. https://doi.org/10.1038/s41467-023-41543-9 » Hicks, J. R. (1942). Consumers’ surplus and index-numbers. Review of Economic Studies 9(2):126–137. » IDB. (2010). Assessing the Effectiveness of Agricultural Interventions. Washington, DC: IDB. » IFPRI. (2023). Food & Fertilizer Export Restrictions Tracker. https://public. tableau.com/app/profile/ifpri.food.security.portal/viz/shared/2CPYTB4G8 » IICA. (2023). Aumenta 137% el valor de las importaciones de fertilizantes químicos de América Latina y el Caribe en 2022. Blog del IICA. https://blog.iica.int/blog/aumenta- 137-valor-las-importaciones-fertilizantes-quimicos-america-latina-caribe-en-2022. » IICA. (2022). Observatory of Public Policies for Agrifood Systems. https://www.iica.int/en/press/news/iica-launches-observatory-public- policies-agrifood-systems-space-dialogue-and-proposals » IICA. (2013). Estudio de las Cadenas de Valor Maíz Blanco y Frijol en Centroamérica. » IMF. (2022). Tracking Economic and Financial Policies During COVID-19: An Announcement- Level Database. https://www.imf.org/en/Publications/WP/Issues/2022/06/03/Tracking- Economic-and-Financial-Policies-During-COVID-19-An-Announcement-Level-Database-518896 » Insight Crime. (2021). Country Profile – Honduras. https://insightcrime. org/honduras-organized-crime-news/honduras/ » International Coffee Organization. (2022). International Coffee Organization - About Statistics of the International Coffee Organization. https://www.ico.org/about_statistics.asp » International Food Policy Research Institute (IFPRI). (2023). Price and Volatility Transmission from International to Domestic Food and Fertilizer Markets in Central America. Draft Report. June 15, 2023. IFPRI. » International Trade Center. (2021). Trademap, list of exporters for the selected products in 2021: 0803 Bananas, incl. plantains, fresh or dried. https://www.trademap.org/. » IPC. (2023). Dominican Republic: IPC Acute Food Insecurity Analysis. October 2022- June 2023. January 2023. https://www.ipcinfo.org/fileadmin/user_upload/ipcinfo/docs/ IPC_DominicanRepublic_AcuteFoodInsecurity_Oct2022June2023_Summary_English.pdf 89 » IPC. (2023b). Honduras: Acute Food Insecurity Situation December 2022 - February 2023 and Projections for March - May 2023 and June - August 2023. https:// www.ipcinfo.org/ipc-country-analysis/details-map/en/c/1156257/?iso3=HND » IPC. (2023c). Honduras: Chronic Food Insecurity Situation 2018-2023. https:// www.ipcinfo.org/ipc-country-analysis/details-map/en/c/1143346/?iso3=HND » Ivanic, M., Martin, W., & Zaman, H. (2011). Estimating the short-run poverty impacts of the 2010–11 surge in food prices. Policy Research Working Paper 5633, The World Bank. » Jacquet, P. (2016). Food Price Volatility: An International Policy Challenge? Chapter in: Pattillo, C.A., Quintyn, M.G., & Zhu, M. (2012). Commodity Price Volatility and Inclusive Growth in Low-Income Countries. Wasgington, DC: International Monetary Fund. » Jayne, T.S. & Rashid, S. (2013). Input Subsidy Programs in Sub-Saharan Africa: A Synthesis of Recent Evidence. IFPRI. Washington, DC https://ebrary.ifpri. org/utils/getfile/collection/p15738coll5/id/4128/filename/4129.pdf » Kalkuhl, M., Von Braun, J., & Torero, M. (2016). Food Price Volatility and Its Implications for Food Security and Policy. Springer International Publishing. » Koop, G., Leon-Gonzalez, R., & Strachan, R.W. (2011). Bayesian inference in a time varying cointegration model. Journal of Econometrics, 165(2), 210-220. » Krivonos, E., & Dawe, D. (2014). Policy Responses to High Food Prices in Latin America and the Caribbean – Country Case Studies. Rome, Italy: FAO. » Krozer, Alice. (2010). A Regional Basic Income: Towards the Eradication of Extreme Poverty in Central America. Economic Commission for Latin America and the Caribbean (ECLAC). Mexico City, Mexico: United Nations. https:// www.cepal.org/sites/default/files/publication/files/25938/lcmexl998.pdf » Lachman, J., & López, A. (2019), “Innovation obstacles in an emerging high tech sector: The case of precision agriculture in Argentina”, Management Research, 17(4), 474-493. https://doi.org/10.1108/MRJIAM-11-2018-0883 » Lachman, J., López, A., Tinghitella, G., & Gómez-Roca, S. (2021). Las Agtech en Argentina: desarrollo reciente, situación actual y perspectivas (N° 2021- 57). Universidad de Buenos Aires, Facultad de Ciencias Económicas, Instituto Interdisciplinario de Economía Política (IIEP-BAIRES). » Liu, Y., Ma, X., Shu, L., Hancke, G., & Abu-Mahfouz, A. (2021). From Industry 4.0 to Agriculture 4.0: current status, enabling technologies, and research challenges. IEEE Transactions on Industrial Informatics, 17(6), 4322-4334. » Lopez, R., & Galinato, G.I. (2007). Should Governments Stop Subsidies to Private Goods? Evidence from Latin America. Journal of Public Economics 91: 1071–94. » Lopez, C.A., Salazar, L., & De Salvo, C.P. (2017). Public Expenditures, Impact Evaluations and Agricultural Productivity. Technical Note IDB- TN-1242, Inter-American Development Bank, Washington, DC. 90 » Lopez, C.A., Salazar, L., & De Salvo, C.P. (2017b). Agricultural Input Subsidies and Productivity: The Case of Paraguayan Farmers. IDB Working Paper Series 802. Inter-American Development Bank. Washington, DC https://publications.iadb.org/publications/english/ viewer/Agricultural-Input-Subsidies-and-Productivity-The-Case-of-Paraguayan-Farmers.pdf » Maffioli, A., & Mullally, C. (2014). The Impact of Agricultural Extension for Improved Management Practices: An Evaluation of the Uruguayan Livestock Program. IDB Working Paper Series 485. Inter-American Development Bank. Washington, DC https://publications. iadb.org/publications/english/viewer/The-Impact-of-Agricultural-Extension-for-Improved- Management-Practices-An-Evaluation-of-the-Uruguayan-Livestock-Program.pdf » Maffioli, A., Ubfal, D., Vazquez-Bare, G., & Cerdan-Infantes, P. (2013). Improving Technology Adoption in Agriculture through Extension Services: Evidence from Uruguay. World Bank. Journal of Development Effectiveness 5(1). https://openknowledge.worldbank. org/server/api/core/bitstreams/898c5ce7-22b3-547b-8701-6aae23d84e97/content » Magrini, E., Morales Opazo, C., & Baile, J. (2015). Price shocks, volatility, and household welfare: A cross-country inquiry. Scientific paper no. 11. ULYSSES project. EU 7th Framework Programme, Project 312182 KBBE.2012.1.4-05. » Ministerio do Desenvolvimento Social e Combate à Fome - Brazil. (2013). PAA: 10 Anos de Aquisição de Alimentos. Brasilia, Brazil. » Ministerio de Economía de la Nación - Argentina. (2023). Argentina Productiva 2030. Misión 6. Adaptar la producción de alimentos a los desafíos del siglo XXI. Plan para el Desarrollo Productivo, Industrial y Tecnológico. » Mittal, S., Hariharan, V.K., & Subash, S. (2018). Price volatility trends and price transmission for major staples in India. Agricultural Economics Research Review, 31(1). » Nzuma, J. M., & Kirui, P.K. (2021). Transmission of global wheat prices to domestic markets in Kenya: A cointegration approach. African Journal of Agricultural and Resource Economics, 16(1), March. » Observatory of Economic Complexity (OEC). (2022). Costa Rica (CRI) Exports, Imports, and Trade Partners. https://oec.world/en/profile/country/cri. » OECD. (2022). Agricultural Policy Monitoring and Evaluation 2022 – Addressing the Challenges Facing Food Systems. Paris, France: OECD. » OECD. (2022b). Agricultural Policy Monitoring and Evaluation 2022 – Reforming Agricultural Policies for Climate Change Mitigation. Paris, France: OECD. » OECD. (2016). OECD’s producer support estimate and related indicators of agricultural support. Concepts, calculations, interpretation and use (The PSE Manual). Trade and Agriculture Directorate. Paris: Organisation for Economic Cooperation and Development. » OECD. (2006). The Bioeconomy to 2030: Designing a Policy Agenda. OECD International Futures Project. OECD, Paris, France. https://doi.org/10.1787/9789264056886-en » OECD, & FAO. (2019). OECD-FAO Agricultural Outlook 2019- 2028. OECD Publishing, Paris/ FAO, Rome. 91 » Okou, C., Spray, J.A., & Unsal, F.D. (2022). Staple Food Prices in Sub- Saharan Africa: An Empirical Assessment. Working Paper No. 2022/135. International Monetary Fund: Washington, DC. » Oxfam. (2016). Livelihoods in Democratic Republic of Congo: Impact Evaluation of the Purchase For Progress (P4P) project. Effectiveness Review Series 2015–16. https:// oxfamilibrary.openrepository.com/bitstream/handle/10546/620414/er-livelihoods-drc- effectiveness-review-160218-en.pdf;jsessionid=15D798EC08DB44C63EED36BC25CECE6F? sequence=1#page=5&zoom=100,109,438 » Poulton, C., Kydd, J., Wiggins, S., A. Dorward, A. (2006). State Intervention for Food Price Stabilization in Africa: Can It Work?” Food Policy, 31: 342–56. » Reyes, B., Camilo, K., & Charry, A. (2022). Estudio de la Cadena de Valor de Café en El Salvador – Informe Final de Resultados. Bioversity / Centro Internacional para la Agricultural Tropical (CIAT). » Robles, M., & Keefe, M. (2011). The effects of changing food prices on welfare and poverty in Guatemala. Development in Practice, Taylor & Francis Journals, 21(4-5), 578-589. » Robles, M., & Torero, M. (2010). Understanding the Impact of High Food Prices in Latin America,. Economia Journal, The Latin American and Caribbean Economic Association - LACEA, (0), 117-164. https://ideas.repec.org/a/col/000425/014300.html » Robles, M., & Torero, M. (2009). Assessing the Impact of Increased Global Food Prices on the Poor in Selected Latin American Countries. Unpublished International Food Policy Research Institute report for the Inter-American Development Bank, Washington, DC. » Rosenberg, A. (2014). Ukraine crisis could mean higher meat prices in the US. CNBC (formerly Consumer News and Business Channel). https://www.cnbc. com/2014/03/05/ukraine-crisis-could-mean-higher-meat-prices-in-the-us.html. » Saikkonen, P., & Choi, I. (2004). Cointegrating Smooth Transition Regressions. Econometric Theory, 20(2), 301–340. http://www.jstor.org/stable/3533382. » Schwarz, G. (1978). Estimating the Dimension of a Model. Annals of Statistics, 6(2), 461-464. » Seale, J.L., Regmi, A., & Bernstein, J. (2003). International evidence on food consumption patterns Technical Bulletin No. 33580. United States Department of Agriculture, Economic Research Service. » SEPSA. (2021). Modelo de Costos de Producción: Maiz. http://www.infoagro.go.cr/ EstadisticasAgropecuarias/CostosProduccion/Documents/MAIZ_PacificoCentral_2021.pdf » Statista. (2021). Mercado Libre no para de crecer. 08/04/2021. https:// es.statista.com/grafico/23471/mercado-libre-en-cifras/ » Srinivasan, S., & Rodriguez, A.G. (2016). Pobreza y desigualdades rurales: Perspectivas de género, juventud y mercado de trabajo. https://www.bivica.org/file/view/id/4924 » Svanidze, M., Götz, L., & Serebrennikov, D. (2022). The influence of Russia’s 2010/2011 wheat export ban on spatial market integration and transaction costs of 92 grain markets. Applied Economic Perspectives and Policy 44(2), 1083-1099. » Tenkorang, F., & Lowenberg-DeBoer, J. (2008). Forecasting long-term global fertilizer demand. Nutrient Cycling in Agroecosystems, 83(3), 233–247. » Timmer, C.P. (2010). Reflections on food crises past. Food Policy, 35(1): 1–11. https://doi.org/10.1016/j.foodpol.2009.09.002 » Trejos Solorzano, J.D. (2012). Políticas Publicas contra la pobreza rural en Costa Rica durante el primer decenio del Siglo XXI. Universidad Nacional de Costa Rica. https://iice.ucr.ac.cr/series/serie_15.pdf » Trocaire. (2014). Feeling the Heat: How climate change is driving extreme weather in the developing world. https://www.yumpu.com/en/ document/read/36387564/trocaire-climate-change-report-2014 » UN Food Systems Summit 2021 Scientific Group. (2021). The True Cost and True Price of Food. Scientific Group of the UN Food Systems Summit. S. Hendriks, A. de Groot Ruiz, M. Herrero Acosta, H. Baumers, P. Galgani, D. Mason-D’Croz, C. Godde, K. Waha, D. Kanidou, J. von Braun, M. Benitez, J. Blanke, P. Caron, J. Fanzo, F. Greb, L. Haddad, A. Herforth, D. Jordaan, W. Masters, C. Sadoff, J.-F. Soussana, M.C. Tirado, M. Torero, & M. Watkins. https://sc-fss2021.org/wp-content/uploads/2021/06/UNFSS_true_cost_of_food.pdf » UNCTAD. (2015). When the Tide Goes Out: Capital Flows and Financial Shocks in Emerging Markets. Policy Brief No. 40. December 2015. https:// unctad.org/system/files/official-document/presspb2015d14_en.pdf » UNODC. (2019). Global Study on Homicide 2019. Vienna, United Nations Office on Drugs and Crime. » USA RICE. (2022). Checking in with Central America. USA RICE Website. https://www.usarice.com/news-and-events/publications/usa-rice-daily/ article/usa-rice-daily/2022/12/07/checking-in-with-central-america. » USA RICE. (2018). Central America. USA RICE Website. https://www.usarice.com/ discover-us-rice/international-markets/international-market/central-america » USAID-RED Programa de Diversificacion Economica Rural. (2005). Costos de Produccion: MAIZ DULCE. Boletion de Produccion #11. https://repositorio.credia.hn/bitstream/ handle/123456789/295/costos_de_produccion_maiz_dulce.pdf?sequence=1&isAllowed=y » USDA Economic Research Service. (2023). Feed Grains Sector at a Glance. https://www. ers.usda.gov/topics/crops/corn-and-other-feed-grains/feed-grains-sector-at-a-glance/ » Unidad Técnica de Seguridad Alimentaria y Nutricional. (2022). Estudio de monitoreo de indicadores seguridad alimentaria y nutricional en los 18 departamentos de Honduras Para el año 2022. Gobierno de la Republica de Honduras, Unidad Técnica de Seguridad Alimentaria y Nutricional. https://www.humanitarianresponse.info/sites/www.humanitarianresponse. info/files/documents/files/informe_estudio_indicadores_san_30_mayo.pdf » Varangis, P., Siegel, P., Giovannucci, D., & Lewin, B. (2003). Dealing with the Coffee Crisis in Central America: Impacts and Strategies. World Bank Policy Research Working Paper 2993. Washington, DC: The World Bank. 93 » Von Braun, J., Afsana, K., Fresco L., & Hag Ali Hassan, M. (eds). (2023). Science and Innovations for Food Systems Transformation. Springer. Switzerland. https://link.springer.com/book/10.1007/978-3-031-15703-5 » Vos, R., Glauber, J., Hernández, M., Laborde, D. (2022). COVID-19 and rising global food prices: What’s really happening? Food Security Portal, Facilitated by IFPRI. Blog Post. February. https://www.foodsecurityportal.org/node/1913 » Williamson, J.M. (2011). The Role of Information and Prices in the Nitrogen Fertilizer Management Decision: New Evidence from the Agricultural Resource Management Survey. Journal of Agricultural and Resource Economics, Western Agricultural Economics Association, 36(3), 1-21. » World Bank. (2013). Improving Nutrition Through Multisectoral Approaches. Washington, D.C.: World Bank Group. https://documents1.worldbank.org/curated/en/625661468329649726/ pdf/75102-REVISED-PUBLIC-MultisectoralApproachestoNutrition.pdf » World Bank. (2023). The LAC Response to Rising Prices. As of March 31, 2023. Washington DC: The World Bank. » World Bank. (2023b). Food Security Update. World Bank Response to Rising Food Insecurity. March 13, 2023. https://www.worldbank.org/en/topic/ agriculture/brief/food-security-update?cid=ECR_GA_worldbank_EN_EXTP_ search&gclid=EAIaIQobChMIvNP-jIiJ_QIVuBPUAR3uGAiXEAMYASAAEgLQLvD_BwE » World Bank. (2023c). Food Security Update. February 13, 2023. https://www.worldbank. org/en/topic/agriculture/brief/food-security-update?cid=ECR_GA_worldbank_EN_EXTP_ search&gclid=EAIaIQobChMIvNP-jIiJ_QIVuBPUAR3uGAiXEAMYASAAEgLQLvD_BwE » World Bank. (2023d). Inflation and Food/Fuel Prices Briefs for Costa Rica, the Dominican Republic, El Salvador, Guatemala, Honduras, Nicaragua and Panama. Mimeos. LAC Poverty Unit, The World Bank. » World Bank. (2023e). Integrated Social Protection, Inclusion and Resilience (INSPIRE) Project. Project Appraisal Document. Report No. PAD5236. May 2, 2023. https://documents.worldbank.org/en/publication/documents-reports/ documentdetail/099050523170528239/bosib09a2b67bd0ab08f4c04455d927307f » World Bank. (2022). Commodity Price Data (The Pink Sheet). September 2022. https:// thedocs.worldbank.org/en/doc/5d903e848db1d1b83e0ec8f744e55570-0350012021/ related/CMO-Pink-Sheet-September-2022.pdf » World Bank. (2022a). The LAC response to rising prices. Power Point presentation. November 2022. » World Bank. (2022b). Agrifood Systems in Northern Central America: Agrologistics for Modern Family Farms. Garcia, A., Perego, V.M.E., Prasann, A., Millán, F., Horton, J., Bueso, F.J., Channa, H., Mora, E., Traverso, M.V., Horst, A.C., and Benjamin, M.P. © World Bank. » World Bank. (2022c). Insumos para el Reporte “Promover la Innovación en el Sistema Agroalimentario de Argentina”. August 2022. 94 » World Bank. (2022d). Belize - Climate Resilient and Sustainable Agriculture (CRESAP) Project. Project Appraisal Document. Report No. PAD3775. February 17, 2022. https://documents. worldbank.org/en/publication/documents-reports/documentdetail/220861647442304917/ belize-climate-resilient-and-sustainable-agriculture-project » World Bank. (2022e). Honduras Rural Competitiveness Project (P101209). Implementation Completion and Results Report. Report No.: ICR00005712. H. Channa. Washington, DC: The World Bank. » World Bank. (2011). Food Price Watch. August 2011. https://documents1.worldbank.org/ curated/en/507241468177841474/pdf/638680NEWS0Foo00Box0361531B0PUBLIC0.pdf » World Bank. (2010). Interim Strategy Note for Honduras. Report No. 56405- HN. September 29, 2010. https://documents1.worldbank.org/curated/ en/446111468254054926/pdf/564050ISN0IDA110Box353755B01PUBLIC1.pdf » World Bank. (2010b). Country Partnership Strategy Progress Report for Guatemala. Report No. 55573-GT. https://documents1.worldbank.org/curated/ en/104701468036362311/pdf/555730CASP0p121Official0Use0Only191.pdf » World Bank, Independent Evaluation Group (IEG). (2011). Impact Evaluations in Agriculture: An Assessment of the Evidence. The World Bank. Washington, DC » World Bank & UNDP. (2021). High-Frequency Phone Survey. » World Food Programme. (2022). Global Report on Food Crises - 2022. https://www.wfp.org/publications/global-report-food-crises-2022 » World Food Programme. (2021). School feeding programmes in low- and lower- middle-income countries. A focused review of recent evidence from impact. Rome. https://api.godocs.wfp.org/api/documents/WFP-0000126779/download/?_ ga=2.163473276.1188681069.1680696432-1172878722.1680696432 » World Food Programme. (2008). Alza de Precios, Mercados e Inseguridad Alimentaria y Nutricional en Centroamérica. https://reliefweb.int/report/el-salvador/alza-de-precios- mercados-e-inseguridad-alimentaria-y-nutricional-en-centroamérica?OpenDocument=&_ gl=1%2A1m32xd3%2A_ga%2AMzk1OTcyNTAuMTY4NzQ1NjQxMw..%2A_ga_ E60ZNX2F68%2AMTY4NzQ1NjQxMy4xLjAuMTY4NzQ1NjQxMy42MC4wLjA. » Zavala-Pineda, et.al. (2015). Las transferencias agrarias en México y su efecto en el sector agropecuario. Economía Agraria y Recursos Naturales. https://ageconsearch.umn.edu/record/229591?ln=en 95 ANNEX 1 - ECONOMETRIC MODELLING AND RESULTS A. METHODOLOGY FOR THE ECONOMETRIC ANALYSES IN CHAPTERS II AND III (i) Conditional Mean and Conditional Variance Equations A1.1 The T-BEKK approach involves modeling both a conditional mean equation and a conditional variance-covariance equation for each price return series considered in the analysis. The T acronym refers to the Student’s t density used in the model estimation in order to better control for the leptokurtic distribution of the price returns series. Price returns are defined as , where is the price of a certain product (commodity) in market m at month t, and m = 1 refers to the domestic market while m = 2 refers to the international market. The logarithmic transformation is a standard measure for net returns in a market and is generally applied in empirical finance to obtain a convenient support for the distribution of the error term in the estimated model. A1.2 To determine the appropriate model for the conditional mean process, the analysis first tests for the presence of cointegration between domestic and international (log) prices using the Johansen trace test, with the number of lags (k) selected based on the Schwarz Bayesian information criterion (SBIC). For those cases where the pair of price returns is not found to be co-integrated, the conditional mean equation is simply modeled as a VAR process such that: (1a) where is a 2 ´ 1 vector of price returns for the corresponding product (commodity) in the domestic and international market at month t; that is, ; is a 2 ´ 1 vector of constants; , s = 1, …, k, are 2 ´ 2 matrices of parameters capturing own and cross lead-lag relationships between markets at the mean level; and is a 2 ´ 1 vector of innovations with zero mean, conditional on past information , and conditional variance-covariance matrix . A1.3 For those cases where the pair of price returns are found to be co-integrated, the conditional mean equation is modeled as a vector-error correction (VEC) model such that: (1b) where is the lagged error correction term resulting from the co-integration relationship; that is, , and is a 2 ´ 1 vector of parameters that measure the adjustment of each (log) price series to deviations from the long-run equilibrium. A1.4 The conditional variance-covariance matrix at time t (with one time lag) is, in turn, given by: (2) 96 where is a 2 ´ 2 upper triangular matrix of constants , A is a 2 ´ 2 matrix whose elements capture the direct effect of an innovation in market on the current price return volatility in market , and is a 2 ´ 2 matrix whose elements measure the direct influence of past volatility in market on the current volatility in market (persistence). Expanding Equation 2, the resulting conditional variance equation for the domestic market is defined as (3) A1.5 This variance-covariance specification allows us to characterize the magnitude and persistence of volatility transmission from international to domestic markets. (ii) Degree of Price and Volatility Transmission A1.6 As the main interest is in the degree of price transmission in the short run, the analysis focuses on the domestic price return response after one period (one month) to a shock (the impulse) in the international price return. The size of the shock is equal to one standard deviation of the international price return and the response is reported as a fraction of the shock for comparison across commodities. Given this standardization, the resulting measure is equivalent to the elasticity of price transmission from the international market to the domestic market. A1.7 Given that the methodology is based on reduced-form models, the estimation errors must be properly orthogonalized so as to obtain valid, causal impulse response functions to a given one-time shock in one market. A Cholesky decomposition is used for this, implicitly assuming away the possibility of contemporaneous transmission of shocks from the domestic market into the international market (while fully allowing for contemporaneous transmission of international shocks into the domestic market). In addition, an impulse-response function for the estimated conditional volatility is derived, to assess how a shock or innovation in the international market transmits to the domestic market and obtain the elasticity of domestic price volatility with respect to international price volatility. To do this, the following two steps are carried out for each estimated model (one per country/commodity): 1. Estimate the size of a shock in the international market such that the steady-state variance of the international price return increases by one percent after one period: 2. Introduce shock in expression (2) and estimate the percentage change in the variance of the domestic price return (with respect to its steady-state value) and compute the volatility transmission VT indicator according to: A1.8 In other words, the volatility transmission indicator shows the reaction (after one period and assuming the system is on steady state) of the domestic price return variance compared to the reaction of the international price return variance to a shock in the international market. If the volatility transmission indicator is equal to one it means that the domestic price return variance increases in one period in the same proportion as the international price return variance, after introducing a shock in the international market. 97 (iii) Conditional Correlations A1.9 The interdependence between international and domestic prices is examined by recovering the time-varying conditional correlations that result from the T-BEKK model. These correlations are obtained using the following formula: (4) where: » is the dynamic conditional correlation of domestic price and international price in period . » is the conditional covariance equation of domestic price and international price in period . » is the conditional variance equation of domestic price in period . » is the conditional variance equation of international price in period . (iv) Compensating Variation A1.10 The compensating variation is calculated using the following matrixial formula per food staple and rural or urban area within each country: (5) where: » represents the direct effect which is the immediate change on a household’s purchasing power after the increase of the international price and depends on whether the household is a net buyer or net seller of the food staple. » represents the substitution effect that occurs when the household exchange the increased price good group for other food staples with unchanged prices. » is the vector with consumption shares according to food staple and area of the country. » is a diagonal matrix with the consumption shares according to food staple and area of the country. » is the vector with production shares according to food staple and area of the country. » is a vector with price changes, which is equal to the domestic price variation resulting from the increase of 20% in the international price and 0% in the case of the food staples not being analyzed. » is the matrix of demand and cross elasticities according to food staple and area of the country. » is the amount of total household expenditure according to the area of the country. A1.11 This formula assumes that supply elasticities are zero, which means that there is no supply response after an increase of international prices. (v) Descriptive Statistics for Price Returns in this Analysis A1.12 Table A.1 below provides basic descriptive statistics for the domestic and international price returns used in the analysis. Several important patterns are worth noting. First, the results from the Jarque-Bera test suggests that none of the domestic or international returns follow a normal distribution. Additionally, the kurtosis of all analyzed markets is greater than 3, indicating 98 that the returns are leptokurtic in nature. Similarly, most skewness coefficients are close to zero, suggesting symmetrical distributions. These results confirm the appropriateness of assuming a Student’s t density for the estimation of the BEKK models. A1.13 Second, the Ljung–Box (LB) and Engle ARCH Lagrange Multiplier (LM) statistics for up to 6 and 12 lags generally reject the null hypothesis that the squared returns have no autocorrelation or no autoregressive conditional heteroscedastic effects. This autocorrelation in squared returns suggests a nonlinear relationship between some of the price returns, which, together with the observed high fluctuations in the returns’ series, motivates the use of MGARCH models that account for heteroskedasticity and volatility clustering in the modeled data. A1.14 Third, the Augmented Dickey-Fuller (ADF) tests show that most domestic and international prices (in natural logarithms) are non-stationary, indicating the need to work with their first difference (i.e., the returns) in the estimation models. Indeed, the same ADF test on domestic and international price returns indicates that all of these are stationary, allowing their conditional means to be modelled through the VAR and VEC approaches. As explained in the methodology section, the choice of one or the other model is determined through a cointegration test, to determine whether a vector error-correction model is needed to account for a potential long-run relationship between the domestic and international prices. 99 Table A1.1: Summary Statistics And Selected Normality, Autocorrelation And Stationary Tests Panel A. Domestic Price Series Costa Dominican   El Salvador Guatemala Honduras Nicaragua Panama Total Rica Republic Number of domestic price series 6 3 2 3 7 2 3 26 Mean price returns 18.24% 0.29% 0.36% 0.38% 3.11% 0.35% 0.30% 5.21% % of series with kurtosis > 3 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% % of series rejecting Jarque-Bera test’s H0 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% % of series rejecting Ljung-box test’s H0 on squared returns (6 lags) 83.3% 100.0% 100.0% 66.7% 42.9% 50.0% 100.0% 73.1% % of series rejecting Ljung-box test’s H0 on squared returns (12 lags) 83.3% 100.0% 50.0% 100.0% 57.1% 50.0% 100.0% 76.9% % of series rejecting AC Q test’s H0 on squared returns (First lag) 83.3% 100.0% 100.0% 66.7% 42.9% 50.0% 100.0% 73.1% % of series rejecting AC Q test’s H0 on squared returns (Second lag) 66.7% 100.0% 50.0% 66.7% 42.9% 50.0% 100.0% 65.4% % of series rejecting ADF test’s H0 - Logarithm of price in levels (6 lags) 33.3% 33.3% 100.0% 0.0% 57.1% 50.0% 100.0% 50.0% % of series rejecting ADF test’s H0 - Price returns (12 lags) 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% Panel B. International Price Series Red Black Rice Maize Banana Coffee Wheat Ammonia Urea DAP beans beans Mean price returns 0.34% -0.01% 0.21% 0.43% -0.81% 0.15% 0.33% 0.76% 0.48% 0.52% Standard deviation of price returns 4.48% 10.22% 11.86% 6.14% 10.55% 5.52% 6.97% 16.57% 9.94% 7.75% Jarque-Bera statistic 180.65* 24.45* 170.44* 25.52* 122.37* 22.02* 486.88* 5961.73* 1734.70* 1169.58* Skewness 0.68 0.09 0.53 -0.10 -0.49 0.58 1.02 -1.76 -0.86 -0.87 Kurtosis 6.73 25.16 6.72 4.48 6.83 3.94 9.41 25.54 15.18 12.95 Ljung-box statistic on squared returns (6 lags) 73.70* 24.45* 26.13* 3.88 22.53* 7.83 29.99* 26.92* 31.13* 48.36* Ljung-box statistic on squared returns (12 lags) 80.34* 25.16* 33.38* 12.75 55.07* 1091.21 30.76* 29.66* 32.35* 61.03* AC Q statistic on squared returns (First lag) 50.97* 14.91* 18.69* 1.66 17.40* 6.88* 3.28* 0.02 4.92* 43.67* AC Q statistic on squared returns (Second lag) 54.07* 14.92* 19.24* 2.00 18.71* 7.02* 28.41* 26.61* 6.19* 47.90* ADF statistic - Logarithm of price in levels (6 lags) -1.89 -3.43* -1.39 -1.68 0.32 -1.56 -2.21 -2.017 -1.84 -2.521 * ADF statistic - Price returns (12 lags) -5.90* -4.80* -8.41* -6.42* -5.31* -5.57* -5.58* -6.086* -5.987* -5.733* Note: This table presents summary statistics and selected normality, autocorrelation, and stationary tests for domestic (Panel A) and international (Panel B) price return series for rice, beans, maize, fertilizers, and other products. An asterisk indicates that the null hypothesis is rejected at a 95% level of confidence. 100 Table A1.2: Selected Domestic And International Price Time Series And Sources of Data Panel A. Selected Domestic Price Series and Data Sources for Central American Countries International Local Country Market Units Price type Start Date End Date Obs. Source International price Commodity product Costa Rica Ammonia Ammonium National CRC / 45kg retail 2008-07 2022-12 174 Consejo Nacional de Ammonia US Gulf nitrate Average Produccion NOLA Costa Rica Black bean Black bean National USD / MT wholesale 2007-09 2016-12 112 CAC - WFP US - Black beans Average Costa Rica Maize Maize (White) National USD / MT wholesale 2007-09 2016-12 112 CAC - WFP Maize (US No. 2, Average Yellow) Costa Rica Rice Rice (80-20) National USD / MT wholesale 2007-09 2016-12 112 CAC - WFP Rice (US Long Grain Average 2.4%) Costa Rica Urea Urea 46% National CRC / 45kg retail 2008-07 2022-12 174 Consejo Nacional de Urea US Gulf NOLA Average Produccion Costa Rica Wheat Wheat (flour) National CRC / 900 retail 2007-01 2022-03 183 Consejo Nacional de Wheat (CWRS) - Average gms Produccion - FAOSTAT Canada Dominican Banana Banana Santo DOP / unit retail 2000-01 2022-12 276 Ministerio de Agricultura. US Banana - East Republic Domingo Depto. de Economía Coast Agropecuaria Dominican Black bean Black bean Santo DOP / pound retail 2000-01 2022-12 276 Secretaria de Estado de US - Black beans Republic Domingo Agricultura - FAOSTAT Dominican Rice Rice (first Santo DOP /pound retail 2000-01 2022-12 276 Secretaria de Estado de Rice (US Long Grain Republic quality) Domingo Agricultura - FAOSTAT 2.4%) El Salvador Maize Maize (White) San USD / wholesale 2006-01 2022-12 204 Dirección General de Maize (US No. 2, Salvador Spanish Economía Agropecuaria - Yellow) quintal (46kg) FAOSTAT El Salvador Rice Rice San USD / wholesale 2006-01 2022-08 200 Dirección General de Rice (US Long Grain Salvador Spanish Economía Agropecuaria - 2.4%) quintal (46kg) FAOSTAT Guatemala Black bean Black bean National GTQ / pound retail 2001-01 2015-12 180 INE - WFP US - Black beans Average Guatemala Maize Maize (White) La Terminal GTQ / 100 wholesale 1998-01 2022-01 289 INE - WFP Maize (US No. 2, pound Yellow) 101 International Local Country Market Units Price type Start Date End Date Obs. Source International price Commodity product Guatemala Rice Rice (first Guatemala GTQ / wholesale 2000-01 2022-12 276 Ministerio de Agricultura, Rice (US Long Grain quality) City Spanish Ganaderia y Alimentacion - 2.4%) quintal (46kg) FAOSTAT Honduras Ammonia Ammonium National HNL / 43kg wholesale 2006-04 2022-12 201 SIMPAH Ammonia US Gulf nitrate Average NOLA Honduras Coffee Coffee National USD / export 2002-10 2021-09 228 Instituto Hondureño del ICO Composite Average Spanish Cafe Coffee Index quintal (46kg) Honduras Maize Maize (White) Tegucigalpa HNL / wholesale 2007-02 2022-12 191 SIMPAH - FAOSTAT Maize (US No. 2, Spanish Yellow) quintal (46kg) Honduras DAP DAP (18-46- National HNL/ 43kg wholesale 2008-06 2022-12 175 SIMPAH DAP US Gulf NOLA 00) Average Honduras Red bean Red bean Tegucigalpa HNL / wholesale 2007-02 2022-06 185 CAC - WFP Nicaragua - Beans Spanish quintal (46kg) Honduras Rice Rice (second Tegucigalpa HNL / wholesale 2007-02 2022-12 191 SIMPAH - FAOSTAT Rice (US Long Grain quality) Spanish 2.4%) quintal (46kg) Honduras Urea Urea 46% National HNL / 43kg wholesale 2008-06 2022-12 175 SIMPAH Urea US Gulf NOLA Average Nicaragua Maize Maize (White) Managua NIO / Spanish wholesale 2000-10 2022-12 267 SIMPAH - FAOSTAT Maize (US No. 2, (oriental) quintal (46kg) Yellow) Nicaragua Rice Rice (first Managua NIO / Spanish wholesale 2000-10 2022-07 262 SIMPAH - FAOSTAT Rice (US Long Grain quality) (oriental) quintal (46kg) 2.4%) Panama Banana Plantains Panama City PAB / 100 wholesale 2013-06 2022-05 108 SIMMAGRO - Instituto de US Banana - East units Mercadeo Agropecuario Coast 102 International Local Country Market Units Price type Start Date End Date Obs. Source International price Commodity product Panama Maize Maize Panama City PAB / wholesale 2006-01 2022-12 204 Instituto de Mercadeo Maize (US No. 2, Spanish Agropecuario - FAOSTAT Yellow) quintal (46KG) Panama Red bean Red bean Panama City PAB / wholesale 2006-01 2022-12 204 Instituto de Mercadeo Nicaragua - Beans Spanish Agropecuario - FAOSTAT quintal (46KG) Table A1.2: Selected Domestic And International Price Time Series And Sources of Data (Cont.) Panel B. Selected International Commodity Prices and Data Sources International Description Country Market Units Source Commodity Rice US Long Grain 2.4% United States United States US$ / Ton FAOSTAT (Primary source: International Grains Council) Red bean Unit value of exports of Nicaragua Nicaragua Worldwide US$ / Kilogram (kg) Central Bank of Nicaragua Black bean Unit value of exports of United States United States Worldwide US$ / Metric Ton (MT) USDA Foreign Agricultural Service Maize US No. 2 Yellow United States U.S. Gulf US$ / Ton FAOSTAT (Primary source: USDA) Banana US East Coast - Main Brands Central America, United States Central America US$ / Box (18.14 kg) FAOSTAT (Primary source: Delivered at Terminal NOTIFAX (CORBANA)) Coffee ICO Composite Index Worldwide Worldwide US$ cents/ Pound (lb) International Coffee Organization Wheat CWRS Canada Canada US$ / Ton FAOSTAT (Primary source: International Grains Council) Ammonia GCFPAMNB Index - Ammonia United States US Gulf NOLA US$ / Metric Ton (MT) Bloomberg Urea GCFPURGI Index - US Gulf NOLA Urea Import United States US Gulf NOLA US$ / short ton137 Bloomberg Prill Spot Price DAP GCFPDANO Index - DAP U.S. Gulf NOLA United States US Gulf NOLA US$ / short ton Bloomberg prices 137  The international price of urea in tons is considered only in the case of Honduras for convergence purposes. 103 Table A1.3: Ranking Of Daily Calorie Contributions To Per Capita Diets By Country Panel A. Costa Rica Panel B. Dominican Republic Rank Rank Product Kcal Product Kcal position position Sugar (Raw Equivalent) 451 1 Rice and products 487 1 Rice and products 436 2 Sugar (Raw Equivalent) 379 2 Wheat and products 408 3 Soyabean Oil 272 3 Milk - Excluding Butter 320 4 Wheat and products 197 4 Soyabean Oil 194 5 Milk - Excluding Butter 184 5 Palm Oil 160 6 Fruits, other 180 6 Maize and products 121 7 Plantains 159 7 Poultry Meat 118 8 Poultry Meat 111 8 Beans 92 9 Coconuts - Incl Copra 97 9 Bananas 90 10 Panel C. El Salvador Palm Oil 82 11 Rank Beans 70 12 Product Kcal position Maize and products 804 1 Panel D. Guatemala Wheat and products 318 2 Rank Product Kcal Sugar (Raw Equivalent) 226 3 position Beans 187 4 Maize and products 832 1 Milk - Excluding Butter 186 5 Sugar (Raw Equivalent) 487 2 Rice and products 124 6 Wheat and products 221 3 Soyabean Oil 121 4 Panel E. Honduras Poultry Meat 101 5 Rank Rice and products 91 6 Product Kcal position Milk - Excluding Butter 74 7 Maize and products 749 1 Beans 66 8 Sugar (Raw Equivalent) 384 2 Palm Oil 340 3 Panel F. Nicaragua Wheat and products 264 4 Product Kcal Rank position Rice and products 139 5 Maize and products 597 1 Beans 113 6 Rice and products 462 2 104 Table A1.3: Ranking Of Daily Calorie Contributions To Per Capita Diets By Country (Cont.) Panel G. Panama Product Kcal Rank position Rice and products 719 1 Wheat and products 317 2 Sugar (Raw Equivalent) 225 3 Maize and products 211 4 Poultry Meat 191 5 Milk - Excluding Butter 172 6 Soyabean Oil 145 7 Fats, Animals, Raw 95 8 Oil crops Oil, Other 85 9 Palm Oil 84 10 Plantains 67 11 Pig meat 65 12 Beer 57 13 Sweeteners, Other 48 14 Bovine Meat 45 15 Butter, Ghee 43 16 Pelagic Fish 27 17 Pulses, Other and products 26 18 Eggs 22 19 Rape and Mustard Oil 21 20 Beverages, Alcoholic 21 21 Potatoes and products 20 22 Soyabeans 18 23 Pineapples and products 16 24 Fruits, other 16 25 Vegetables, other 14 26 Oats 13 27 Cocoa Beans and products 13 28 Beans 12 29 Note: This table presents the kilocalorie contribution of each food product per day and per capita.138 138  The data are derived from the food balances information available at FAO (2023). 105 Table A1.4: Selected Model Results And Residuals Tests COSTA RICA AMMONIUM BLACK RICE WHEAT   WHITE MAIZE UREA 46% NITRATE BEANS 80-20 FLOUR Conditional Mean Equation  Model VEC VEC VAR VAR VEC VEC No. of lags 1 1 1 2 2 1 α0 0.0025 0.0016 0.0086 0.0026 0.0010 -0.0005   (0.008) (0.010) (0.008) (0.003) (0.009) (0.005) α1,11 -0.186* 0.300* -0.308*   (0.094) (0.097) (0.071) α1,12 -0.1103 0.0433 0.0961   (0.107) (0.075) (0.086) Conditional Variance Equation c11 2.998 5.350 0.826* -0.563* 6.594* 1.558*   (2.491) (3.174) (0.397) (0.120) (1.955) (0.465) a11 0.726 0.799* -0.115* -0.055* 0.833* 0.467*   (0.406) (0.361) (0.007) (0.007) (0.239) (0.168) a21 -0.107* 0.202 0.035* -0.078* -0.292 -0.081   (0.000) (0.140) (0.005) (0.014) (0.248) (0.077) g11 0.763* 0.341 -0.064* -0.603* 0.388 0.791*   (0.059) (0.990) (0.016) (0.049) (0.207) (0.061) g21 -0.406* -0.062 1.143* 0.495* 0.173 0.139* (0.107) (0.189) (0.567) (0.053) (0.129) (0.065) ν 2.452* 4.719* 4000168.347* 4000052.415* 3.674* 4.265*   (0.378) (1.282) (27783.487) (2802.219) (0.744) (1.077) Wald test for presence of innovation and persistence effects from international to domestic market (H0: a21 = g21 = 0) Chi-squared 3691515831482580.000* 2.469 2526.176* 171.983* 2.374 4.592 p-Value 0.000 0.291 0.000 0.000 0.305 0.101 Ljung-Box test for autocorrelation (H0: no autocorrelation in squared residuals) LB(6) 16.788* 15.551* 14.845* 9.979 10.328 5.063 p-Value 0.010 0.016 0.021 0.126 0.112 0.536 LB(12) 20.170 16.424 19.432 14.410 16.325 6.401 p-Value 0.064 0.173 0.079 0.275 0.177 0.895 Lagrange multiplier (LM) test for ARCH residuals (H0: no ARCH effects) LM(6) 7.544 4.225 6.899 3.767 11.403 1.573 p-Value 0.273 0.646 0.330 0.708 0.077 0.954 LM(12) 34.531* 5.880 10.983 7.120 15.547 1.928 p-Value 0.001 0.922 0.530 0.850 0.213 1.000 Hosking Multivariate Portmanteau test for cross-correlation (H0: no cross-correlation in squared residuals) M(6) 31.125 25.852 17.929 23.348 34.527 14.070 p-Value 0.150 0.361 0.806 0.499 0.076 0.945 M(12) 44.550 40.233 34.899 51.018 57.308 38.030 106 COSTA RICA AMMONIUM BLACK RICE WHEAT   WHITE MAIZE UREA 46% NITRATE BEANS 80-20 FLOUR p-Value 0.615 0.780 0.921 0.356 0.168 0.848 Log Likelihood -1300.6 -784.9 -728.0 -577.0 -1184.0 -1023.9 No. of Obs. 172.0 110.0 109.0 108.0 171.0 168.0 Table A1.4: Selected Model Results And Residuals Tests (Cont.)   DOMINICAN REPUBLIC EL SALVADOR   BANANA BLACK BEANS FIRST QUALITY RICE WHITE MAIZE RICE Conditional Mean Equation Model VEC VEC VAR VAR VEC No. of lags 1 2 2 2 2 α0 0.0029 0.005* 0.0010 0.0013 0.0014   (0.006) (0.002) (0.003) (0.005) (0.003) α1,11 0.162* -0.0912 0.294* 0.0081   (0.060) (0.060) (0.077) (0.069) α1,12 -0.0058 -0.0028 0.0669 0.190*   (0.022) (0.075) (0.077) (0.061) Conditional Variance Equation c11 4.340* 1.552* -0.157 0.000 0.154   (1.502) (0.378) (0.477) (0.000) (0.159) a11 0.575* 0.720* 0.328* 0.251* 0.332*   (0.202) (0.122) (0.073) (0.060) (0.159) a21 -0.016 -0.039 0.119 -0.094 0.132   (0.127) (0.075) (0.077) (0.085) (0.070) g11 0.708* 0.666* 0.934* -0.241* 0.943*   (0.183) (0.079) (0.018) (0.073) (0.019) g21 0.015 -0.003 -0.084 -0.933* -0.122*   (0.058) (0.075) (0.048) (0.083) (0.016) v 3.121* 4.350* 3.456* 4.364* 2.487*   (0.484) (0.836) (0.924) (0.963) (0.498) Wald test for presence of innovation and persistence effects from international to domestic market (H0: a21 = g21 = 0) Chi-squared 0.158 3.363 3.326 127.583* 56.786* p-Value 0.924 0.186 0.190 0.000 0.000 Ljung-Box test for autocorrelation (H0: no autocorrelation in squared residuals) LB(6) 4.637 1.991 6.917 0.897 8.761 p-Value 0.591 0.920 0.329 0.989 0.187 LB(12) 14.690 3.422 21.816* 8.827 15.087 p-Value 0.259 0.992 0.040 0.718 0.237 Lagrange multiplier (LM) test for ARCH residuals (H0: no ARCH effects) LM(6) 0.071 1.531 0.068 22.449* 1.372 p-Value 1.000 0.957 1.000 0.001 0.968 LM(12) 0.252 4.968 0.405 23.439* 4.360 107   DOMINICAN REPUBLIC EL SALVADOR   BANANA BLACK BEANS FIRST QUALITY RICE WHITE MAIZE RICE p-Value 1.000 0.959 1.000 0.024 0.976 Hosking Multivariate Portmanteau test for cross-correlation (H0: no cross-correlation in squared residuals) M(6) 22.877 10.449 18.108 27.642 29.122 p-Value 0.527 0.992 0.798 0.275 0.216 M(12) 44.351 26.892 43.571 71.475 46.555 p-Value 0.623 0.994 0.655 0.016 0.532 Log Likelihood -1542.2 -1702.6 -1361.0 -1284.9 -946.9 No. of Obs. 223.0 272.0 271.0 200.0 197.0 Table A1.4: Selected Model Results And Residuals Tests (Cont.) GUATEMALA   BLACK BEANS WHITE MAIZE FIRST QUALITY RICE Conditional Mean Equation Model VEC VEC VEC No. of lags 1 2 2 α0 0.005* 0.0000 0.0019   (0.002) (0.005) (0.001) α1,11 0.329* 0.179*   (0.058) (0.058) α1,12 0.0500 0.144*   (0.069) (0.032) Conditional Variance Equation   c11 1.485* -0.079 -0.655*   (0.367) (0.385) (0.327) a11 0.802* 0.328* 0.291*   (0.157) (0.161) (0.127) a21 0.044* -0.049 0.139*   (0.021) (0.315) (0.044) g11 -0.196 0.691* 0.532*   (0.326) (0.151) (0.074) g21 0.059 -0.816* -0.462* (0.043) (0.178) (0.050) ν 4.337* 8.038* 4.014* (1.001) (2.341) (0.794) Wald test for presence of innovation and persistence effects from international to domestic market (H0: a21 = g21 = 0) Chi-squared 10.173* 25.091* 88.603* p-Value 0.006 0.000 0.000 Ljung-Box test for autocorrelation (H0: no autocorrelation in squared residuals) LB(6) 4.764 10.931 8.740 p-Value 0.574 0.091 0.189 LB(12) 8.321 76.975* 12.693 p-Value 0.760 0.000 0.392 108 GUATEMALA   BLACK BEANS WHITE MAIZE FIRST QUALITY RICE Lagrange multiplier (LM) test for ARCH residuals (H0: no ARCH effects) LM(6) 0.439 2.828 6.075 p-Value 0.999 0.830 0.415 LM(12) 1.089 17.202 10.222 p-Value 1.000 0.142 0.596 Hosking Multivariate Portmanteau test for cross-correlation (H0: no cross-correlation in squared residuals) M(6) 27.658 30.542 16.797 p-Value 0.275 0.167 0.857 M(12) 46.722 108.330 32.791 p-Value 0.525 0.000 0.954 Log Likelihood -1085.0 -1699.4 -1291.1 No. of Obs. 178.0 262.0 273.0 Table A1.4: Selected Model Results And Residuals Tests (Cont.) HONDURAS SECOND AMMONIUM   COFFEE RED BEANS DAP RED BEANS QUALITY UREA 46% NITRATE RICE  Conditional Mean Equation Model VEC VEC VEC VEC VEC VEC VEC No. of lags 1 2 2 1 1 2 2 α0 0.0072 0.0010 0.0004 0.0011 0.0020 0.0015 -0.0009   (0.004) (0.004) (0.006) (0.003) (0.009) (0.002) (0.004) α1,11 0.164* 0.489* -0.0275 0.241*   (0.069) (0.068) (0.073) (0.066) α1,12 0.211* 0.240* 0.250* 0.096*   (0.079) (0.094) (0.052) (0.042) Conditional Variance Equation  c11 3.063* 3.503 1.986 -0.938* 6.051* 1.455* -1.803   (1.442) (3.452) (1.726) (0.085) (1.688) (0.394) (2.603) a11 0.740* 0.846* 0.142 0.217* 0.300* 0.833* -0.545   (0.302) (0.176) (0.088) (0.016) (0.121) (0.259) (2.064) a21 -0.121* -0.165 -0.102 0.148* -0.045 -0.379* 0.409   (0.055) (0.108) (0.105) (0.010) (0.144) (0.099) (2.745) g11 0.819* 0.136 -0.246 -0.012* 0.791* 0.446* -0.466   (0.073) (0.331) (0.173) (0.002) (0.084) (0.182) (1.785) g21 -0.064 0.182 1.031* 0.456* 0.030 0.004 0.208 (0.040) (1.886) (0.175) (0.045) (0.102) (0.009) (2.280) ν 2.434* 5.724* 4.825* 3999986.162* 3.865* 3.365* 4000363.260*   (0.355) (1.473) (1.164) (2515.248) (0.798) (0.720) (105497.606)  Wald test for presence of innovation and persistence effects from international to domestic market (H0: a21 = g21 = 0) Chi-squared 5.912 3.296 40.898* 218.172* 0.098 15.286* 14.182* 109 HONDURAS SECOND AMMONIUM   COFFEE RED BEANS DAP RED BEANS QUALITY UREA 46% NITRATE RICE p-Value 0.052 0.192 0.000 0.000 0.952 0.000 0.001 Ljung-Box test for autocorrelation (H0: no autocorrelation in squared residuals)  LB(6) 5.453 6.373 2.969 2.978 14.795* 4.737 15.260* p-Value 0.487 0.383 0.813 0.812 0.022 0.578 0.018 LB(12) 9.442 28.281* 19.205 7.692 23.894* 11.378 21.425* p-Value 0.665 0.005 0.084 0.809 0.021 0.497 0.044  Lagrange multiplier (LM) test for ARCH residuals (H0: no ARCH effects) LM(6) 3.339 4.404 1.987 11.745 7.899 2.539 4.402 p-Value 0.765 0.622 0.921 0.068 0.246 0.864 0.623 LM(12) 6.056 17.918 6.709 14.920 12.057 3.850 16.211 p-Value 0.913 0.118 0.876 0.246 0.441 0.986 0.182  Hosking Multivariate Portmanteau test for cross-correlation (H0: no cross-correlation in squared residuals)  M(6) 19.180 21.496 24.103 42.643* 49.702* 24.481 32.553 p-Value 0.742 0.609 0.456 0.011 0.002 0.434 0.114 M(12) 35.902 54.076 51.896 58.021 82.871 36.924 52.466 p-Value 0.901 0.254 0.325 0.152 0.001 0.877 0.305 Log Likelihood -1389.2 -1237.4 -1240.6 -1053.8 -1219.4 -896.6 -1079.9 No. of Obs. 199.0 204.0 188.0 173.0 172.0 188.0 172.0 NICARAGUA PANAMA   WHITE MAIZE FIRST QUALITY RICE MAIZE PLANTAINS RED BEANS Conditional Mean Equation Model VEC VEC VAR VAR VAR No. of lags 2 2 2 1 2 α0 0.0001 0.0012 0.0033 0.0017 0.0031   (0.008) (0.001) (0.003) (0.006) (0.005) α1,11 0.292* 0.218* 0.308* -0.0898 0.483*   (0.061) (0.058) (0.077) (0.099) (0.073) α1,12 0.0607 0.0144 0.0268 0.0709 0.0524   (0.121) (0.034) (0.048) (0.100) (0.047) Conditional Variance Equation c11 9.583* 1.433* 0.293 5.542 3.179*   (4.194) (0.329) (0.272) (28.757) (0.864) a11 -0.251 1.052* 0.233* 3.188 0.621*   (0.151) (0.388) (0.075) (17.502) (0.159) a21 -0.404 0.058 -0.190* -0.356 -0.050   (0.252) (0.120) (0.052) (1.908) (0.045) g11 -0.492 0.200 0.914* 0.331* 0.664* 110 NICARAGUA PANAMA   WHITE MAIZE FIRST QUALITY RICE MAIZE PLANTAINS RED BEANS   (0.609) (0.131) (0.024) (0.169) (0.114) g21 -0.432 -0.042 -0.020* -0.155 0.131* (0.255) (0.050) (0.007) (0.276) (0.055) ν 8.728* 3.005* 5.376* 2.100 3.817*   (2.788) (0.523) (1.484) (1.167) (0.676) Wald test for presence of innovation and persistence effects from international to domestic market (H0: a21 = g21 = 0) Chi-squared 6.830* 0.729 18.389* 0.390 6.133* p-Value 0.033 0.694 0.000 0.823 0.047 Ljung-Box test for autocorrelation (H0: no autocorrelation in squared residuals) LB(6) 5.426 7.170 6.284 7.635 1.091 p-Value 0.490 0.305 0.392 0.266 0.982 LB(12) 22.621* 11.460 15.887 11.683 6.169 p-Value 0.031 0.490 0.196 0.471 0.907 Lagrange multiplier (LM) test for ARCH residuals (H0: no ARCH effects) LM(6) 1.508 11.252 0.920 2.181 1.840 p-Value 0.959 0.081 0.988 0.902 0.934 LM(12) 3.268 13.388 1.025 15.316 5.444 p-Value 0.993 0.341 1.000 0.225 0.941 Hosking Multivariate Portmanteau test for cross-correlation (H0: no cross-correlation in squared residuals) M(6) 21.840 18.711 27.915 30.633 23.160 p-Value 0.589 0.767 0.264 0.165 0.510 M(12) 50.717 48.536 63.017 44.644 58.161 p-Value 0.367 0.451 0.072 0.611 0.149 Log Likelihood -1845.0 -1187.6 -1185.1 -577.2 -1165.2 No. of Obs. 263.0 259.0 200.0 100.0 175.0 Note: This table presents selected coefficients from the estimated conditional mean and conditional variance equations for each available country-commodity series, together with goodness of fit tests. Numbers in parentheses are standard errors. An asterisk indicates that the null hypo- thesis is rejected at the 95 percent level of confidence. Table A1.5: Net Food Imports As A Share Of Domestic Availability Country Banana Beans Coffee Maize Rice Wheat Costa Rica   76% 99% 43% 105% Dominican Republic -35% 51%   1%   El Salvador     39% 76%   Guatemala   6% 35% 76%   Honduras   4% -825% 47% 72%   Nicaragua     32% 24%   Panama -4% 56% 79%     Note: This table shows the degree of dependence on food imports for each of the countries in our sample. The degree of dependence on food im- ports is calculated as (M - X)/A, where M is the volume of imports, X is the volume of exports, and A is total domestic availability, defined as production plus net imports plus change in stocks. All quantities reflect 2010–2020 averages and are derived from FAO (2023). 111 Figure A1.1: Evolution Of Domestic And International Prices By Commodity And By Country Panel A. Costa Rica Panel B. Dominican Republic 112 Figure A1.1: Evolution Of Domestic And International Prices By Commodity And By Country (Cont.) Panel B. Dominican Republic (cont.) Panel C. El Salvador Panel D. Guatemala 113 Figure A1.1: Evolution Of Domestic And International Prices By Commodity And By Country (Cont.) Panel E. Honduras 114 Figure A1.1: Evolution Of Domestic And International Prices By Commodity And By Country (Cont.) Panel F. Nicaragua Panel G. Panama Note: This figure illustrates the monthly evolution of the assigned pairs of domestic and international prices in levels. The figures with the blue lines depicts the domestic price evolution for a given commodity in a given Central American country, and the figures below them with the green lines represent the international price evolution for the same commodity. 115 Figure A1.2: Evolution Of Domestic And International Price Returns By Commodity And By Country Panel A. Costa Rica Panel B. Dominican Republic 116 Figure A1.2: Evolution Of Domestic And International Price Returns By Commodity And By Country (Cont.) Panel B. Dominican Republic Panel C. El Salvador Panel D. Guatemala 117 Figure A1.2: Evolution Of Domestic And International Price Returns By Commodity And By Country (Cont.) Panel E. Honduras 118 Figure A1.2: Evolution Of Domestic And International Price Returns By Commodity And By Country (Cont.) Panel F. Nicaragua Panel G. Panama Note: This figure illustrates the monthly evolution of the returns (month-to-month percentage changes in prices) for the assigned pairs of domestic and international prices. The blue lines in each figure depict domestic price returns for a given commodity in a given Central American country, while the green lines in the figures below them represent the international price returns for the same commodity. 119 Figure A1.3: Conditional Correlations Between International And Domestic Markets Panel A. Costa Rica Panel B. Dominican Republic 120 Figure A1.3: Conditional Correlations Between International And Domestic Markets (Cont.) Panel B. Dominican Republic (Cont.) Panel C. El Salvador Panel D. Guatemala 121 Figure A1.3: Conditional Correlations Between International And Domestic Markets (Cont.) Panel E. Honduras 122 Figure A1.3: Conditional Correlations Between International And Domestic Markets (Cont.) Panel F. Nicaragua Panel G. Panama Note: The figures show the conditional correlations between international and domestic markets that result from the estimated T-BEKK models. The horizontal red line depicts the average correlation over each corres- ponding period of analysis and the dotted lines are the 95 percent confidence intervals. The three vertical green lines mark (when applicable) the Food Price Crisis of 2007-2008 (January 2008), the COVID-19 Pandemic (March 2020), and the Russia-Ukraine war (February 2022), respectively. 123 Figure A1.4: Three-Month Elasticities of Price Return Transmissions Panel. A Costa Rica Panel. B. Dominican Republic 1.0 1.0 0.8 0.8 0.6 0.442^** 0.6 0.471^** 0.369^** 0.4 0.229^ 0.4 0.159^** 0.161^ 0.2 -0.003 0.2 -0.012 0.004 0.0 0.0 BLACK BEANS RICE 80-20 AMMONIUN UREA 46% MAIZE WHEAT FLOUR FIRST QUALITY BLACK BEANS BANANA TNITRATE -0.2 -0.2 RICE Panel C. El Salvador Panel D. Guatemala 1.0 1.0 0.932^** 0.8 0.694^** 0.8 0.6 0.6 0.4 0.4 0.348^ 0.191^** 0.2 0.072 0.2 0.0 0.0 BLACK BEANS MAIZE MAIZE RICE QUALITY FIRST RICE Panel E. Honduras Panel F. Nicaragua 1.0 0.879^** 0.859^** 1.0 0.8 0.737^** 0.8 0.655^** 0.6 0.558^** 0.6 0.4 0.323^ 0.4 0.2 0.157^ 0.2 0.141^ 0.0 0.0 MAIZE DAP RED BEANS MAIZE QUALITY RICE UREA 46% QUALITY COFFEE AMMONIUN -0.2 -0.2 FIRST NITRATE RICE SECOND Panel G. Panama 1.0 0.8 0.6 0.4 0.2 0.002 0.001 0.010** 0.0 MAIZE PLANTAINS RED BEANS Note: ^ denotes that the elasticity is derived from a VEC model with no lags (long-term relationship only) and ** denotes a statistically significant estimate at the 5 percent level. Source: Authors’ elaboration. 124 Figure A1.5: Three-Month Elasticities of Price Volatility Transmissions Panel. A Costa Rica Panel. B. Dominican Republic 1.0 1.0 0.8 0.8 0.6 0.6 0.365** 0.4 0.4 0.2 0.075** 0.2 0.029 0.000** 0.008 0.016 0.000 0.022 0.000 0.0 0.0 RICE 80-20 BANANA AMMONIUN UREA 46% MAIZE BEANS BEANS WHEAT FLOUR BLACK BLACK QUALITY TNITRATE RICE FIRST Panel C. El Salvador Panel D. Guatemala 1.0 0.916** 1.0 0.8 0.8 0.641** 0.6 0.6 0.4 0.4 0.267^ 0.111^^ 0.2 0.071 0.2 0.0 0.0 BEANS MAIZE BLACK QUALITY RICE MAIZE FIRST RICE Panel E. Honduras Panel F. Nicaragua 1.0 0.8 0.828 0.8 0.6 0.6 0.377 0.4 0.4 0.2 0.062** 0.2 0.017 0.003** 0.001 0.018 0.002 0.008 0.0 0.0 MAIZE DAP RED BEANS QUALITY RICE UREA 46% COFFEE AMMONIUN MAIZE QUALITY NITRATE SECOND FIRST RICE Panel G. Panama 1.0 0.8 0.6 0.4 0.2 0.144 0.049** 0.0 MAIZE PLANTAINS RED BEANS Note: The figure is truncated to preserve scale, with outlier values indicated in bold. ** denotes a statistically significant estimate at the 5% level. Source: Authors’ elaboration. 125 ANNEX 2 - FINDINGS FROM AGRICULTURAL IMPACT EVALUATIONS139 A2.1 This Annex presents the findings of four meta-analyses of a large number of impact analyses of food security and agriculture programs around the world.140 The first analysis, conducted in 2011, is by the World Bank’s Independent Evaluation Group (IEG). The second analysis, conducted in 2010, is by the Inter-American Development Bank (IDB), while the third one, conducted in 2022, is by FAO and Innovations for Poverty Action (IPA). The fourth meta-analysis was conducted as part of a background paper prepared by FAO specifically for this Review.141 While most of the underlying operations included in these four meta-analyses of agricultural impact evaluations were not aimed directly at increasing food security, the operations had a clear bearing on food security in light of their focus on increasing yields, incomes and production for farmers and reducing production risks. A. THE WORLD BANK INDEPENDENT EVALUATION GROUP META-ANALYSIS A2.2 IEG’s meta-analysis of agricultural impact evaluations around the world provides a comprehensive overview of the effectiveness of agricultural interventions. It reviews 86 impact evaluations of agricultural interventions conducted between 2000 and 2009. The underlying operations were conducted in various regions of the world, with 17 percent of them implemented in LAC. Most interventions date from the 1990s, and the time elapsed between the end of the interventions and the Impact Evaluations (IEs) ranged from less than six months to as much as two decades.142 The IEG analysis cautions that results for given types of operations should not be taken as representative of these types of operations globally and may be subject to some biases (e.g., evaluations of best projects being published). A2.3 Improvement in input technology, access to financial services, land reform, access to or improved irrigation, and improved linkages for smallholders to markets, yielded positive results for the key analysed variables in more than 60 percent of the analysed interventions (Figure A2.1): » Improved input technologies: The distribution of improved (i.e., more productive and/or disease resistant) varieties, especially at an affordable cost, generally yielded positive results. 139  This Annex is drawn from a research paper prepared for this Review by FAO. See FAO (2023d). 140  A meta-analysis is the analysis of a large collection of individual studies for the purpose of integrating the findings. A meta-analysis can make use of statistical techniques for amalgamating, summarizing, and reviewing quantitative research to overcome limits of size or scope in individual studies (IEG, 2011). 141  See FAO (2023d). 142  The evaluations were selected from a total of 278 identified in the literature as they met a number of standard criteria: (i) they pertained to a well identified intervention (with clear design, objectives, and implementation strategy), (ii) they focused on farm performance and were based on clear indicators (including yields, incomes, production, and profits), (iii) they included a clearly defined counter-factual, (iv) the results could be attributed to the intervention, (v) the evaluations were deemed well designed and rigorous by the authors. IEG’s meta-analysis faced several limitations: (i) experimental designs, such as randomized control trials, were rarely used to evaluate agricultural interventions (only 6 percent did so, while around 60 percent of the IEs used quasi-experimental designs to construct plausible counterfactual groups); (ii) the heterogeneity of interventions made it difficult to use common and comparable units of explanatory and impact variables; (iii) it was not possible to compare the cost effectiveness or the magnitude of impacts across different intervention results or across groups within the same intervention; and (iv) In the absence of a common denominator, the analysis were compared based on the sign of the impact (positive or negative). IEG (2011) warns that the small number of studies, the potential for selection bias, and the focus on only IE studies (not evaluations using other approaches) are limitations and the results from the sample are not a reliable guide to intervention success rates in general. Additionally, impact magnitudes (on yields, incomes, production, and profits) vary widely across country and intervention context, thus limiting feasible conclusions on the basis of aggregated evidence. 126 Most analysed interventions promoted the adoption of improved crop varieties, improved seed technology, and innovative fertilizer. They all also include extension activities. Only farmers applying the right quantities of fertilizers for their land were found to have increased their profits. An impact evaluation in Kenya illustrated the importance of field experimentation and demonstration for the success of fertilizer distribution programmes from 2000-2003. » Financial services: Good targeting and complementarity of services (e.g. credit plus financial advice) were contributors to success. In value chain projects, incentives for creditors to lend are more apparent toward the end of production cycles, when repayment is more likely. This suggests that more effort can usefully be put into reducing perceived risks (and/or transaction costs) for financial institutions when dealing with family farmers. Large access to credit programs in Bangladesh were found to have raised per capita consumption and to have had positive spillover effects. While savings promotion based on production such as buffalo and rice banks in Thailand was found to have harmed household consumption, women’s savings and credit groups had positive impacts. Figure A2.1: Results Per Intervention Category As Classified By IEG 100% 90% 80% 4 8 13 70% 13 7 9 68 8 6 60% 50% 40% 30% 5 12 5 20% 1 5 3 3 37 3 10% 2 2 2 1 2 10 0% 0 1 0 er n l g rm RM n y ce ta io og th tin io To an fo at ns O N ke ol re ig fin te hn ar Irr nd Ex ro ec M ic La tt M pu In Negative no significant Positive Source: IEG (2011) » Land reform: Evaluations of land reform projects suggest that property rights imperfections can remain serious constraints even after land privatization processes, if regulatory, infrastructure or knowledge constraints remain unaddressed. Depending on context, land reform projects can increase the value and tradability of farmers’ assets, and positively impact the efficiency of their farming systems and their resilience to shocks. The analyzed operations aimed to increase landowner confidence in their property rights, improve access to credit, and lower land transfer costs. Projects in Nicaragua suggested that landowners perceived the value of land had increased and pointed to an increase in long-term agricultural investment. In Guatemala, the effects of land titling on the efficiency of plot use depended on the conflict context of the community, while in Peru there were shifts in investments from non-agricultural to agricultural self-employment activities, as well as a significant increase in the market value for titled plots. However, the intervention in Peru did not produce clear positive results in terms of increased access to credit. » Improved irrigation: Results were most consistently positive for on-farm irrigation improvements, followed by improved systems management. Drip irrigation in Peru increased 127 yields and profits with positive spillovers, while the introduction of water user associations in the Philippines resulted in improved water resources management and increased rice production. However, investment in infrastructure revealed mixed results. For example, a canal rehabilitation in Peru did not yield significant results in the short term, while dam constructions in Ethiopia and India had large negative spillovers. Still, the construction of irrigation systems (as in the case of India) increased land use intensity and average yields (contributing to 60 percent increases in net farm income). » Smallholders marketing: Contract farming was found to have had largely positive impacts, but it is a complex area, with few clear examples for LAC: In Peru, a lack of investment in accompanying infrastructure and regulatory reforms resulted in potato contracts failing to produce the expected results. On the other hand, in Costa Rica, average farm gate price improved considerably as a result of participation in both specialty and cooperative marketing channels. Examples from five countries on contract farming are positive,143 but they are not risk- free and require strong involvement of private firms, as well as accompanying extension, credit, and input provision. A2.4 Impacts with regard to natural resources management operations and extension services interventions were more mixed: » Natural resources management (NRM): This is a broad thematic area with a limited number of available impact evaluations. Results were found to be largely positive for practices that improve soil structure (conservation practices, hedgerows, and soil fertility), but only positive 40 percent of the time when operations involved conservation structures. In El Salvador, operations to promote mulching, minimum tillage, crop rotation, and green manure had a positive effect on incomes, but support for terracing, ditches, live barriers, and stone walls was not as successful. Impact evaluations concluded that infrastructure investments need to be tailored to each specific environment and that generic solutions normally fail to produce positive results. » Extension: Interventions in this area were divided into three subcategories: (i) Farmer field schools were more effective in decreasing the use of pesticides (e.g., in Peru) than in increasing yields. Results were produced on trainer’s farms but were inconclusive for the trainees; (ii) advisory services in the sample were more successful, with positive results in Argentina (in terms of yields and quality for grapes) and Uruguay (livestock productivity); and (iii) the provision of market information produced positive results on sales prices in all analysed cases (none of which were in LAC). However, another review of cases in Colombia and India found that knowledge of market prices (through SMS) did not change farmers bargaining capacity or prices.144 A2.5 Adequate consideration of context and strong operational design appear to be more important determinants of success than the particular type of intervention undertaken. The results show that some aspects of the interventions—mainly the constraints addressed, the design, and implementation characteristics—can influence the likelihood of positive outcomes. In various cases, complex interventions involving multiple components were undertaken in order to address various complementary factors that contribute to improving productivity. However, the evidence 143  FAO and IPA (2022) reviewed more recent literature with success cases of contract farming in 20 countries noting, for instance, that the contracts established with a mix of extension, credit and input provision in Kenya failed one year after, as the European Union import requirements changed. Information asymmetries and fraud were also reported risks in the literature, although independent contract monitoring can help mitigate these risks. 144  See FAO and IPA (2022). 128 shows that when interventions were too complex, and the capacity to implement them was low, negative results followed. Several IEs demonstrated that most types of interventions, especially those that involved new technologies, needed to be complemented by knowledge and credit- related activities to deliver their full impact. The IEG analysis demonstrates that only a subgroup of interventions was effective when implemented alone, while most interventions did best when complemented by other activities. B. THE INTER-AMERICAN DEVELOPMENT BANK META- ANALYSIS A2.6 The IDB meta-analysis found that the effectiveness of different interventions in agriculture varies considerably across projects.145 The IDB analysis covered both survey papers and stand-alone papers that provided robust evaluations using appropriate methodologies and quality data. Preference was given to those studies and surveys that attempted to create a proper counterfactual, or to control for the endogeneity of key explanatory variables. Papers that did not have ideal data, but that at least used secondary or tertiary data to correct this limitation, were also considered. And, when no studies were found that used data or econometric approaches to evaluation, the IDB report included reviews from reputable sources and authors as a way to fill in the gaps. As in the case of the IEG analysis, the underlying operations included in the IDB meta-analysis did not aim directly at increasing food security, but had strong implications for long- term access to food, through increased and more stable incomes and productive capacity for producers. The main findings of IDB’s meta-analysis are the following: » Land titling: Consistent positive results were found regarding the link between land titling and increased investment and property values. Another consistent result was the lack of effect that land titling had on facilitating access to credit. Further, while titling leads to the activation of land rentals, it does not appear to significantly influence the buying and selling of land. The results suggest that land titling on its own cannot deliver all the anticipated outcomes if operations are not accompanied by other types of support, especially credit and technical support. » Watershed management: When watershed management projects have clearly defined objectives, they can have positive effects in terms of conservation, productivity, and agricultural diversification. » Export promotion: The empirical evidence seems to confirm that export promotion policies, by attenuating information problems, can reduce transaction costs and foster trade. These policies may also have a positive effect on product diversification, and there is some evidence that export promotion may have a greater impact on smaller or less-experienced enterprises. » Guarantee funds: Although the results are mixed (mainly in terms of magnitude) and different methodologies were used in assessing these schemes, guarantee funds seem to have an effect on additionality, productivity, employment, and economic growth (at the local level). However, most of the results came from developed countries and more research is needed for developing economies, such as those in LAC. » Agricultural research and development: While only one competitive grant project for R&D clearly identified a contribution to productivity, several projects appeared to have made 145  See IDB (2010). 129 important contributions to increases in sales, product diversification and process innovation. Competitive grant projects also seem to have had a consistent, although not conclusive, effect on increases in R&D budgets and additionality, with no signs of crowding-out effects. » Agricultural extension: Results on the impact of extension programs on productivity are mixed. Some studies found evidence of improvements in yields while others did not find a clear impact. Others still found mixed results, depending on the sample or crop and period under investigation. The same is true for impacts on technology adoption: some studies found positive impacts while others found less conclusive impacts. The results point to a need to understand better the reasons for significant variations in results across operations. » Investment in infrastructure: A reasonable amount of evidence was found to point to a positive relationship between rural road investments and productivity gains. Moreover, the result holds independently of the country or region under study. Generally consistent positive results were also found with regard to the impacts of such investments on improvements in access to transportation, greater employment opportunities and in some cases better pay, increased access to input and output markets, and increased production and incomes. » Market information: The IDB analysis found that having access to a mobile phone has the potential to decrease information asymmetries and price dispersion and improve consumer and producer welfare. However, the provision of valuable market information can also be achieved via other means. In sum, the results provided significant support for the role of access to information, whether this is acquired by mobile phone, radio, or via entities that disseminate market prices. » Institutional strengthening: While the results were mixed, a reasonable number of empirical papers found that strong public institutions play an important role in improving income, growth, and productivity. This is also true when looking at the role of institutions in agricultural productivity, based on the scarce sources found in the IDB analysis. C. THE FAO AND IPA META-ANALYSIS A2.7 The meta-analysis conducted by Innovations for Poverty Action (IPA) is the most recent of the three meta-analyses, and reviewed 40 studies on the effectiveness of commodity pricing policies on farmer livelihoods and farmer investment decisions.146 The review sheds light about interventions that are often not captured by program impact evaluations, notably price support initiatives, trade policies and price transmission mechanisms for smallholder farmers, and quality upgrading schemes: » Price support initiatives: In Ghana, the government’s National Buffer Stock Programme offers price support through buffer stock operations (BSOs) to protect smallholders’ incomes. BSOs were found to have increased incomes by 12 percent through effects on farm gate prices, with income, by unit of output, increasing by 17 percent. However, the impacts on income depended on complementary factors, such as the gender and age of producers, the use of extension services, the availability of transport services, and packaging costs. In Madhya Prradesh, India, the Price Deficiency Payments scheme was found to have caused a fall in the minimum price for farmers, who earned less revenue because the government paid the difference between the floor and average sales price rather than the farmer’s sale price, even if it decreased more than 146  See FAO and IPA (2022). 130 the average sales price. Thus, Price Deficiency Payments need to be well calibrated to produce positive results, as they may promote over-supply and decrease farm gate prices, either putting an onus on the government or resulting in lower income for farmers. » Trade policy and price transmission mechanisms for smallholder farmers: The impact of trade on poverty and income distribution varies across countries and by policy instrument. Various trade policy dimensions, such as safety or quality standards and tariffs, shape price transmission to smallholder farmers differently. In theory, they raise prices and incomes for domestic farmers, but in practice this may not occur due to certain barriers. In Senegal, for example, the increase in food safety standards was found to have raised incomes and reduced poverty by 14 percent. In Kenya, a program for helping smallholder farmers to switch to export crops, by offering in-kind loans for inputs and marketing services for rice, cassava, and maize, was found to have increased export crop production and lowered marketing costs, resulting in a 32 percent income gain for new adopters. However, farmers’ inability to satisfy European Union export requirements over the longer term led to the cancellation of the programme. In Côte d’Ivoire, Cocoa farmers in northern provinces experienced higher living standards after the implementation of lower export tax rates. Export production was found to reduce poverty via price increases for products, even when product standards are high. However, outcomes vary considerably depending on the commodity; the labor market; the type of support policy; the market structure (notably whether it is competitive or not), and the country. » Quality upgrading schemes: Schemes that offer guaranteed price premia for improved quality, or quality recognition mechanisms applied during market transactions, have shown that farmers can be responsive to prices and can change their production patterns and achieve better incomes, when these schemes are matched with credit, training, and access to inputs for adopting the new technology. Fair Trade certification, for example, has resulted in increased incomes for farmers in Colombia (coffee), Costa Rica (coffee), Vietnam (tea), and Thailand (rice). However, studies in Ethiopia demonstrate that price premia are not always transmitted perfectly to producers, and that the premia may not be sufficient to offset the costs of certification for smallholder farmers. Thus, specific enabling conditions need to be in place for these schemes to work. A2.8 The FAO and IPA analysis highlights seven key messages for effectively facilitating price transmission to smallholders: » Public sector price support initiatives reduce poverty, although they can have distortionary effects (e.g., wealthier farmers benefit more from supports). » Quality upgrading schemes can affect farmer incomes positively, but only under certain conditions. » Contracts do not have to be complicated to increase prices and reduce uncertainty for smallholder farmers, but institutional capacity is key to enforcing contracts. » Inadequate, expensive or non-existent transport infrastructure can limit price transmission and depress farmer livelihoods. This varies by commodity, and by the availability of related storage facilities. » Enhancing competition between intermediaries and improving farmers’ access to wholesale 131 markets can reduce barriers and facilitate price transmission to smallholders. » Commodity price information services are important but are rarely sufficient for smallholder farmers to increase revenues. » A key area for further study is the time frame between price transmission and smallholder investment behaviour. Existing studies do not distinguish between the short and long-term effects of price changes on smallholder behaviour and welfare. These effects can differ significantly and should be considered in planning any intervention, though more rigorous research in needed in this area. D. THE FAO META-ANALYSIS OF ITS AGRICULTURAL POLICY REPOSITORY A2.9 The agricultural policy repository analyzed in the meta-analysis prepared for this Review includes 35 interventions in agriculture whose impact assessments were published between 2010 and 2023. Eighty percent of the interventions refer to LAC countries, with Brazil having the largest number of assessments (6), followed by Nicaragua (4), Argentina (3), Bolivia (3) and Mexico (3). The remaining the program assessments mainly refers to countries in Africa, with the exception of one intervention (in Greece). Many of the assessed interventions were financed by the IDB (34 percent); with World Bank and IFAD programs each accounting for 9 percent of the total interventions financed. The interventions were classified according to OECD nomenclature for Producer Support and General Services Support (Table A2.1). In light of the focus of the underlying operations on agriculture, rather than food security for consumers, interventions for Consumer Support were not included in the analysis. Table A2.1: OECD Producer And General Service Categories Producer Suport (PS) Categories General Services Support (GSS) Categories A1 Market price support H Agricultural knowledge and innovation system A2 Payments based on output I Inspection and control B Payments based on input use J Development and maintenance of infrastructure C Payments based on area K Marketing and promotion O Other L Cost of public stockholding M Miscellaneous Note: This table presents a subset of all OECD categories (the ones utilized for the reviewed interventions). Source: OECD (2016) A2.10 As in the previous three meta-analyses, the interventions were not specific to food security, and presented challenges due to the heterogeneity of their objectives and targets. Nevertheless, most of the operations aimed to increase access to, availability of, or utilization of food, or to improve the stability of food supplies in the geographical areas of intervention. The 35 interventions were almost equally distributed between Producer Support (54 percent) and General Services Support (46 percent) programs (Figure A2.2). Producer Support interventions were highly concentrated on the provision of inputs (directly or through vouchers), and on non- reimbursable financial support for on-farm investments (with 95 percent of all PS programs in these two areas). General Services Support, on the other hand, covered several operational modalities, with agricultural research and extension being the most common (accounting for 31 percent of the total number of GSS interventions), followed by investments in infrastructure (25 percent), miscellaneous (19 percent), and marketing and promotion (13 percent). 132 Figure A2.2: Agricultural Policy Repository: Distribution Of 35 Agricultural Interventions By Type Of Support And Operational Modality (OECD Nomenclature) H. Ag research & extension 31% J. Infrastructure 25% M. Miscellaneous 19% K. Marketing & promotion 13% I. Inspection & control GSS 6% 46% Other 6% PS 54% B. Payments based on input use 95% A2. Payments based on output 5% Source: FAO (2023d) A2.11 The results of the impact assessments were found to be positive for most of the interventions, although comparisons across assessments are difficult to make. An overwhelming 83 percent of the interventions had positive results, while only 17 percent were found to be negative, mixed or to have non-statistically significant positive results. One should be cautious regarding the representativeness of these figures, as there could be a “publication bias” in favor of publishing positive results. In order to attribute causality in the evaluations, most assessments used Randomized Control Trial approaches. The evaluations covered a broad range of impacts, including with regard to: agricultural production; yields; farm household income; planted area; market sales; value of production; food security; use of improved technologies and good agricultural practices (GAP); profits; participation in water use associations; poverty; assets; social mobility; dietary diversity; food quality; food consumption; prices received by farmers; rural youth migration; land diversification; employment; resilience; gender and women’s empowerment; perception of land tenure security and land values, and pest prevention. In several cases, the programs assessed covered more than one OECD operational category, such as programs including a component on extension (GSS) coupled with input provision (PS). In fact, some programs had several components such as investment in infrastructure (whether regional, local, and/or on-farm); capacity building; extension services; input provision, etc. The meta-analysis of the impact assessments yielded the following principal findings: » Payments made to promote the use of preferred inputs (including for on-farm investments) yielded positive results in 83 percent of the interventions. Out of the 35 programs included in the policy repository, 18 dealt with payments for input use and on-farm investments. These programs were implemented through the provision of non-reimbursable financial resources, vouchers, or the financing of inputs and infrastructure (such as seeds, fertilizers, hail-resistant nets, wire, irrigation equipment, etc.). The interventions were at times also implemented through the provision of subsidized credit to support on-farm investments and input purchases, and were often accompanied by training and technical assistance. Through these types of interventions, programs sought to encourage farmers to adopt technological innovations. Most programs yielded positive results, generally in terms of increased yields, production and/ or farmers’ incomes (as in the case of programs in Argentina, Bolivia, Brazil, the Dominican Republic, Mexico, and Nicaragua, inter alia). In those cases where there were mixed results 133 or no statistically significant impacts, such as in Mozambique’s fertilizer and seeds voucher program for maize farmers,147 Paraguay’s input donation program,148 and various input subsidy programs in Sub-Saharan countries,149 asymmetries in the distribution of support (e.g., resource- poor farmers receiving proportionally less subsidy than wealthier farmers), as well as liquidity and/or information issues, may have constituted important constraints that prevented the programs from achieving their expected results. Some evaluations highlighted the importance of accompanying input support with other interventions, such as investments in infrastructure. The existence of complementarities appears to have generated larger impacts than the sum of individual interventions.150 » Output market interventions were found to be very common in the LAC region, but no recent policy assessment was found for them. Only one assessment of an output market intervention was identified, namely an intervention in Ghana.151 The Ghanaian government provided a price support to producers via a buffer stock mechanism for a full decade, in order to protect smallholders’ incomes from volatility in maize prices. The assessment found positive results in terms of farmers’ incomes and poverty alleviation. Nevertheless, this type of intervention can also create market distortions. » Agricultural innovation and extension programs are highly rated in the assessments because of their non-market-distorting nature and of the expected positive long-term productivity and sustainability impacts. The reviewed assessments reported positive results for these types of interventions. In particular, Farmer Field Schools were found to have enhanced the livelihoods of farmers in eastern and southern Africa, increasing crop income by 80 percent in Kenya and by 100 percent in Tanzania.152 In Mexico, extension on livestock production and animal health management, via the Ministry of Agriculture’s Technical Assistance and Training Program, was found to have increased farmers’ profits by 8 percent.153 In Uruguay, the publicly-funded and privately-delivered Livestock Program was found to have increased calf production and calf sales.154 The Nicaraguan Institute for Agricultural Technology’s National Strategy for the Development and Strengthening of Community Seed Banks, was found to have enabled participating farmers to maintain production levels during the COVID-19 pandemic.155 However, an assessment of Uruguay’s Farm Modernization and Development Program found mixed results (with higher adoption of new fruit varieties but no statistically significant impact on yields). The authors of this assessment cautioned about timeframe considerations and the importance of promoting all relevant agricultural practices to obtain the desired impact.156 » Investment in infrastructure (notably irrigation) had positive impacts in the assessments reviewed. Argentina’s Provincial Agricultural Services Program (PROSAP), Bolivia’s National Irrigation Program with a Watershed Approach, and Brazil’s National Program for the Strengthening of Family Farming (PRONAF), as well as Brazil’s Rural Communities Development 147  See Carter et.al. (2013). 148  See López, Salazar and De Salvo (2017b). 149  See Jayne and Rashid (2013). 150  See Gibbons et.al. (2016). 151  See Abokyi et.al. (2020). 152  See Davis et.al. (2010). 153  See Cuevas-Reyes (2018) 154  See Maffioli and Mullally (2014). 155  See FAO and Agencia Mexicana de Cooperación Internacional para el Desarrollo (2021). 156  See Maffioli et.al. (2013). 134 Project in the State of Bahia, contributed to the adoption of new technologies and to increased crop yields, value of production, market sales and farm incomes. However, in the case of the Bahia project, while outcome-level results were achieved, higher-level impacts (notably on household income) were not, probably due to the presence of one of the worst droughts of the century.157 Thus, unexpected or force majeure factors can affect the intended impact. » Marketing and promotion are intervention areas that have unexplored potential. Two interventions were found in this regard: one was publicly driven and the other was private sector- led. The publicly driven intervention refers to a program on food procurement from small-scale family farmers in Brazil. This program involves food purchases (fruits, vegetables, beans, rice, meat, milk, etc.) from family farmers at prices that are no higher than those prevailing in regional markets. Even though this intervention has not had a rigorous impact assessment, three studies have identified positive results in terms of the strengthening of farmer associations, higher prices received by farmers, higher food quality, and a more balanced relationship between farmers and intermediaries.158 The private sector-led intervention refers to contract farming involving smallholder rice producers in Benin. In particular, a private rice processing and marketing firm offered production contracts to local smallholder farmers, with an established price for rice that met a stipulated quality threshold, together with the provision of inputs (on loan), and agricultural training and technical assistance throughout the growing season.159 This intervention is highly relevant as it shows how a mechanism involving little or no public resources can support small farmers in deal with volatility of food prices. 157  See Garbero and Paliwal (2018). 158  See FAO (2015); FAO (2016b); and Ministerio do Desenvolvimento Social e Combate à Fome (2013). 159  See Arouna et al. (2021). 135 ANNEX 3 – SETTING FOOD SECURITY WITHIN THE BROADER CONTEXT OF SUSTAINABLE DEVELOPMENT A3.1 It is helpful to situate the current challenges and future threats to the Central America region’s food security within the broader worldwide context of the first three United Nations Sustainable Development Goals (SDGs).160 Policies that strengthen the resilience of food security not only benefit the direct recipients but also help to establish a firm foundation for balanced, equitable, sustainable socioeconomic growth. A3.2 SDG 1: “End poverty in all its forms everywhere.” Access to food is inextricably tied to levels of poverty and purchasing power. The backdrop worldwide is that COVID-19 has erased more than 4 years of progress against poverty, further exacerbated by rising inflation and the impacts of the Russian invasion of Ukraine. The most recent data available show that approximately 38 percent of households in the Central America countries (23.3 million people) have been living below the national poverty lines in their respective countries, ranging from 21 percent in the Dominican Republic and in Panama to 59 percent in Guatemala (Table A3.1).161 A3.3 SDG 2: “End hunger, achieve food security and improved nutrition, and promote sustainable agriculture.” (SDG 2 is often referred to as “Zero Hunger.”) Ukraine and Russia typically produce 30 percent of the world’s wheat, 20 percent of global maize, and 80 percent of sunflower oil seed products. The ongoing war in Ukraine has, therefore, triggered food shortages for the world’s poorest, including in Central America, which is highly dependent on imports of grains, as a result of the food price inflation exacerbated by worldwide shortages of key food products (see Chapters II and III). Table A3.1: Poverty Rates In Central America Population living below Poverty rate, based on Country Population national poverty line national poverty lines Costa Rica 5.2 1.6 30.0 Dominican Republic 11.1 2.3 21.0 El Salvador 6.3 1.7 26.2 Guatemala 17.1 10.1 59.3 Honduras 10.3 4.9 48.0 Nicaragua 6.9 1.7 24.9 Panama 4.4 0.9 21.5 Total 61.3 23.3 38.0 Note: Poverty rates based on national poverty lines are for 2019-20, except for Guatemala (2014) and Nicaragua (2016) Source: World Bank - World Development Indicators Database (https://wdi.worldbank.org/tables) 160  See research on this by the International Food Policy Research Institute (IFPRI): https://www.ifpri.org/topic/food-security. 161  See World Bank - World Development Indicators at https://wdi.worldbank.org/tables. 136 A3.4 SDG 3: “Ensure healthy lives and promote well-being for all at all ages.” The most dramatic drops in worldwide health indices since 2020 derive directly and indirectly from COVID-19, including disruptions in health services and in nutrition patterns, particularly among the lowest-income households. By mid-2021, almost half of households in the Central America region reported suffering income losses relative to pre-COVID-19 incomes, with proportions ranging from 39 percent in El Salvador to 58 percent of households in Panama. Public programs to a certain extent mitigated the pandemic’s food access and nutritional impacts on vulnerable households, however, health indicators were impacted, not only directly by the COVID-19 pandemic but also due to greater food insecurity. 137 ANNEX 4 – HOW A CHANGING CLIMATE AND NATURAL DISASTERS ARE EXACERBATING FOOD INSECURITY IN CENTRAL AMERICA A4.1 The Central America region is highly exposed and vulnerable to extreme climate-induced natural hazards. The Global Climate Risk Index ranked Guatemala and El Salvador as the 16th and 25th countries most vulnerable in the world to extreme weather events for the 1999-2018 period, taking into consideration fatalities due to climate-related loss events, both in absolute number and as a share of the population; the absolute value of economic losses and their share of GDP, and the number of major events to which the countries have been exposed.162 Honduras, which is also ranked in the first quartile of the global vulnerability distribution, has been struck by half of the dozen severest hurricanes of the twentieth century.163 More recently, in November 2020, two back-to-back Category 4 Hurricanes (tropical storms Eta and Iota) brought heavy rains and severe flooding affecting millions of people, amid severe damages to key infrastructure (e.g. irrigation and storage infrastructure, rural roads, among others), land and crops. A4.2 Irregular rainfall patterns and extreme weather events continue to create complex interactions that have the potential to substantially affect food security stability and agricultural activities. In Honduras, monthly rainfall data for the 2018-2022 primera and postrera grains planting seasons show that rainfall was below average in 2019 and 2021, whereas excess rainfall was observed in 2020 and 2022 (up to 300 percent in October-November 2020 due to Hurricanes Eta and Iota). This situation has disrupted the growing cycle of seasonal crops, especially maize, rice and beans. It resulted in lower-than-average harvests in 2020-2021 (by approximately 4 percent for corn and 8 percent for rice) and stagnant beans production compared to 2018-2019. It was estimated that maize production would drop by up to 20 percent and rice production by up to 50 percent in 2022 due to excess rain, lower access to fertilizers, and Central America Free Trade Agreement (CAFTA) tariff-rate quota phaseouts.164 A4.3 Hurricanes Eta and Iota in 2021 exacerbated the already challenging economic and social conditions prevailing during the COVID-19 pandemic, generating increased vulnerability.165 In Honduras, the hurricanes affected about 4.7 million people (48 percent of the population), with social and economic costs estimated at US$1.8 billion (7.5 percent of 2020 GDP). In El Salvador, the WFP estimates that the hurricanes severely impacted the food security of more than 300,000 Salvadorans and affected 22,000 farmers, through substantial harm to the country’s infrastructure and agricultural production, as well as damage or destruction of 2,800 hectares of crops. A4.4 The crisis resulting from the COVID-19 pandemic and tropical storms is impacting the agri-food sector severely. The two crises compounded have produced disruptions in distribution chains, destruction of key infrastructure, contraction of supply, impact on demand, reduction of traditional marketing spaces, price variability, decrease in the liquidity of agricultural micro-, small- 162  See Eckstein, Künzel and Schäfer (2021). 163  See Trocaire (2014). 164  https://www.laprensa.hn/premium/produccion-frijol-caera-arroz-insumos-agricolas-honduras-EH8009336 165  See Global Network against Food Crises and Food Security Information Network (2022). 138 and medium-scale enterprises (MSMEs), and contractions in employment due to substantial exit of MSMEs from the market (the Honduran Council of Private Enterprise estimated a permanent closure of more than 40 percent of MSMEs). Based on World Bank high frequency phone surveys, between 40 and 60 percent of households across the Central America region reported income losses in mid-2021 relative to pre-COVID levels (Figure A4.1), and between 25 and 35 percent of households reported running out of food in the month previous to the survey, with higher incidence among lower-educated and poorer families. The combined damage to the agri-food sector resulting from the pandemic and tropical storms was estimated by Honduras’ Ministry of Agriculture (SAG) at more than US$2 billion. In Guatemala, Eta and Iota affected 90 municipalities and more than 72,000 households: damages affected more than 137 million hectares of crops, producing economic losses worth more than US$115 million, especially for food security crops such as maize and beans (Figure A4.2). Figure A4.1: Changes In Reported Household Incomes, Mid-2021 Vs. Pre-Pandemic (Percent Of Households) 100 90 80 36.8 44.5 38.3 70 40.6 52.1 49.2 46.2 52.1 60 50 40 30 58.0 50.4 51.6 47.9 45.4 41.1 44.5 20 39.0 10 0 El Salvador Guatemala Honduras Costa Rica Nicaragua LAC Panama Dominican Republic Reduced Samee Increased Source: World Bank and UNDP (2021) Figure A4.2: Agricultural Losses And Damages In Guatemala Due To Tropical Storms Eta And Iota 200 180 160 Millions of quetzals 140 120 Agricultural losses due to impapct on productivity 100 80 Agricultural damages - crop 60 40 Agricultural losses dueto 20 surface reduction 0 li s tle e o om on ll co n n na an er e at hi or or at off ni th m C oc m na Be C C O C da O To C Br Ba ar C Source: Bello and Peralta (2021) 139 ANNEX 5 – DETAILED POLICIES ADOPTED IN CENTRAL AMERICA IN RESPONSE TO FOOD PRICE CRISES Table A5.1: Policy Measures Adopted In Central America Since 2008 Country Consumer Support Producer Support General Services Support Red: trade measures and market interventions Amber: government budget recurrent expenditures Green: government budget investment expenditures Costa Rica COVID-19 Pandemic 2022 » Family Farming Fund financed, and improvements » Food kits delivered to » Import tariffs eliminated introduced in the agricultural supplement schoolchildren’s for 5,231 tons of beans innovation/research system meals at home* (negative support)** (with World Bank and IDB loans of US$200 million and 2022 » New agriculture subsidy US$42 million, respectively, introduced for small scale that also included producer » Fuel price controls**** rice farmers (less than 100 supports) hectares)**** » Import tariffs eliminated for 5,231 tons of beans** » Cash transfers implemented for households most affected by inflation » New reduction on VAT (from 13 percent to 1 percent) for a range of products in the basic basket of goods 140 Country Consumer Support Producer Support General Services Support El Salvador 2008-2011 2008-2011 2008-2011 » Elimination of import » Elimination of import tariffs » Reinforcement of the tariffs from neighboring from neighboring countries Special Programme For countries** (negative support)** Food Security, which also includes producer support166 » Introduction or » Support for community reinforcement of school seed banks COVID-19 Pandemic meals and related legal framework (with FAO and COVID-19 Pandemic » Adoption of local and WFP) regional food procurement » Support to coffee farmers policy, and improvement of 2022 via a scheme of payments the efficiency of smallholder for results. market systems (with WFP » Fuel price controls**** support)* 2022 » Elimination of import tariffs on food and fertilizers**** » Elimination of import tariffs on food (negative support)**** » Elimination of import tariffs on fertilizers**** 166  The Special Programme for Food Security (SPFS) was in place from 2000 until 2014. Its main donor, Spain, invested US$32.2 million in El Salvador, Guatemala, Honduras and Nicaragua, including US$14.8 million during 2009-14. The European Food Facility and the Initiative on Soaring Food Prices (ISFP), established in 2009, channeled additional resources to this programme. The SPFS was multidimensional and included interventions in institutional strengthening at all government levels. It also included training and on-farm investments, with a focus on agroforestry and maize production as well as diet diversification through the installation of vegetable gardens. Program evaluations revealed that it increased maize yields in all countries, although the scale of the results varied across locations. Maize production increased by 24 percent in El Salvador, 51 percent in Guatemala and 23 percent in Honduras from 2001 to 2015, likely as the result of a number of a number of complementary policies. Beans production also increased by 26, 45 and 77 percent over the same period in El Salvador, Guatemala and Honduras, respectively, in the same period. However, Nicaragua showed a sharp decrease in maize production from 2013 to 2014 and in beans from 2012 to 2015. 141 Country Consumer Support Producer Support General Services Support Guatemala 2008-2011 2008-2011 2008-2001 » Elimination of import » Elimination of import tariffs » Conditional transfers to tariffs from neighboring from neighboring countries improve health and nutrition countries** (negative support)** services (with US$120 million in World Bank » Introduction or » Payments for soil recovery financing, that also included reinforcement of school and fertility improvements consumer support) meals and related legal with FAO and IICA support, framework (with FAO and reaching 180,000 farmers. » Reinforcement of the WFP support) Special Programme For » Pilot project implemented Food Security (which also COVID-19 Pandemic for agricultural insurance. included producer supports) » School Feeding Program » Program on strategic food COVID-19 Pandemic strengthened (with support reserves implemented. from FAO, WFP, and IFAD)* » Adoption of local and » Participation of family regional food procurement 2022 farmers as suppliers to the policy and improvement of School Feeding Program the efficiency of smallholder » Propane gas subsidy (until (US$ 180 million), supported market systems (with WFP)* January 2023) by Food Security Law. » Conditional transfers to improve health and nutrition services (with World Bank US$120 million support, that included General Services Support) 142 Country Consumer Support Producer Support General Services Support Honduras 2008-2011 2008-2011 2008-2011 » Elimination of import » Elimination of import tariffs » Reinforcement of the tariffs from neighboring from neighboring countries Special Programme For countries** (negative support)** Food Security (which also included producer supports) » Ban on maize exports** » Ban on maize exports (negative support)** Between crises » Introduction or reinforcement of school » Reduction of production » Global Agriculture and meals and related legal taxes on grains** Food Security Program framework (with FAO and (GAFSP) for the design and WFP support) implementation of business plans and incremental food COVID-19 Pandemic security plans to support small-scale rural producer » Food kits at home (in lieu of organizations and small school meals)* enterprises (US$30 million) 2022 COVID-19 Pandemic » Electricity and fuel subsidies » Adoption of local and (with BCIE) regional food procurement policy, and improvement of » Price control of 40 products the efficiency of smallholder (staple food) during market systems (with WFP December 2008* support)* 2022 » Cooking gas and fuel subsidies 143 Country Consumer Support Producer Support General Services Support Nicaragua 2008-2011 2008-2011 2008-2011 » Emergency food aid** (15 » Distribution of inputs and » Training on nutrition and million meals distributed in livestock to female farmers farm management (with 2009)*** for increased food security US$20 million in IDB and commercialization of support, that included » Wheat price controls** surplus (Bono productivo) – producer and consumer US$30 million.*** supports) » Investment in storage and retail network with regulated » Conditional cash transfers » Promotion of the producer and consumer*** (on technology adoption) for consumption of cheaper prices and food distribution improved production. food (potato bread)** in needy areas (including consumer and producer » Training on nutrition and » Investment in the storage supports)*** farm management (with and retail network with US$20 million in IDB regulated prices and food COVID-19 Pandemic support, that included distribution in needy areas consumer and general from 2008-12 – (US$32 (and hurricanes) services supports, and million, including consumer used in part to fund the and producer supports)*** » Food kits were delivered to programme above). supplement schoolchildren’s » Reinforcement of the meals at home* » Certified seeds (for beans, Special Programme For maize, rice and sorghum) Food Security (which 2022 and inputs (urea and cattle included producer » Fuel price controls**** feed) distribution in 2010 supports)- (US$19.2 million)*** » Liquified propane gas price » Access to credit - various controls » Support for food production funding mechanisms for families with children supported to improve (with US$7.3 million in World farmers’ access to Bank support) financial resources for productive activities, inputs and intermediate goods purchase, and technology adoption.*** 144 Country Consumer Support Producer Support General Services Support » - Recovery of communities Between crises affected by Hurricane Felix in the North Atlantic » Global Agriculture and Food Autonomous Region (with Security Program (GAFSP) US$17 million in World Bank support to build resistance support) to coffee rust – IFC (US$30 million) » - Finance program “Usura Cero”: subsidized credit for COVID-19 Pandemic small-scale entrepreneurs » Adoption of local and (mainly family farmers). regional food procurement » - Support for community policy and improvement of seed banks. the efficiency of smallholder market systems through infrastructure, equipment and training (with WFP support)* » Entrepreneurship Program: training on finance, digital marketing, etc. Central 2022 2022 2022 American countries not » Food import tariff reduction » Cash transfers to farmers » Support for the production specified* (in 4 countries) and use of organic fertilizers » Trade facilitation (in 2 (in 4 countries) countries) » Price subsidies for fertilizers (in 3 countries) » Support for increased » Price controls of food efficiency of input use (in 3 and energy prices (in 4 » Supply of inputs and countries) countries) services (in 2 countries) » Promotion of local » Support for food stores (in 1 » Import tariff reductions for consumption (in 1 country) country) fertilizer (in 2 countries). » Food and energy subsidies (in 4 countries) » Food and public services vouchers or discounts (in 3 countries) » Provision of food baskets (in 2 countries) Sources: *ECLAC/FAO/WFP (2022); ** FAO (2009); *** Krivonos and Dawe (2014) ****World Bank (2023d) 145 ANNEX 6 - GLOBAL TRENDS IN POLICY SUPPORTS TO ADDRESS FOOD PRICE CRISES A6.1 Available information suggests that most of the policies put in place globally during recent food crises fall under the OECD’s consumer support category. Although there is no comprehensive repository of measures worldwide, a survey of 77 countries’ policy responses to the 2007-2008 food crisis found that about half of the countries, globally, reduced cereal import duties; 55 percent of them introduced price controls or consumer subsidies, one-quarter of the countries applied export restrictions, including export taxes; and around the same proportion took measures to increase food supplies, drawing on cereal stocks.167 Furthermore, the survey identified that responses varied considerably by region: countries in East Asia, South Asia and the Middle East and North Africa undertook significant activities in all four areas of intervention (import tax reductions, releasing grain stocks, increasing export restrictions and implementing price controls/consumer subsidies). In every geographical region except Sub-Saharan Africa, 50 percent or more of the countries reported using price controls or consumer subsidies. On the other hand, countries in Sub-Saharan Africa and in Latin America and the Caribbean showed the lowest levels of policy interventions, with roughly 20 percent and 30 percent of their countries, respectively, reporting no activity in any of the four policy categories (Figure A6.1 below).168 A6.2 In 2011, the G20 created the Agricultural Market Information System (AMIS) as an inter-agency platform to enhance food market transparency and policy responses for food security.169 The AMIS system assesses global food supplies (focusing on wheat, maize, rice, and soybeans) and provides a platform to coordinate policy action in times of market uncertainty. The system is composed of G20 members plus Spain and seven additional major exporting and importing countries of agricultural commodities,170 thereby representing around 80-90 percent of global production, consumption, and trade volumes of the targeted crops.171 The aim has been to work collaboratively to prevent unexpected price hikes and strengthen global food security, although the current food crisis suggests that this objective is not straightforward to achieve, and that, in times of crisis, individual countries prioritize policies that favor their domestic consumers and producers. A6.3 The policy response to the current global food crisis by countries around the world is similar, in terms of the measures adopted, to the policy responses to the 2007-08 and 2011 food crises. Notwithstanding WTO policy recommendations to avoid trade restrictions that can intensify food price spikes, various countries have imposed strong protectionist measures that have indeed exacerbated the increase in global food prices. According to IFPRI, during 2022, 32 countries imposed 77 export restrictions in the form of export licenses, export taxes or duties, export bans, or a combination of measures to protect their domestic consumers. By May 2022, restrictions had been imposed on more than 16 percent of all exports of key food and feed products, before the share declined to close to 8 percent of these exports in August 2022, as 167  See FAO (2009). 168  See FAO (2009). 169  The G20 is an intergovernmental forum for international economic cooperation that comprises 19 leading economies and the European Union. 170  The seven countries are Egypt, Kazakhstan, Nigeria, The Philippines, Thailand, Ukraine and Vietnam. 171  See https://www.amis-outlook.org/amis-about/en/ 146 restrictions notably on palm oil exports were lifted (Figure A6.2). However, if trade in food and fertilizer continues to be restricted (and in particular if earlier restrictions are reimposed or new restrictions are imposed), prices would be expected to continue to rise, threatening food security in more countries that depend on food imports. Figure A6.1: Policy Actions Adopted Around The Globe To Address The 2007-08 Global Food Crisis Percent 100 80 60 40 20 0 Africa East Asia Europe and Lat. Am. and Middle East South Asia Central Asia Caribean and N. Africa Reduce Taxes 44% 80% 33% 44% 60% 78% on foodgrains Increase Suply Using 22% 80% 0% 0% 60% 56% foodgrain stocks Export Restrictions 19% 40% 33% 19% 20% 44% Price Controls/ 80% 67% 33% 50% 100% 67% consumer subsidies None 22% 0% 0% 31% 0% 11% Note: A total of 77 countries were surveyed. The Y axis refers to the percentage of countries in each region that implemented policy actions, grouped into four categories of policy actions. Some countries adopted no actions while other countries adopted policy actions in various categories. Source: FAO (2009). Figure A6.2: Share Of Global Food And Feed Exports Affected By Restrictions In 2022 (Percentage By Product) Percent of global food and feed esports (calorie basis) Coconut and palm kernel oil Maize (corn) Others Palm oil (crude or refined) Rice Soya-bean oil Sun-flower oil Wheat and meslin 16 14 12 10 8 6 4 2 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan 2022 2023 Source: https://www.ifpri.org/blog/food-export-restrictions-have-eased-russia-ukraine-war-continues-concer- 147 ns-remain-key A. THE POLICY RESPONSES OF LATIN AMERICAN COUNTRIES TO THE 2022 FOOD AND FERTILIZER PRICE INCREASES A6.4 In response to the 2022 global food price increases, countries in Latin America and the Caribbean have implemented a wide range of policy measures. These include reductions in import tariffs and trade facilitation; promotion of local food consumption, local food markets and short agri-food chains; close monitored food prices and supplies; cash transfers, food vouchers and food baskets for vulnerable households, as well as food and energy subsidies, inter alia (Figure A6.3). Almost half of the measures adopted in LAC in response to the 2022 food crisis have involved subsidies and social assistance, while tax measures have accounted for around one-quarter of the measures adopted across the region.172 Countries in South America have placed most emphasis on tariff reductions, price controls and transfers to vulnerable households. Countries in the Caribbean have particularly emphasized promotion of local foods and short agri-food chains, and provision of food baskets and vouchers, and are the most likely to have implemented new school feeding programs and soup kitchens. Countries in Central America and Mexico have focused on monitoring food prices and supplies and on providing food and energy subsidies, as well as coupons or discounts for food purchases and for the payment of public services.173 Figure A6.3: Latin America And The Caribbean: Number Of Countries That Implemented Measures In Response To Rising Food Prices, By Subregion (During February To May, 2022) 12 10 8 2 6 3 1 1 2 6 4 4 1 4 4 3 2 2 4 2 2 5 2 1 1 3 1 2 3 2 3 2 3 3 3 3 1 1 1 2 2 2 1 1 1 1 0 Promotion of local consumption, food markets and short agrifood chains Provision of food baskets Improving distribution infrastructure Tariff reduction or trade facilitation Dialogue with supply chain actors to maintain prices Transferr of funds to households, minium wage and benefits Building up food stores Reduced consumption taxes Food or enerrgy subsidies Studying and monitoring food prices and supply Promorion of oeri-urban and backyard agricultute Control or stabilzation of food and energy prices Regional cooperation Other measures School feeding programmes and soup kitchens Vouchers or discounts for food and public services South America Central America and Mexico The Caribbean Source: ECLAC, FAO and WFP (2022) A6.5 Countries in LAC also have responded to the massive spike in global fertilizer prices in 2022 with a range of additional policy measures (Figure A6.4). According to FAOSTAT data, Latin American and Caribbean countries import about 85 percent of the fertilizers that they use, (including intraregional imports). In response to the doubling of global prices for commonly used fertilizers, countries in South America have placed particular emphasis on producer supports via the transfer of funds to farmers, as well as on reducing import tariffs on fertilizers. Countries in the Caribbean have attempted to subsidize fertilizer prices and increase the provision of agricultural supplies and services. Mexico and countries in Central America have focused on production 172  See World Bank (2023d). 173  See ECLAC, FAO and WFP (2022). 148 support, through cash transfers to farmers and price subsidies for fertilizers. Central American countries and Mexico have also invested in general services support, including by supporting the production and use of organic fertilizers and programs to increase the efficiency of input use.174 Figure A6.4: Latin America and the Caribbean: Number of Countries that Implemented Measures in Response to Rising Fertilizer Prices, by Subregion (During February to May, 2022) 12 10 2 8 6 3 5 3 6 2 2 4 1 4 1 2 3 4 4 3 1 1 6 5 1 1 1 2 2 2 3 1 1 3 2 2 2 2 2 2 1 1 1 1 1 1 0 Support for the production and use of organic fertilizers Vouchers or discounts for public services Price subsidies Domestic production of fertilizers and other inputs Agricultural insurance Techical assistance and training Monitoring prices and demand for fertilizers and inputs Transfering funds to farmers Grace periods, soft loans and guarantees Multi-stakeholder round tables Agroecological or circular production Providing agricultural supplies and services Other measures Tariff reduction or trade ffacilitation Dialogue with agrifood chain actors to maintain prices Programmes to increase efficiency in input use South America Central America and Mexico The Caribbean Source: ECLAC, FAO and WFP (2022) 174  See ECLAC, FAO and WFP (2022). 149 2024