Report No. 27523-CO Colombia Agricultural and Rural Competitiveness December 19, 2003 Colombia and Mexico Country Management Unit Environmentally and Socially Sustainable Development Unit Latin America and the Caribbean Region Document of the World Bank A STUDY OF COLOMBIA'S AGRICULTURALAND RURALCOMPETITIVENESS Table of Contents Acronyms and Abbreviations ................................................................................................................... iv Executive Summary .................................................................................................................................... v 1 . Introduction: Origins andPurposes of the Study ....................................................................... 1 2. The Approach ................................................................................................................................. 1 2.1. Methodological Issues ........................................................................................................ Scope of the Study .............................................................................................................. 1 2.2. 2 2.2.1. Concepts and Indicators of Competitiveness......................................................... 2 2.2.2. 3 4 Factors for Competitiveness at the Country Level................................................. FiscalConstraints................................................................................................... The Violence.......................................................................................................... 2.2.3. 2.2.4. 5 3. Evolutionof the Agricultural Sector andI t s Policies ................................................................. 6 3.1. 6 An Outline of Agricultural Policy Reforms inthe PastDecade......................................... A Sector inTransition......................................................................................................... 3.2. 3.2.1. Product Chains: A Competitiveness Policy Instrument......................................... 8 10 4. The Competitiveness of Colombian Agriculture by Product ................................................... 12 4.1. 12 4.2. Comparative Advantage Analysis for Grains and Short-Cycle Oilseeds.......................... Methods of MeasuringCompetitiveness........................................................................... 14 4.3. Comparative Advantage Analysis for FruitsandVegetables ........................................... 15 4.4. 16 4.5. Comparative Advantage Analysis for Tubers. Tobacco. and Cotton................................ Comparative Advantage Analysis for Other Long-Cycle Crops ...................................... 17 4.6. Comparative Advantage Analysis for Livestock Products ............................................... 19 4.7. 20 4.8. Comments on Other Products........................................................................................... Comparative Advantage Analysis for Forestry................................................................. 21 5. The Role of Exportsinand the Effect of Policies on Competitiveness .................................... 22 5.1 22 5.2 The Effects of Policies on Competitiveness ..................................................................... The Role of Exports inCompetitiveness .......................................................................... 23 Agricultural Resource BasePolicy Options............................................................ 5.2.1. Incentives for Agricultural Growth........................................................................ 5.3. .........23 28 5.3.1. LandTenure Policy................................................................................................ : 28 5.3.2. Irrigation Policy ..................................................................................................... 30 5.4. Policies for Access to Technology and Markets............................................................... 5.4.1. Agricultural Technology Policy............................................................................. 30 30 5.4.2. Rural Infrastructure andMarketing........................................................................ 33 5.4.3. Rural Finance Issuesand Options .......................................................................... 34 6. Competitiveness and the Campesino Economy. Approaches to Poverty Reduction. andSustainable Competitiveness ............................................................................................... 6.1. The Role of Campesino Production.................................................................................. 35 35 6.2. Policies to Improve the Competitiveness of Campesino Producers.................................. 36 6.2.1. ImprovingIncentives for Campesino Production................................................... 37 6.2.3. ImprovingCampesino Access to Markets, Inputs,and Technology ...................... 6.2.2. ImprovingCampesino Access to Basic Resources................................................. 38 39 6.3 Approaches to Poverty Reduction .................................................................................... 39 6.4 Sustainable Competitiveness ............................................................................................ 41 7. Conclusion .................................................................................................................................... 42 7.1. Strategic Considerations ................................................................................................... 42 7.2. Meeting InternationalStandardsinthe Agricultural Sector ............................................. 43 7.3 Recommendations for a Transition Program................................................................... .44 Annex 1: Tables, Figures, andBox CitedinText .............................................................. 47 Table A.1.1. Distributionof LandArea andEmploymentinCrop Agriculture Table A.1.2. Family and Local Agroprocessing Activities andSmallholder Marketing Channels Table A.1.3. NominalEconomic Protection Providedto 11PrincipalCrops, 1991-2001 Table A.1.4. Product Chain Organizations Table A.1.5. Colombian Agricultural andAgroindustrial Exports Table A.1.6. Export Diversification Table A.1.7. Destinationof Agricultural andAgroindustrial Exports Table A.1.8. Effects on Employment and Incomeof a Shift inIncentives toward Products with Comparative Advantage Table A.1.9. Effects on the Livestock Sector of Variations inTariffs on Corn and Soybeans Table A.1.10. Committed andDisbursed Aggregate Measures of Support to Agriculture Table A.1.11.Budget Appropriations for Agriculture, 1993-2002 Table A.1.12. Value of ICR Subsidies Paidby Type of Investment Table A.1.13. Impact of 10Percent Exchange Rate Change on Some RegionalAgricultural Products Table A.1.14. Distribution of Rural LandinColombia, 1998 Table A.1.15. Agricultural Lending, 1998-2001, Disbursements inMillionUS$ Table A.1.16. Distribution of FINAGRO Credits by Economic Size of Beneficiary Table A.1.17 Participation inAgricultural Productionof Smallholders and CommercialProducers: Crop Classificationby Area Table A.l. 18. Evolution of the Productivity Gap between Smallholders and CommercialFarmers Figure A.l.l InternationalPrices, Tariffs, and Exchange Rate Effects for 10Crops, 1991-2001 Figure A.1.2. InternationalPrices, Tariffs, and Exchange Rate Effects for 4 Livestock Products, 1991- 2001 Figure A.1.3. Illustration of a LandTax and Direct Payment SystemTogether (vertical scale inUS$) Box A.1.1. Product Chains Annex 2: Comparative Advantage Analysis of SelectedColombianAgricultural Products ............65 Table A.2.1. NominalProtectionRates for Grains and Soybeans Table A.2.2. Colombian InternationalTrade in Grains Table A.2.3. Mean and Median Tariffs for Grains by Region of the World, 2002 Table A.2.4. Revealed Comparative Advantage for ProcessedProducts of Grains, Short-Cycle Oilseeds (export values inmillion US$) Table A.2.5. Comparative Advantage Indicators for Grains and Soybeans, 2001 Table A.2.6. Comparative Advantage Indicators for Grains and Soybeans, 2001, with a 20 Percent Increase in InternationalPrices Table A.2.7. National Average Coefficients of Labor Use by Crop (inperson-days/ha.) Table A.2.8. Exports of Fresh and ProcessedFruitOther Than Bananas and Plantains, 1997-2001 Table A.2.9. RevealedComparative Advantage for Fruitsand Vegetables Table A.2.10. Comparative Advantage Indicators for Fruits and Vegetables, 2001 ll Table A.2.11. RelativeTrade Balance of Vegetable Oils Table A.2.12. RelativeTrade Balance for the Cacao Chain Table A.2.13. RevealedComparative Advantage for Other Long-Cycle Crops and Their Products Table A.2.14. Comparative Advantage Indicators for Other Long-Cycle Crops, 2001 Table A.2.15. RelativeTrade Balance for the Potato Chain Table A.2.16. RelativeTrade Balance for the CottonChain Table A.2.17. RelativeTrade Balance for the Tobacco Chain Table A.2.18. RevealedComparative Advantage for Tubers, Tobacco, andCotton Table A.2.19. Comparative Advantage Indicators for Tubers, Tobacco, andCotton, 2001 Table A.2.20. DRCs for Cotton, Calculated with a 15 Percent IncreaseinInternationalPrices Table A.2.21. RelativeTrade Balance for Beef and Beef Products Table A.2.22. RelativeTrade Balance for Dairy Products Table A.2.23. Comparative Advantage Indicators for Livestock Products vis-84s the U.S. Market, 2001 Table A.2.24. Comparative Advantage Indicators for Poultry vis-8-vis Brazil Table A.2.25. RelativeTrade Balance for Forestry Table A.2.26. Comparative Advantage Indicators for Forest Species Table A.2.27. Exports of Principal SeafoodProducts Table A.2.28. BogotB, Comparison of Prices for Conventionaland Organic Agriculture Annex 3: FormulasUsedinthe Analysis of the Competitiveness of ColombianAgriculture .........98 StatisticalAnnex ..................................................................................................................................... 100 Bibliography .......................................................................................................................................... ,137 ... lll Acronymsand Abbreviations ALADI Asociaci6nLatinoamericanade ICA Instituto Colombiano Agropecuario Integraci6n ICR Incentivefor RuralCapitalization ASOCA~~A Asociaci6nde Cultivadoresde CaRade IDEAM Instituteof Hydrological, Azlicar de Colombia Metereologicaland Environmental ATPDEA Andean Trade PromotionandDrug Studies, Instituto de Hidrologia, EradicationAct Meteorologiay EstudiosAmbientales AUGURA Asociaci6nde Bananerosde Colombia IDEMA NationalAgricultural Marketing Bancoldex Bancode ComercioExterior de Institute Colombia IFAD IntemationalFundfor Agricultural CCI Corporaci6nColombiaIntemacional Development CERTs The traditionaltax-rebatemechanism IGAC Instituto Geogdfico Augustin CIAL LocalAgricultural Research Codazzi, NationalGeographic Institute Committees IICA Inter-AmericanInstitutefor CIAT IntemationalCenterfor Tropical Cooperationon Agriculture, Instituto Research,CentroIntemacionalde Interamericanode Cooperaci6nparala Agricultura Tropical Agricultura CLAYUCA ConsorcioLatinoamericanoy delCaribe IMF IntemationalMonetaryFund de Apoyo a la Investigaci6ny el INAT Instituto Nacionalde Adecuaci6nde Desarrollode la Yuca Tierras, the IrrigationInstitute CONPES NationalCommissionfor Economic INCORA NationalColombianInstituteof and SocialPolicy, Comisi6nNacional de AgrarianReform, Instituto Nacional PoliticaEcon6micay Social Colombiano de la ReformaAgraria CORABASTOS The traditionalwholesale marketfor INDERENA NationalInstituteof NaturalResources the country INPA Instituto Nacionalde Pescay Acuicultura CORPOICA Corporaci6nColombianade ISNAR IntemationalService for National Investigaci6nAgropecuaria Agricultural Research DANE NationalDepartmentof Statistics, MADR Ministeriode Agricultura y Desarrollo DepartamentoNacionalde Estadisticas Rural DEAGRO Direcci6nde DesarrolloAgrario NGO Nongovemmental organization DNP Departmentof NationalPlanning, OECD Organisation for Economic Co- DepartamentoNacionalde Planeaci6n Operation andDevelopment DRCs Domesticresourcecost coefficients PADEMER Programfor ruralmicroenterprise DRI Integratedrural development PIPAA of Guatemalaindeveloping afood DTF Dep6sitosaT6rmino Fijo safety certification EU EuropeanUnion PRAN NationalProgramfor Agricultural FA0 FoodandAgriculture Organization Reactivation FEDECAFE Federaci6nNacionalde Cafeterosde PRONATTA ProgramaNacionalde Transferencia Colombia de Tecnologia Agropecuaria FEDEPALMA FederacidnNacionalde Cultivadores RCA Revealedcomparativeadvantage de Palmade Aceite RTB Relativetrade balance FEDERACAFE FederacidnNacionalde Cafeterosde SAC Sociedadde Agricultoresde Colombia Colombia TDAs Titulos de DesarrolloAgropecuario FINAGRO Fondode Financiamientoparael TRQs Tariff-rate quotas Sector Agropecuario UAF Unidades agricolas familiares FIP Fondode Inversionesparala Paz UMATAs MunicipalUnits for Agricultural DRI Desarrollo Rural Integral TechnicalAssistance FIP Fondode Inversionesparala Paz USAID United States Agency for Intemational FORAGRO ForoRegionalparala Investigacibny Development el Desarrollo Tecnol6gico Agropecuaio USDA UnitedStates Depaamentof GDP Grossdomesticproduct Agriculture GMS Globalmeasureof support WTO World TradeOrganization HACCP Good ManufacturingPractices A STUDY OF COLOMBIA'SAGRICULTURALAND RURAL COMPETITIVENESS Executive Summary The purpose of this study is to assess agriculture's competitiveness in Colombia. Duringthe past 12 years, Colombia's agricultural sector has performed poorly, resulting in the continuation of extensive rural poverty. Improving the sector's competitiveness i s the only sure and lasting way to improve its growth performanceand reduce poverty. Thus, the main objectives of this study are to assess: (a) the sector's current and potential competitiveness, (b) the factors that determine or inhibit its competitiveness, (c) the extent to which the campesino economy i s competitive and could become more so, and (d) the role of policies in determining the sector's competitiveness. The study outlines strategic options that would strengthen the sector's competitiveness, with emphasis on development of transition measures to ease the path to adoption of new orientations. The work has been organized around three main themes: (a) the competitiveness of products and product chains, (b) underlying factors that influence competitiveness, and (c) policy orientations that influence competitiveness. Requirements for Competitiveness and Its Measurement Competitiveness means being able to export at a profit or compete profitability against imports. Comparative advantage implies being able to do it without subsidies. Profitability i s a basic indicator of competitiveness, but the most central requirement for maintaining competitiveness over time i s increasing productivity. Agricultural researchi s a key to improvingproductivity. A starting point for measuring competitiveness in the international context is reviewing the net trade balance by product group or chain. The second measure of competitiveness, revealed comparative advantage (RCA), provides a dynamic perspective on competitiveness. It shows whether competitive products are increasing their competitiveness or losing it, and whether currently noncompetitive products are moving toward competitiveness over time. Inherent comparative advantage is measuredby the methodof domestic resource cost coefficients (DRCs). A DRC represents the cost in domestic resources of earning or saving (a net) dollar of foreign exchange, through exporting or substituting for imported products, and therefore it requires information onproductioncosts. Constraints on the Agricultural Sector Violence in Colombia has displaced an estimated 2 million people from their homes, mostly lower-income families inrural areas, and has significantly reduced willingness to invest in modemization of agriculture by banks, individuals, and businesses. The majority of producers pay a cost for continuing to operate under these circumstances, including payment of extortion demands, provision of additional security measures, and less administrative attention to the farms for lack of personal security. Smallholders are particularly harmedby the effects of violence. The violence also inhibitsthe delivery of agricultural services. It discouragesbanks from lending in rural areas, and restricts the scope for agricultural extension and visits by phytosanitary inspectors, V whose approvals are essentialfor exporting. This i s especially prejudicial to the ruralpoor because export products are almost invariably the most labor intensive. On the positive side, however, it appears that the level of violence has abated somewhat in the last year. The study concludes that ending the violence will not be sufficient to restore the sector's dynamism. In spite of its moderate rebound in 2002, sustaining an agricultural revival will require changes in some basic orientations of economic policy for agriculture and also some deep institutional reforms. The agricultural sector has also suffered from fiscal constraints. Expenditures have fallen from an already low level of 8.0 percent of agricultural GDP in 1996 to 6.6 percent of agricultural GDP in 2001. In dollar terms, total fiscal expenditure in the sector fell from US$830 million in 1996 to US$328 million in 2002. Concomitantly, real expenditures on agriculturalresearchdeclined by 10percent during 1997-2001. As a proportion of agricultural GDP, Colombia's expenditure on agricultural researchi s one- sixteenth the level of agricultural researchexpenditures in developed countries, and less than one-seventh of the levelrecommended for countries inits stage of economic development. Colombia's InherentAdvantages Colombia has abundant amounts of unused and underused land. Of the more than 18 million hectares with agricultural potential, less than 4 million are cultivated, and Colombia's water resourcesper square kilometer are three times the South American average and six times the world average. The country i s richly endowed with soils and altitudes that are appropriate for coffee, the world's second- largest export product after petroleum. Unfortunately, however, Colombia's agronomic conditions are also suitable for cocaine andother illicit crops. Colombia is geographically relatively close to the largest market in the world, the United States, andit has adequateport facilities for participatingininternationaltrade. A disadvantage, however, is that not all producing areas of the country are well connected to ports or to major domestic markets, and therefore internal transport costs can be high. However, Colombia possesses a long tradition of entrepreneurship and agroprocessing, and has a relatively well-educated labor force and considerable talent inthe scientific and professional areas. A Sector inTransition Marked structural changes have occurred within the agricultural sector in the last 10to 15 years. Between 1985 and 2001, more than 1million hectares of agricultural land were converted from cropland to grazing areas. Now the land area devoted to livestock i s about double the area appropriate for that purpose, while crops occupy about one-third of the land that is appropriate for them. Within cropped areas, perennial crops have gained ground at the expense of most annual crops, and some export crops at the expenseof import substitutes. Although the sector was particularly hard hit by collapses in the coffee and cotton markets, other crops have expanded in area and production. For example, in 1991 plantains and other fruits (except bananas) representedonly about 60 percent of the value of coffee production, but in 2001 their production value was greater than that of coffee. During 1991-200 1, cereal crops, short-cycle oilseeds, and cotton experienced notable declines in production. The only major short-cycle crops to have escaped the negative trend of the 1990s were rice and potatoes. There was growth in a few lines of production, mainly some of the perennial crops (including fruit, oil palm, yams, sugar cane, and flowers). Some other perennial crops either did not vi experience growth or actually declined during the decade (including, in addition to coffee, bananas, cacao, jute, and dark tobacco). These broad trends continued in 2002 except for an upswing in coffee. A major question for the study is whether the trends reflect the workings of comparative advantage or other economic forces. There remains an acute dualism between a smallholder sector with insufficient land, often- backward production technology, and little access to ancillary services, and a modem entrepreneurial sector of medium-size and large farms with modern technology, which receives most of the production and marketing services and policy incentives. More than 60 percent of the rural population still live below the poverty line. Estimates are that 74 percent of the extreme poverty, and 57 percent of all poverty in the country, are found in rural areas. Poverty in rural areas has been exacerbated in recent years by the collapse of coffee prices, which eliminated more than 100,000 jobs. Without the shift in agricultural production toward the more labor-intensive products, rural poverty would be even worse today. Inthe 1990s Colombia took an important step toward institutionaldecentralization by devolving extension systems to the municipal level, in the form of Municipal Units for Agricultural Technical Assistance (UMATAs). The original idea behind the formation of the UMATAs was that smallholders would be able to choose their own extension agents, through local producers' committees. However, in practice the management and even staffing of UMATAs has become politicized at the local level in many cases, and it i s estimated that, of the 800-odd UMATAs originally created, only about half have performedsatisfactorily. PrincipalFindings on Comparative Advantage This report assesses comparative advantage for six groups of products, and comments are offered on additional products based on other studies. The six groups are (a) grains and soybeans; (b) fruit and vegetables; (c) coffee, sugar, palmoil, and cacao; (d) cassava, potatoes, tobacco, and cotton; (e) livestock products; and (f)commercial forestry. The first group of products, grains and soybeans, were found to exhibit a comparative disadvantage. One indication of this i s that, inthe past decade, imports of these products have been rising in relationto domestic production, in spite of the highand increasing levels of border protection enjoyed by most of them. On the other hand, a number of processedgrain products have demonstrated increasing international competitiveness, especially in the Andean regional context. These products include flours, breakfast cereals, animal foods, and baked products, and often their production i s carried out with imported primary products as inputs. Fruit and vegetable crops are found to have considerable comparative advantage, without exception. Their role in generating employment i s proportionately greater, and hence they are valuable crops for a poverty-alleviation strategy. They also enjoy the advantage of continuously increasing world demand. Hurdles to more rapid expansion of fruit and vegetable production for export include: (a) low rates of use of improved technologies, (b) a low priority for these crops in the national research system, (c) plant disease risk faced by producers, (d) a consistent downward trend inworld market prices for most fresh fruits and vegetables, and (e) lack of an integrated policy to stimulate and promote fruit and vegetable exports. The thirdgroup of products includes two of the four largest agricultural exports, coffee and sugar, and one of the fastest-growing exports, palm oil, plus cacao. All four of these crops have a comparative advantage, with the possible exception of cacao in part of the country. These crops are highly labor- intensive and therefore have considerable capacity for creatingemployment. Vii The fourth group of crops covers cassava, potatoes, tobacco, and cotton. In the case of all of them, processed forms of the products have been increasing in importance in Colombia, relative to the raw forms. Colombia has a comparative advantage in all livestock products. It is strongest for poultry, eggs, and milk, and least evident for extensive livestock raising. Beef-fattening operations and pork raising for the complete cycle fall in between. When the calculations are made with current rates of protection on feed grains, pork loses its competitiveness andpoultry's becomes marginal. Structural and sanitary barriers in world markets for poultry, beef, and milk inhibit the possibilities for expansion of Colombia's exports of these products, but the potential for it exists were world conditions to improve. Inthe caseofcommercial forestry, costdatasuggestthat Colombiahasacomparative advantage inseveral broadleafspecies, andto alesserextent insome pines andone species of eucalyptus. On the basis of other studies, it appears that Colombia can also be very competitive in cultivated shrimp andorganic agriculture. In overall terms, Colombia has comparative advantage, often strong, in export products as opposed to import substitutes, in perennial crops as opposed to annual crops, and in the more labor- intensive crops. All regions have a comparative advantage in some lines of production. Inmany cases realizing this comparative advantage will require concerted efforts in areas such as agricultural technology development, marketing programs, and food safety programs. It can also be promoted by modifications in priorities in incentives policies and by changes in the way in which some programs are carried out, including researchand extension and irrigation management. Incentives Policies and Competitiveness Trade and pricing policies are major instruments of production incentives, along with exchange rate policy and fiscal incentives. Inthe past, incentives provided through trade and pricing mechanisms and fiscal policies have been characterized by: (a) providing significantly more incentives to products with a comparative disadvantagethan those with a comparative advantage; (b) providing more incentives to import substitutes than to export products (with the principal exceptions of palm oil and sugar); (c) providing more incentives to annual crops than to perennial crops (although the government i s working to change this on the fiscal side); and (d) providing more incentives to large-scale farmers than to medium- scale farmers and smallholders. Priorities in the allocation of agricultural research resources are not aligned with comparative advantage. Inmany cases this pattern of incentives works against the sector's comparative advantage, and therefore i s not providing maximum benefits in terms of employment creation and sector growth. In addition, although there are a number of programs aimed at promoting exports, most of them are limited in scope and they are not integrated into a consistent strategy for agroexports. Developing and implementingthis kindof a strategic thrust would help realize Colombia's agricultural potential. LandandWater Policies The study finds that 40 years of agrarian reform in Colombia has not substantially altered the concentration of landholdings, and explores some of the reasons why this has occurred. It also discusses Viii the importance of developing land rental markets in order to promote access to land for the poor, and ways to strengthen such markets. Irrigation plays an important role in Colombian agriculture, but the national irrigation system has suffered a number of problems, especially the part of it that belongs to the public sector. For example, about a third of the more than 500 small irrigation districts are inoperative. For those districts, large and small, that continue to function, frequently observed problems include excessive applications of water, salinization, and waterlogging of soils, inappropriate cropping patterns inrelation to the value of water as a resource, and negative environmental consequences. A wide-ranging review of the country's irrigation systems i s warranted. Agricultural Technology Policy Colombia has registered significant accomplishments in agricultural technology. It has strong product-based research units in which producers' organizations play the principal role in defining researchpriorities, it has been a leader indeveloping participatory researchwith smallholders, it has given a prominent role to producer groups in defining national research priorities, and it has implemented a successful model of acompetitive fundfor allocatingresourcesfor technology transfer. Nevertheless, on a national level yields have not increased in line with expectations, real funding for research i s declining, research priorities have been questioned, there are important operational deficiencies in the extension service, and smallholders tend to receive few benefits from the technology system. A mission of International Service for National Agricultural Research (ISNAR) recently found that research was mainly concentrated on traditional, import-substitute crops, with relatively little attention given to crops inwhich Colombiahas a comparative advantage. The value of participatory research, in which farmers play an active role, should be incorporated into policy and institutionalized. In addition, the excessiveuse of pesticides, and the toxic nature of some of the compounds that are permitted need to be addressed, because the excessive use of pesticides weakens the competitiveness of Colombian agriculture and causes significant health problems among field workers. Finally, Colombia needs a clearly defined policy with respect to biotechnology, or genetic modification of plants. Work in this area could have high payoffs for Colombian farmers, especially if genetically modified coffees that are resistant to rust could be developed, and cotton made less vulnerable to pests. The linkage to markets is crucial. RuralInfrastructure, Marketing, andFinance Colombia is well endowed with ports and airports, and airfreight i s the preferred mode of export shipment for some types of producers, such as flower growers. The ports of Buenaventura, Santa Marta, and Barranquilla have adequate depth for ships of deep draught to enter, and port infrastructure and administration have been improved dramatically since 1990. Thanks to this progress, the costs of port- handlingserviceshavefallen from around US$50or more per ton in 1990to US$8 per ton today. The main transportation bottleneck in many rural areas i s inland, in the form of lack of adequate secondary and tertiary roads. In addition, the primary road network i s poorly developed in an east-west direction. Agriculture would benefit considerably from expansion of the network in other directions and from improvement of rural feeder roads. ix One of the most important steps policy can take to strengthenthe linkages between producers and modern retail outlets i s to develop legislation that requires that payment be made to growers of fruits and vegetables within 30 days of delivery of the product. Such a law was adopted in Argentina in 2001, and similar legislation has beenunder consideration inCostaRica and Brazil. Drastic changes have occurred in the rural financial sector in recent years, and the framework of bank supervision and regulation should be reviewed in order to encourage more private banking in agriculture, strengthen rural microfinance, concentrate rediscount lines more in areas of Colombia's comparative advantage, and develop mechanismsto provide finance for land acquisition. Competitiveness andthe Campesino Economy Smallholders provide about two-thirds of total agricultural production, a percentagethat i s higher when illegal crops are included. Campesino production often i s characterized by lower physical yields but much lower monetary costs, compared with larger-scale farms. Nevertheless, there i s some evidence that over time campesinos have been quick to adapt technologies and have been more successful than larger farmers inraising their yields. A principal indicator of the competitiveness of campesino production is its extensive participation in exports, such as in coffee, milk and milk products, plantain, raw tobacco, chocolate products, and certain fruits, vegetables, and fruitjuices. Campesino production can be competitive, and when it is, it i s founded on efficient use of factors of production, especially of family labor, and careful control of cash costs. In spite of these indications of productivity and entrepreneurial ability among campesinos, poverty remains widespread in rural areas, and an integral, consistent strategy i s needed for the smallholder sector. More training i s neededfor producers inregard to quality norms and marketing. The government might explore ways to partially guarantee long-term sales contracts to the buyers. Campesinoshave a reputation for breaking contracts when they find a temporarily higher price elsewhere, and a priority should be training about the long-term benefits of becoming a reliable supplier and fulfilling contracts. In the area of rural finance, the Fondo de Financiamiento para el Sector Agropecuario (FINAGRO)should give smallholders a higher priority inprograms. MeetingInternational Standards inthe Agricultural Sector Making the Colombian agricultural sector more competitive requires ensuring it and its institutionsmeet internationalstandards. This requires attention to the following three factors: 0 Quality. In both international markets and domestic markets in developing countries, quality considerations for agricultural products are rapidly becoming more prominent. Price differentials have emerged according to product quality, and international markets increasingly are becoming closed to products that do not satisfy basic standards, particularly with respect to food safety and plant and animal hygiene. Achieving better quality requires both training producers and institutional support. 0 Confidence. Confidence i s also integral to competitiveness. It i s required in all commercial and financial relationships. Smallholders need to demonstrate reliability in fulfilling contracts in order to be able to gain access to more stable, commercial markets. Equally, exporters and middlemen need to make timely payments to producers in order to gain their confidence for X future purchases. Confidence is basic to agricultural lending, particularly to smallholders, where social capital substitutes for physicalcapital. Consistency. A principalrequirement for effective agricultural policy i s that it be consistent, over time, among regions, and over products. Uniformity of the rules of the game, and consistency in their application, i s essential in order for policy to be effective. Consistency in exchange rate policy i s perhapsthe most basic requirement for successful agriculturaldevelopment, but equally, inthe private sector, the more consistent producers can become inmeetingquality requirements and delivery times, the greater their opportunities to participate inmore commercial agriculture. Strategic ConsiderationsandTransitionPrograms Although most of Colombia's products are already competitive by international standards, many steps can be taken to improve the sector and to achieve the goal of alleviating poverty. Inaddressing the issues, a transition program is recommended-one that would be consistent with an overall strategic vision, including: Giving agriculturea higher priority inthe nationalbudget. Reallocating some expenditures in the sector with greater emphasis on supporting products with comparative advantageand programs to assistthe rural poor. Strengthening and making more efficient the programs in areas such as quality control and food safety, agricultural researchand extension, and irrigation development and management. Development and implementation of a full, integrated agroexport program and associated measuresfor improving and guaranteeingproduct quality. Emphasizing institutional decentralization and reforms and improvements in areas such as some of the parafiscal funds, the UMATAs, and the agricultural finance system. Farmers need to have the right to choose their extension agents, and dismiss ones that do not deal adequately with their concerns. Strengthening land markets with emphasis on mechanisms for promoting greater access to land on the part of the rural poor, via both rental and purchase. Implementinga comprehensive strategy for reduction inrural poverty, based in part on improving the competitiveness of the campesinosector. Gradually reducing the current high levels of protection for products without a comparative advantage, butnot eliminating those tariffs as long as internationalprice distortions exist. Providing new forms of support to producers that reach export crops and import substitutes, and smallholders and more commercialfarmers. Linkingincentives to performance andto compliance with environmental conditions. Greater strategic emphasis on realizing Colombia's potential in aquaculture and sustainable forestry, as well as crops and livestock products that have comparative advantage. Greater emphasis on organic cultivation, for both domestic and external markets. Maintaininga competitiverealexchangerate. Public policy i s already moving to incorporate these recommendations, but a comprehensive outline of concrete agricultural development policy, or an agricultural development strategy, i s not yet available to producers or the public. Developing it, with considerable policy specificity in its proposals and in consultation with the private sector, could help guide decisionmaking throughout the sector and help ensure consistency inthe many detailedpolicy decisions that are made on a day-to-day basis. xi Recommendations Shifts in priorities for fiscal expenditures and trade policy always result in helping or hurting various players. In the long term, however, the result is a more competitive sector. The following recommendations are suggestedfor inclusion in a transition programto make the Colombian agricultural sector more competitive, and thus help alleviate poverty: e Immediate increases in funding for the agricultural sector, particularly for agricultural research, as a matter of the highest urgency, along with development of mechanisms for establishing regional research agendas and development of a policy framework for supporting participatory research. Reconversion tarzffs, through which funding i s provided for investments to either improve the productivity of currently noncompetitive lines of production, or to support the transition to other types of production, while an import tariff i s being scaled downward. Under existing fiscal legislation it may be difficult to tie the proceeds from those tariffs directly to expenditures for the subsectors concerned, but approximate equivalences could be developed. An essential aspect of the program would be to require investment plans linked to credible increases in productivity at the farm level, in an existing line or production or a new one, as a condition for approval of the funding. Another aspect would be a plan for progressive tariff reduction that would acquire the status of an international treaty, so it would be binding. Strongerfiscal supportfor the sector during an extended transition period, perhaps provided in part by international entities. Negotiations with the international agencies (including the International Monetary Fund) over such support should underscore the fundamental role of agricultural development inreducing poverty (and inreducingproductionof illicit crops), and the nature of institutional and policy reforms that would accompany the expenditures. e Initiation on apilot basis of linkedprograms of direct support per hectare and area-based land taxes. Some funding now directed to other incentives could be reallocated for this purpose. Work would begin immediately in pilot locations in developing local institutions and participatory procedures for allocating some of the land tax receipts to infrastructure and training programs, under central government monitoring and supervision. While most producers would not perceive the economic effects of these programs right away, they could see the direction of changes in incentives that would flow from the programs and begin to adapt their production accordingly. Institution of across-the-board requirementsfor demonstrated productivity increases in order to continuereceivingfiscal incentives directedtowardproducts. e Development of an immediate agreement with producers to review and restructure the parafiscal funds whose rates of compliance with quotas have been very low and/or whose expenditures on administration have been high. e Development of an outline of a land fund and consultativeprocesses for formulation of the definitive structureandproceduresfor such afund. Again, the effects on producer expectations andbehavior would beginto be registeredduringa transitionperiod. Xii Developmentand disseminationof model contractsfor land rental and sharecropping. This is a measure that could be carried out quickly, in consultation with producer groups in different parts of the country. Provision of funding for export marketing on an exploratory basis and visits by producers to international tradefairs and marketing associations, under clear, published rules for selection of those who participate insuch activities. Establishment of thefirst Rural Centersfor Entrepreneurial Development on a pilot basis in selectedareas. Developmentof draft legislationto clarifv responsibilitiesfor ensuringfood safety. Immediate increases in the funding of Znstituto Colombiano Agropecuario (ZCA) with a mandate to develop a program for certification of the acceptability of export products according to internationalnorms of food safety and plant andanimal hygiene. Taken together, a package of short-term measures such as these would send an unmistakable signal that priorities and incentives in the sector are changing, and would provide some short-term assistanceto groups most affected by the transition. Xiii A STUDY OF COLOMBIA'S AGRICULTURAL COMPETITIVENESS 1. Introduction: Origins and Purposes of the Study The purpose of this study i s to assess agriculture's competitiveness in Colombia. Duringthe past 12 years, Colombia's agricultural sector has performed poorly, resultingin negative effects on the rest of the economy, and the continuation of extensive rural poverty. Improving the sector's competitiveness i s the only sure andlasting way to improve its growth performance and reduce poverty. Agriculture traditionally has been one of the cornerstones of Colombia's economy. Crop and livestock production expanded rapidly during the 1970s and the 1980s, growing at annual rates of 4.6 percent and 2.9 percent, respectively, in those decades. Inaddition, agroindustry always has represented the bulk of industrial activity, providing 64 percent of total industrial output in 1970 and 62 percent in 1990 (Balcdzar, Vargas, and Orozco 1998%). Agricultural and agroindustrial exports together generated more than $2.9 billion in foreign exchange in 2001, about 36 percent of total legal exports. Primary agriculture generates more employment than any other sector-seven times the employment in the entire industrialsector (DANEHousehold survey). However, the sector lost its customary vigor in the 1990s. Between 1990 and 2000 crop production expanded by only 0.7 percent per year and livestock production grew by only 2 percent per year. In 2001 and 2002 agricultural growth remained unacceptably low, in spite of sharp movements inthe real exchangerate that provided greater incentives to the sector. Today, after a decade of unsatisfactory performance throughout the economy, Colombia is at a crossroads. Illegal activities and armed civil conflict continue, and rural areas are still especially affected by the crisis. If violence and social disruption do not abate, the economy couldfall deeper into crisis, and this in turn could compound social unrest and civil strife. The weakening economy has seriously inhibited efforts to reduce poverty, which still affects more than 60 percent of the rural population. This study was conceived in conversations between the Government of Colombia, the World Bank, and the Food and Agriculture Organization (FAO) in September 2002. In October 2002 a study team was assembled, consisting of Colombian experts and an international coordinator at the technical level. The World Bank andthe FA0 provided overall supervision of the study, and UnitedStates Agency for InternationalDevelopment (USAID) supplied additional technical and financial support. The United States Department of Agriculture (USDA) provided comments and suggestions. Government staff and private sector leadersalso have provideduseful feedback. Throughout the endeavor, coordination has been maintained with relevant government agencies, particularly staff of the Ministry of Agriculture and Rural Development (MADR) and of the National PlanningDepartment (DNP),and with representatives of the private sector and other Colombian experts. The authors are grateful to all of them for their helpful inputs. This draft of the study i s designed to stimulate feedback and dialogue on the issuesraised. 2. The Approach 2.1. Scope of the Study The purpose of this study is to assess agriculture's competitiveness in Colombia. Specifically, it assesses: (a) the sector's current and potential competitiveness, (b) the factors that determine or inhibitits competitiveness, (c) the extent to which the campesino economy i s competitive and could become more so, and (d) the role of policies in determining the sector's competitiveness. The study outlines strategic 1 options that would strengthen the sector's competitiveness in the long term, with emphasis on development of transition measures to ease the path to adoption of new orientations. Requirements for Competitiveness andIts Measurement Competitiveness means being able to export at a profit or compete profitability against imports. Comparative advantage implies being able to do it without subsidies in net terms. Profitability i s a basic indicator of competitiveness, but the most central requirement for maintaining competitiveness over time i s increasing productivity. Agricultural researchi s a key to improving productivity. In light of the heterogeneity of the sector and the diversity of issues confronted, the study is organized around three main themes: (a) the competitiveness of products and product chains, (b) underlying factors that influence competitiveness, and (c) policy orientations that influence competitiveness. The specific issues covered in the study reflect the complexity of the sector, and include: food safety and quality issues for export markets; the need for marketing links and input supply channels; agroclimatic conditions and irrigation; research and extension efforts; the agricultural financial system, including access to credit; human capital and levels of rural education; institutional aspects, particularly food safety and quality control; the effects of the continuing violence in ruralareas; land tenure; the role of farmer organizations incompetitiveness; andpossible effects of internationaltrade negotiations. 2.2. Methodological Issues 2.2.1. Concepts and Indicators of Competitiveness Profitability i s a basic indicator of competitiveness, and the most central requirement for maintaining competitiveness over time i s increasing productivity, both in terms of yields per unit of land and in cost reductions per unit of output. As markets become more sophisticated and demanding, sustaining competitiveness increasingly refers to the quality dimension. Maintaining and increasing quality in turn requires investment in human capital and connectivity, and the ability to adapt continuously to changing conditions. Agricultural research i s a key to improving productivity and it generally has shown very high returns. Changing the sector's product composition toward higher-value products i s fundamental for increasing productivity at the sectorwide level. For this study, competitiveness i s measured quantitatively, and qualitative assessments are presented as well, some of them on the basis of other studies. In the study's quantitative work, the benchmark i s the internationalmarket, because as globalization advances, worldwide, the reference point for competitiveness is becoming the international market. Furthermore, competitiveness vis-&vis international markets i s a basic guideline for more efficient resource allocation and therefore for achieving higher economic growth rates. A starting point for measuring competitiveness in the international context is reviewing the net trade balance by product group or chain. Colombia's agricultural trade balance as a whole i s still strongly positive, but agricultural imports have been risingmuch more rapidly than exports over the past 10 years. (See Statistical Annex tables.) The second measure of competitiveness, revealed comparative advantage (RCA), provides a dynamic perspective on competitiveness. It shows whether competitive products are increasing their 2 competitiveness or losing it, and whether currently noncompetitive products are moving toward competitiveness. Inherent comparative advantageis measuredby the method of domestic resource cost coefficients (DRCs). A DRC representsthe cost in domestic resources of earning or saving (a net) dollar of foreign exchange, through exporting or substitutingfor imported products, and therefore it requires information on production costs. For this study, 27 crops and 6 livestock products were analyzed. In addition, comparative advantage calculations were made for 12commercial forestry species, allowing a quantitative assessment of the types of production lines inwhich Colombia has a comparative advantage. For its part, the revealed comparative advantage approach indicates trends over time in competitiveness and extends the analysis to a larger number of products, including to a significant number of agroindustrial products for which DRCs could not be calculated owing to lack of availability of information on cost structures. To strengthen the dynamic aspect of the analysis, qualitative information has been gathered on conditions specific to the production of each product. On that basis it has been possible to characterize the agroclimatic conditions and agricultural research strategies under which, for example, production of cacao, cassava, and cottonmightbecome more competitive. Another dimension of competitiveness i s the international one. Here the principal factors are agricultural policies inOrganisation for Economic Co-operation andDevelopment (OECD) countries and the competition in export markets coming from other developing countries. The international price distortions arisingfrom agriculturalpolicies indeveloped countries are taken into account in this study. 2.2.2. The Violence Violence in Colombia has displaced an estimated 2 million mostly lower-income people from their homes, and has significantly reduced willingness to invest in modernization of agriculture by banks, individuals, and businesses, undermining Colombia's agricultural competitiveness. The majority of producers pay a cost for continuing to operate under these circumstances, including payment of extortion demands, provision of additional security measures, and less administrative attentionto the farms for lack of personal security. Many lower-income farmers who cannot afford to pay the costs of the violence are driven off the land or worse. Inaddition to major threats, they routinely suffer theft of farm animals and crops (World Bank 2003:40). Smallholders are particularly harmed by the effects of violence. The violence also inhibitsthe delivery of agricultural services. It discourages banks from lending in rural areas, and restricts the scope for agricultural extension and visits by phytosanitary inspectors, whose approvals are essentialfor exporting. This is especially prejudicialto the rural poor becauseexport products are almost invariably the most labor intensive. The violence has also undermined basic societal institutions. In addition, the combination of violence and unemployment has spawned an attitude of growing acceptance of illegal activities (ibid.). According to Misidn Paz (2001:54-55)' guerrillas or paramilitaries operate in two-thirds of the country's municipios. The violence also inhibits the delivery of agricultural services of various kinds (ibid.), and raisescosts of production, especially for livestock. On the positive side, however, it appears that the level of violence has abated somewhat in the last year, and a regular annual survey of about 400 commercial agricultural producers found that the 3 concern over lack of security fell from second rank in importance to sixth within the past year (SAC 2003). However, ending the violence will not be sufficient to restore the sector's dynamism. That will require changes in some basic orientations of economic policy for agriculture, and deep institutional reforms. 2.2.3. Fiscal Constraints The agricultural sector has also suffered from fiscal constraints. Expenditures have fallen from an already low level of 8.0 percent of agriculturalGDP in 1996to 6.6 percent of agricultural GDP in 2001 (MADR2003).' Indollar terms, total fiscal expenditure inthe sector fell from US$830 millionin 1996to US$328 million in 2002 (see Table A.1.11). Concomitantly, real expenditures on agricultural research declined by 10percent during 1997-2001 (see Statistical Annex tables). Now Colombia's expenditure on agricultural research is very low by any standard. As a proportion of agricultural GDP, Colombia's expenditure on agricultural research is one-sixteenth the level of agricultural research expenditures in developed countries, and less than one-seventh of the level recommended for countries at its stage of economic development.2 Along with the declines in research expenditures, fiscal stringency has had the most deleterious effects on the sector by weakening the performance of a few key institutions such as the Znstituto Colombiano Agropecuario (ICA) in the area of phytosanitary controls, and the Corporacidn Colombia Znternacional (CCI) in the area of support for export crops. It also has weakened the Ministry of Agriculture's ability to oversee programs and institutions such as the parafiscal funds.3 Although some forms of fiscal support to the sector are not consistent with the sector's comparative advantage, given the magnitude of the problem of rural poverty, there i s an urgent need for fiscal support to small farmers and the rural landless that i s more effectively targeted on those groups. A fiscal retrenchment has been unavoidable because of the sharp increase in public debt as a share of GDP-it has more than doubled since 1995-and because it appears that the fiscal deficits are not cyclical but rather are structural. This study concludes that a more viable fiscal strategy for Colombian agriculture would includethe followingelements: 1. Fiscalsupport that is moreeffectively targetedon the ruralpoor 2. Incentives that are reoriented toward crops with comparative advantage rather than toward those with poor future prospects 3. Stronger fiscal support for a few institutions that are crucial for the sector's development 4. Measuresto increasegovernment revenue collection inrural areas 5. Greater priority for agriculture in general in the government budget, given the unique contribution of the sector to reducingpoverty. Arguments for generalized support for agriculture, and for makingit the last sector to suffer the consequences of fiscal cutbacks, are the prevalence of rural poverty and the fact that agriculture has suffered the effects of "Dutch disease" with regard to the exchange rate, because of the large volume of foreign exchange earnings generatedby illegal crops: This has raisedthe real exchange rate over what it would have been in the absence of the illegal trade, and therefore it has depressedreal agricultural prices 'The percentagesare calculatedonthe basis of budgetdisbursementsfor MADR, ICA, INCORA, INAT, INPA, IDEMA, BuncoAgrurio, and the Fondo DRI, excluding debt payments but including investmentexpenditures. * Intemationalbenchmarks for agriculturalresearchexpenditureare providedinSecretariado Te`cnicodel FORAGRO IICA (2000). Communicationsbetweenthe MADR andthe Contralodu General, June 2001, onthe topic of the parafiscalfunds. Thoumi (2002:186) estimates that the value of illegal exportearnings inColombiawas around US4 billion to US$5 billion per year inthe late 1990s. that is, around25 percentof total export earningsor more. While there are other estimatesalso, it is clear that the magnitudeof the illegal trade is very substantial. 4 within the economy, which in turn has represented a disincentive to g r ~ w t h .In addition, in the past ~ decade solid evidence has emerged that agricultural growth i s not only effective in alleviating rural poverty, but it i s more effective than industrial growth in reducing urban poverty (Ravallion and Datt 1996:19). As valid as these arguments for support to agriculture are, this study also emphasizes the needto make fiscal expenditure in Colombian agriculture more efficient and to find ways for agriculture to contribute more effectively to fiscal revenues. 2.2.4. Factorsfor Competitivenessat the Country Level Colombia's Inherent Advantages and Disadvantages Colombia has abundant amounts of unused and underused land. Of the more than 18 million hectares with agricultural potential, less than 4 million are cultivated, andColombia's water resources per squarekilometer are three times the South American average and six times the world average (MisibnPaz 2001:19). Another advantage that Colombia possesses i s a marked diversity of microclimates and ecological zones, which enable it to supply to the market the same kind of fruits and vegetables 12 months of the year. Colombia's abundance of ecosystems also endows it with great biodiversity-more thanany other country inthe hemisphereexcept Brazil, which has six times more territory. On the other hand, Colombia does not possess the type of broad, fertile plains that are suitable for efficient production of crops like corn and soybeans, which are subject to economies of scale and hence to large-scale mechanization, with the possible exception of parts of the Orinoquia. The country i s richly endowed with soils and altitudes that are appropriate for coffee, the world's second-largest export product after petroleum. However, this can expose the country to economic and social risks since coffee prices have had a long history of volatility. The risks can be somewhat attenuated by appropriate strategies, such as concentrating on coffees of high quality and emphasizing product diversification. Unfortunately, however, Colombia's agronomic conditions are also suitable for cocaine and other illicit crops. Colombia is geographically relatively close to the largest market in the world, the United States. This gives it an advantage in transportation costs relative to other major agricultural exporters such as Argentina, Brazil, Chile, Ecuador, and New Zealand. It also has adequateport facilities for participating in internationaltrade. This geographic advantage has not been sufficiently exploited. A disadvantage, however, i s that not all producing areas of the country are well connected to ports or to major domestic markets, and therefore internaltransport costs can be high. Another advantage i s that Colombia possesses a long tradition of entrepreneurship and agroprocessing, and has a relatively well-educated labor force and considerable talent inthe scientific and professional areas. Until the mid-1950s, Colombia's agriculture was more biased toward primary production, and less toward agroindustries, than the typical developing country of its income level. But since the 1950s, the Colombian agricultural pattern has become increasingly more industrialized than that of the rest of the developing world (Berry and Bejarano 1990). This, too, i s an advantage that has not been fully exploited. Today Colombia has not only large-scale industries based on its own primary products (oil palm, coffee, bananas, processed fruits, dairying, sugar, cacao, aquaculture, tuna fishing, rice, and so forth), but also export-oriented agroindustries whose principal source of raw material i s imported (for example, wheat millingand confectionaries, poultry, and leather goods6). The effects of appreciating real exchange rates on real agricultural prices have been well documented throughout the world. For an early summary of the most systematicresearch onthe topic, see Krueger, Schiff, and Vald6s (1988). The leather goods industry uses primarily imported hides not because of the lack of national hides but rather because of their frequently poor quality. 5 A disadvantage of a historical nature that Colombia shares with many other Latin American countries i s the tradition of large landholdings operatedat low levels of intensity, and the resulting social schisms and distrust in rural areas. In addition to the inequitable land distribution that this tradition has bequeathed to present generations, it has inhibited, for example, the development of active land rental markets, which potentially could constitute a major channel of access to land for the rural poor. The history of landlessness, land invasions, and land reforms has created a situation in which landowners are hesitant to rent out their land for fear of losing it. Overcoming the negative aspects of this legacy i s a major challenge for agriculturalandrural policy. 3. Evolution of the Agricultural Sector and Its Policies 3.1. A Sector inTransition Inadditionto the weakening of overall economic performance in agriculture, markedstructural changes have occurred within the sector in the last 10 to 15 years, in response to external and internal economic trends, new economic policies, and the pervasive violence. According to recent estimates of Colombia's National Geographic Institute (IGAC), more than 1millionhectares of agricultural land were converted from cropland to grazing areas between 1985 and 2001. This has given rise to a situation in which the land area devoted to livestock i s about double the area that i s appropriate for that purpose by agronomic criteria, while crops occupy about one-third of the land that i s appropriate for them. Within cropped areas, perennial crops have gained ground at the expense of most annual crops, and some export crops at the expenseof import substitutes. Although the sector was particularly hard hit by collapses in the markets for coffee and cotton, other crops have expanded in area and production. For example, in 1991 plantains and other fruits (except bananas) representedonly about 60 percent of the value of coffee production, but in 2001 their production value was greater than that of coffee. To put it another way, in 1991coffee exports, in value, were approximately equal to the value of all other agricultural and agroindustrial exports taken together. By 2001 these other exports had increased by about 50 percent and representedthree times the declining value of coffee export^.^ During 1991-2001, notable declines in production were experienced by cereal crops (barley, sorghum, wheat and, to a lesser degree, corn), short-cycle oilseeds (soybeans, peanuts), and cotton. The only major short-cycle crops to have escaped the negative trend of the 1990s were rice and potatoes. There was growth ina few lines of production, mainly some of the perennial crops such as fruit (a growth rate of 5.9 percent during 1991-2001), oil palm (8.0 percent), plantain for export (3.4 percent), yams (17.3 percent), sugarcane (2.6 percent), panelu (2.9 percent), flowers (5.0 percent), and cassava (1.9 percent). Some other perennial crops either did not experience growth or actually declined during the decade including, in addition to coffee, bananas (-1.0 percent), cacao (-2.8 percent), jute or 3que (-6.2 percent), coconuts (-2.7 percent), and dark tobacco (-5.5 percent). These trends continued in 2002, with production of annual crops declining by 1.5 percent and production of perennials increasing by 4.1 percent. Coffee experienced a substantial rebound. Including livestock, which expanded at 0.8 percent, sectoral production grew by 1.8 percent in 2002.' Perennial crops generatemore than double the employment and income per hectarethat annual crops do. Therefore this trend toward structural change has been positive in economic terms, although the overall sectoral growth rate remains unsatisfactory. The structural change also brings the production pattern more in line '*Preliminaryestimates It shouldbenotedthat coffee experienceda slight recovery in2002, expandingits productionfaster thanmost other crops. for 2002 from DANE. 6 with the sector's comparative advantage, as discussedbelow, which implies that future growth prospects have been strengthened. The substitutionof permanentcrops for annual crops, and the decline in total land area cultivated, has occurred in all regions of the country. In the Western Region, for example, the area under annual crops declined from 578,000 hectares in 1990 to 306,000 hectares in 1997, while the area planted in permanent crops rose from 507,000 to 569,000 hectares. Likewise, in the Caribbean Region in the same period, the area in annual crops fell from 716,000 hectares to 435,000 hectares, and perennials increased slightly from 198,000 hectaresto 202,000 hectares (BalcBzar, Vargas, and Orozco 1998:97). These changes in cropping patterns have increasedthe employment intensity of agriculture. For example, in 2001 annual crops accounted for 41.6 percent of acreage but only 20.9 percent of employment inthe crop sector. Conversely, inthe same year perennial crops other than coffee, which are expanding at the expense of annual crops, accounted for 38 percent of the planted area and 52.1 percent of employment incrops (Table A.1.1). Inpart, the trends incroppingmay representareaction to the opening of the economy that was put into effect in the early 1990s, since production has fallen most sharply in importable crops such as grains and annual oilseed crops. Simultaneously, the sector was buffeted by the collapse of the world markets for coffee and cotton. However, care must be taken in attributing cropping changes to the economic policy opening, because tariffs for short-cycle crops have remained highon average. The maineffect of the opening was to eliminate direct controls on prices and trade and other non- tariff barriers, and not so muchto bringtariffs down to low levels, as i s documented later inthis report. A central feature of the policy regime of the 1990s that was clearly unfavorable for agriculture was the increasing appreciation of the real exchange rate, a trend not reversed untilafter the crisis of 1999. Also, international prices for all grains and oilseeds and sugar declined significantly between 1991and 1997. For any crop, it i s difficult for average unit yields to rise while the area planted in the crop is expanding, because the newly cultivated areas often do not possess the same soil qualities or microclimates as the areas that were first sown with the crop. Nevertheless, inthe case of oil palm, yields increased by 5 percent per year between 1990/91 and 2000/01,9 in spite of a large increase in its area planted. Other crops that have stood out with respect to yield increases in the past decade are dark tobacco for export and corn grown with the most capital-intensive technology. Cotton recorded an annual average increase in yields of 4.5 percent, but this was undoubtedly due in large part to the drastic reduction in the area cultivated in that crop, which left cotton growing only in the soils most appropriate for it. Inthe majority of the other crops, yields have not increasedat a satisfactory pace since 1990. Thus deep structural changes are underway in Colombian agriculture. A central question i s what kinds of policy reforms can help acceleratethe process of positive changes and at the same time promote a sizeable contribution to reducing poverty. A major concern i s that the expansion of livestock at the expense of crops signifies a substantial underutilization of the country's agricultural potential. It i s widely recognized that this trend i s due in large measure to the lack of personal security in the countryside and to the influence of funds generatedby illicit activities. Encouraging more intensive and profitable use of these lands remains an important challenge for agricultural policy. There remains an acute dualism between a smallholder sector with insufficient land, often- backward production technology, and little or no access to ancillary services, and a modern entrepreneurial sector of medium-size and large farms with modern technology, which benefits from most Since yields often experiencesharp fluctuations from year to year, here the growth rates for the periodwere calculated from two-year averages for the beginningand end of the period. 7 of the production and marketing services and policy incentives. More than 60 percent of the rural population still live below the poverty line (Ocampo and others 1998). It has been estimated that 74 percent of the extreme poverty, and 57 percent of all poverty in the country, i s found in rural areas (Misi6n Paz 200153). Poverty in rural areas has been exacerbated in recent years by the collapse of coffee prices, which eliminated more than 100,000 jobs. Within the campesino subsector, while poverty continues to be prevalent, a positive sign is the increasing emergence of small-scale entrepreneurial producers, particularly in products such as potatoes, milk, panela, and some fruits and vegetables. This phenomenon has been accompanied by increasing linkages between smallholders and agroindustries, plus strengthening of small-scale rural processing activities, both on-farm and in villages. The links between smallholders and agroindustry are most notable in the cases of coffee, milk, and oil palm, but they have been intensifying also for berries, passion fruit, other fruit, panela, and potatoes. Direct links between small-scale producers and exporters are especially strong for plantain, granadilla, and uchuva. Table A.1.2 presents some of the principal local processing activities and the types of marketing channels that are emerging for the more entrepreneurial smallholders." They include those of illegal crops, which constitute a perverse demonstration of the adaptability of smallholders. In spite of these developments, rural poverty remains serious and widespread, and it is increasingly becoming urbanized because of the displacement of rural populations by the violence. Therefore the problem confronted by poverty reductionpolicy has two facets: (a) to respond to the short- term basic needs of displaced people while fostering an economic environment in which they can find urbanemployment, and (b)to put into place policies for assisting the rural poor to develop self-sustaining paths to economic betterment inthe countryside. 3.2. An OutlineofAgriculturalPolicyReformsinthePastDecade Like many countries in Latin America, Colombia historically pursued economic policies that favored import substitution and regulated the domestic market." A significant part of agricultural trade was controlled by the State. Over the years, there were variations in the net incentive effects of these policies for producers, as a result of changes in internationalprices, swings inthe real exchange rate, and cyclical changes infiscal expenditures, but the policy framework remained the same. In 1990Colombian policymakers took the decision to break with that tradition and deregulate domestic markets and liberalize internationaltrade for all sectors of the economy. For agriculture, this meant reducing crop support prices by replacing them with lower "floor" prices based on international quotations, eliminating the State monopoly in agricultural marketing, and changing financial sector policies to encourage private banks to lend to farmers and for agricultural exports. The amounts of government purchases of crops were reduced substantially, andfinally eliminated in 1996. In the financial sector, a new approach was developed to oblige private banks to lend rediscounted funds to agriculture, under which banks are free to offset the higher intermediation costs through a combination of slightly lower interest rates to savers and slightly higher interest rates to borrowers. The rediscount programhas been managedby the second-story financial institution Fondo de Financiamiento para el Sector Agropecuario (FINAGRO). However, the overall level of interest rates increased sharply between 1993 and 1999, and this trend discouraged farmers from borrowing for on- farm improvements. Real interest rates on the rediscounted funds reachedthe extremely highlevel of 23 percent in 1998 and 1999. ''This lo table and most of this discussion are taken from Forero Alvarez (2003). Basic references for this section are Jaramillo (1998, chapter 3); Balcdzar (2001); and Balcizar, Vargas, and Orozco (1998); and work done under the aegis of the present study. 8 Since then, real interest rates have come down substantially, and the government has emphasized lending to agriculture on the basis of contracts for sale of the harvest, often for groups of farmers ("associative credits"). This mechanism not only provides a form of collateral, but also stimulates stronger linkages between producers andmarketingagents and processing industries. Import quotas and licenses were practically eliminated in the early 1990s, and all tariffs were to be reduced gradually, although at a slower pace in agriculture. Price bands were instituted for a number of crops declared to be sensitive, in order to smooth international price fluctuations before they were transmitted to the domestic market, and then the system was extended to additional agricultural commodities. It i s a regional system shared with Ecuador and Venezuela and now covers 154 customs classifications of traded commodities. Inaddition to the price bands, inNovember 1994 (through Decree 2439) the government reintroduced a requirement that the Ministry of Agriculture approve imports, inthis case for 68 agricultural products, and for each product approval was made conditional on the domestic agroindustry absorbing the country's own production. In2003 (Decree 047), the number of commodities covered by this mechanism was reduced to 35, to comply with World Trade Organization (WTO) dispositions. Insum,the originalintentionstoreducetariffprotectionfor agriculture were notrealized, inlarge part becauseof the functioning of the price bands to offset movements ininternational prices and because of the "absorption" agreements requiring purchase of domestic production as a condition for permission to import. For 11major crops, the average level of border protection increased very significantly between 1991 and 2001, as shown in Table A.1.3. Ininterpreting this table, it should be borne in mind that the system of price bands makes automatic increases in tariffs when international prices decline below their long-run trends. As part of the wave of reforms of a decade ago, regional trade agreements were negotiated between 1992 and 1994 and export subsidies via the traditional tax-rebate mechanism (CERT) were modified, with the net effect of a reduction inits benefits for agriculture. In 1992 the CCIwas established to promote exports. It has concentrated its functions mainly in the area of providing information on prices for nontraditionalproducts in internationalmarkets and advice on farm management and marketing to producers who wish to participate in those markets. Since the close of the decade of the 1990s, a major new source of financing for the sector has emergedinthe form of the NationalAgricultural Commodities Exchange (BolsaNucional Agropecuaria), which has developed a variety of instruments for financing production, domestic marketing, and export.12 Inaddition, forward sales contracts through the Exchange have become an important incentive for the production of rice and corn. However, access to medium- or long-term investment finance has remained very scarce for the vast majority of producers, in spite of the priority accorded to such finance in the National Development Plan (Secretaria de laPresidencia ...200255). Other fundamental changes in sectoral policies made in the early 1990s included a reorganization of the institutions devoted to agricultural research, a widening of the scope for participation in research to include semipublic and private entities, a decentralization of agricultural extension to the county (municipio) level, and more investment in irrigation and drainage with subsidies on the construction costs of the systems for low-income producers. Inthe field of agricultural research, the process of generating researchproposals has been significantly decentralized at the same time that a greater range of institutions has begun to participate in carrying out the research. Land reform policy was changed markedly, toward supporting negotiations between landless families and landowners over the plots of land to be The development of these instruments i s describedinArias Puerta (2001). 9 redistributed, and financing the land purchase and initial on-farm investments with a significant subsidy element. Recently, the policy emphasis has included, among other topics, developing forestry and agroforestry projects for campesino groups and rural communities (Presidencia de la Repdblica ... 2002:55); special assistance for the coffee sector; credit and incentives like the ICR; connectivity through communications and community Internet access (ibid. 102-103); and agricultural technology development including biotechnology. However, as noted, budgetary restrictions have limited the effort inthe area of technology-and also inimgation development. Decentralization and development of local institutions has been another central element of the policy thrust. At present the government is negotiating increased foreign access for Colombian agricultural exports, and its overall agricultural trade policy i s consistent with the positions of the Cairns Group, of which Colombia i s an active member, in favor of the elimination of international agricultural subsidies. At the macroeconomic level, the real effective exchange rate appreciated by about 32 percent between 1991 and 1997, andthen it offset most of that change with a real depreciation between 1997 and 2002.13 However, for most major traded products (grains, soybean oil, crude palm oil, sugar, chicken parts, pork, milk), international prices showed an upward trend from 1991 until 1996 or 1997, and then declined until 2001 or 2002. Therefore the price movements approximately nullified the exchange rate movements from a viewpoint of incentives to Colombian producers, but since the price bands are geared to international prices, the net effect of the three factors-international prices, price bands, exchange rate-was to increase producer incentives between 1996 and 2001. In spite of this positive trend, production growth remained very slow. Figures A.1.1 and A.1.2 show these three factors at work for 14 major agricultural andlivestock products. Perhaps the most enduring and successful of the reforms initiated in the early 1990s were the reorganization of the agricultural research system, with introduction of competitive bidding for research projects (through the Programa Nacional de Transferenciade TecnologiaAgropecuaria [PRONATTA]), and the removal of government intervention in agricultural marketing. The price bands functioned well, as intended, but it will be shown later in this report that they have had the effect in recent years of protectingthe crops with the weakest long-termgrowth perspectives. 3.2.1. Product Chains: A CompetitivenessPolicyInstrument Agricultural policy has centered on the signing of competitiveness agreementsbetween members of product chains (agribusiness value chains) to address the increased competition brought about by the liberalization of the economy in the early 1990s. The goal of this policy i s to establish commitments and an action plan that improve the overall competitive position of agribusiness while increasing exports, adding value to goods, and creating jobs. Technological innovation and an improved organizational environment are the key drivers of this policy. The role of the government inthis initiative i s to facilitate interaction between the different members of each industry and coordinate policy so that government incentives and infrastructure investment support private initiatives that generate competitive advantage (see Box A.l.l). Viewing agricultural policy through the demands of the entire product chain has altered negotiations between the public and private sectors, as policy i s no longer reduced to the lobbying power of only one industry or producer organization. This approach has also expanded the scope of policy instruments in research, environment, and commerce while increasing the planning horizon of l3Banco Central de la Repdblicu, under the definition "ITCRIPC(T)" of the alternative calculations of the real exchange rate used by that institution. This variant of the calculations covers all external trade, deflates exchange rates by consumer price indexes in all trading partners, and weights trade flows ineach year by Colombia's total exports and imports inthe base year. 10 agribusiness. As this policy evolves, the government i s promoting regional competitiveness agreements to strengthen the competitive position of clusters. Since a key objective of this policy i s to increase exports, the Ministry of Commerce, Industry and Tourism has encouraged strategic export plans as a mechanism to facilitate dialogue between government and the private sector. These plans, which seek to improve the competitiveness of Colombian goods and services internationally while adding value to exports, are detailed in Exporting Competitiveness Agreements that are coordinated by the Ministry of Commerce, Industry and Tourism. While there have been no in-depth impact studies of the chain policy, there i s general consensus among participants that it has improved coordination among different government agencies and between government and the private sector, improved monitoring of public policy, and reduced distortions created by individual negotiation by isolated sectors with the government. Businesses have benefited from the sharing of information, which has allowed them to understand the competitiveness of various industries and regions, encouraging productivity improvements. Further, the policy has improved access to financing because members of the chain have entered into purchase contracts (that serve as collateral) or entered into forward contracts based on the commitments established in the competitiveness arrangements. However, the planninghorizon of business i s limitedor lacks sufficient detail to be translated into projects or activities, and most of the requests for government intervention remain oriented to generating short-term rents. Some chains leave out specific members, especially those downstream in the production process, and simply establish prices and standards for raw materials. The level of distrust between members of the chain remains high, impeding the creation of social capital that would further propel competitiveness. In addition, the policy has focused on crops with lesser advantages seeking to compensatetheir loss of competitiveness rather than promoting promissory sectors. Table A.1.4 provides a list of the competitiveness agreementssigned to date. Another challenge to the chain policy relates to the enforceability of the competitiveness agreements. Attempts have been made to regulate the process, although these have been met with opposition by producer organizations. While some suggest that those organizations that can successfully lobby government to extract rents fear that these agreements will diminish their power, elevating the agreements to the contractual level may not encouragethe development of the trust between parties that i s required for competition to flourish, nor allow them sufficient flexibility to adapt to changes in business conditions. The unenforceability of the agreements thus creates an even greater challenge for government, whose role as facilitator requires that it not only coordinate the meetings of members of the chain and provide financing for incentives, but that it mediate among parties while ensuring an adequate legal framework to guaranteelegal commitments (for example, contracts and forwards). Improving the competitiveness of agribusiness through the chain approach adopted by the government i s a learning process that requires a cooperative environment that encourages innovation and competition. This policy encouragesthe establishment of rules by participants inthe chain that are meant to guide transactions among them; however, the degree to which this cooperation limits rivalry among competitors, thus stifling innovation, i s not understood. Economies of scales may be obtained in the fields of researchand development and in the provision of infrastructure and marketing services, but care must be taken to avoid complacency by agribusiness as each should be responsible for creating its own competitive advantage. Also, the process should be open to all members of the chain including downstream industries, retailers, and cumpesinoorganizations. It is the participation of smallholders that poses one of the most formidable challenges to an agricultural policy based on the chain approach. Cumpesinos could benefit by being integrated into the 11 chain, because the main obstacle to their productive growth i s access to production factors, which competitive agreements can facilitate. Scattered private and public projects have provided access for campesinos to financing (commercial and bank), technology, and land. These partnerships have covered a wide variety of crops, although those where small farmers have a comparative advantage, that is, are labor intensive, and high value crops (for example, organics) may be more suited to smallholders. However, the chain approach to productive growth and competitive advantage cannot be the main policy instrument to combat rural poverty because low levels of education, weak or nonexistent producer organizations, and a lack of capital and business experience lessenthe bargainingposition of campesinos. To prevent the participation of campesinos from becoming an obstacle to the development of competitive advantage in the countryside, the government must develop the appropriate rules of the game for the inclusion and empowerment of campesinos, and actively mediate on their behalf while other actions in support of social and human capital formation are implemented. The government has a wide variety of incentives available to promote chains; however, funding of each itemi s limited and there i s little coordinationbetween programs and incentives. Many are tied to formal financing where it i s not always clear what benefit is perceived by smallholders or other members of the chain. Other areas of concern are research and development and technology transfer because the product chains are not yet fully integrated to the investment decisionmaking process. Additionally, for this policy to achieve its goals, agribusiness requires basic and commercial infrastructure to reach markets. The provision of these goods and the improvement of information systems by the government should be basedpartly on the potentialto strengthenclusters. While the provision of government incentives and investment are an important component of this policy, the meetings between it and members of the chain cannot be seen only as an opportunity to lobby government for preferences. Nor should industry limit its participation to the negotiation of inexpensive raw materials. Rather, this should be an opportunity to evaluate agribusiness and develop clear medium- and long-term objectives that create a win-win situation for all participants. These plans should then be translated into specific tasks, and conflict resolution mechanisms established agreeable to all parties to enforce the commitments. Finally, the parties should establish systems to monitor and evaluate the impact of incentives and agreements on productivity, competitiveness, and the construction of social capital. The government can support these efforts by providing a space for discussion, clarifying the incentive structure, including small farmers in the strategy, unifying export and competitiveness strategy, and developing a trade policy based on potential growth rather than compensation of competitive disadvantage. 4. The Competitiveness of ColombianAgriculture by Product 4.1. Methods of Measuring Competitiveness Relative prices and internationaltrade patterns tell a good deal about comparative ad~antage.'~If a product's domestic price i s significantly higher than the international price equivalent (after adjusting for transportation cost differentials and port handling costs), the suspicion arises that it may not have a comparative advantageand that only the price differential allows it to remain competitive. However, that i s not always the case. If the product enjoys a favorable price differential in this sense, i s not exported, and significant quantities of it are imported, then it almost certainly does not have comparative advantage in production, because its imported variant would be shown to be competitive in spite of a price disadvantage. l4The calculations reportedin this section are taken mostly from Balcizar, Orozco, and Samach (2003), and Espinal(2003), a report prepared for this study. 12 On the other side, if a product's domestic price is close to the internationalprice, or below it, and (for importables) there are no imports of it, then it is likely to have a comparative advantage. If it i s exported under these conditions, then it is almost certain to enjoy a comparative advantage. Fiscal subsidies to producers can change this picture. Inthe first case, of a price differential that i s favorable for a domestic product, if the product receives substantial fiscal support, then the doubts about its potential for comparative advantageincrease. Inthe second case, of unfavorable relative prices, a product's receipt of fiscal subsidies can weaken the presumption that it i s likely to have a comparative advantage. It i s known that in the Latin American region, generally, the support goes mainly to import- competingproducts, andinparticular to grains. The analysis of comparative advantage begins with a review of relative prices and with the net trade balance of each product. The quantitative measure for the latter i s called the relative trade balance (RTB) and is calculated only for product chains. In the next step, the analysis of trade patterns is deepened through use of the revealed comparative advantage (RCA) at the level of individual products. This concept is a measure of the degree of Colombia's specialization in a product on the international scene.15 RCA values are calculated for five successive years (1997-2001), so they also tend to show whether Colombia's degree of international specialization in a product i s increasing or decreasing, another indicator of the likelihood of comparative advantage. However, the RCAs often fluctuate considerably from year to year, so it is not always possible to identify a clear trend. The final step for each product is the calculation of the domestic resource costs coefficients (DRCs). To avoid the problem of not knowing fiscal support levels by product, the DRCs are calculated with long-term market prices (or shadow prices) of inputsthat may be currently subsidized, such as credit and agricultural machinery.16 The principal input for which this was not possible is irrigation, and therefore true irrigation costs in the public irrigation districts (the opportunity cost of water) are not used inthe analysis. Accordingly, to adegree, the DRCspresentedhere may overstate comparative advantage, although the bias is not likely to be large inquantitative terms.17 This study presents analysis by product chains and groups of the following products: grains and oilseeds; fruits and vegetables; other long-cycle crops; tubers, tobacco, and cotton; livestock products; and fisheries and forestry products. Observations are also made about organic products, and analysis i s presented of some areas in which campesino production displays comparative advantage or has the potential for it. Results will be summarized here; details can be found in Annex 2. Annex 3 presents the formulas usedinthe analysis. In overall terms, Colombia has comparative advantage, often strong, in export products as opposed to import substitutes, in perennial crops as opposed to annual crops, and in the more labor- intensive crops. All regions have a comparative advantage in some lines of production. In many cases realizing this comparative advantage will require concerted efforts in areas such as agricultural Is To illustrate it with completely fictitious numbers, if Colombia's total exports represented 10 percent of all world exports, but Colombia's coffee exports represented30 percent of world coffee exports, then the RCA for coffee would be 3.0 (30 percent40 percent), a relatively high value. Thehigher the RCA, the morelikely that the productpossessesa comparative advantage although, as noted, the RCA measurealoneis not sufficientto say definitely whethercomparativeadvantageis present. It does, however, indicatethe degree of competitivenessof a productunder the presentpolicy regime. l6 Following standard procedures, an effort has been made to price nontradableinputs at their long-run opportunity cost, in order to best reflect the underlyingstructure of theColombianeconomyinthe calculations. As a methodologicalnote, laborhasbeen pricedat the prevailing ruralwage rates. The reasons are: (a) for a given productiontechnology, the breakdownof labor inputs betweenhired labor and family labor is not known, becausethat will dependon farmsize amongother considerations; and (b) from a viewpoint of the supply function for labor, at a zero or very low wage rate, the supply,of labor will be zero. Sometimes it is fallaciously assumed that excess supply of labor means the opportunity cost of labor is zero, but people always value their time at a non-zero worth. if even for noneconomicuses of their time. 13 technology development, marketing programs, and food safety programs. It can also be promoted by modifications in priorities in incentives policies and by changes in the way in which some programs are carriedout, includingresearchandextension andirrigation management. 4.2. ComparativeAdvantageAnalysisfor GrainsandShort-CycleOilseeds Nominal protectionrates increasedbetween 1991-93 and 1999-2001 for five of the six principal grains including the main short-cycle oilseed. Protection levels are currently highand, with the exception of barley, they represent one-third to one-half of producer income in those crops (Table A.2.1). The economic protection afforded to grains i s stronger than indicatedby the nominal protection rates because of the implementation of the "absorption agreements" that oblige agroindustry to purchase the domestic crop inexchange for beingallowed to import grains. Barley is the one exception, but it i s by far the least important grain produced in Colombia. The trend in Colombia's international trade in grains has been toward an increase in imports as a share of total supply, in spite of the high levels of protection. This strongly suggests that domestic grain production does not possess a comparative advantage. The import share of total supply was significantly higher during 1996-2001 than during 1992-1995 (Table A.2.2). It might have increased further but for the effect of the absorption agreements. There always are exceptions to general statements, and Colombia has been able to export small quantities of a few high-valuedgrains (Table A.2.2). To place Colombian tariff protection levels in internationalperspective, of 13 world regions only four (South Asia, sub-Saharan Africa, Caribbean Islands, and non-EU Western Europe) have grain protection generally as highas Colombia's if barley i s excluded from the latter (Table A.2.3). The more important issue, explored later, i s whether the prevailingtariffs on grains inColombiarepresent the policy most likely to promote growth of income and employment in agriculture. Calculation of the RCA suggests a future potential in some processedforms of grains and short- cycle oilseeds, though not in all. By far, the largest share of processed grains and short-cycle oilseeds products show an increasing export relative importance in Colombia's basket of export goods (Table A.2.4). The trend in these exports represents another indication that Colombia's comparative advantage inthe areaof grains andshort-cycle edible oils lies inprocessedproducts rather than primary products. If that i s so, then an export-oriented strategy would lean toward lower tariffs on the imports of raw grains and oils, inorder to makethe processedproducts more competitive. Analysis reveals Colombia does not have a comparative advantage in the grains and soybeans subsector (Table A.2.5). There appears to be a possibility that sorghum could become competitive in Cundinamarca with reasonable increases in productivity and decreases in costs, but only with great difficulty. Zones that possess relatively high levels of productivity in yellow corn do not escape the general comparative disadvantage of Colombia in corn and other grains. White corn has an even more marked comparative disadvantage than yellow corn. Production of feedgrains in Colombia evidently i s sustained by the high tariff levels. The weakest of Colombian grains in international competitiveness i s rice (Tables A.2.5 and A.2.6). In general Colombia's strengths in agriculture do not extend to grains and short-cycle oilseeds. From the viewpoint of long-term national development policy in Colombia, the wisdom of continuing to provide the sector's strongest incentives for the cultivation of grains i s put into doubt not only by the results provided above, but also by the fact that those crops generate relatively little employment per hectare (Table A.2.7; grains rank among the lowest crops interms of capacity to create employment). 14 From a perspective of product chains, some of the most competitive components of the cereal chains appear to be breads and cookies. Their exports increased from US$6 million in 1991 to U S 3 5 million in 2001. Efficiency of irrigation, improving phytosanitary controls, reducing the use of agrochemicals, and increasing the use of certified seeds are issues that needto be addressedinthis sector. 4.3. ComparativeAdvantage Analysisfor Fruitsand Vegetables There i s increasing demand for fresh, healthful fruits and vegetables, and growing international markets for these products (Mission Paz 2001:160; Secretariado TCcnico.. . 2002:17). Fruits and vegetables are important in Colombia's external trade balance, and they also are among the crops that generate the most employment per hectare. Bananas, Colombia's third-largest agricultural export, after coffee and flowers [Statistical Annex tables], alone generate an estimated 34,543 direct jobs. Tonnage produced of other fruit such as plaintain, mandarin oranges, lemons, and strawberries has increased rapidly, and accounts for about US$30 million in exports. Beans and asparagus are other significant export crops. Uchuva i s now Colombia's third-most-important fruit export, after bananas and plantains, andgranadilla, lulo, passion fruit, berries, and papaya, are among other fruit crops that have potential in the internationalmarket. Industries usingfruit as an input account for about 17 percent of total industrial employment. Colombia has negotiated preferential access for some fruit exports in parts of Latin America, although fruit exports have received little incentives from the government. On the import side, these categories of fruit enjoy tariff protection levels (standard for Colombia) of 15 percent and 20 percent. Citrus fruit does not participate much in international trade, therefore the citrus expansion has been directed at the domestic market (andthe same has beentrue of mango). Hurdles that fruit and vegetables face for continuing expansion are inadequate production technology and a high risk of plant diseases owing in part to the use of traditional technologies. Other risks include volatile prices and a long-term downward trend of prices in world markets. These negative price trends can be expected to continue for most fruit and vegetablecrops. Factors favorable to Colombia are its variety of climates and growing seasons, and that, for the time being, Colombia can export many fruit and vegetable products to the U.S. without paying any tariffs (through an agreement known as the Andean Trade Promotion and DrugEradication Act [ATPDEA]).18 However, generally, fruits and vegetables are quite important for generating both income and employment. The principal reasons for Colombia's lag in promoting fruit and vegetable exports include (a) phytosanitary problems; (b) lack of a system to facilitate the establishment of links between producers, exporters, and buyers in other countries; (c) weak technological development, reflecting insufficient commitment to research in this area; (d) insufficient awareness among producers of the importance of product quality, including absence of chemical residues, adequate flavor, uniformity of appearance, and timeliness of delivery; (e) weakness of producer organizations for fruits and vegetables; (0 inadequate attention to the needto open up internationalmarkets, and (g) in some areas, violence. Trade data suggest a possible comparative advantage for most fruits and vegetables in Colombia, but the existence of a degree of tariff protection in Colombia for these products, although less than for grains, means the trade data alone are insufficient to determine whether they have a comparative advantage. Table A.2.9 shows the RCA coefficients for principal fruit and vegetable line items in the l8This paragraphand part of the following one are basedoninformation andcommentary inFonseca(2003). 15 customs classifications. Table A.2.10 shows that the DRC coefficients indicate that Colombia has a notable comparative advantage in fruit and vegetable crops. Inaddition, their high rate of employment creation per hectare means they have significant benefits in terms of poverty reduction and social equity inrural areas. 4.4. Comparative Advantage Analysisfor Other Long-Cycle Crops This category of crops includes two of the four largest agricultural exports, coffee and sugar, one of the fastest-growing exports, palm oil, andcacao. All four crops are linkedto important agroprocessing industries in Colombia. The sugar and palm oil subsectors also have been leaders in technological and institutional development in the last 10 years, and coffee has long had a strong institutional base. Palm oil production has beenincreasing while cottonseedoil and soybeanoil have beendeclining. Nominal tariff protection for palmoil, cacao, and sugar has fluctuated markedly over the past 15 years (Statistical Annex tables). In the case of palm oil, this i s due in part to the functioning of the Andean price bandsystem. The cacao product chain has tariff levels of 10, 15, and 20 percent, depending on the degree of processing of the product, but market forces often have depressed domestic price levels below what the tariffs would indicate. On average, it has had little or no nominal protection. The sugar chain has tariff rates between 5 percent and 20 percent, and its domestic price i s managedthrough a stabilization fund. The net result i s a highly protected subse~tor.'~ Imports of coffee are trivial, and imports of sugar are less than 10 percent of sugar exports. For vegetable oils, imports are about four times larger than exports (Table A.2.11). In the case of cacao, exports have not increased, but their composition has shifted toward processed products, and imports have increased rapidly, so the relative trade balance has deteriorated (Table A.2.12). The revealed comparative advantage coefficients for these products are presented in Table A.2.13. The picture i s mixed. However, the DRC calculations presented in Table A.2.14 show that all four categories of long-cycle crops do possess a considerable comparative advantage. All these crops are highly intensive in labor use and therefore have considerable capacity for creating employment. In addition, employment i s relatively stable and well remunerated, and therefore it i s employment of high quality. The same i s true of bananas andplantains. Inthe cases of sugarcane, oil palm, and coffee, producers also are well organized. The presence of smallholders, incombination with the economies of scale inprocessing and marketing, provides strong incentives for social cooperation and organization inthe rural areas where the crops are dominant. Coffee i s especially important in all these respects, and i s of overwhelming importance for rural society as well as for the nationaleconomy. Coffee's comparative advantage has slipped in recent years with the entry of producers such as Vietnam and the consequent decline in international coffee prices. This points to the need to improve productivity (control costs) and place greater emphasis on high-quality coffees, in order to differentiate Colombia's product in world markets. However, while some producers are going out of coffee, those that have been able to control costs (usually smaller producers) remain competitive internationally. In addition, the importance of quality cannot be sufficiently emphasized. ~~ l9However, it shouldbe recognized that hardly any countries inthe world are competitive insugar when the free market price i s used as the benchmark. 16 However, compared with some Central American countries, Colombia has been slow to improve its coffee grades and develop new, high-quality brands. The opportunities for Colombia in coffee are still there, but seizing them will require a concerted effort in research, extension, and marketing, all aimed at positioning Colombia at the highend of the evolving market. With respect to cacao, technical problems at the farm level needto be resolvedbefore cacao can realize its apparently high potential. The comparative advantage of cacao would appear to imply that a strengthenedtechnology development effort is warranted for that crop. A favorable circumstance for palm oil is that world demand is expected to grow at least in the range of 3.3 percent to 3.9 percent annually for the next 20 years. Beyond areas currently planted in palm, there are literally millions of hectaresof additional lands with appropriate soils for the crop. For palm oil, the DRC calculations indicate that the crop can be competitive on world markets without the support provided through tariffs and pricing arrangements, but it may be argued that these incentives are conferring important dynamic benefits to Colombia by spurring rapid expansion and technological development inthis subsector. The issue i s one of balancing those benefits with the interest of domestic consumer welfare. 4.5. ComparativeAdvantage Analysis for Tubers, Tobacco, and Cotton Potato originated in the Andes. Today it i s the 10" most important crop in Colombia in area planted, and the sixth most important in value of production. Close to 90 percent of potato producers have less than 5 hectares, and produce 35 percent of the potato harvest. The potato-processing industry i s extensive, and exports of potatoes and other tubers earned more than US$20 million in 2001. Potatoes and their products have import tariffs of 10and 20 percent, with preferences granted to a number of Latin American countries. Cassava i s both consumed fresh and used as an animal feed, in milled form, and work i s underway to exploit its potentialfor makingstarch and industrial glues. Colombia grows both light and dark tobacco. The tobacco product chain includes extensive processing facilities and a substantial export industry. However, in recent years tobacco imports have increased more rapidly than exports. Tobacco i s one of the most labor-intensive crops to grow, and it i s often grown on very small holdings. The production of cotton fell in the 1990s, but has recovered slightly. The nominal rate of protection was negative during 1989-91, then rose to 21 percent and 41 percent in 1992 and 1993, respectively, then fell to 5 percent in 1994. It remained at slightly positive or negative levels until 2001, when it rose abruptly to 22 percent, only to drop again to 3 percent in 2002. Its profitability has been undermined by falling world prices, the rising real value of the Colombia peso in the 1990s, and rising costs of plantprotection at the farmlevel. The cotton product chain includes important subsectors producing cotton thread and yam, fabric, and clothing. Colombia i s a net importer of thread and yam and fabric but also exports them and i s a significant exporter of clothing. The RCA coefficients in Table A.2.18 confirm Colombia's growing potential in fresh and processed potatoes and in processed tobacco. Leaf tobacco appears to be losing ground in relative importance. The export values reported in Table A.2.16 confirm the robustness of the textile sector in 17 spite of the decline of cotton cultivation. (It should be noted that cotton i s not an especially employment- intensive crop; see Table A.2.7). The comparative advantagecalculations for these crops are shown inTable A.2.19, inthe form of DRC coefficients. These results confirm the comparative advantage, at the farm level, of potatoes and tobaccos in Colombia. Dark tobacco i s the most competitive of these crops. The comparative advantage i s almost as strong as that of fruit and vegetables, and representsa solid basis for long-run development of these crops. Potatoes also have a comparative advantage when produced by smallholders, in spite of their generally lower yields. A major hurdle for potato growers i s to reduce their dependency on agrochemicals. It i s widely known that potato growers regularly use excessive quantities of fungicides, herbicides, insecticides, and antibacterial compounds. In the future, increasing numbers of export markets will be closed to products that do not meet food safety standards. Potato exports also face phytosanitary barriers, andovercomingthe phytosanitary barriers to fresh potato exports will require greater producer organization, and close coordination with research and extension agencies. Interms ofcomparative advantage,cottonpresentsmixedresults. Itdemonstratesacomparative advantage in four departments, Cbrdoba, Valle del Cauca, Huila, and Meta, although only marginally so in Cbrdoba. It has a comparative disadvantage in Cesar, Sucre, Magdalena, and Tolima, although it is close to being competitive in the first two of these departments. These regional differences spring from differences incost structures, inparticularthe costs of pest control, and from degradation of soils. Cotton i s another crop that requires a change in techniques of cultivation to be competitive and sustainable over the long run. Given its susceptibility to pests, it i s also a crop that clearly can benefit from biotechnology, provided the appropriate institutional structures for biotechnology research and controls can be put into place. Cotton's profitability undoubtedly has been affected by subsidies in industrialized nations. To test this hypothesis, the DRCs for cotton were recalculated with a 15 percent increase in international prices. The results (from BalcBzar, Orozco, and SamacB 2003) are shown inTable A.2.20. It can be seen that if it were possible to consistently offset the effects of international subsidies on prices (say through tariff policy), cotton would become viable in more regions of the country. Cotton textiles create much more employment and export earnings than raw cotton does. Cassava has a comparative disadvantage everywhere except in Meta. The most productive use for cassava appears to be in animal feeds and in diverse industrialapplications, especially as starch and glue. Despite its comparative disadvantage, Colombia i s already the world's third-largest producer of it, mainly for supplying the domestic market. The country's favorable conditions for cassavain the long run include: (a) CIAT has the greatest knowledge of cassava and the largest gene bank for the crop inthe world; (b) cassava i s readily adaptable to a variety of soils, including poor and acid soils; and (c) Colombia already has advanced relatively far down the road of developing and testing commercial varieties, and mechanization of some cultivation tasks (MisibnPaz 2001:144). It i s also adaptableto acid soils. Nevertheless, cassava i s the one crop with a comparative disadvantage that may have potential over the long run, and warrants closer attention for that reason. The basis for its potential is that its productivity ceiling i s very highinrelationto actual performance-higher than for any other crop. 18 Cassava's main hurdles include susceptibility to pests; the high water content of the root, which makes it expensive to transport; and the lack of processing facilities that would purchase the crop on a significant scale. From a viewpoint of public policy, more rapid exploration and development of cassava's potential has been held back by the lack of commitment of sufficient research resources to the crop. While the outcome of such a commitment cannot be predicted with certainty, all indications are that this is apromising area. 4.6. Comparative Advantage Analysis for Livestock Products Livestock products reviewed in this study include poultry and eggs, pork, and beef and milk. Over the past 20 years or so poultry production increased by almost two and one half times, and milk production almost tripled. Over the last 10 years, pork production increasedby about 10percent (DANE, MADR). Beef production remained around the same level from 1982 to 1997, then dropped abruptly during 1998-2001, possibly due to the rural violence. In addition, beef exports are prohibited from entering the U.S. market because of hoof-and-mouth disease. However, Colombia i s still the ninth-largest producer of beef inthe world, and the largest inthe Andean region. Exports of dairy products have grown very rapidly over the last 10 years (Tables A.2.21 and A.2.22). Dairy exports are almost entirely of milk, and their principal markets are Ecuador and Venezuela. Because of the presence of international subsidies on milk and the efficiency of dairy producers such as New Zealand and Denmark, it would not be realistic for Colombia to aspire to dairy exports to the wider international market. However, continuing to export to the Andean region, in the framework of the Andean Community, appears to be a viable route. In the poultry subsector, eggs are the principal export, although trade i s hampered in part by the lack of sanitary clearancefor poultry meat to enter the United States. The domestic resource cost coefficients (Table A.2.23) show a comparative advantage for all livestock products in Colombia. It i s strongest for poultry and eggs, followed by milk. However, when the calculations are made with prevailingrates of protection on imports of corn, sorghum, and soya, pork loses its competitiveness and poultry's competitiveness becomes only marginal. Milk's competitiveness also diminishes somewhat. Thus the protection policy for feedgrains turns out to be an important factor indeterminingthe international competitiveness of livestock products, especially pork andpoultry. Concerns have been raised about the competitiveness of Colombia's poultry industry in the face of competition from Brazil. However, two factors attenuate this competition. One i s the high cost of shipping poultry from Brazilian ports to Colombia. The other has been the recent Brazilian currency devaluation. The DRCs reported for poultry in Table A.2.23 refer to comparative advantage vis-&vis the United States market. To address the question about Brazil, the DRCs were recalculated with Brazilian prices as of June 2003. The results of those calculations are shown in Table A.2.24, which shows that poultry has a marginally lower comparative advantage when confronted with competition from Brazil, but it still retains a comparative advantage. In the case of cattle, analysis of the costs of production of 17 representative livestock operations in various regions of the country revealed that the comparative advantage is strongest for the most intensive operations. Infact, for extensive cattle raising, the DRC indicator i s approximately unity, which indicates that its comparative advantage i s nil. Because many of these ranchers do not take into account the opportunity cost of their land, nor of the financial capital they put into the operation, they perceive a profit in cash-flow terms (financial terms) when in fact they may be losing in economic terms, that is, in 19 comparison with alternative forms of investing the resources. Nevertheless, the comparative advantage for intensive operations, both for beef andfor beef plus milk, i s real and substantial. The analysis for all livestock products taken tQgetherconfirm Colombia's comparative advantage in subsectors that embody more product processing. Because of certain structural barriers in world markets-sanitary certification for beef and poultry; the heavy international subsidies on milk- Colombia's comparative advantage in these products may not have an opportunity to be expressed in the form of more exports inthe near future. However, should the sanitary hurdles be overcome, the potential for export growth in these products i s considerable. In the meantime, until better international arrangements can be negotiated, the distorted nature of international trade in poultry products appears to warrant continuationof a significant import tariff on chicken parts. 4.7. Comparative Advantage Analysis for Forestry The forestry sector comprises products including logs, boards, plywood, and wood manufactures. It is estimated that in Colombia about 25 million hectares of land are appropriated for forestry (Misi6n Paz 2001:120). Currently the country has about 8 million hectares of natural forests and forest plantations, of which only a little over 100,000 hectares are plantations. The timber industry i s poorly regulated, with a significant component of illegal, unregistered activity. Frequently the wood felled in Colombia i s characterized by low yields and poor quality, and often it i s not delivered in a timely fashion to sawmills. In spite of these deficiencies, exports of wood products have begun to increase at an accelerated pace (Table A.2.25). However, in species like pine and eucalyptus the natural growth of wood in cubic meters per year i s higher than in exporting countries such as the United States, Argentina, and Chile (ibid.), so Colombia possessesa physicalbasis for competitiveness inthe sector. Ininternationalmarketswoodproducts havebeenincreasingat asubstantialpacefor the past20 years, and it is expected that growth will continue to be strong. Thus the market opportunities exist for an expansion of the forest industry. Among its benefits are considerably greater employment per hectare than cattle raising, and environmental benefits in controlling soil erosion and protecting watersheds. Watershed benefits, normally considered to be economic externalities, can be internalized in financial terms. Forestscan also provide the setting for the provision of environmental services that have a market value, including sequestrationof greenhouse gases and bioprospecting. As of mid-1999,46 countries had projects for the capture of greenhouse gases, of which 14 are inLatin America, including Colombia. The potential market for the findings of bioprospecting i s also very large; international sales of medicinal plants amount to about US$14 billion. Therefore, there are a wide variety of potential economic benefits from adequate forest management; however, Colombia has been slow to exploit them. Increasingly, success in exporting forest products will require certification of sustainable forest management, and that inturnrequires appropriate institutional development to support the sector. For the present study, comparative advantage calculations were made for 12 forest species, with the results shown in Table A.2.26. Most of the species analyzed have a clear comparative advantage, more so inbroad-leafed species than inconifers or eucalyptus. Economic analysis thus points to potential strengths in the forestry area. Thus, it would be worth rethinking the incentives for reforestation, and for proper management of natural forests. Currently, the fiscal incentive for reforestation i s tied to the availability of funding to back it, and there have been reports of viable reforestationprojects being put on hold in the expectation that eventually sufficient funding will become available to allow the project to access the incentive provision. 20 4.8. CommentsonOther Products The seafood sector i s an important export earner in Colombia, as shown in Table A.2.27. The most dynamic product inthe sector has beencultivatedshrimp, mainly produced along the Atlantic Coast. World demand for fish continues to expand rapidly, and ocean and lake fish catches will fall short of demand by an increasing margin, owing to depletion of natural fish stocks. Consequently, there will be no limitation on the demand side to expansion of output. On the supply side, Colombia has natural advantages in shrimp cultivation on the Atlantic Coast because the area i s less susceptible to diseases of the species than i s the Pacific Coast, and Colombia's yields, at about 4 tons per hectare of pond, are far greater than those of Ecuador. Colombia's highyields result from the fact that it i s a world leader indeveloping shrimp technology. Its use of its own laboratory larvae is the reason why it has been able to avoid the viruses that have severely affected the cultivated shrimp industry in many countries (Contraloria de la Rep6blica 2002:55). Inshort, the opportunities for Colombia in this field appear to be great, and studies on the cost-competitiveness of Colombia's production and the relevant policy options should be carried out.20 Organic agriculture i s another area in which world demand i s increasing rapidly. Colombia's exports of organic products were valued at US$4 million in 1998, US$9 million in 2000, and US$11 million in 2001. The most important organic products to date are coffee, palmoil, bananas, mango pulp, and sugar, although several others also are certified for sales as organic products.21 The area under certified organic cultivation is still small, about 9,000 hectares of crops and 6,500 hectares of cattle and buffalo in 2001, but it i s expanding rapidly. In general, organic farming is attractive to many small farmers because they do not apply extensive amounts of agrochemicals in the first place, so the shift to chemical-free cultivation requires little or no change in production techniques, and to the extent change i s needed, it implies a savings in monetary costs (Damiani 2002). Insome areas inColombia, smallholders who already apply low dosages of agrochemicals are beginningto make the switch to organic production. The financial potential for farming with organic methods can be seen in Table A.2.28, which presents observed price differentials in the Bogota market for 28 products cultivated under the two different methods. The price differentials are very substantial, and this i s undoubtedly an area with considerable growth potential, in both domestic and international markets. Forero (2003) analyzed the profitability of organic production compared to conventional production in specific cases. He calculated that the costs per kilo for organic mango were lower, the yields higher, and the profit higher than for conventional production. A similar outcome was found for organic punelu. The future prospects for organic farming are undoubtedly strong, especially for smallholders. However, continuing development in this area will require considerable effort in the areas of information and training, producer organization, and market organization. For very optimistic views of Colombia's potentials inthis area, see Misi6n Paz (2001); and Acero (2000). The information reportedhere on organic productionis taken from Ministerio de Agricultura y Desarrollo Rural y Instifuto Alexander von Humboldt, with the support of World Bank (2002). 21 5. The Role of Exports inandthe Effect of Policies on Competitiveness 5.1. The Role of Exports in Competitiveness In all countries and all sectors exporting represents an important avenue for increasing production and income. This i s especially true for agriculture. By the operation of Engel's Law, aggregate domestic demand for food products will grow by only about two-thirds of the growth of incomes. Therefore accelerated growth of production can occur only through exports or import substitution. In the case of Colombia, the principal crops that compete with imports do not have a comparative advantage, so expansion of their output would have to come at the expense of higher domestic prices to consumers or greater fiscal outlays, or both. Most export crops do have a comparative advantage, and therefore increasesintheir output and sales can occur without large incentives for the most part. Exports also play a role in increasing the quality of domestic production, since markets in industrialized countries tend to be more demanding of product quality than the domestic market in developing countries. Export markets transmit information on standardsof product quality infour senses: plant and animal sanitary norms, food safety requirements from the viewpoint of human consumption, and consumer tastes, and delivery requirements interms of volume and timing. The need to keep abreast of these requirements in diverse markets i s one reason, along with the need to improve technologies of production, that export agriculture is increasingly a knowledge-intensive industry. Largely because of the role of exports, duringthe decade of the 1990sagriculture was the fastest-growing sector in Chile and the sector that created the most new jobs in scientific, technical, managerial, and administrative areas. In order for Colombia to reap full benefits from its comparative advantage in export crops, its agriculture will have to undergo a similar transformation inacquisition of skills and knowledge. Table A.1.5. shows that Colombia has a solid agricultural export sector that has been able to increase the volume and value of its exports in spite of the collapse of world coffee prices and declines in other agricultural prices. Exporting, of course, brings with itjust this risk, of fluctuating or falling prices on world markets. This risk can be attenuated in part through use of futures contracts, but such arrangements are not available for very many agricultural products. Another responseover the longer run i s to continually increase productivity, so that profit margins are not affected so much by reductions in prices. This i s already beingdone in some exporting sectors such as oil palm. A third response to price risk on international markets is diversification of the basket of export goods. This i s accomplished mainly at the national level, but to a degree it can be done by some enterprises as well. FEDEPALMA,for example, i s seeking to market palmkernel oil as well as palm oil itself, and to develop new products basedon processing of African palm. At the nationallevel, the degreeof diversificationof agricultural exports over the past 10years is impressive. The number of products with an export value of more than US$5 million was 32 in 1991, but this number hadgrownto 64 in2001 (excluding flowers andcounting bananas and plantains as one item). Table A.1.6 provides the detailed accounting of this diversification of exports. As can be seen, virtually all the increased diversification has arisen in the agroindustrial product lines, with very few exceptions such as eggs and root crops. On one hand, this i s another indication of Colombia's relative strength in agricultural processing, but on the other hand it i s another confirmation that the country i s not exploiting its comparative advantage in fruits and vegetables-and its probable comparative advantage in some nuts and spices. Promoting continuing diversification should be an integral part of a national agricultural export strategy. It implies providing support and/or incentives for both small and large products, encouraging export of a wider variety of fruits, vegetables, and nuts, both fresh and processed, organic products, and 22 other specialized products. Since these products tend to be highly labor intensive, a national diversification strategy goes hand inhand with employment creation and poverty reduction. The other side of diversification is the markets for destination of export products. Here also the trend has been toward less concentration. Traditionally, Colombia's principal markets have been the United States, the European Union, Venezuela, Central America and the Caribbean, and Peru. Inthe last 10 years, however, although the U.S. share of Colombia's exports has grown slightly, the exports have expanded rapidly to countries such as Chile, Mexico, Ecuador, and Canada. As Table A.1.7 shows, in 1991, only five destinations each accountedfor more than 1percent of Colombia's total agricultural and agroindustrialexports. In2002, nine countries each received more than 1percent of Colombia's exports. Between 1991and 2002, exports to seven countries of destination increased by more than 10percent per year. This i s a healthy trendfor the long run. 5.2. The Effects of Policies on Competitiveness This chapter examines government policy options for promotingthe products inwhich Colombia enjoys a comparative advantage: perennial (as opposed to annual) crops, products for export (as opposed to import substitution), livestock, and certain agroindustrial products. Recent experience in Chile, Brazil, and elsewhere indicates that agriculture can still be a source of substantial economic growth for developing countries, and Colombia's comparative advantage in these particular agricultural subsectors has the potential to reduce rural poverty and benefit small and large producers. For this potential to be realized in Colombia, a favorable policy environment i s needed, particularly including adequate (a) production incentives; (b) resource base assurances; and (c) access to markets, input supplies, and production technologies. 5.2.1. Incentives for Agricultural Growth Sound policies on international trade, fiscal measures, and exchange rates are the major instruments for an effective incentives policy. Fiscal measures should be targeted as much as possible at low-income producers-particularly for Colombia in light of the urgency of reducing the rate of poverty and violent conflict in rural areas. With that exception, incentives policies should be applied relatively uniformly across products, sectors, and production factors, recognizing that markets are more effective than governments at "picking winners." Achieving suchuniformity i s a major challenge for Colombia, as itis for other developing countries. Agricultural Tradeand Pricing Policy Colombian government economic reforms instituted in the 1990s to enhance agricultural competitiveness have seriously eroded under the pressures of international trade negotiations and domestic funding shortages. This study and others support the conclusion that trade protections and incentives such as price bands, import restrictions, and high tariffs have not supported the products with greatest comparative advantages.22 Net effective protection has been highly unequal among products. ** Sugar and palm oil are the principal products whose prices are still artificially manipulated. Sugar is in the paradoxicalsituationof being a highly competitiveproduct that nonetheless receivessubstantialdomestic price support. Consequently, it is overproduced(althoughthe level of productionmight be appropriate if intemational price distortions were eliminated). Government plans to requireethanol use as an automotive fuel andto devote significant sugarcane acreageto producingethanol shouldpermit a reductionindomestic sugar prices and exports into volatile world markets. Shouldthat plan fail, considerationmight be given to a gradualreductioninthe sugar prices chargedto domestic consumers. In the case of palm oil, intemational price distortion is less than for sugar, leaving little justification for maintaining high domestic prices, particularly since edible oils account for a significant portion of the expenditures of low-income households. Gradual price reductions to internationallevels would have the added advantagesof putting downwardpressure on vegetable oil import tariffs and improvingthe livestock andother oil-consumingindustries. 23 Few approved subsidies have actually been disbursed to producers. Such a situation does not represent the best strategy for promoting agricultural sector growth inColombia. One remedial strategy worth considering i s to modify trade (and fiscal) policies inthe direction of increasing the price of less competitive crops relative to the prices of crops with a comparative advantage (other than coffee). Using information on supply response elasticities for those crops, along with information on costs and employment, this study projects a modestly favorable shift in prices could be expected from a 10percent modification-about a 5 percent decrease in prices for highly protected crops and a 5 percent increase in those with greater competitive advantages. Such a shift could result in a 3.6 percent increase intotal crop income and create an additional40,000 jobs (see Table A.1A). The largest income increases would occur for perennial crops and others with comparative advantages, such as potatoes and cotton, while corn, rice, sorghum, wheat and barley would decrease in production and, consequently, income generated. These benefits could be approximately doubled if accompanied by just a 20 percent change in relative tariff incentives between the two classes of crops, even after takinginto account likely market limitations inthe form of downward-sloping demand curves. Tariff changes by Colombia designed to move away from unproductive protections would constitute "reconversion tariffs" acceptable to the WTO. To be effective, tariffs would need to be reduced gradually over a rigidly fixed time period (perhaps as long as 10 to 15 years). Most important, tariff revenues generated would need to be allocated to domestic producers based upon field-tested productivity improvements, or to help inefficient producers convert to producing other commodities. Any short-term economic disruptions such changes might cause could be temporarily offset with appropriate fiscal measures (discussedinthe next section). Table A.1.9 projects the approximate effects of a hypothetical 10 percent reduction in tariffs on corn and soybeans on the livestock sector. It suggests that both consumers and producers would benefit from lower prices leadingto increasedconsumption. An effective internationaltrade negotiating strategy will be necessary to accomplish Colombia's need for more aggressive export promotion programs, direct agricultural subsidies, structural economic changes, and reduction of foreign barriers to Colombian agricultural exports. Despite a strong bias in the WTO framework in favor of import substitution measures rather than export promotion for developing countries,23 Colombia has successfully negotiated rather substantial levels of "aggregate support measures." However, few of the authorized support funds have actually been disbursed. Table A.1.10 compares recent WTO notification levels for the global measure of support (GMS) to Colombian agriculture with actual disbursements. Notifications for "green box" expenditure^^^ fell to only US$79 million in 1999, while proposed "special and differentiated treatment" expenditure^^^ fell to US$40 million for that year. Similarly, utilization of WTO-approved expenditures for production-related export subsidies26fell to zero for 1999 and 2000, casting doubt that official notifications of subsidy levels of US$297 million and US$287 for 2003 and2004 will actually be spent. European "structural funds" established a precedent for international assistance for needed structural changes in developing agricultural economies.27 For Colombia, greater investment in human 23For example, export subsidies, which in developing countries tend to favor the most labor-intensive products, are not permitted at the same levels of magnitude as protectivetariffs, which stimulate import substitution. Theseinclude research and extension, infrastructure, rural development programs, andother expenditures not tied to production or trade levels. 25Theseinclude reforestation and ruralcapitalization incentives. These were to be supported through the Plan Vulleo, export credits, and duty drawbacks (CERTSs). CERTs were determined not to constitute export subsidies. Nevertheless, Decree 1989 in2002 reducedCERT benefits to zero. *'Ireland,Spain, Portugal, and Greece, for example, benefited from these intheir transition to EUmembership. 24 and physical capital, and measures to assure that export subsidies reach small producers in rural areas, require structural and administrative solutions, which will be discussedinthe next section of this study. With respect to foreign barriers to its agriculturalexports, Colombia is currently negotiating with the United States for greater market access.28 The regional trade organizations in which Colombia participates (Cairns Group, Andean Community, ALADI), and its bilateral trade agreements, either exclude agriculture or contain significant loopholes. Colombian negotiating priorities should include closing these gaps and expanding agreements with the United States. Foreign phytosanitary requirements for Colombian agricultural market access are a challenge that particularly deserves greater negotiating and structural reform efforts. New restrictions for food imports into the EU scheduled to go into effect in 2006 will make greater food safety compliance even more important for agricultural exporters. In recognition of the vital importance of product quality for success in exporting, the National Program for Quality Assurance provides incentives to enterprises to meet international sanitary, safety, and quality standard^.^' But existing institutions?' while often staffed with qualified experts, generally lack sufficient funding. New legislation needs to be enacted to overcome fragmentation of nationalhealth responsibilities. An improved programshould include expanded training of producers, laboratories capable of testing and certifying food safety compliance, and enhanced research on pest management to reduce reliance on agrochemicals banned or restricted inforeign markets. Sooner or later an export certification system will be needed, along with a fee system to pay for it and to subsidize smallholders. Group certification procedures, if implemented, could help cut the cost. At the same time, international negotiations should seek to develop areas of agreement on phytosanitary issues between Colombia and its principal export markets.31 Agricultural Fiscal Policv Agricultural fiscal policy consists of (a) government expenditures for agricultural programs, including funding allocations andprogramadministration; and (b) methods of raising revenues to support such programexpenditures. This section examines practices and options for both sides of this equation. Table A.1.11provides an overview of fiscal expenditures on Colombian agriculture over the past 10years. After risingfrom 1993 to 1996, expenditures declined sharply thereafter, averaging in2001-02 only 55 percent of the 1993-94 average. The only institution to receive increased funding was the Ministry of Agriculture and Rural Development itself (less than half of the increase being attributable to accounting changes). Table A.1.11 also reveals that expenditures have been allocated heavily to subsidizing specific products or "product chains," rather than supporting farmers generally or basic farming requirements such as irrigation, landtenure and improvement, or agricultural technology. Cuts in irrigation expenditures were particularly severe47 percent over the past decade. Fundingfor integrated ruraldevelopment (DRI) through the BancoAgrario andother institutionsalso declined sharply. These levels of funding, as well as many of the programs to which funding was allocated, are not consistent with Colombia's national priority of increasing agricultural GDP, expanding exports, and reducing rural poverty. Most of the participants in the Rural Capitalization (ICR) and product storage '*US.tariff-rate quotasaffectingColombiahavebeencompiledby Monge-Gonzalez,Conzalez-Vega,andMonge-Amo(2002: Table 10). 29Accepted norms include I S 0 9000, I S 0 14000, OS 9000,andthe GoodManufacturingPractices(HACCP) seal. 30These includethe Corporucidn Colombia Znfemacionul, PROEXPORT, Colciencias, Bancoldex, and inthe phytosanitary area, ICA. 31Subject to confirmationby more detailedcalculations,this study concludesthat, for example, both Colombia andthe UnitedStates would gain from the removal of phytosanitary barriers to poultry by both sides. To meet differences in consumer preferences for chicken pms in the two countries, Colombia could export chicken breasts to the U.S.,and the US. the remaining parts to Colombia. Inthe U.S., prices for chicken breasts would fall and for other parts would rise, with the reverse occurringin Colombia. Colombia should also accelerateefforts to negotiate broad certification of freedomfrom plant diseases alongthe lines of the agreement CostaRica has reachedwith foreign markets for its fresh fruit exports. 25 programs, for example, have been medium- and large-scale farmers.32 The requirement of commercial financing under these programs tends to exclude the poor, and to favor capital-intensive production and import substitutioncrops. The ICR i s symptomatic of the problems associated with expenditures on product chains whose purpose i s to promote greater cooperation and sharedvision among growers, processors, and marketers of particular products or product categories. Although in some cases they may achieve that purpose, producer chains also tend to become organized lobbies for government support of products lacking comparative advantage, thereby draining support away from more efficient sectors. Most are dominated by agroindustrial interests, leaving primary agricultural interests without awareness of or an effective voice in the concerted actions of the chain. Such inequity also occurs in subsidy programs such as recently approved tax exemptions for investments in long-cycle crops, which, by definition, exclude small-scale producers whose incomes are too low to pay taxes. A small farmer who produces an export crop often sells it to an intermediary or processor, who then sells it to an exporter (transactions which are often informal and undocumented). So the bulk of any export subsidies are absorbed by processors and marketers large enough to manageexports, but rarely tickle down to the farmlevel. Given these concerns, it would appear appropriate to review product chains as mechanisms for fiscal support of the agricultural sector, and the Colombian government appears already to be moving somewhat in that direction. Its support for cotton production, for example, i s conditioned on demonstrated productivity improvements-a linkage rarely seen in other Colombian programs, but a technique that has beenkey to much of East Asia's rapid economic growth. With respect to exports, two remedial options deserve consideration: (a) export incentive mechanisms that explicitly conform to links inthe marketingprocess, and (b) direct incentives to farmers, with emphasis on small producers.33The following are illustrations of each of these approaches: Illustration I: Marketing Coupon Program. Through a producers' association, coffee growers are given coupons worth X monetary units per kilo of coffee.34 Coupons would have no value unless and untilsold to an exporter. The same coupon would be worth more to the exporter than to the grower (say, X(I + s), where s is, say, 0.20 to 0.50, reflectingthe "value added" by the exporter). Exporters would buy coupons from farmers only for the quantity of coffee they could sell to international markets. Exporters could then cash in the coupons for exports that are verified by appropriate documentation, receiving sX monetary units per kilo (the grower having already received X monetary units). Such a mechanism assures that benefits reach the grower. It i s easily monitoredby governments using, for example, a three- segment coupon (one segment being retained by the grower, another retained by the exporter, and the last retained by the remitting bank or government agency). The scheme can be easily modified to include processors and other "links" (intermediaries) that perform valuable functions in the total process from farmto final consumer. Illustration 2: Uniform Direct Supports. The PROCAMPO program in Mexico provides direct incentives to all participating producers (farmers) in the amount of US$lOO per hectare in production (regardless of production PROCAMPO has been found to redistribute income to the rural poor 32Although the govemmenthas recently taken steps to increaseaccess to the ICR for small- and medium-scalefarmers and perennialcrops, it is estimated that only about 2 percent of ICR disbursements in 1999-2000 reached small farms (Forero 2003:55). (See Table A.1.12 fpr ICR subsidiespaidby type of investment.) 33The WTO prohibition against high levels of direct export subsidies, and the general intemationaltrend discouragingtheir use, must be taken into account in designing and implementing these approaches. For developing countries, however, it may be possible to negotiate WTO acceptanceof such measures, at leaston atemporaryandtransitionalbasis. 34This example is takenfrom Norton(2003). 35Import-competingcrops would receivebenefits as well, which could assist farmers in making a transition necessitatedby lower import tariffs. No such transition has happened in Mexico, but as yet there has not been an opportunity to observe the combined effects over time of PROCAMPOsupport anddecliningborderprotections. 26 more effectively than either price support or agrarian reform programs (Valdivia and others 2000:93- 127). This approach can replace state-owned development banks, which have generally proven unsustainable because loan arrearages eventually reachunmanageable levels. Land-based subsidies offer the added benefit of encouraging titling, explicit rental contracts, and other rural land reforms. Above all, however, they reach smallholders and export crops that are otherwise difficult to reach, and do so in a reasonably equitable manner. Careful monitoring by the Ministry of Agriculture would be required to assure system integrity, but Colombia already has a basis for determining eligibility of farmers to participate in such a program (including number of hectares under cultivation) through its unidudes agricolus fumiliares (UAF) program.36 Key determinations to be made in implementing such a program would include settingappropriate acreage limits andenvironmental restriction^.^^ Raising revenues for agricultural development-the other side of the fiscal policy equation- poses equity and efficiency problems similar to those on the expenditure side. Typically, constraints on general revenues and real estate tax collections dictate that some funding must come from within the agricultural sector itself.38 These constraints can be eased by a threefold approach: (a) land taxes based on area (flat rates) rather than valuations, (b) collection of taxes jointly by national and local govemments (even though most revenues go to the latter), and (c) establishment of local infrastructure investment and training funds (parafiscal trusts) providing for local citizen participation infunding allocation decisions. Area-based tax systems, which have been implemented in India, Malaysia, and elsewhere (see Khan 2001), would not discourage investments in farm improvements as valuation-based systems often do, and would encourage landuse by avoiding intrusive administrative procedures and penalties. Flat tax rates support land redistribution through market mechanisms rather than government coercion by removingthe distorting effects of differential tax burdens. Should such a revenue system be implemented in conjunction with direct payment incentives, smaller farms would become net beneficiaries and larger farms net contributors. Figure A.1.3 shows a hypothetical illustration in which farms above 8 UAFs become net contributors. That threshold could be set at any level by adjusting applicable tax rates. A flat area-basedtax rate i s one of the options for increasing incentives for effective landuse in Colombia and it-together with the best way of distributingthe revenues accruing through such a measure-should be exploredin more detail and with a goal of developing a very concrete proposal to that effect. Parafiscal agricultural development funds (recommendation (c) above) are not without problems. Rates of compliance with producer contribution requirements vary widely, from 100 percent in the case of poultry, to 4 to 5 percent for fruits and vegetables (Contraloria General.. . 2002, 2003). Furthermore, according to the same source, administrative expenses of some of the parafiscals are excessive. Only 6 percent of the revenues of the National Livestock Fund, for example, were expended for its intended purposes-research, technology transfer, and training. Finally, parafiscals often direct a disproportionate share of benefits to larger-scale farms. A study of the cacao fund, for example, shows that the technology package it developed was too costly for most smallholders, who are the majority of cacao producers (Contraloria General.. .2001). Despite these problems, parafiscal funds are already an important force in the Colombian economy. Almost one-fourth of total public sector expenditures in 2000 were managed by parafiscal financial entities. Seventy percent of those funds went to the National Price Stabilization Fundfor Palm Oil and Related Products, and the National Cattle Fund. The rice, poultry, cereals, and cacao industries 36Descriptions of the UAFmethodology are found in Direcci6nde Desarrollo Agrario .(2000) andDirecci6n de Desarrollo Social .. .(2002). . 37Mexico has no upper limit on eligible acreage, but Colombia should set one. Environmental restrictions might include, for example, excluding from subsidies any cultivation inthe pcirumo, which has fragile soils and is vital to national water supplies. Similarly, steep hillside cultivation and excessive use of agrochemicalsmight be restricted from eligibility for subsidies. '*Realestate tax collections are generally low because (a) cadastral values of land and real estate appraisals in developing countries are usually undervalued and out of date, and (b) local govemments usually are not strong enough to administer and enforce effective tax collection, especially on larger properties. 27 also participate in parafiscals. Studies indicate that those that function well (such as the coffee fund which has operated successfully since 1940) deliver the benefits of fiscal decentralization. Those that represent the most homogeneous products seem to function best, while those that represent highly disparate products more often flounder. These findings suggest that a single national parafiscal fund to support agricultural development i s unlikely to succeed, and that local parafiscal funds for particular products or homogeneous product groups are likely to generate greater compliance and effectiveness. Both localized control of parafiscals, and strong oversight by national ministries, are essential to their success. ExchangeRate Policv and Anricultural Incentives Unless offset by international price trends or tariff changes, exchange rate changes have a powerful depressing effect on agricultural production and real rural incomes. Having been confirmed in many countries, this linkage argues for avoidingexchangerate fluctuations. InColombia, exchangerateshavemoveddecisively inthe direction of devaluationinrecent years after appreciating in the 1990~.~'The Central Bank calculates that the exchange rate i s currently undervalued by about 10 percent, although this could be an underestimation due to high levels of illegal export earnings. This study reachedthe conclusion that modest (circa 10percent) exchange rate changes have no significant effect on the comparative advantage of most agricultural products. Noncompetitive products found their comparative disadvantageworsened in a 10percent revaluation scenario, while products with a competitive advantage generally retained it. In a few borderline cases, however, exchange rate movement had significant effects. Table A.1.13 shows the shifts in DRC values for certain regional products. These data confirm that currency revaluations should generally be avoided in the interest of maintaining agricultural growth andpoverty reduction. 5.3. Agricultural Resource Base Policy Options 5.3.1. LandTenure Policy Agricultural landholding inColombia has remained highly concentrated for centuries. More than US$3.5 billion was spent on land reform from 1961 to 1999, resulting in only 1.5 millions hectares redistributed to 102,000 rural families (Balcazar 2001). Worse, significant reconcentration of land has occurred. Rural violence has driven many smallholders off the land. Lack of capital and credit has made it difficult for smallholders to hang on to redistributedland, much of which is in agriculturally marginal areas:o Unequal concentration of both land and land value in larger holdings (see Table A.1.14) has produced widespread land underutilization. Larger holdings are extensively used for livestock operations-often not the most intensive or productive use. Such low productivity prevents Colombia's comparative advantage from being fully realized. Inefficient rural land markets, combined with the nonfarming benefits of large-scale holdings, such as tax advantages, access to credit, and inflation protection, have perpetuatedthis pattern of inequality and misuse.41While Colombia has relatively active landmarkets, they tend to be segmented, with larger holders dealingwith other large holders, and smaller owners similarly dealing mostly with each other. 39"The appreciationof the exchangerate hasbeena critical factor inexplainingdecreasingretums for tradables and the structuraltransformation of Colombianagricultureinthe 1990's" (JaramiUo1998:151). 40For an in-depthanalysis of land policy in Colombia, see the World Bank-financedstudy, "Colombia: Land Policy in Transition," by Klaus Deiningerand IsabelLavandenz(2004). 41These benefitsare discussedinGrusczynskiand Jaramillo (2002). 28 Considerable work remains to be done to improve the functioning of land markets and achieve greater land distribution equality inColombia. It i s now widely acceptedthat lower-income rural families generally cannot earn enough from working the landto pay full market prices for it and meet other living costs (Binswanger, Deininger, and Feder 1995). So subsidies are required, preferably through landfunds in the form of second-story financial institutions that would operate through qualified financial intermediaries. Experience suggests that land funds are more effective when combined with a player who will assume the risk associatedwith providing product marketingchannels for new landowners-an approach that might be called "entrepreneurial land distribution." Some examples of this approach already occurring inColombia show that it i s well suitedto agricultural situations: 0 Banana companies in the Urabi sold land from their extensive holdings to families already cultivating bananas on it, and provided financing, technical advice, and purchase contracts for the fruit that enabledthe newly independentgrowers to pay for the landover time. 0 The owner of a large grape and fruit plantation in La Unih, el Valle del Cauca, set up a fruit processing plant and sold the land in small parcels to interested rural families. The processor (former landowner) provided mortgage financing and fruit purchase guarantees, and assumed all marketing risks for the processed products, thereby virtually eliminating risks to the small growers. Both growers and the processorhave prosperedunder the plan. 0 Indupalma used support from Fondo de Znversiones para la Paz (FP) to finance sales of more than 2,000 hectares of forestland to smallholders in Cesar Department. The new landowners entered into preharvest contracts for the sale of their output and received technical support from the purchasers. This approach should bereviewedwith the aimof replicating it inmorelocations and agricultural communities. Its use can be expanded more rapidly if governments provide financing support, which should be accompanied by new legislation that absolutely precludes expropriation (except in cases of purchases with illegally earned funds). Wider adoption could mark a new era of land acquisition by the rural poor basedon secure landrights!' Important as it is, landownership i s not the only route to improved rural living standards. Land rental i s the more feasible avenue for rapidly increasing earning possibilities for campesinos,and serious consideration should be given to policies to improve landrental markets. Like ownership, rental access i s likely to benefit from programs combining finance, tenure security, acreage-basedtax systems, and risk- reducing purchase guarantees. Rental segmentation and inequity, which arise from mutual distrust between large landowners and the rural poor, can be reduced by formulating and widely publicizing model contracts for rental and sharecropping that are based on terms that are fair to both landlords and tenants. Attempts to control landrents directly can give rise to other market distortions, and rarely work. Finally, building social trust in land and rental markets at the local level i s critical for effective and lasting reform. Establishing locally managed infrastructure funds can contribute to such trust. But the importance of education cannot be overstated. Ultimately, productivity and competitiveness depend most of all on education levels. Inthe 12-to-17-year age group inColombia, the rate of school attendance in 2000 was 82 percent in urban areas, but only 60 percent in rural areas (DNP n.d.). Clearly, rural 42Seethe backgroundstudy, "Elementos paraunaReformaAgraria Efectiva," by R. Suirez and K. Vinha (2003), Bogotd, Colombia, which, basedonabigger sample, identifiessome of the conditions for success. 29 education needs strengthening on both the supply and demand sides if land reform and other rural programs are to succeed. 5.3.2. Irrigation Policy Irrigation is potentially one of the most productive investments in agriculture. Yet a number of recent studies have documented the inadequate design and deficient operation of public sector irrigation systems in Colombia.43 Excessive water use i s common, causing soil salinization, erosion, and waterlogging. Especially in smaller districts there i s little consideration of crop features prior to construction of irrigation systems, and water releases to secondary and tertiary channels often deviate from accepted technical norms. Perhaps most important, the country has no management training or monitoring and evaluation systems for irrigation. Together, these shortcomings contribute to environmental problems, including hydrological damage, reduction of fish stocks in coastal areas near irrigation systems, and loss of biodiversity. Colombia has 6.6 million hectares of irrigable land, of which 842,000 hectares have irrigation and drainage improvements. Of that area, 62 percent i s in private irrigation districts, and the remaining 38 percent in public districts (INAT 1998). In2000 there were 26 large- or medium-scale districts covering 246,962 hectares. Small farming families were servedby 568 districts covering 39,472 hectares, but only 353 (66 percent) were in operation, servicing a mere 24,500 hectares and 17,000 families. A growing number of smallholders engage in"micro-irrigation" usinggarden hoses, which can be very productive. It is beyond the scope of this paper to analyze the problems of Colombian irrigation in detail. However, a thorough review of this important factor of production would be worthwhile. Worldwide experience should be taken into account in formulating a new irrigation strategy for Colombia. In brief, that experience suggests that irrigation systems should be constructed or reconstructed so that they are simpler to use. Criteria for water releases should be established on the basis of careful watershed studies that include all land uses and the conjunctive nature of water supplies (surface and groundwater). Economic studies should be done to determine crop patterns that are appropriate to each irrigation district. Public irrigation authorities should be decentralized and accountability to users strengthened. Indeed, transferring management, maintenance responsibilities, and even property rights for irrigation systems to water user associations can be productive.44 Governments, however, should provide adequate fundingto rehabilitate systems before shifting responsibilities to users, or find suitable arrangements to that effect. Water user associations should be representative (including women and younger farmers), and users should be trained inappropriate local water management. 5.4. Policies for Access to Technology and Markets 5.4.1. Agricultural Technology Policy Continuous development andadoption of improved agricultural technology i s central to achieving sustained competitiveness. Colombia has significant accomplishments in this area, which will be discussedin this section. However, on a national level, research fundingi s declining. Researchpriorities are questionable and there are operational difficulties in the extension service. Not surprisingly, smallholders tend to receive few benefits from technology programs. This study, therefore, offers some suggestions on how to reform and improve agricultural research, technology transfer, and technology support. ~ ~~ 43See, for example, Herrera (2002); Martinez (2001); CONSULTPLAN(1998); and INAT (1998). Water users associations are explored extensively in Subramanian, Jagannathan, and Meinzen-Dick (1997). In regardto transfer of systems, see Vermillion (1999). 30 Issues inAnricultural Research Worldwide studies confirm that agricultural research generally yields high returns, and that technology plays an important role in poverty reduction. Yet research funding in Colombia in recent years has fallen from a peak of 0.40 percent of agricultural GDPto 0.14 percent-a decline of about two- thirds. It was estimated in 1999 that 80 percent of existing Ph.D. researchers would soon be retiring (Arango Nieto, Rivera, and Junguito 1999:83). An international team (Misidn ISNAR) pointed out that "CORPOICA in recent years has lost an important part of its most qualified human resources" (Misibn ISNAR 2002:4). The fact that products with comparative advantage are, in many cases, receiving the least research support has raised doubts about the appropriateness of the research agenda. A Sociedudde Agricultores de Colombia (SAC) workshop devoted to identifying research needs for 11crops and crop groups revealed deficiencies for cacao, fruits and vegetables, cotton, and potatoes-all strong comparative advantage crops (Arango Nieto, Rivera, and Junguito 199957). Misidn ISNAR confirmed these findings in 2002.45 To be relevant, research agendas cannot be defined only at the national level. They need to reflect regional and subregional interests. Producer organizations have helped identify Colombia's research priorities. Agroindustry has called attention to product quality issues. Stronger producer organization couldhelp reflect consumer demands inthe researchagenda-a necessity inan era of greater integration or production and markets.46 But this mechanism represents only a partial answer to developing more appropriate research agendas. The value of participator research in which farmers play an active role is increasingly recognized. Although Colombiahas been at the forefront of this movement, it has not formulated a policy framework for promoting and supporting the participatory research as an integral part of the national research program. The participatory approach has been recognized by CORPOICA, but institutions for implementing it more widely are not yet operational. Indeed, CORPOICA may not be the best agency for a participatory research program that i s national in scope. Participatory research is, by nature, a highly decentralized activity. Over the long run it could prove difficult for a specialized national institution accustomed to a top-down approach to avoid trying to dictate local research agendas. Such a program would be more appropriately located in a rural development institution, such as PBA and the CIALs. The Ministry of Agriculture, CORPOICA, and four universities are members of PBA's board of directors. Fifty-two local committees have been formed. They elect regional committees that help choose biotechnology research projects for campesinos to carry out. After successful trials are completed, participating producers often form small enterprises to market genetic materials and other inputs. CIALs undertake activities similar to PBAs. The use of facilitators and the establishment of small local funds, managed by the participants, for research materials and other inputs, have been key to forming successful and lasting CIALs. This approach i s popular with farmers, who have demonstrated a capacity to teach one another about participatory research. Inboth types of local agencies, the role of extension agents is performed by participating farmers. If a national policy were to be formulated for participatory research, it would be important to incorporate these features, and to certify and carefully define the role of facilitators vis-5-vis supportingproducers and researchinstitutions. Excessive pesticide use deserves particular attention inColombia's research agenda. Colombia's use of pesticides averages 6 kilograms of active ingredients per hectare, but i s as highas 16 kilograms in 45Morerecently, the Misidn ISNAR pointed out that, "The analysis of the allocation of resources for agricultural research, both for the entire system . . and for CORPOICA, suggests a high researchconcentration on traditional crops . . . to the detriment of more promising crops in , which Colombia has a naturalcomparativeadvantages such as fruit and vegetables which, ingeneral, have receivedless attentioninagricultural policy" (Misi6n ISNAR2002:lO-11). 46This point hasbeenmadeby Cano (2002). 31 some locations. Of the 307 active ingredients used in commercial pesticides, 52 are prohibited, severely restricted, or considered highrisk (Niva 2001). Some that are prohibitedor restricted on the Consolidated List of the United Nations are still used in Colombia. This weakens Colombian agricultural competitiveness by raising costs of production, causing health problems among field workers, degrading the ecosystem(including soil and water contamination), and threatening to bar export products from some international markets.47 Controlling this problem will require research on integrated pest management techniques, restrictions on the importation and sale of toxic materials, and a high-level campaign to raise awareness about the alternatives to pesticide overdependence. Finally, Colombia needs a clearly defined policy with respect to biotechnology generally, and genetic modification of plants in particular. Research on plant resistance to stress, droughts, toxic soils, and pests could have high payoffs for Colombian farmers if properly managed. There i s hope for developing coffees that are resistant to rust, and cotton could be modified to be less vulnerable to pests, for example. A policy for genetic research should set priorities, specify appropriate practices and restrictions, and take account of market requirements. To accomplish that, both grass-roots education and the Colombian legal framework dealing with biotechnology researchneed strengthening. Issues in Agricultural Technologv Transfer and Suuuort Although, ultimately, participatory researchand agricultural extension services are on convergent paths, buildingresearchcapacity and the institutionsto sustain it takes time. So extension services will be needed for the foreseeable future. In the 1990s, recognizing that top-down systems were not meeting farmers' needs andhad done little to reduce rural poverty, and for budgetary reasons, Colombia devolved extension systems to the municipal level in the form of Municipal Units for Agricultural Technical Assistance (UMATAs). The original idea behind UMATAs was to enable smallholders to choose their extension agents through local producers' committees. Inpractice, becauseof politicization, funding limits, high rates of turnover, lack of links with research programs, and various management and structural problems, only about half of about 800 UMATAs have performedsatisfactorily. The Colombian government i s aware of these problems and i s working on plans to refine this approach to extension services. One option would be to shift managementto subregional (rather than municipal) levels. No matter how well organized and managed the system, however, if it i s hierarchical the tendency will be for fields agents to provide services that satisfy superiors inhome offices rather than client farmers' needs. The desirable goal of providing farmers with extension service choices has been addressed in Estonia, the former East Germany, and elsewhere by privatizing these services. This requires subsidies, becauselow-income farmers cannot afford to pay for private services. Subsidies, however, guaranteethe market for the services of private extension agents. Other approaches include providing vouchers to groups of smallholders for the purchase of extension services. Alternatively, federations of farmers could enter into contracts with governments to provide extension services underwritten with government financing. All of these approaches, which are well worth considering, require adequate fiscal support, and appropriate controls and supervision. Inareas where no choice of extension services i s yet available, care should be taken to avoid undermining the prospects for the emergence of private, subsidized services. For better or worse, a small farmer i s an entrepreneur, and extensions must increasingly provide the business and management skills needed for entrepreneurial success-leadership, project planning, financial and information management, communications and marketing strategies, and the like (Can0 47 No adequate system of epidemiological monitoring for pesticide exposure yet exists. In 2000, the Ministry of Health Reported 106 people affected byjust one substance, Endosulfan, usedincoffee. 32 2002). To provide these skills, and link extension and researchsystems, local agricultural communication centers would be helpful. Internet communications have been established for some CIALs, apparently with good results. Rural Centers for Entrepreneurial Development could be established-each equipped with computers, access to the Internet, classrooms, and a permanent facilitator who could demonstrate how to download technical and real-time market price information. Such centers (which could be used evenings for adult literacy and farm management instruction) would constitute a decisive step toward overcoming the lack of communications that holdrural areas back from growth and development. Finally, a national agricultural technology system i s more than researchand extension services. It i s a body and system of knowledge that must be organized, expanded, and preserved (as well as applied). As such, it requires a legal and policy framework for intellectual property rights that takes into account the special needs and conditions of agricultural research, including protecting information on product quality, food safety, plant and animal health, export markets, and other competitive factors discussed in this chapter. Colombia's legal framework for intellectual property consists mainly of international conventions the government has signed. Implementing regulations, however, are lacking. ICA i s charged with the enforcement responsibilities, but currently does not have the capacity to fulfill the task. Although much valuable agricultural information i s inthe public domain, privately funded research plays an increasingly important role. Effectively enforced intellectual property regulations are particularly important for creating a climate that will encourage and maximize privately funded research. 5.4.2. RuralInfrastructureandMarketing The main internal bottleneck in many rural areas i s lack of adequate secondary and tertiary roads. Inaddition, the east-west primary road network is poorly developed. The government is moving to improve roads in poorly connected areas, and also to widen coverage of electricity and telecommunications services in rural areas. The government has also moved decisively to improve security on the highways under the National Development Plan (Presidencia de la Republica 2002:39) :* . . . Such improvements would benefit agriculture andhelp reverserural poverty. Supermarkets have transformed marketing in Colombia, especially for fruits and vegetables. The volume of domestic sales of fruits and vegetables by supermarkets i s about two-and-one-half times greater than that of export sales (three times greater if bananas are excluded) (Reardon and Berdegue 2002:385). Forero (2003) found evidence that supermarkets are displacing some traditional marketing intermediaries, with positive effects on product quality, new product development, and packaging. CORABASTOS, Colombia's traditional wholesale market, has seen its share of the market in Bogota and other cities and regions diminish. But it handled about the same tonnage in2001 as it did in 1987, and so remains an important nationalagricultural marketinglink.49 On the export side, marketing is carried out mostly by companies that purchase directly from producers (often smallholders), producer associations, and (for organic products) nongovernmental organizations. These marketing channels are often noncontractual, depending instead on mutual producer-marketer confidence and marketing agents who supply technical assistance, credit, and other services. The case of plantains is noteworthy for the technical assistance and other inputs supplied by exporters. 48 River transport is by far the least expensive way to move freight (especially bulk goods) in Colombia, and the country has an unexploited resource inthe MagdalenaRiver. Fulfilling the govemment's priority goal to makethe Magdalenanavigable for most of its length will improve export prospectsfor some agriculturalproducts. 49 Oneof the most important policy steps that canbe taken to strengthenproducer-to-retaillinkages is to develop legislationrequiring payment to growers of fruits and vegetables within 30 days, as has been done in Argentina and was under consideration recently in Costa Rica and Brazil (ReardonandBerdegue 2002:386). 33 To make export marketing more effective, the government could provide more technical assistance on quality issues, food safety, export finance, incentives for trial shipments, and funding for participation in trade fairs for producers committed to exporting. Some elements of such a program are already inplace, but they need strengthening especially for nontraditionalproducts. ICA has a network of 26 livestock and 14 agricultural crop diagnostic centers which function well and would constitute the basis for a national certification system for food safety and plant and animal hygiene. Product quality improvements generated by supermarkets have had the additional benefit of better preparing producers for entry into export markets. 5.4.3. RuralFinanceIssues andOptions Rural finance policy has focused on agricultural lending at the expense of providing more integrated financial services for the rural sector.50 In the past decade, policy has revolved around FINAGRO, a second-tier lender, with the CajaAgraria, later replaced by the Banco Agrario, being the primary lender charged with executing said policy. FINAGRO i s funded through forced investments required of banks in Titulos de Desarrollo Agropecuario (TDA), which are then rediscounted to banks that provide loans for agriculture at below-market rates. Loans rediscounted by FINAGRO are eligible for an investment subsidy (ICR) of up to 40 percent, and a credit guarantee rangingfrom 50 percent of principal on loans for large farmers to 100 percent for small farmers. These instruments create an artificial demand for FINAGRO's credit operations, but have not kept the outstanding loan balance from declining in real terms since 1998, when the economic downturn greatly reduced the demand for agricultural lending. Loans to small producers plummeted duringthis period as the Caja Agraria was liquidated and financial cooperatives entered into a crisis brought on by aggressive expansion and poor supervision (Table A.1.15). Private banks, disinclined to lend to smallholders due to highrisk and elevated costs of lending in the Colombian countryside, restricted agricultural lending further during this period. The Banco Agrario that took the place of the CajaAgraria implemented stricter controls than its predecessor, and the board authorized conservative risk analysis criteria, distancing the bank from its client base comprised of small producers and poorer rural dwellers. The cooperative sector was placed under much closer supervision and its presence has yet to recover to pre-crisis levels. Stricter regulation has increased the paperwork required to show the ability of borrowers to pay, and reduced the admissibility of rural collateral to the point that smaller parcels are often not accepted. On the demand side, small borrowers are dissuaded from participating in the formal financial sector by the paperwork demanded by banks that are out of sync with the informality of the rural economy, delays in loan processing, the unsuitable credit lines, and excessive transaction costs that include taxes and fees that result innegative returns to savings. The decline in bank lending has been countered in part by progress in access to capital markets through the Bolsa Nacional Agropecuaria, which provides liquidity through forward contracts, collateralized obligations, the discounting of contracts, and factoring. However, innovation has partially been hamperedby what the Bolsa deems unfair competition from FINAGRO that provides medium- and long-term financing guarantees and subsidies at the taxpayer's expense (Table A.1.16). As in bank lending, access to capital markets i s largely limited to large commodity producers, although some brokers have begun to operate at the local level, bringing small producers together to participate in forward contracts. The participation of small farmers in capital markets, and the expansion of instruments to campesinos crops, should be closely monitoredto improve access. 50For a comprehensiveanalysis of rural finance in Colombia, see the World Bank-financedstudy, "Colombia Rural Finance: Access Issues, ChallengesandOpportunities," World BankReportNo. 27269-C0, November2003. 34 The rural conflict will continue to limit the participationof private lenders in agriculture for some time to come. As such, the Bunco Agrario remains the most likely lender for smallholders, although some cooperatives are poised to take a more active role in this sector. The Banco Agrurio should reorganize itself in line with the principles of good-practice microfinance, a change that would require a sustained effort to adopt lending technologies that evaluate the debt capacity of the entire borrowing unit instead of the feasibility of a particular crop. Caps on interest rates should be removed or at least increased to reflect the risks and costs of rural lending. Collateral restrictions should be lifted, and the system for executing guarantees improved. The presentation of paperwork should reflect the informality of the rural economy. Finally, the provision of credit guarantees and subsidies should be untied from FINAGRO's credit lines. Rural lending should take a business approach to developing product lines adjusted to the needs of its clientele. Lendingdecisions should be devolved to the local level and basedon ability to pay rather than use of funds. Both the Bunco Agrurio and cooperatives could benefit from new technologies that reduce the risks in the transportation of cash and allow for closer monitoring of loan performance. A business approach to rural finance and a more flexible regulatory approach will require improved risk management techniques. A move from risk management to risk aversion will allow rural financial institutions to increase their client base, guaranteeing their future viability while increasing access to financial services by the rural poor. 6. Competitivenessandthe CampesinoEconomy,Approaches to Poverty Reduction, and Sustainable Competitiveness 6.1. The Roleof CampesinoProduction Smallholders participate extensively in crops that represent about two-thirds of total agricultural production, a percentage that is higher when illegal crops are included (see Table A.1.17). The association of smallholders with proscribed crops, which may have increased in the last three years with the increasing unemployment in coffee-growing regions, illustrates the social dimension of the problem of illegal drugproduction. The campesino share of output i s quite high for some crops, representing all of jute production, exceeding 90 percent for tobacco, exceeding 80 percent for panela and cassava, and rising above 70 percent for vegetables, fruits, plantain, and wheat. Rice and cotton are two of the crops least cultivatedby smallholders, with around 10 percent of their output coming from the campesino sector (Jaramillo 1998:21). Campesino production often i s characterized by lower physical yields but much lower monetary costs compared with larger-scale farms. Nevertheless, there i s some evidence that over time cumpesinos have been more successful than larger farmers in raising their yields. Table A.1.18 shows how yield differences between smallholders and commercial farmers narrowed during 1973-76 and 1988. Some observers, such as Forero (2003), feel these data exaggerate the progress of cumpesinos relative to commercial producers, but it seems clear that at least some movement has been madetoward reducing the productivity gap. Nevertheless,concrete examples may be cited from more recent years that illustrate the ability of smallholders to adapt technologies and innovate. These include (Forero 2003): 0 The construction of small hillside irrigation systems for fruit and vegetable production, based on use of gardenhoses. 0 The growth of production of some fruit vegetables, especially tomute de hbol, in small greenhouses. 35 Technological improvements in milk production by smallholders who have become more integrated into the product chain for milk. These improvements include pasture improvement, use of electric fences and pasture rotation, some use of forage crops produced on farm, genetic herdimprovements, and use of nutritionalsupplements. The great rate of survival of the coffee crisis among smaller producers. The production of uchuva for export, which has required meeting exporters' specifications. However, there still i s a tendency to apply excessive amounts of pesticides to this crop, a practice that may endanger its access to export markets in the future if not controlled. Much of the technology transfer for uchuva has occurred among producers themselves. Adaptation of better technologies for potato production, again to a considerable extent by producer-to-producer contact. Improvements intechnology for productionand local processing of panela. These have consisted primarily of better seeding and fertilization techniques, adoption of chemical methods of weed control, and adoption of improved seeds. In recent years some of the m i l l s (trapiches) have adopted more environmentally friendly processes and have achieved greater purity and cleanliness of their product. Another characteristic of campesino production i s that it often consists of production systems rather than monoculture. Campesinocoffee producers, for example, have relied on plantain production, inassociation with coffee, to tide them over the coffee crisis. Producers of uchuva and other vegetables in Sumapaz have continued their production of potatoes, beef, and corn. The production of potatoes in Cundimarca often i s combined with cattle for both beef and milk, family vegetable plots, small parcels of corn, and barnyardpoultry for home consumption. A principal indicator of the competitiveness of campesino production is its extensive participation in exports. Export products with significant campesino participation include the following: coffee, milk and concentrated milk and cream, cheese and cottage cheese, plantain, raw tobacco, cigarettes and cigars, chocolate products, uchuva, pitahaya, granadilla, passion fruit, cassava, yams, fresh potatoes, various fruitjuices andpreserves,cacaoinits semiprocessedforms, andorganic products. The DRC calculations for campesino production of uchuva and panela were mentioned above, and they also suggest competitiveness of the campesino sector. When campesino production i s competitive, it i s founded on efficient use of factors of production, especially of family labor, and careful control of cash costs. Since campesinos' access to institutional credit i s very limited, use of purchased inputs is very costly for them once the high interest rates on the informal credit market are taken into account. Likewise, they often do not have the access to modern marketing channels that larger producers do, so this and difficulties in accessing credit reinforce their reliance on efficient utilization of factors at the farm level to maintaintheir competitiveness. 6.2 Policies to Improve the Competitiveness of Campesino Producers Campesino producers may be defined in different ways-by farm size, by their reliance on own and family labor, by their low rates of use of purchased inputs, and the like. However they are defined, it i s clear that many of them are competitive producers, and that their competitiveness extends to a number of crops and livestock products. They are also strengthening their linkages with marketing networks, for both the domestic and export markets, as indicated in an earlier section of this report. However, it i s not possible to estimate how many of these producers rise significantly above subsistence levels of production, and it is evident that there still i s a considerable amount of poverty in rural areas. Therefore, an important policy question is how a greater number of campesino producers can be brought into the ambit of competitive production, and how that process can be accelerated. Attaining competitiveness in production i s one of the main means for achieving sustainable reduction of poverty. Campesino 36 production i s largely consistent with Colombia's comparative advantage, so assisting this group of producers to improve their livelihoods would also contribute to creating a more efficient agricultural sector, in addition to beingjustified on social grounds. The strategies for improving the competitiveness of the campesino economy can be divided into the same three basic areas mentioned above for competitiveness of the sector as a whole: adequate incentives for production, access to an adequate resource base for production, and access to markets for outputs and inputs and to technologies of production. These three areas are taken up in turn, drawing in part on the discussions in the foregoing sections of this study. In the three areas, in total 17 policy approaches are identified that can helpimprove the competitiveness of smallholders. An integral, consistent strategy is neededfor the campesino sector. Inthe past, there has not been a strategy aimed at addressingthat sector's needs, only individual projects here and there unsupported by a targeted policy framework. Inthe words of the Contraloria General (2002:47), "the State's attention to the campesino sector has been residual." Development and implementation of an agroexport strategy would assist campesinos,given their significant participation inexport crops, but taken alone it would not represent a complete strategy for raising production and incomes among smallholders. The following paragraphs suggest some possible elements of sucha strategy. 6.2.1. ImprovingIncentives for Campesino Production Pricing and trade policies are not differentiatedby type of producer, so prices tend to be the same for smallholders and large-scale producers in a given area. The main exceptions to this statementconcern quality of products, degree of access to markets, and timeliness of deliveries. A number of entities, both governmental and nongovernmental, are involved in training producers in regard to quality norms and marketing requirements in general. However, this process can be accelerated with policy support, including more emphasis on product quality and marketing in the labors of the agricultural extension service. The government also may wish to explore ways to partially guarantee long-term sales contracts to the buyers, so that they would be more willing to enter into them in spite of the reputation of many campesinos for breaking contracts when they find a temporarily higher price elsewhere. This topic also should be a priority in training: the long-run benefits of becoming a reliable supplier and fulfilling contracts. By the same token, it could be helpful to have legislation that requires payment on agricultural purchasecontracts within 30 days. This measure alone would provide greater incentives to campesinos to enter into contractual relationships. For all marketing relationships, the importance of establishing confidence needs to be underscored. Legislation also should stipulate severe penalties for using violent means to control segments of agricultural markets and illegal appropriation of public spaces, such as the unauthorized subleasing of warehouse space in public agricultural crop collection centers. In addition, new marketing legislation should strongly penalize the frequent practice of chemically adulterating panela for the sake of improving its color, which has negative effects on the health of consumers and hinders access to export markets. In the case of rice, the most important marketing improvement that could be made would be to change the grain handling system to bulk handling, rather than handling in sacks. Perhaps some of the funding that currently subsidizes rice storage could be redirected to support the expenditures necessary to make this kindof conversion. This conversion already has been made inanumber of South American countries. Campesino incentives also could be improved through a policy that makes the incidence of fiscal support more nearly uniform over farm sizes. The ICR, for example, i s highly skewed toward larger- scale producers. Monitoring mechanisms are needed to review the incidence of policy by farm size group, along with corrective mechanismsfor ensuring that the incidence i s reasonably uniform. 37 Inthisregard,direct supportpaymentswith acaponeligibleacreagewouldconstituteoneofthe most powerful mechanisms for improving the incentives of smallholder producers. Although the administrative requirements for such a system would be demanding, sooner or later the agricultural authorities will need a stronger local capacity for carrying out and monitoring programs. The framework of land classification by UAF i s by no means complete throughout the country, but workable approximations or substitutescould be conceived, such as defining a few categoriesof landquality. In the marketing area, larger farmers often obtain better prices because they can truck their products to a more distant market, whereas smaller farmers tend to sell the crops locally, sometimes even before they are harvested. The formation of marketing cooperatives can help smaller producers achieve some of the marketing benefits enjoyed by larger producers. However, this i s an undertaking that requires careful planning and considerable farmer training, for the agricultural sectors of all countries have experienced failures of such cooperatives. Finally, establishment of an effective, real-time national price and market information system could assist smaller producers to achieve better average prices for their products. 6.2.2 Improving Campesino Access to Basic Resources The most basic physical resourcesfor production are land and water, and inboth areas the access of poor rural families could be substantially improved. First, there i s little doubt that the functioning of land markets could be improved through the systematic application of an area-based land tax, with an exemption for the first X hectares of land (according to aggregate categories of land quality). Such a tax would provide an incentive for larger landholders to either work their land more intensively, which would create more rural employment, or rent out or.sell part of their land. Inorder for the tax to have these effects, its rate should be perceptibly higher than the equivalent rate in prevailing rural real estate taxes, which in any case are applied inc~nsistently.~~ At the same time, to encourage more active land rental markets, it would be essential to clarify that the mechanism of land expropriation i s being abjured forever, except for the case of lands acquired through illegal earnings. In the case of land rental markets, it will be important to establish an atmosphere of mutual confidence among all social groups in rural areas. In the current atmosphere of violence, this will be difficult, but it may be feasible to do so gradually, proceeding area by area, starting on a pilot basis, and giving emphasis to the strengthening of local institutions, both technically and financially. A helpful measure could be to develop and widely disseminate model contracts for the different forms of landrental. To enhance land sales markets, the existence of financing will be critical. While some mortgage finance i s available at present, it i s far too little compared with the needs. Consideration may be given to establishing a land fund, as mentioned above, from which financing could be solicited by willing sellers or by intermediaries to land transactions. The approach of entrepreneurial land redistribution also has the potentialto substantially improvecampesino access to land. A subsidy for land purchaseby poor families could be extended through this approach or through the operations of a land fund, since it i s widely recognized that such families normally cannot pay a market price for land.52The important principles are targeting the benefits and concentrating on breakingthe bottleneck of lack of long-term financing. '' It would be important to avoid the trap of trying to tax only "idle land," since the pervasive difficulties of making an administrative determination of whether a parcel of land is truly idle (for example, how long in fallow? how many head of livestock per hectare?)constitute an ''invitationissuei to corruption inthe implementationof such atax. This s discussed in Binswanger, Deininger, and Feder (1995). The authors point out that the prestige and speculative motives for acquiring rural land often raise its price above what pure rate-of-retum considerations would indicate, and therefore put it out of reach of most rural poor. 38 The deteriorated state of the national irrigation system especially affects the competitiveness of campesino producers, since it is the small irrigation districts that have suffered the highest rate of abandonment. Reviewing national irrigation policy and taking steps to improve the functioning of irrigation districts would significantly strengthenthe competitiveness of smallholders. The role of farmer training, inboth irrigation managementand planning croppingpatterns, and farmer participation, inboth the design and operation of the systems, merits emphasis. 6.2.3. Improving Campesino Accessto Markets, Inputs, and Technology Market access considerations have been discussed above in the context of production incentives. For access to bothinputsand technology, strong producer organization i s vital, but the functioning of the nationalagricultural technology systemand the rural financial system are also key elements. On the researchside, puttinginto place a national policy for participatory researchand extending the reach of participatory efforts would represent a major assist to campesino producers. Experiences to date with participatory research in Colombia have left no doubt as to the benefits of this approach for smallholders. At the same time, it would be important to reconsider the structure and functioning of some of the parafiscal funds, particularly those for panela, milk, and fruits and vegetables, and to reformulate them. In the case of fruits and vegetables, as mentioned, it is unlikely that a single national fund can function well, given the heterogeneity of those crops and the wide geographical dispersion and generally small scale of the producers that are involved. The transactions (communications) costs involved in makingthat kindof organization function well are simply too high. In the area of technology transfer, finding a way to continue the successful experience of PRONATTA would be a highpriority. At the same time, efforts could be undertaken to make its project selection criteria more consistent, without undermining its decentralized character. Also, a key step would be the further reform of the agricultural extension system along the lines suggested above, with emphasis on creating better incentives for extension agents to involve producers in both identifying problems and suggesting lines of solution, and on providing long-term subsidies to enable smallholders to acquire extension services. Under the privatization option, such subsidies could be financed out of the budgetary savings realized by privatization. The complementary step of establishing Rural Centers for Entrepreneurial Development could assist farmers in receiving training in a broad range of topics, not only crop cultivation techniques but also farm managementand marketingconsiderations. They would mark an important step in overcoming the lack of connectivity that characterizes many poor rural areas, and they would provide venues for specialized forms of education in topics relevant to rural development. There are precedents for centers like this both inColombia and inother countries. For rural finance there will be neither a single solution, nor an immediate one. Nevertheless, strengthening microfinance and providing (modest) incentives for it to move into rural areas could be an important step. Equally or more important would be giving greater priority to the smallholder sector in the funding provided through programs of FINAGRO. Moreover, a solution to the credit needs of the smallholder sector, indeed of agriculture in general, needs to come from strengthened rural financial institutions, and not from attempting to channel more public funds to the sector in the form of production credit. The latter approachhas provenunsustainable in countries throughout the hemisphere. 6.3 Approaches to Poverty Reduction The origins of rural poverty are complex, partly historical, partly social as well as economic, and partly rooted in low rates of human capital formation. Accordingly, solutions are also multifaceted and 39 involve the development of nonagricultural sources of employment and income as well as growth of agriculture itself. Poverty reduction programs have long recognized the need for provision of potable water and health services to the poor and food aid in some cases. In recent years they also have emphasized upgrading educational services for the poor, since in the long run better human capital is essential to sustainable increases in living standards. In addition, in rural areas in many countries some projects have tried to improve the physical productive capital and technical farming knowledge of the rural poor, through small-scale irrigation, better access to land, hillside farming techniques, provision of improved seed, and the like. What i s sometimes missing i s a policy framework that improves the opportunities and earning power of the ruralpoor, and which can make all investment projects more productive. The findings of this study suggest four orientations that may be useful in developing a policy framework conducive to reduction of ruralpoverty: e Policiesfor improving the competitiveness of campesinoproducers,as outlined in section 7.2.1, covering the areas of production incentives, access to basic resources, and access to markets, inputs, andtechnologies. e Policiesaimed at encouragingtheproduction of labor-intensiveproducts. These are largely the products in which Colombia has a comparative advantage (for example, export crops, perennial crops). As shown by the analysis presented in Table A.1.7, a shift in incentives policies toward these products could generate significant gains in both rural employment and rural income, and the employment factor is especially relevant for reducing poverty. The range of policies that can be oriented in this direction goes beyond incentives and includes the priorities in agricultural research, planning for cropping patterns under irrigation, and the elements of an agroexport strategy. e Targeting agricultural subsidies on low-incomefarmers. This study documents the fact that the overwhelming share of subsidies in the sector i s biased in a regressive direction. A partial reorientation of the subsidies toward poor farmers could increase their incomes in the short run and stimulate stronger supply response, and hence yet more income, in the medium and longer run. The study also makes suggestions for new kindsof agricultural subsidies that would be less regressive in their incidence. They could be financed through a reallocation of the funds currently spent on other subsidies or through new sources of tax income such as those mentioned in the study. These new kinds of subsidies are not the only options, and it is recommended that efforts be devoted to identifying options for fiscal expenditures in the sector that would have proportionately greaterbenefits for the poor. e Strengthened programs of rural development. In the end, sustainable and significant improvement in the well-being of the rural poor cannot take place without improvements in several areas, including roads, telecommunications, energy, education (from the primary level on), production incentives, and better access to markets, technology, and resources. Successful rural development requires improvement in several kinds of capital: human, social (organizational), institutional, infrastructural, andprivate (on-farm) physicalcapital. As emphasized throughout this study, improving the institutional environment for the rural poor-and for all of agriculture-requires greater decentralization, a process already initiated but not yet complete. This in turn requires attention to both fiscal and technical capacities at local levels. Without greater capabilities in this sense, the possibilities of assisting the rural poor to become self-sustaining on their own development paths will remain limited. 40 The requirements for each of these types of capital differ among communities. For example, where literacy rates are low, priority should be accorded to strengthening primary education and adult literacy training. Where literacy rates are higher, emphasis can be placed on training in areas such as marketing and farm management. Where road communications are poor, emphasis on the axis of infrastructure capital has to be placed on road improvement. Where the road network i s more adequate, emphasis can be placed on improving telecommunications. A successful rural development program would start with diagnosing the status and needs of each community along these axes, and defining local priorities accordingly. In the process of diagnosis, review of the status of women should not be overlooked, for there i s very strong international evidence that indicates that educating women and improving their income-earning opportunities has a greater effect on family welfare than making similar strides for men.53 Above all, community participation in both the design and implementation of rural development programs i s a key to their success. In northeast Brazil, in Paraiba, more community participation and improved coordination of central agencies' efforts, with more flexibility in defining responsibilities, have played a key role in creating successfulruraldevelopment projects (Cox 2001). 6.4 Sustainable Competitiveness Achieving competitiveness does not necessarily mean it will be sustained. The experience of higher-cost producers of coffee in Colombia in recent years i s ample illustration of this point. But, equally, the experience of many credit cooperatives that collapsed in 1998 and 1999 illustrates another kindof lack of sustainability, inthis case of institutional sustainability. Maintaining long-term competitiveness in the sector requires that sustainability be achieved in five senses: 0 Economic sustainability. The sector needs to maintain comparative advantage in as much of its production as possible. This requires above all continuous improvements in productivity, and therefore it requires a strong agricultural technology system. e Social sustainability. The extensive poverty needs to be reduced, or the sector's development path will become socially unviable. As noted, this may require new emphases in the policy framework. e Fiscal sustainability. It i s important to identify new funding sources for some of the proposed outlays to further the sector's development. However, some new expenditures can be funded with a reallocation of existing expenditures. Also, it should be borne in mind that agriculture's current share of the fiscal budget i s very low by any standard. 0 Institutional sustainability. Institutional structures also need to be durable. Inthis regard, there are evident questions concerning some of the parafiscal funds, some of the UMATAs, and s3 For example, in the case of China, "our analysis provides empirical evidence to support the claim that the gender-focused programs that emphasize promotingwomen participationinboth economic and socialcommunityactivitiescan generate significant benefits" (Coady, Dai, and Wang 2001). For Bangladesh, athorough evaluation of microfinanceprograms found that "annual householdconsumption expenditure increases Tk 18 for every Tk 100of additionalborrowing by women and Tk 11for every additional Tk 100 of borrowingby men," and "women have proved to be excellent credit risks, with loan default rates-of only 3 percent-significantly lower than the 10 percent default rate for men" (Khandker 1998). For Africa, in regardto farm decisionmaking, it was found that "in householdsin Burkina Faso, Cameroon, and Kenya more equalcontrolof inputs and farm incomeby women and mencould raise farm yields by as muchas a fifth of current output" (World Bank 2002). And for developing countries in general, FAD (2001) has found that "female schooling does much more at the margin for income, poverty reduction, and child healthandnutrition thanextramaleeducation." 41 financial structures such as savings and credit cooperatives. In regard to rural finance, appropriate bank regulation and supervision modalities are vital for sustainability, and subsidies in the financial system need to be used with extreme caution, since generally they tend to undermine the sustainability of financial institutions. Institutions that are subject to short-term political influences (state-owned banks, UMATAs in the way they are currently structured) are not likely to be sustainable. In the case of the parafiscal funds, mechanisms are needed for adequate oversight and for promoting genuine and widespread producer involvement in their decisionmaking. 0 Environmental sustainability. InColombia the main threats to the environmental sustainability of agriculture are: (a) the excessive use of pesticides in some crops, which persist in soils and water supplies as well as directly affecting humans and animals; (b) erosion, salinization, and desertification of soils; it i s estimated that almost 600,000 hectares have been seriously affected by salinization (Le6n 200358); and (c) compactation of soils becauseof loss of organic material, reduction of biological activity in the soil, and excessive exposure to solar radiation because of lack of adequate soil management. The most alarming example of soil degradation i s that of the soils formerly used for cotton inCesar, and it stands as a warning of the dangers of inappropriate soil management. To reduce pesticide use, steps should be taken to enhance farmer education on the subject and to control the importation and sale of toxic products that are banned elsewhere. Also, an important first step would be to improve the registers of rates of pesticide use by crop and type of chemical product. Avoiding overapplication of irrigation and developing adequate drainage facilities for irrigation districts are essential elements of a program to reduce the rate of salinization. Ensuring sustainability in all these dimensions is not an easy challenge, but successful agricultural sectors are finding ways to meet it. The five sustainability requirements should be an integral part of any national agriculturaldevelopment strategy. 7. Conclusion 7.1. Strategic Considerations Although this study has identified a number of issues in Colombian agriculture that need to be addressed to improve and sustain its competitiveness, it i s important to reiterate that Colombian agriculture has many strengths, and that most of its products are already competitive by international standards. It also should be reiterated that abrupt changes in policy directions are not recommended and should be avoided. For that reason a transition program would be important. Such a program should have rapid impacts on producer incentives and expectations, and should be consistent with an overall strategic vision. Some of the main overall strategic directions that emerge from the study include: 0 Giving agriculture a higher priority inthe nationalbudget. 0 Reallocating some expenditures in the sector with greater emphasis on supporting products with comparative advantage and programs to assist the rural poor. 0 Strengthening and making more efficient the programs in areas such as quality control and food safety, agricultural researchand extension, and irrigation development and management. 0 Development and implementation of a full, integrated agroexport program and associated measures for improving and guaranteeing product quality. 0 Emphasizing institutional decentralization and reforms and improvements in areas such as some of the parafiscal funds, the UMATAs, and the agricultural finance system. 42 Strengthening land markets with emphasis on mechanisms for promoting greater access to land on the part of the ruralpoor, via both rental and purchase. Implementing a comprehensive strategy for reduction inrural poverty, based inpart on improving the competitiveness of the campesino sector. Gradually reducing the current high levels of protection for products without a comparative advantage, but not eliminating those tariffs as long as intemationalprice distortions exist. Providing new forms of support to producers that reach export crops and import substitutes, and smallholders and more commercial farmers. Linkingincentives to performance andto compliance with environmental conditions. Greater strategic emphasis on realizing Colombia's potential in aquaculture and sustainable forestry, as well as crops and livestock products that have comparative advantage. Greater emphasis on organic cultivation, for both domestic and external markets. Maintaining a competitive real exchange rate. For a number of these topics, the analysis in the foregoing sections has been fairly specific, although not pretending to be definitive. For example, the analysis points inthe directions of the need for a policy framework for participatory research and extension, and relatively more support for those activities, attention to questions of institutional incentives in the reform of the extension system, new forms of agricultural taxation to increase revenue availability and strengthen local governments, new legislation and monitoring capacities in the food safety area, legislation to improve the functioning of marketing systems and agriculturalcontracts, and a number of other specific issues. Public policy i s already moving in some of these overall directions, but a comprehensive outline of concrete agricultural development policy, or an agricultural development strategy, i s not yet available to producers or the public. Developing it, with considerable policy specificity in its proposals and in consultation with the private sector, could help guide decisionmaking throughout the sector and help ensure consistency in the many detailed policy decisions that are made on a day-to-day basis in any government. Part of such a strategy would be a set of measures to promote and ease the transition to a more competitive sector. 7.2. MeetingInternationalStandardsinthe Agricultural Sector Experience throughout the world has demonstrated that the most important factor for achieving competitiveness and sustainable agricultural development i s education, an issue beyond the scope of this study. It should be noted, however, that in Colombia, the quality of education in rural areas i s inferior to that of urban areas. As long as this situation persists, accelerating the rate of agricultural development will be more difficult. In additionto education, however, making the Colombian agricultural sector more competitive requires ensuring it and its institutions meet international standards. This requires attention to the followingthree factors: Quality. In both intemational markets and domestic markets in developing countries, quality considerations for agricultural products are rapidly becoming more prominent. Price differentials have emerged according to product quality, and international markets increasingly are becoming closed to products that do not satisfy basic standards, particularly with respect to food safety and plant and animal hygiene. In Colombia, agroindustry and retail outlets are becoming more demanding in regard to quality. There i s little doubt that improving and sustaining product quality will be one of the keys to agricultural growth in the future, and countries as diverse as Chile, the Dominican Republic, Costa Rica, and Mexico have established international reputations for quality in some major product areas. Achieving better quality i s not only a question of training producers; it also requires institutional support. A viable strategy for agriculturalcompetitiveness requires a substrategy for product quality. 43 0 Confidence. Confidence i s also integralto competitiveness. It is required in all commercial and financial relationships. Smallholders need to demonstrate reliability in fulfilling contracts in order to be able to gain access to more stable, commercial markets. Equally, exporters and middlemen need to make timely payments to producers in order to gain their confidence for future purchases. Confidence is basic to agricultural lending, and in particular it underpins associative forms of lending to smallholders, where social capital substitutes for physical capital. Confidence i s a comerstone to the willingness to pay taxes, and accordingly efforts to construct transparent, local fiscal mechanismsthat permit taxpayers to perceive tangible benefits from their contributions, and to participate in making decisions on them, can repay their initial investment many times over. Confidence is needed among different social strata in rural areas so that land rental markets can function, providing greater access to land for poor families. Inthe area of public services, extension agents need to convince farmers they understand their concems and gain their confidence, inorder to deliver effective assistance. Mutualconfidence and partnerships are also one of the building blocks of participatory research. In terms of public administration, buildingconfidence requires transparencyand effectiveness. Inprivatecommercialrelationships, experience shows the value of developing relationships of confidence, but training programs also can help inthat regard. 0 Consistency. A principal requirement for effective agricultural policy i s that it be consistent- over time, among regions, and over products. Uniformity of the rules of the game, and consistency in their application, are essential for policy to be effective. Abrupt swings in border protection, as has happened for some crops in Colombia, are damaging to incentives for investment and production because of the uncertainty they create. (Price bands, on the contrary, tend to reduce price fluctuations andtherefore reduce uncertainty.) Consistency in exchange rate policy i s perhaps the most basic requirement for successful agricultural development. Equally, in the private sector, the more consistent producers can become in meeting quality requirements and delivery times, the greater their opportunities to participate in more commercial agriculture. In brief, although agricultural strategies are built around topics like research and extension, irrigation, land tenure, and the like, quality, confidence, and consistency are three pervasive requirements for the ways in which both the public and private sector function. In order to promote the sector's competitiveness, these three themes deserve careful examination inthe development and application of all policies, and inthe private sector's development of commercial and financial relationships. 7.3 Recommendationsfor a TransitionProgram Shifts in priorities for fiscal expenditures and trade policy always result in some losers as well as gainers, even though in the longer run the result i s a more competitive sector, and therefore a transition program i s neededto support those shifts. Based on the analyses carried out for this study, the following elements would warrant consideration for possible inclusion in a transition program. They-or other measures judged to be better for the purpose-could be considered to be some of the initial steps in the implementation of longer-term activities and policies: 0 Immediate increases in funding for the agricultural sector, particularly for agricultural research, as a matter of the highest urgency, along with development of mechanisms for establishing regional research agendas and development of a policy framework for supporting participatory research. 0 Reconversion tariffs, through which funding is provided for investments to either improve the productivity of currently noncompetitive lines of production, or to support the transition to other 44 types of production, while an import tariff i s being scaled downward. Under existing fiscal legislation it may be difficult to tie the proceeds from those tariffs directly to expenditures for the subsectors concerned, but approximate equivalences could be developed. An essential aspect of the program would be to require investment plans linked to credible increases in productivity at the farm level, in an existing line or production or a new one, as a condition for approval of the funding. Another aspect would be a plan for progressive tariff reduction that would acquire the status o f an international treaty, so that it would be binding. (As emphasized earlier, this does not mean complete elimination of tariff protection in all cases.) a Strongerfiscal supportfor the sector duringan extended transition period, perhaps provided in part by international entities. Negotiations with the international agencies (including the IMF) over such support should underscore the fundamental role of agricultural development in reducing poverty (and in reducing productionof illicit crops), as well as the nature of institutional andpolicy reforms that would accompany the expenditures. a Initiation on apilot basis of linkedprograms of direct supportper hectareand area-based land taxes. Some funding now directed to other incentives could be reallocated for this purpose. Work would begin immediately in pilot locations in developing local institutions and participatory procedures for allocating some of the land tax receipts to infrastructure and training programs, under central government monitoring and supervision. While most producers would not perceive the economic effects of these programs right away, they could see the direction of changes in incentives that would flow from the programs and begin to adapt their production accordingly. In this sense, initiation of the programs would have an effect during a transition period. a Institution of across-the-boardrequirementsfor demonstratedproductivity increases in order to continue receivingfiscal incentivesdirectedtowardproducts. a Development of an immediate agreement with producers to review and restructure the parafiscal funds whose rates of compliance with quotas have been very low and/or whose expenditureson administration have been high. a Development of an outline of a landfund and consultativeprocesses for formulation of the definitive structure and procedures for such a fund. Formulation of an entrepreneurial approachto land redistribution, in which market accessfor campesinoharvests isfacilitated at thesame time they areprovided with access to land. Again, the effects on producer expectations and behavior would beginto be registered duringa transition period. a Developmentand disseminationof model contractsfor land rental and sharecropping. This i s a measure that could be carried out quickly, in consultation with producer groups in different parts of the country. a Provision of funding for export marketing on an exploratory basis and visits by producers to international tradefairs and marketing associations, under clear, published rules for selection of those who participate insuch activities. a Establishment of thefirst Rural Centersfor Entrepreneurial Development on a pilot basis in selectedareas. a Development of draft legislationto clarifv responsibilitiesfor ensuringfood safety. 45 Immediate increases in the funding of ZCA with a mandate to develop a program for certification of the acceptability of export products according to international norms of food safety and plant and animalhygiene. Taken together, a package of short-term measures such as these would both send an unmistakable signal that priorities and incentives in the sector are changing, and provide some short-term assistanceto groups most affectedby the transition. 46 Annex 1 Tables,Figures,and Box CitedinText Table A.l.l. Distributionof LandArea andEmploymentinCropAgriculture ----41.6% of CroppedArea % of Employment inCropping H 1980 1990 2001 1980 1990 2001 48.8 52.6 27.0 28.3 20.9 Cereals 32.0 36.4 29.1 10.6 13.1 8.9 10.4 10.8 5.1 6.5 6.5 2.3 6.4 I 5.4 I 7.4 I 10.0 I 8.6 I 9.6 11 27.6 26.1 38.0 39.5 39.2 52.1 11Perennials except coffee Exportables 5.5 7.9 11.9 8.4 13.4 17.9 22.0 18.1 26.0 31.1 25.8 34.1 76.4 78.6 79.5 66.5 67.4 72.9 23.6 21.4 20.5 33.5 32.6 27.1 I1 Source: BalcBzar V., OrozcoA., and Henry Samac6 (2003). TableA.1.2. Family andLocalAgroprocessingActivitiesand Smallholder Marketing Channels Tropical fruit (uchuva, bananito,pitahaya, Product selection on farmand Stable arrangements with exporters. granadilla) locally. Sales to middlemen. Relations of confidence with large- 47 Table A.1.3. NominalEconomic ProtectionProvidedto 11PrincipalCrops, 1991-2001" (millions of dollars) Value of Import Value Value of Protection as % of Protection as % of _Year Production Before Tariffs Protection ProductionValue Import Value 1991 2,522 2,35 1 171 7 7 1992 2,645 2,205 440 17 20 1993 2,678 2,085 594 22 28 1994 3,216 2,249 967 30 43 1995 3,377 2,707 670 20 25 1996 3,611 2,854 757 21 27 1997 3,646 2,521 1,125 31 45 1998 3,869 2,35 1 1,517 39 65 1999 3,352 2,108 1,244 37 59 2000 3,325 2,029 1,296 39 64 * Cotton,rice, sugar, cacao, barley, milk,corn, palm oil, sorghum, soybeans, and wheat. Source; Balcsizar, Orozco, and Samach(2003). 48 Industry Competitiveness Export Technical Agreement Competitiveness Coordinator Sources: RedColombiaCompite; andEspinaland SamacL(2003). 49 Table A.1.5. Colombian Agricultural and Agroindustrial Exports /b ear Value (mt> price Value (mt) Price Value (mt) Price 1 2,725,273 2,931,527 930 1,336,506 734,048 1,821 1,388,767 2,197,479 632 1992 2,735,496 3,411,010 802 1,261,280 968,243 1,303 1,474,2 16 2,442,767 604 1993 2,568,930 3,570,342 720 1,152,339 788,036 1,462 1,41639 1 2,782,306 509 1994 3,738,380 3,738,679 1.OOO 1,998,664 674,969 2,96 1 1,739,7 16 3,063,7 10 568 1995 3,696,077 3,277,315 1.128 1,840,985 559,630 3,290 1,855,092 2,717,685 683 1996 3,431,972 3,346,211 1.026 1,581,294 601,024 2,63 1 1,850,678 2,745,187 674 1997 4,288,902 3,672,325 1.168 2,264,670 618,216 3,663 2,024,232 3,054,109 663 1998 4,044,727 4,053,068 998 1,895,530 637,197 2,975 2,149,197 3,415,871 629 1999 3,409,659 4,347,583 784 1,348,537 568,697 2,37 1 2,061,122 3,778,886 545 2000 3,151,651 4,356,743 723 1,069,823 508,619 2,103 2,081,828 3,848,124 541 2001 2,9 15,900 4,021,635 725 769,386 560,248 1,373 2,146,514 3,461,387 620 2,947,726 4,448,738 663 782,180 579,083 1,351 2,165,546 3,869,655 560 nnual growth rates (%): -2.1 -4.3 -3.9 -0.5 I 4.3 4.1 -0.4 -3.3 1.o -4.4 II -15.7 Note: Agricultural exports include agroindustrial exports. Source: DANE and calculations of the ObsewatorioAgrocadenas. 50 Table A.1.6. ExportDiversification (productlines with more than US$5 millioninexports, values inUS$ millions) Jhapter Code ProductLine 1991 1992 2001 2002 01 01029090 Other live cattle 28.622 20.562 5.230 n.a. 01 01051100 Live chickens * * 10.698 5.144 02 02011000 Fresh, refriger. beef carcasses 34.010 11.200 7.416 * ~~ 03 0303213 Frozen trout & like fish 49.693 47.246 ** 03 03034200 Frozen yellowtail tuna * * 41.902 39.353 03 03054000 Smoked fish incl. fillets 11.552 27.332 * * 03 03061100 Lobsters 9.207 8.584 8.623 10.622 03 03061310 Frozen crayfish 17.478 8.223 10.696 6.887 03 03061390 Shrimp 58.692 57.149 69.434 61.182 04 04022111 Granulated andpowdered milk 11.248 7.740 04 04022119 Other milk and cream 37.284 27.586 04 04029110 Evaporatedmilk 6.849 5.132 04 04070010 Poultry eggs **** **** 9.809 4.134 06 06031 Cut flowers and buds 279.611 339.641 06 06031010 Freshcarnations 184.521 189.691 06 06031020 Freshchrysanthemums 9.918 12.266 06 06031040 Freshroses 182.788 193.351 06 06031090 Other flowers and buds 232.263 268.779 07 07019000 Potatoes,fresh or refrigerated 5.459 12.062 8.603 6.357 07 07133090 Other beans 9.733 6.670 6.846 n.a. 07 07149000 Other root crops * * 11.963 n.a. 08 0803000 Bananasand plantains 404.874 407.325 08 08030011 Plantains 40.356 36.770 08 08030012 Cavendish bananas 364.867 401.090 08 08101000 Fresh strawberries 6.862 6.081 * 08 0810900 Other fruit 4.920 5.540 ** * 08 08109050 Freshuchuvas * * 8.713 8.37 1 09 09011100 Unroastednon-decaf coffee 1,336.429 1,258.850 764.168 772.199 09 09011200 Unroasteddecaf coffee * * 4.406 9.127 10 10063000 White or semi-white rice 24.264 .014 * * 11 11072000 Malt of barley, other cereals 15 15111000 Crude palmoil 15 15119000 Other palmoils & derivatives 15 15132110 Crude palmkernel oil 15 15162000 Other vegetable oils & cake Edible mixes & preparations of 15 15179000 vegetable andanimal oils & fats 16 16041400 Preservedfish 17 17011190 Other raw sugars 17 17019900 Other sugars 26.652 87.089 66.201 102.070 17 17023020 Glucose syrup 17 17031000 Cane molasses 17 17041010 Sugar-coatedchewing gums * * 18.437 16.768 17 17041090 Other chewing gums 17 17049010 Candies andcaramels 51 18 18010010 Raw cacao 6.920 4.379 * * 18 18040000 Cacao oil and cake 9.967 5.831 5.519 5.272 18 18061000 Powdered cacao, sugar added * 4.956 6.095 18 18063200 Chocolate inblocks 8.735 9.389 18 18069000 Other chocolate & cacao prep. 4.926 5.446 19 19011010 Baby food andprepared milk 6.706 4.445 19 19053000/1 Cookies **** ***** 23.268 18.215 19 19059000 Other bakery products 6.368 7.308 10.274 11.185 20 20089990 Other fruits and fruit mixtures * * 6.940 11.302 20 20098019 Other fruitjuices 30.970 31.015 * * 21 21011000 Coffee extracts, concentrates 62.142 63.217 89.253 81.598 21 21041010 Soup andbroth preparations * * 4.310 7.213 Powderedingredients for 21 21069101 puddings, ice creams, gelatins 5.904 5.932 21 21069090/9 Other food preparations ** ** 10.545 15.366 Other nonalcoholic beverages 22 22029000 excluding fruitjuices 10.010 14.452 22 22084000 Rumand other cane liqueurs ** ** 6.67 1 4.949 Soybeancake andother solid 23 23040000 soy residues * * 11.242 2.557 24 24011010 Leaftobacco (dark) 17.719 23.214 5.321 5.785 24 24012020 Partly processedleaf tobacco 5.481 6.680 * * (light) 24 24022020 Cigarettes (from light tobacco) 10.874 25.332 21.695 40.176 35 35030010 Gelatins and their derivatives 13.638 14.859 15.714 15.969 41 41032000 Raw leather and skins of reptiles 12.824 10.412 41 41042200 Pre-cured cattle hides ** 28.636 n.a. 41 41043100 Driedhorse and cattle hides * *** 10.446 n.a. 41 41043900 Other dried horse and cattle 5.429 5.469 18.625 n.a. hides 41 41072900 Other reptile skins * * 8.218 n.a. 41 4109000 Polished leather and skins 14.630 17.636 9.745 n.a. 44 44101900 Other wood boards * * 8.325 n.a. 52 52010000 Uncarded, uncombed cotton 69.846 56.380 * * 52 52093200 Dyedwoven serge or twill cloth * * 8.613 7.534 Woven denimcloth with colored 52 52094200 yarns 6.615 13.655 n.a. 52 56060000 Braided yarns ** * 19.942 20.810 *Includes canned tuna. Indicates exports of lessthan US$5 million. Notes: Chapter and code refer to the customs codes. Source: Adapted from Espinal and SamacA (2003). 52 Table A.1.7. Destination of Agricultural and Agroindustrial Exports (except coffee; values inmillions of US$) Annual % GrowthRate, 1991-2002 2 14 4 13 11.670 63.217 53.155 88.767 98.561 16 3.263 8.225 13.575 22.101 22.343 20 46.937 40.122 32.288 48.361 45.795 -2 141.668 185.995 255.214 398.774 239.453 9 6.395 2.260 2.260 16 8.603 18.673 34.857 21.442 27.614 10 523.951 704.167 811.630 813.369 897.002 4 Central America & 54.222 70.326 109.871 119.731 152.630 10 410.869 545.375 487.966 413.005 435.426 1 3 4 Source: DANEand calculations of the Obsewatorio Agrocadenas. Table A.1.8. Effects on Employment and Income of a Shift inIncentives toward Products with Comparative Advantage (shiftof 10%inrelativeprices toward crops with comparative advantage) Production C 37,507.1 174,121.6 *Inmillions of 1994pesos. Source: BalcBzar, Orozco, Samaci (2003). 53 Table A.1.9. Effects onthe LivestockSector of Variations inTariffs on Corn and Soybeans (simulated effects of a 10percent decrease inthose tariffs) Source: Adapted from Balcfizar, Orozco, and Samacfi(2003). Table A.l.lO. Committedand DisbursedAggregateMeasuresof Supportto Agriculture (totallevels, 1995-2001, inthousand US$) Year GMSNotification GMS Disbursement Utilization Rate 1995 392,465 58,109 15% 1996 387,165 3,961 1% 1997 381,858 14,393 4% 1998 376,555 9,77 1 3% 1999 371,25 1 6,805 2% 2000 365,947 n.a. n.a 2001 360,644 n.a. n.a. Note: Disbursementfigures were not yet provided to the WTO for 2000 and 2001. Source: Balchzar, Orozco, and Samaci (2003), on the basis of WTO (2002). 54 Table A.1.12. Value of ICR Subsidies Paid by Type of Investment (millionsof pesosat 1998constant prices) Investment Area 1994 1995 1996 1997 1998 1999 2000 2001 2002 Irrigation 210 9,938 7,776 11,798 6,726 6,884 2,997 4,376 4,239 Machinery & 0 205 2,846 4,636 2,940 4,212 1,614 4,457 6,201 equip. Infrastructure 0 11 0 156 77 171 471 895 1,023 Planting 0 0 0 0 0 0 3,251 8,000 9,448 perennials Transport infrastr. &equipment 14 3,142 2,775 5,637 3,627 8,552 3,453 5,515 5,949 Technological modernization 0 0 70 264 25 27 47 0 0 Purebred cattle 0 0 0 0 0 0 0 0 16 Total 223 13,297 13,467 22,490 13,396 19,847 11,833 23,242 26,875 Note: (1) Tota P may not add up because of rounding errors. (2) The outlays on purebred cattle began to increase sharply inearly 2003. Source: Balcfizar, Orozco, and Samacfi (2003). Table A.1.13. Impactof 10PercentExchangeRate Changeon Some RegionalAgricultural Products 56 TableA.1.14. Distributionof RuralLandinColombia, 1998 Farms LandArea 1998Appraisal Value FarmSize Group (ha.) Number % Hectares % MillionPesos % Less than 1 754,822 35.4 272,082 0.5 471,580 6.5 1to lessthan 3 492,505 23.1 848,210 1.6 587,759 8.1 3 to less than 5 210,731 9.9 794,549 1.5 402,806 5.6 5 to less than 10 222,668 10.4 1,549,985 3.O 621,025 8.6 10to less than 20 165,581 7.8 2,305,440 4.4 718,393 9.9 20 to less than 50 156,916 7.4 4,935,566 9.5 1,059,923 14.7 50 to less than 100 70,804 3.3 4,834,056 9.3 842,353 11.6 100to lessthan 200 34,463 1.6 4,672,245 9.0 742,868 10.3 200 to less than 500 16,987 0.8 5,006,676 9.6 757,590 10.5 500 to less than 1,000 4,298 0.2 2,983,854 5.7 334,003 4.6 1,000 to lessthan2,000 1,575 0.1 2,114,675 4.1 158,355 2.2 Greater than 2,000 1.458 0.1 21.793.028 a 41.8 536.324 7.4 I , Total 2,132,808 100.0 7,232,979 100.0 Source: lnstituto Geogrdfi > Augustin Codaui, cadastral records for 1998. 52,110,366 Table A.1.15. AgriculturalLending, 1998-2001, DisbursementsinMillionUS$ (% accounted for by FINAGRO rediscountsinparentheses) Privatefinancial intermediaries* * 472.77 (78.8) 457.88 (75.1) 387.65 (68.6) 367.15 (68.7) *CajaAgraria, Banco Agrario, and Bancafe'. *** Bancoldex ** Banks and finance companies. and the Instituto de Fomento Industrial. Source: Compiled by Carlos Cuevas and Lisa Taber for the World Bank. Financial activities of the National Agricultural Commodity Exchange are not included. * Definedas a family (husband and wife) with total net worth less than US$15,250. Source: Contraloria General de la Repdblica (2002:92). 57 Table A.1.17 Participation inAgricultural Production of Smallholdersand Commercial Producers: Crop Classification by Area (average 1999-2000) Note: The estimates in the table derive from classification of crops as primarily of smallholders or primarily of commercial farmers, according to the list below. Primarily smallholder crops: coffee, traditional corn, plantains, panela cane, cassava, potatoes, fruit, beans, vegetables, cacao, rainfed under nonmechanized technology, yams, jute, wheat, coconuts, arracacha, light tobacco, sesame, dark tobacco, and barley. Crops primarily of commercial farmers: irrigated rice, coffee, sugarcane, mechanized rainfed rice, corn under modern technologies, oil palm, sorghum, cotton, export bananas, and soybeans. Source: Forero (2003), calculated from statistics of the Ministry of Agriculture and FEDECAFE. For coca and amapola: Areas Policia Nacional, Tavera (2000). 58 Table A.1.18. Evolutionof the Productivity Gap between Smallholders and Commercial Farmers (Ratio of yields of the latter to yields of the former) I/ PRODUCT 1973-1976 1988 Sesame 0.97 1.04 Cotton 3.58 1.03 Rice 2.60 1.40 Export bananas n.a. 1.27 Cacao 2.19 1.35 Panela cane 2.22 1.70 Barley 1.59 1.07 Beans 2.01 1.29 Corn 2.40 1.61 Potatoes 6.08 1.07 Plantain 1.60 0.78 Sorghum 0.97 1.34 Soybeans n.a. 1.30 Tobacco 1.54 1.10 Wheat 2.42 1.17 (/Cassava 1.46 1.25 Note: The criteria for distinguishing smallholders from other producers were not necessarily consistent, but apparently were based on own cultivation of the plot, use of family labor only, and in some cases possessionof not more than 20 hectares inthe Andean zones and 50 hectares inthe Atlantic and Orinoquian zones. Source: Berry and Bejarano (1990:135). 59 Figure A.l.l International Prices, Tariffs, and Exchange Rate Effects for 10Crops, 1991-2001 WHITE RICE WHEAT 400, , 140.00 I 140.00 350 12000 200 12000 E 300 10000 10000 - 3 ' 250 8000 150 200 5s 6000 5 , 150 6000 100 6000 3 100 40 00 40 00 50 20 00 50 20 00 n -nnn 19 19 19 19 19 19 19 19 19 20 20 20 -I 0 0 00 19 19 19 19 19 19 19 19 19 20 20 20 91 92 93 94 95 96 97 98 99 00 01 02 1 91 92 93 94 95 96 97 98 99 00 01 02 I Tariff e rice FOB Bangkok 10% Part. ATCR (6) ~ YELLOW CORN - WHITE CORN 180t I 140.00 2501 I 140.00 160 - 1;140 120 100 80 60 40 20 0 19 19 19 19 19 19 19 19 19 20 20 20 19 19 I 9 19 19 19 19 19 19 20 20 20 91 92 93 94 95 96 97 96 99 00 01 02 1 1 91 92 93 94 95 96 97 98 99 00 01 02 60 BARLEY I SOYA 250, , 140.00 1 350, 140.00 12000 c 250 10000 8000 5 s 3 2oo e 100 150 6000 x 100 40 00 50 50 20 00 0 0 19 19 19 19 19 19 19 19 20 20 20 0 00 19 19 19 19 19 19 19 20 20 20 91 92 93 94 95 96 97 98 99 00 01 02 91 92 93 94 95 96 97 98 99 00 01 02 White corn FOB KC Boardof ITCR (6) *Tariff Trade lTCR I -Tariff Soybeans FOB Gulf #2 - RAW SUGAR 1 WHITE SUGAR 4501 I 140.00 12000 10000 8000 5 6000 X B 40 00 20 00 0 00 19 19 19 19 19 19 19 19 19 20 20 20 19 19 19 19 19 19 19 19 19 20 20 20 91 92 93 94 95 96 97 98 99 00 01 02 91 92 93 94 95 96 97 98 99 00 01 02 Raw sugar FOB N.Y. Contract White sugar FOB ITCR (6) a aTariff No11 L T C R (6) maTariff London CRUDE PALM OIL CRUDE SOYBEAN OIL 8001 I_ 140.00 700, 140.00 700 12000 12000 600 35 10000 10000 500 8000 8000 5 400 B a 300 6000 6000 X 200 40 00 40 00 100 20 00 20 00 O ' 19 '19 ' 19' 19 ' 19 ' 19 ' 19 19 ' 19 ' 20 ' 20 ' 20 0 00 0 00 ~ 19 19 19 19 19 19 19 19 19 20 20 20 91 92 93 94 95 96 97 98 99 00 01 02 91 92 93 94 95 96 97 98 99 00 01 02 ITCR (6) Crude soybean oil FOB ITCR Tariff Crude palm oil CIF Rotterdam l_l a m ) Tariff Argentina "- (6) 61 Figure A.1.2. International Prices, Tariffs, andExchange Rate Effects for 4 Livestock Products, 1991-2001 CHICKEN PARTS BEEF 1,000, , 140.00 2,800 140.00 2,600 12000 2,400 120.00 2,200 10000 2,000 100.00 1.800 8000 5 80.00 L 0 6000 6 1,200 60.00 1.000 X 40 00 800 40 00 600 20 00 400 20.00 200 0 00 0 0.00 91 92 93 94 95 96 97 98 99 00 01 02 91 94 95 96 97 98 99 00 01 02 Tariff -Tariff Chicken parts FOB ITCR (6) Frozendeboned US beef importsfrom Australia, N. Z. Gulf -ITCR(6) I PORK POWDERED MILK 2.000> , 140.00 2,500 18" 12000 1600 2,000 12000 1400 10000 c 10000 8000 L f 80 00 5 1,500 800 6000 1.000 6000 600 40 00 40 00 400 500 200 20 00 20 00 0 0 00 0 0 00 91 92 93 94 95 96 97 98 99 00 01 02 I1 92 93 94 95 96 97 98 99 00 01 02 Powderedmilk FOB ITCR (6) -Tariff PorkFOB Omaha -ITCR(6) -Tariff N. Zealand _I 62 Figure A.1.3. Illustration of a LandTax and Direct Payment System Together (vertical scale inUS$) 1800 1600 1400 1200 1000 800 600 400 200 0 0 UAF 2 UAF 4 UAF 6 UAF 8 UAF 10UAF 63 Box A.l.l: ProductChains As of January 2003, 21 product chains had been formed and Competitiveness Agreements (Acuerdos de Competitividud) have been reached for most of them.* These agreements are made among representatives of all the principal participants in the chains and the government, and they cover measures to improve productivity and competitiveness. Among other things, these agreements coordinate the provision of 42 different kinds of incentives for increasedproductivity, investment, production, and export. FINAGROhas supported the chains with associativecredits, which may be allocated to groups of producers of any scale that have forward contracts or other specified guarantees. It also provides a subsidy of up to 30 percent of the cost of investment projects carried out by producers, in a programknown as the Incentive for Rural Capitalization (ICR). Disbursements of ICR funds amounted to 64 billion Colombian'pesos between 1998and 2001. *Cereals for human consumption; poultry and pork; potatoes; dairy products; tuna; poultry; rice; oilseeds, oils and fats; cultivated shrimp; cotton for the textile industry; forest products for wood; forest products for paper; plantain; cacao; citrus; rubber; tobacco; artisan sugar; export fruit; fruit for processing; and fish cultivation. Note: The descriptivematerialon the productchains i s basedonEspinalG. and Samach(2003), astudy preparedfor FA0 and the World Bank for the uresenturoiect. 64 Annex 2 ComparativeAdvantage Analysis of SelectedColombianAgriculturalProducts 1. Grainsand Short-Cycle Oilseeds Nominal protection rates increased between 1991-93 and 1999-2001 for five of the six principal grains including the main short-cycle oilseed. Protection levels are currently high and, with the exception of barley, they represent one-third to one-half of producer income in those crops (Table A.2.1). The economic protection afforded to grains is stronger than indicatedby the nominal protection rates because of the implementation of the "absorption agreements" that oblige agroindustry to purchase the domestic crop in exchange for being allowed to import grains. Barley is the one exception, but it is by far the least important grain produced inColombia, with only slightly more than 5,000 hectaresplantedin it in 2001. TableA.2.1. NominalProtectionRatesfor Grainsand Soybeans Barlev 49 12 33 11 White rice II 8 II 64 II 8 II 39 IIII ~~ ~ ~ Corn 45 93 31 48 Sorghum 27 47 22 32 Soybeans 32 62 24 38 The trend in Colombia's international trade in grains has been toward an increase in imports as a share of total supply, in spite of the high levels of protection. This strongly suggests that domestic grain production does not possess a comparative advantage. Table A.2.2 shows that the import share of total supply was significantly higher during 1996-2001 than during 1992-1995. It might have increased further but for the effect of the absorption agreements. There always are exceptions to general statements, and in this case Table A.2.2 shows that Colombia has been able to export small quantities of a few high-valued grains, inthis case, mostly pearlbarley. It illustrates that successful development of niche products can run counter to general trends. To place Colombian tariff protection levels in international perspective, world agricultural tariffs averaged 62 percent in 2000 and for grains the average was 69 percent (Gibson and others 2001). Average tariff rates differ substantially by region of the world, as shown by the data for grains in Table A.2.3, and they are often high.s4 Still, of 13 world regions only four (South Asia, sub-Saharan Africa, Caribbean Islands, and non-EU Western Europe) have grain protection generally as high as Colombia's if barley i s excluded from the latter. Thus Colombia's protection rates for this subsector are relatively high by world standards. The more important issue i s whether the prevailing tariffs on grains in Colombia represent the policy most likely to promote growth of income and employment in agriculture. 54Another perspective on agricultural protection is provided by the fact that in 2002 farmers in OECD countries received on average prices 31 percent above world prices (OECD2003:4). 65 TableA.2.2. ColombianInternationalTradeinGrains InValue (000US$) InVolume (MT) Production Imports/ Year Exports Imports Exports Imports inMT TotalSupply 1992 511 241,571 903 1,494,566 3,769,666 .28 1993 1,504 264,494 4,270 1,718,045 3,635,186 .32 1994 1,353 390,95 1 3,298 2,375,819 3,739,897 .39 1995 567 450,627 746 2,507,653 3,572,190 .41 1996 169 694,6 10 186 3,192,3 12 3,235,407 S O 1997 416 588,045 336 3,203,850 3,297,390 .49 1998 1,229 613,535 1,025 3,794,907 2,965,180 .56 1999 471 405,850 268 3,091,857 3,435,810 .47 2000 476 418,335 438 3,256,152 3,777,544 .46 2001 311 - 466.149 363 3,428,328 > 3,854,816 .47 Source: Statistical Annex tables. - TableA.2.3. MeanandMedianTariffsfor GrainsbyRegionof the World, 2002 - Region Mean P Median Region Mean Median Southern Africa 37 33 Non-EUWestern Europe 100 50 Asia-Pacific 60 25 European Union - 15 53 63 South Asia 103 100 South America 46 35 Sub-SaharanAfrica 75 80 Central America 55 45 North Africa 84 34 Caribbean Islands 86 100 Middle East 40 - 15 NorthAmerica 25 3 Eastern E u r o p e 47 20 Source: Gibson and others (2001). Calculation of the revealed comparative advantage (RCA) suggests a future potential in some processedforms of grains and short-cycle oilseeds, though not in all. As Table A.2.4 illustrates, by far the largest share of these processed products show an increasing export relative importance in Colombia's basket of export goods in recent years. Many of these exports go to the Andean regional market. Agroindustrial exports have until recently benefited from a duty drawback (CERT) of 2.5 percent, but nevertheless, the trend in these exports represents another indication that Colombia's comparative advantage in the area of grains and short-cycle edible oils lies in processed products rather than primary products. If that i s so, then an export-oriented strategy would lean toward lower tariffs on the imports of raw grains andoils, inorder to make the processedproducts more competitive. 66 Table A.2.4. RevealedComparative Advantagefor ProcessedProductsof Grains, Short-CycleOilseeds(export values inmillionUS$) - Export RevealedComparative Advantage F A 0 Tariff Value, 2001 Product Chain - 1997 1998 1999 2000 2001 0.132 Cereals for human 0.64 2.84 10.19 14.89 12.94 consumption Bread, cookies, 22 1905 33.548 Cereals for human 2.02 2.39 2.04 2.06 2.30 wafers consumption Corn flour 58 1102 0.891 Cereals for human 0.64 0.66 0.87 2.75 3.67 consumption Baby food 109 1901 6.845 Cereals for human 1.57 1.15 0.57 1.47 2.38 I consumption Soybeancake II 238 2304 11.242 Cereals, poultry, pork 0.02 0.02 0.26 0.62 0.70 Foodfor domestic 843 2309 3.491 Cereals, poultry, pork 0.05 0.05 0.10 0.13 0.19 animals Breakfast cereals 41 1904 5.448 Cerealsfor human 0.36 0.33 0.53 0.67 1.54 consumption Malt extract, other food preparations 115 1901 3.892 Cerealsfor human 0.51 0.56 0.53 0.54 1.04 consumption Ground oats 76 1104 0.471 Cereals for human 0.09 0.04 0.77 0.33 1.91 consumption Dough for bakeries, pastries 114 1901 1.102 Cereals for human 0.03 0.04 0.01 0.07 0.58 consumption Soybeanoil 237 1507 3.069 Oilseeds, fats and oils 0.01 0.03 0.07 0.67 0.53 Nuts, sesame & 289 1207 0.176 Oilseeds, fats and oils 0.00 0.00 0.50 0.09 0.04 palm seed Soya 236 1201 -- Cereals, poultry, pork 0.00 0.01 -- 0.00 -- 0.904 Cereals for human 0.09 1.15 0.85 0.28 0.52 consumption 0.633 Oilseeds, fats and oils 1.02 0.60 0.19 0.12 0.61 Note: (1) The first row section of the ta:includesproductswithanRCA>1 ) (on average) and generally increasing; the second section, products with an RCA < 1.0 and generally increasing; and the third, those with an RCA generally decreasing. (2) RCAs for some other categories of edible oils were not included here because they may include oils from perennialcrops. (3) The tariff codes are those applied inthe Andean Common Market. Sources: Adapted from Espinal and Samac6 (2003), and official trade statistics. The domestic resource cost coefficients (RC) calculations for grains and soybeans are presented in Table A.2.5, and they confirm that Colombia does not have a comparative advantage inthis subsector. The only crop/department in which a marginal comparative advantage i s indicated i s for sorghum in Huila. There appears to be a possibility that sorghum could become competitive in Cundinamarca with reasonable increasesin productivity and decreases in costs, but the general conclusion i s that it will be quite a struggle for grains to acquire underlyingcompetitiveness inColombia. Inthe case of yellow corn, for example, each dollar saved by producing instead of importing implies an opportunity cost of domestic resources, which ranges from 1.3 dollars in Cesar to 2.4 dollars in Quindio. Zones that possess relatively high levels of 67 productivity in corn, such as the Valley of the Cauca, Cbrdoba, and Tolima, do not escape the general comparative disadvantage of Colombia incorn and other grains. Production of feedgrains in Colombia evidently is sustainedby the hightariff levels. Inturn, those tariffs reduce the competitiveness of poultry and pork, as well as processed grain products. This topic i s reviewedinrelation to poultry and pork ina subsequent section. White corn has an even more marked comparative disadvantagethan yellow corn. Its DRC ranges from 1.4 to almost 5.0. This situation reflects the relatively low productivity of labor in white corn. It i s a crop relatively intensive in the use of labor, but the opportunity cost of labor i s high in relation to its low productivity in white corn and the low world prices of that crop. The calculations reported here are made for corn as a grain. White corn sold and consumed as a vegetable (on the cob) may be an exception to this study's conclusions, because its higher international transport costs, relative to value, would confer a greater advantage to domestic producers. That advantage would be strongest in rural areas. Under current conditions, the weakest of Colombian grains ininternational competitiveness i s rice. None of its DRCs fall below 2.0, and some of them are extremely high. Table A.2.5. ComparativeAdvantage Indicatorsfor Grains and Soybeans,2001 Santander 2.31 Tolima 1.59 2.12 1.75 3.12 C6rdoba 2.26 1.38 2.18 2.01 1.75 Magdalena 1.80 2.19 1.80 Cesar 1.30 1.80 1.81 2.40 Meta 1.37 2.59 1.90 7.97 5.63 1.79 Nariiio 3.34 2.88 Sucre 1.90 9.34 Bolivar 2.23 2.30 1.26 2.01 2.52 Boyacfi 4.98 2.07 8.98 Antioquia 2.46 Huila 1.82 1.78 0.96 3.41 Norte de Santander 4.61 2.24 2.11 Quindio 2.41 *Traditional technology of production. Notes: The national average for each crop is weighted by acreage in corresponding departments in that crop. For white corn, the production technology is the traditional, least-mechanized one. Source: BalcBzar, Orozco, and SamacB (2003). To what extent is the lack of comparative advantage in grains attributable to the fact that these products are subsidized in major exporting countries such as the United States and the European Union? 68 The extent of those subsidies i s well documented and i s a major, recurring topic on the agenda of international trade negotiations. The U.S. Department of Agriculture, among other agencies, has made estimates of how world market prices would be affected by a removal of those subsidies and other impediments to international trade in agriculture. It has been estimated that, as of 2000, a complete liberalization of international trade and elimination of agricultural subsidies would result in the following percentageprice increases on world markets for agricultural products: wheat, 18.1; rice, 10.1; other grains, 15.2; fats and oils, 11.2; sugar, 16.4; and meat and milk, 22.3 (Burfisher 2001).55 Itis validto inquire why a developing country should transmit the distortingeffects of international subsidies to its own agriculture. The question acquires more force in light of the pervasiveness of rural poverty in most developing countries, and also the quasi-irreversibility of labor flows between agriculture and the rest of the economy. If families are driven out of agriculture inpart because of low prices on world commodity markets caused by policy decisions in richer countries, then if those decisions are changed someday it would be virtually impossible to induce those families to return to rural areas and participate in reviving agriculture. Thus a long-term structural shift inlower-income economies would be takingplace on the basis of policy decisions madeel~ewhere.~~ To analyze this issue in the Colombian context, the DRC coefficients were recalculated with internationalprices 20 percent higher for grains and soybeans. Such an increase could arise from changes in world market conditions or from changes in subsidy policies in OECD countries. The results are shown in Table A.2.6. It can be seen that on the whole Colombian grain production still would not have a comparative advantage even under an optimistic scenarioabout liberalization of world agricultural policy. Under this scenario, sorghum would strengthenits comparative advantage inHuilaandwould move into the competitive category in Bolivar, and yellow corn would attain a marginal comparative advantagein Meta. Nevertheless, these products in these regions would be vulnerable to fluctuations in the exchange rate, and they would not have a comparative advantage in most regions of the country. Colombia's comparative disadvantage in rice, wheat, barley, and soybeans would not be reversed by a 20 percent increase ininternational prices. The conclusions of these analyses are clear: in general Colombia's strengths in agriculture do not extend to grains and short-cycle oilseeds-with the possible exception of corn in Meta-but that the country may have potential for increasing its exports of processed grain products. For those producers whose productivitykost relationship i s better than the average, the best hope lies in continuing to increase productivity. Inthe case of corn as a feedgrain, sorghum, and soybeans, the highplateau of the Eastern part of the country, in the Orinoquia (ultillunuru del Oriente), offers intriguing possibilities for obtaining high yields. Meta already i s the most competitive area in Colombia in soybeans, as well as one of the two most competitive in yellow corn. However, cultivation of the ultillunuru would have to take place on a large scale, in plots of thousands of hectares, in order to take advantage of the economies of scale offered by mechanization, and large initial investments would have to be made inimproving the soil quality in light of the relatively high quantities of aluminum content of the soils of that region and their high acidity. In addition, a substantial, long-term commitment would have to be made for agricultural research on those crops under those conditions, and the feasibility of such a project might not be known until 10 years or more of experience i s accumulated. Misi6n Paz (2001:179) is optimistic about the potential of the Orinoquia for corn, soybeans, and cassava, and states that recently hybrid varieties have become available that are resistant to aluminum and acidity in the soil. However, private sector field trials in the region have produced disappointing results so far. Thus, an important potential may exist in that region, but the economics of it are as yet uncertain. 55 Estimates of this nature were first made by the OECDin 1993 and have been made by that agency at intervals since then, and by others as well. '`TheThis Burfisher estimates are cited here becausethey are among the most recent. point is discussedinNorton (2003, forthcoming). 69 Table A.2.6. ComparativeAdvantage Indicatorsfor Grainsand Soybeans, 2001, with a 20 Percent Increase inInternationalPrices Santander 1.92 Tolima 1.33 1.27 1.27 1.53 C6rdoba 1.51 1.13 1.33 1.48 1.24 Magdalena 1.30 1.83 1.32 Cesar 1.oo 1S O 1.25 1.56 Meta 0.96 1.82 1.20 2.79 2.55 1.28 Nariiio 2.65 1.88 Sucre 1.41 3.37 Bolivar 1.59 1.87 0.91 1.48 1.75 Boyach 3.82 1.52 3.92 Antioquia 1.55 Huila 1.14 1.37 0.69 2.01 Norte de Santander 3.07 1.55 1.72 -Cundinamarca Quindio 1.84 2.24 0.84 1.19 1.11 1.32 --__t ~ - Minimumvalue Maximumvalue 2.4 1 4.98 national average 2.18 9.34 - - 5.63 - - 3.84 2.88 8.98 _c 1.61 P 5.33 2.49 2.92 2.13 5.33 P *Traditional technology of production. Notes: The national average for each crop is weighted by acreage in corresponding departments in that crop. For white corn, the production technology i s the traditional, least-mechanized one. Source: Balctizar, Orozco, and Samacti(2003). Inthe caseofcorn, itmay bearguedthat itsimportance liesinthe fact that itis a subsistencecrop for many rural families. About one-fourth of Colombia's corn i s grown under mechanized conditions with high doses of fertilizers and pesticides. The other three-fourths of the crop is planted mainly on hillsides with traditional techniques of cultivation, using mainly family labor with very little application of agrochemicals. However, between 1991 and 2001 the area planted in corn under traditional cultivation techniques fell by about one-third. It i s unclear how much of this decline represented a move to more modem cultivation techniques and how much was attributable to families abandoning the land because of the violence. The area of corn under more modern technologies did increaserapidly duringthis period, and average yields increased from about 2.8 tons per hectare to about 3.7 tons per hectare, but still they fall short of competitive levels inrelation to costs. In regard to corn's role as a subsistence crop, to the extent that it is used purely for home consumption on small farms, that role would not be affected by a reduction intariff protection. Inaddition, inColombia, the range of crops used for home consumption on small farms is wide and includes plantain, cassava, beans, and some fruits and vegetables. Along with sugar and milk, rice i s one of the three products with the largest monetary value of tariff protection in Colombia (Statistical Annex tables). Inaddition, it benefits from absorption agreements and 70 receives substantial fiscal subsidies including large subsidies for storage costs. Therefore its comparative disadvantage i s even greater than indicatedinTables A.2.5 and A.2.6. It may be argued that the position of rice could be strengthened through productivity improvements, but they would have to be very large. Also, it should be borne in mind that productivity continues to improve in other rice-producing countries, so it would be very difficult for Colombia to enhance its position vis-84s other countries in that crop. During 1991-2001, rice yields grew by only 2.0 percent per year.57 That i s a decent rate of increase for many crops, but not nearly enough to overcome rice's comparative disadvantage (on average) in the foreseeable future. For the more efficient producers who wish to cultivate rice in spite of its strong comparative disadvantage, the most important avenues to pursue would include widening the range of varieties available, improving the efficiency of irrigation, improving phytosanitary controls and reducing the use of agrochemicals, and increasing the use of certified seeds. For many crops, rates of irrigation water application often are far in excess of what i s needed and contribute to degradation of soil quality in addition to higher costs of production. It has been estimated that for rice irrigation, applications could be reduced by as much as 30 percent (CONSULTPLAN 1998). The best zones for irrigated rice are currently more competitive than those producing rainfed rice, and a rationalization of irrigation practices would increase that advantage. However, rainfed yields could be improved in C6rdoba through, for example, use of certified seed and the construction of small dams to capture surface runoff from rain. Inregardto use of agrochemicals, pesticides represent35 percent of the total cost of production of rainfed rice and 27 percent of the total cost of production of irrigated rice, for Colombia as a whole (Sicard 2003). These percentageshave almost doubled since 1981, which i s an indicationthat the pesticide strategy in rice is unsustainable. The high humidity in Meta means that zone is an especially heavy user of pesticides for rice. Hence, if rice in Colombia i s to become internationally competitive, fundamental changes will have to be made in the way it i s produced. One advantage of this subsector i s that the milling anddrying technology is relatively advancedcompared with that of rice producers such as Brazil, Uruguay, Venezuela, and the UnitedStates. Under some circumstances, when a country has a comparative advantage in the absence of international subsidies, their presence can be an argument for not eliminating all tariff protection on those products. However, from the viewpoint of long-term national development policy inColombia, the wisdom of continuing to provide the sector's strongest incentives for the cultivation of grains i s put into doubt not only by the results provided above, but also by the fact that those crops generate relatively little employment per hectare. Table A.2.7, which shows the average amount of employment, inperson-days per hectare, generated by 27 major crops, reveals that the grains rank among the lowest crops in terms of capacity to create employment. From a perspective of product chains, some of the most competitive components of the cereal chains appear to be breads and cookies. Their exports increased from US$6 million in 1991 to US$35 million in 2001. Their principal markets are in Venezuela and Ecuador, which means their export strength may be partly dependent on the arrangements of the Andean Common Market. It i s noteworthy that the main raw material for these products, wheat, i s entirely imported, except for very small quantities of specialty wheats grown at highaltitudes inNariiio. The competitive potential of processedflour products i s shown by their increasing RCA value in Table A.2.4. DRC calculations, which are the most solid indicators of comparative advantage, could not be made for processed foods because their information on costs i s proprietary. 57Calculated as a geometric rate of growthbetweenaverage yields for 1991/1992and for 2000/2001. 71 Table A.2.7. NationalAverage Coefficientsof LaborUse by Crop (inperson-daydha.) - - Crop Labor U s , Labor Use Crop Labor Use Isabellagrapes 520.0 Lulo 116.4 White corn 39.8 Green asparagus 339.4 Potatoes 104.6 Oil palm 35.2 Coffee 249.7 Plantain 97.4 Irrigatedrice 33.4 Granadilla 220.8 Citrus 96.9 Yellow corn 23.3 Light tobacco 214.2 Industrialcassava 87.9 Rainfed rice 22.0 Pitahaya 208.9 Mango 84.4 Wheat 20.6 Darktobacco 196.3 Beans 75.9 Soybeans 18.0 Passionfruit 166.2 Sugarcane 65.1 Barley 15.0 Cacao 121.7 Cotton 49.9 Sorghum - 13.4 121.1 Sources: (1) Isabella grapes, Misi6n Paz (2001). (2) Other products, Balchzar, Orozco, and Samach (2003), on le basis of data for 2001 from the MADR. 2. FruitsandVegetables Misi6n Paz (2001:160) has pointed out that "the increasing demand for products that are fresh and rich in vitamins and other nutrients, and that do not contain either fats or calories, has causedthe market for fruit and vegetables to grow at an accelerated pace in the world. . . . In the United States, imports of vegetables and fresh roots and tubers grew at 9.1 percent annually from 1995 to 1999." FORAGRO has provided another perspective on future international market prospects for fruits and vegetables for all of Latin America andthe Caribbean: highincome levels, are leadingto demandsfor more quality products such as fruit and vegetables. . The new requirements for health and the progressive ageing of populations, principally those of ..In this regard, the potentialfor tropical and subtropical fruits is great, and this is confirmedinthe statistics on internationaltrade. Of the total growth in the value of agricultural and livestock exports between 1980/82 and 1993/95, 86 percent was achieved by fruit (41.4 percent), vegetables (16.9 percent), soybeans (15.9 percent), and meats (11.8 percent). The production of tropical fruit (especially bananas and citrus) offers magnificent opportunities in particular for the Andean Region, Brazil and Central America, while in temperate climate fruit Chile and Argentina have the greatest advantagesinproduction. (Secretariado TCcnico...2002:17) Fruits and vegetables are important in Colombia's external trade balance, and they also figure among the crops that generate the most employment per hectare. It i s estimated bananas alone generate 34,543 direct jobs. Bananas are Colombia's third-largest agricultural export, after coffee and flowers (Statistical Annex tables). The value of banana exports has fluctuated around US$400 million per year in recent years, and plantains contribute an additional US$40 million. Other fruits, in both fresh and processedforms, now add about US$30 million in exports, and beans and asparagus are other significant export crops. In 2001 Colombia's total tonnage produced of fresh fruits other than bananas and plantains increased by 4.5 percent and, according to preliminary data for 2002 from the MADR, the total production 72 grew by another 14 percent. Of the 10 fruit crops that accounted for the most acreage in 2000, the area planted expanded most rapidly during 1999-2001 for mandarin oranges (61 percent), and for lemons and strawberries (24 percent). Fruit production i s concentrated mainly in the Departments of Santander, Cundinamarca, Valle del Cauca, and Tolima. Santander leads in production of pineapple, mandarin oranges, and guava (guayaba);Valle del Cauca in bananas for the domestic market, granadilla, pitahaya, and grapes (in spite of that region's high internal transport costs to access export markets); Tolima in soursop (guandibana),lemons, and mango; and Cundinamarca incitrus, strawberries, berries, and u c h u v ~ . ~ ~ Uchuvaexports have beengrowing rapidly, mainly to Europe (principally to Germany), and in spite of its production and marketing being handled in an artisan manner, that crop i s now Colombia's third- most-important fruit export, after bananas and plantains. Uchuva registered nearly US$9 million in exports in 2001 and 2002. Like several other fruits, it is mainly a smallholder crop. Granadilla is another fruit whose main export market i s Europe. Lulo, passionfruit, berries, papaya, and other fruit crops have shown potential as well, although it has not been possible to analyze the comparative advantage of all of them in the context of this study. Lulo and oranges have a considerable domestic market in the form ofjuices. The importance of fruit as an input to agroindustry can be seen inthe fact that in 1998 two-thirds (65.3 percent) of the country's 1,680 food processing enterprises required fruit as an input into their pr0cesses.5~Fruit i s used in products as diverse as processed dairy products, starches, breakfast cereals, cookies, soft drinks, mineral waters, preserves, baby foods, and other prepared meals. Industries usingfruit as an input account for about 17 percent of total industrialemployment. Bananasare cultivated mainly inthe Antioquian UrabB, andto a lesser extent inthe northern part of the Department of Magdalena. These are areas especially affected by the violence, and banana production has been notable for sustaining its levels in spite of the lack of security. However, there has been a slight downward trend in production, more in Magdalena than in UrabB, because of phytosanitary problems. Banana exports depend in good measure on preferential access to Europe, and in the long run this dependency implies a certain fragility of the export position of bananas. Plantain exports have increased considerably since the early 1990s. Its production i s more diversified throughout the country. The largest producing areas are the Andean zones and the inter-Andean valleys, and the largest producer for export i s the Antioquian UrabB. Plantainalso has spawned an incipient processing industry in Bogota and the Cauca Valley. A key to plantain's competitiveness i s provision of genetic material of good quality. Inthis case, materials obtained in vitro are not acceptable because they are easily contaminated with viruses and are difficult to manage in the field. Inareas such as Caldas and Quindio, for example, plantain has good access to research centers and high-quality genetic material, as well as good soils. Where plantain is not very successful it i s due to poor soils, a humid climate, weak access to healthy genetic materials, and the need to apply greater amounts of fertilizer and pesticides because of the soils and climate. The acreage planted in fruit other than bananas and plantains more than doubled from 1991 to 2001, increasing from 93,072 hectares to 190,197 hectares in that period. Table A.2.8 presents data on exports in recent years of fruit other than bananas and plantains. Export growth in these products is very strong, at nearly 15 percent per year, and would be even stronger had not a large part of the passion fruit processing industry collapsed for reasons unrelated to the product's inherent competitiveness. Colombia has negotiated preferential access for some fruit exports in parts of Latin America. For example, bananito (and bananas) do not pay tariffs when they are exported to Central America and Chile, and fruit such as pitahaya and uchuva can enter Mexico under a 3.9 percent tariff. Fruit exports have received few incentives from the Government, mainly duty drawbacks of 2.5 to 3.0 percent and, for some, access to credits from Banco de Comercio Exterior de Colombia (Bancoldex). On the import side, these categories 58These and some other data inthis subsectionare taken from Forero (2003). 59Datafrom the manufacturingsurveysof the Corporacidn Colombia Znremacional, reportedinCCI (2000). 73 of fruit enjoy tariff protection levels (standard for Colombia) of 15 percent and 20 percent, according to the degree of processing. Citrus production expanded at more than 8 percent per year during the 1990s, and it i s estimated that citrus trees covered 41,555 hectares in 1998. The main producing area i s the center of the country (Santander, Cundinamarca, and Tolima, as mentioned, and also Boyach) but the most rapid growth of production has occurred in the Orinoquia and the Atlantic Coast. Citrus fruit does not participate much in international trade, either in exports or imports. Therefore the citrus expansion has been directed at the domestic market (and the same has been true of mango). It would appear there i s considerable scope for further growth of this market since, for example, Colombian levels of consumption of orangejuice are still well below those of other countries: 5 liters per person per year compared to 12liters in Venezuela, 20 to 30 liters in most European countries, 44 liters inthe United States, and 60 liters inGermany. Citrus production receives tariff protection of 15 and 20 percent and citrus exports can enter many Latin American countries with low tariff barriers. Based onexport performance and informed opinion, lemons and limes appear to be the most competitive of the citrus fruits for fresh sales, and oranges have demonstrated potentialfor juice on the domestic market. TableA.2.8. ExportsofFreshandProcessedFruitOther ThanBananasandPlantains, 1997-2001 (000US$ fob) - % Growth 1997 1998 1999 2000 2001 Rate 3 7 19 34 64 114.9 6 0 28 230 210 143.3 485 996 1,707 1,527 1,995 42.5 28 2,093 1,459 212 496 105.2 138 50 77 10 12 -45.7 oupe, watermelon, andpapaya 8 143 31 38 264 140.0 8 0 66 254 128 100.0 11 6 28 61 72 60.0 7,956 9,507 9,854 11,290 13,525 14.2 rawberries, berries, etc.) 591 595 397 357 386 -10.1 (strawberry, peach, etc.) 3 5 4 10 1 -24.0 120 6 42 73 124 3.3 eets of vegetables and fruit and fruit skins 57 69 80 59 27 -17.0 480 791 574 827 1,327 28.9 3,159 3,831 3,362 3,103 8,007 26.2 4,021 4,289 4,323 2,365 2,686 -9.6 9,357 13,408 13,712 14,096 17,277 16.6 7.717 8.980 8.339 6.354 12.047 11.8 17,074 22,388 22,051 20,450 29,32C 14.5 Source: MADR (2002, adapted fromForero [2003]). Trade data suggest the possibility of a comparative advantage for most fruits and vegetables in Colombia. Inthe case of citrus, it i s worth repeating that to date the traded quantities are small, and hence they may not be indicative of true potentials. Table A.2.9 shows the RCA coefficients for principal fruit and vegetable line items inthe customs classifications. 74 Table A.2.9. RevealedComparativeAdvantagefor Fruitsand Vegetables (export values in million US$) Export RevealedComparative Advant. F A 0 Tariff Value, Product Code --Code 2001 Product 1997 1998 1999 2000 2001 Chain Bananas 486 0803 364.868 Bananas Figs 569 0804 0.376 Fruit 5.38 6.43 6.32 7.00 8.11 Mangoes 571 0804 1.339 Fruit 0.15 0.76 1.64 1.35 1.61 Uchuvaand fresh fruit n.e.c. 610 0810 13.528 Fruit 7.30 13.24 10.24 11.61 13.12 Onions 403 0703 1.218 Vegetables 0.56 0.01 0.84 5.60 0.87 Other vegetables including temporarily preserved 474 0711 1.045 Vegetables 1.24 1.23 2.45 1.12 1.54 mixtures Drybeans 176 0713 7.004 Vegetables 0.20 1.41 5.23 2.91 2.82 Peppers, whole and ground 689 0904 2.598 vegetables 1.10 2.75 3.29 1.56 3.56 Canned mushrooms 451 2003 2.429 Vegetables 1.32 1.34 2.13 2.23 2.06 Fruitjuice n. e. c. 622 2009 2.355 Fruit 1.81 1.97 1.86 0.98 1.04 Asparagus 367 0709 1.930 Vegetables 4.53 5.48 3.09 1.85 2.32 Lemons and limes 497 0805 0.374 Citrus 0.00 0.08 0.00 0.09 0.26 Freshpineapple 574 0804 0.267 Fruit 0.17 0.18 0.15 0.24 0.27 Fruitpreparations 623 0811 0.386 Fruit 0.45 0.55 0.46 0.47 1.10 Fruitwith rindsn.e. c. 234 0802 0.210 Fruit 0.01 -- 0.04 0.35 0.34 Other dry fruit 620 0813 0.124 Fruit 0.18 0.01 0.08 0.11 0.20 Processedpeanuts 246 2008 0.134 Fruit 0.02 0.09 0.12 0.11 0.16 Vegetables in vinegar 471 2001 0.390 Vegetables 0.07 0.15 0.30 0.14 0.28 Fresh mushrooms 449 0709 0.472 Vegetables 0.00 -- 0.12 0.28 0.31 Garlic 406 0703 0.149 Vegetables -- 0.01 -- 0.47 0.14 Carrots 426 0706 0.162 Vegetables 0.04 0.18 0.01 0.07 0.19 Lettuce 372 0705 0.165 Vegetables 0.00 -- 0.02 0.16 0.08 Processedtomatoes 391 2002 0.116 Vegetables 0.04 0.03 0.04 0.04 0.05 Oranges 490 0805 0.121 Citrus 0.00 0.47 0.38 0.02 0.03 Jellies, marmaladesetc. 626 2007 1.320 Fruit 0.21 0.35 0.04 0.05 0.08 Tomatoes 388 0702 0.217 Vegetables 0.01 0.22 0.06 0.17 0.03 c P P Note: (1) The first section of the table includes products with an RCA > 1.0 (on average) and generally increasing; the second section, products with an RCA > 1.0 and generally decreasing; the third, products with an RCA < 1.0 and generally increasing; and the fourth, those with an RCA < 1.0 and generally decreasing. (2) The tariff codes are those applied in the Andean Common Market. Source: Adapted from Espinal and Samacfi (2003), and official trade statistics. In regard to Isabella grapes, Misidn Paz points out that producers in the Cauca Valley plan to increaseproduction from 11,000 tons to 32,000 tons per year, for export to Ecuador and Venezuela during the counter-season when Chilean grapes are not available, and for the domestic market (Misi6n Paz 2001:164). Thus there are a number of indications of potential competitiveness in the area of fruits and vegetables, but the existence of a degree of tariff protection in Colombia for these products, although less than for grains, means the trade data alone are insufficient to determine whether they have a comparative advantage. Therefore, DRC calculations were carried out for the present study in order to determine how 75 solid their competitiveness might be. These calculations are summarized for selected products in Table A.2.10. As Table A.2.10 shows, the DRC coefficients indicate that Colombia has a notable comparative advantage in fruit and vegetable crops. In addition, their high rate of employment creation per hectare means they have significant benefits in terms of poverty reduction and social equity in rural areas. (If labor's shadow price were set at a lower level, as some analysts prefer, then the comparative advantage of fruit and vegetables would be even greater, and the comparative disadvantage of grains would be even stronger.) One of the hurdles that fruits and vegetables face for continuing expansion i s inadequate production technology. Inits National Plan for Technology Transfer, ICA has pointed out that 92.4 percent of fruit farms use little or no improvedtechnology, 5.3 percent use some improved technology, and only 2.3 percent use improvedtechnology intensively. Another hurdle i s a highrisk of plant diseases, owing in part to the use of traditional technologies. For example, in 1998 Antioquia was the principal producing area of granadilla, but then its plantings of that crop were devastated by an infestation (of the disease Nectria haematococca Berk.), and now the Cauca Valley i s the main producer of that crop (Exdtica 2002:5:18). Banana and plantain struggle against sigatoka. Another risk faced by producers of these crops is volatile prices and a long-termdownward trend of prices in world markets. As more producers enter these profitable internationalmarkets, prices fall. For example, international prices declined at the following annual rates between 1995 and 2000 for this sample of products, all of which experienced rapid increases in world export volumes: mangoes, 5.2 percent; pineapples, 1.9 percent; lemons and limes, 4.7 percent; cantaloupes and other melons, 5.0 percent; canned mushrooms, 7.3 percent; fresh mushrooms, 9.1 percent; fresh onions and shallots, 8.4 percent; eggplant, 5.6 percent; papayas, 1.3 percent; figs, 6 percent; and asparagus, 5.6 percent (CCI). These negative price trends can be expected to continue for most fruit and vegetable crops. Nevertheless, other countries in Latin America have overcome these obstacles and have expanded their exports of fruits and vegetables much more than Colombia has. Today Peru annually earns only US$15 million in mango exports, Costa Rica about US$lOO million in pineapple exports, Argentina US$83 million in lemon and lime exports, and Mexico US$124 million in mangoes and more than US$50 million inavocadoes. The wide market potentialfor fresh fruit is illustratedby the experience of berries, of which there are 40 varieties native to Colombia. In 1992, only 189 metric tons were exported, but they went to external markets as diverse as Germany, Finland, the Antilles, Aruba, Curacao, Belgium, Spain, Australia, Canada, the United States, andthe United Arab Emirates. Table A.2.10. -- ComparativeAdvantage Indicatorsfor FruitsandVegetables,2001 - Passion Green Department Granadilla Citrus Fruit Lulo P i t a h a d Asparagus Valle del Cauca 0.33 0.26 0.18 0.76 0.26 Tolima 0.29 0.22 0.30 C6rdoba 0.19 0.15 Meta 0.21 0.19 0.20 Risaralda 0.28 0.19 Antioquia 0.72 0.22 0.88 0.15 0.36 0.29 0.73 Norte de Santander 0.23 Caldas 0.26 0.14 0.17 0.34 -Minimum Cesar 0.32 Quindio 0.85 0.27 0.14 0.46 0.20 0.26 - P value 0.33 0.21 0.14 0.23 0.76 0.15 0.22 0.29 0.73 Maximum value 0.85 0.29 0.19 0.46 0.88 0.20 0.36 0.30 0.73 - - National average 0.64 0.26 0.17 0.35 0.82 0.18 0.29 0.29 -0.73 Notes: These calculations take into account transportation and marketing margins and rates of product spoilage. The national average for each crop is weighted by acreage in corresponding departments in that crop. Source: BalcBzar, Orozco, and SamacB (2003). Brazil has successfully based its transformation of a smallholder sector in part of the Northeast on the irrigated production of fruits and vegetables. Inthe district of Petrolina-Juazeiro, in the north of Minas Gerais, 83 percent of the cropping pattern of smallholder participants comprises fruit and vegetables. The leadingcrops include mangos (27 percent of the area), grapes (19 percent), industrial tomatoes (13 percent), onions (13 percent), and plantain (11 percent). The governmental development agency for the district successfully stimulated agricultural research and the involvement of processing industries, and with these ingredients fruit and vegetable crops became a major vehicle for raising the incomes of the rural poor.60 Colombia's variety of climates and growing seasons means over the long run it i s in a favorable position to take advantage of windows of opportunity inthe huge market of the UnitedStates. For example, normally cantaloupe exports to the U.S. face a tariff of 29.7 percent, but during August 1to September 15 their tariff is reduced to 12.8 percent. Inany case, for the time being Colombia has the additional benefit of being able to export many fruit and vegetable products to the U.S. without paying any tariffs, through an agreement known as the Andean Trade Promotion and Drug Eradication Act (ATPDEA). The danger of not taking advantage of export opportunities for these crops is that their international trade i s increasingly becomingconcentratedin a few exporting countries which are developing sophisticated marketing links and technolo ical know-how, so in the future it will be even more difficult than it i s now to break into these markets. 8 Thisexperienceis analyzed inDamiani(2001). This paragraphandpart of the following onearebasedon information andcommentary inFonseca(2003). 77 Some observers dismissthe importance of fruits and vegetables on the grounds that each occupies a small amount of acreage. As seen, however, taken together they are quite important for generating both income and employment. According to MADR official statistics, in 2001, at constant 1994 prices, fruits and vegetables, except plantains and bananas, accounted for 23 percent of the value of total crop output without coffee. Bananas and plantains added another 19 percent, so that all together fruits and vegetables represented more than 40 percent of the value of agricultural output. Their proportionate contribution to agricultural employment i s considerably higher. The principal reasons for Colombia's lag in promoting fruit and vegetable exports appear to be the following: 1. Colombiahas attained certification of admissibility to the US.market for very few products on phytosanitary grounds, and that means the market for fresh produce there i s almost closed to Colombian exporters. During the past 10 years such certification has been obtained for only one product (though it i s close to being obtained for another one). In a much shorter period of time Costa Rica obtained phytosanitary certification for three products. Costa Rica and New Zealand now export uchuvu to the United States, although it i s a product native to Colombia. Three years ago Argentina gained admissibility to the U.S. market for fresh citrus, something Colombia has not achieved. Although the violence inrural areas may complicatethe process of obtaining additional certifications, what i s most needed i s an appropriate institutional response inthe phytosanitary areaandpublic sector leadership inmeetingcertification standards. 2. Although these products have a comparative advantage, they lack a system of market intelligence that would facilitate the establishment of links between producers, exporters, and buyers inother countries. 3. Technological development i s weak in many fruits and vegetables, reflecting the fact that the commitment to agricultural research in this area i s insufficient. As has been pointed out for Latin America in general, "Research in tropical countries focused more on traditional food crops that offer fewer competitive advantages in national and international trade, neglecting products such as fruits and vegetables, where the region enjoys clear advantages" (Kondo 2003). 4. There is insufficient awareness among producers of the importance of product quality in all senses: absence of chemical residues, adequate flavor, uniformity of appearance, timeliness of delivery. Achieving the desired quality in some cases may require supplementary irrigation. However, consensus building among product chains has helped promote a greater awareness among all producers of the importance of product quality for future growth prospects. 5. The deficiencies in the area of product quality are in part a result of the weakness of producer organizations for fruits and vegetables. It i s noteworthy that among all the parafiscal funds- which are funded inpart by producers and are used to support research and technology transfer, marketing, phytosanitary improvement, infrastructure development, and other forms of production support-the parafiscal fund for fruits and vegetables has experienced the highest rate of producer evasion of their responsibilities for making contributions to the fund, at 95 percent, of all such funds. Incontrast, the evasion rate among producers i s zero for poultry, 4 percent for oil palm, 5 percent for cacao, 6 percent for beef cattle, and 20 percent for rice (Contraloria General... 2001: 15). Inpart, this situation i s attributable to the spatial dispersion of producers of fruits and vegetables and the large number of intermediaries who participate in 78 marketing those products and who are legally responsible for collection of the parafiscal fund contributions (set by law at 1percent of the sales value inthe case of fruits and vegetables).62 6. In the past, a high priority has not been assigned in the country's international trade negotiations to opening up markets for Colombian exports in this area and to providing high- quality marketing assistance. Rather, emphasis has been placed on maintaining tariffs for import substitutes, and there has not been a coordinated strategy for export of fruits and vegetables, or indeed an agroexport strategy ingeneral. Overcoming these drawbacks i s a considerable challenge, but it i s one that promises benefits throughout rural society if it can be achieved. 3. Other Long-Cycle Crops This category of crops includes two of the four largest agricultural exports, coffee and sugar, and one of the fastest-growing exports, palm oil. Exports of raw cacao products have decreased sharply since 1994, but exports of chocolates have increasedby a larger amount. All four crops are linked to important agroprocessing industries in Colombia. The sugar and palm oil subsectors also have been leaders in technological and institutional development in the last 10 years, and coffee has long had a strong institutional base. Palm oil productionhas been increasing while cottonseed oil and soybean oil have been declining. Nominal tariff protection for palm oil, cacao, and sugar has fluctuated markedly over the past 15 years (Statistical Annex tables). Inthe case of palmoil, this i s due inpart to the functioning of the Andean price band system. Nominal protection was less than 5 percent in 1994 and 1995, jumped to about 20 percent in 1996, then fell to minus 30.5 percent in 1997, rose to about 2 percent in 1998, and since then has fluctuated inthe range of 45 to 77 percent. Domestic prices have varied by smaller amounts since the price bands offset part of international price fluctuations. The cacao product chain has tariff levels of 10, 15, and 20 percent, depending on the degree of processing of the product, but market forces often have depressed domestic price levels below what the tariffs would indicate. Nominal protection was negative during 1990-94 and 1996, and slightly positive in 1995, 1997, and 1998. Then it rose to about 23 and 35 percent in 1999 and 2000, respectively, before falling to negative levels again in 2001 and 2002. On average, it has had little or no nominal protection. The sugar chain has tariff rates between 5 percent and 20 percent, but its tariffs often move out of that range because it belongs to the Andean price band system. Its domestic price is managed through a stabilization fund, taking as reference prices the level indicated by the price bands and the export prices. The net result i s a highly protected subsector with respect to the free intemational market for sugar. During 1998-2002, the nominal protectionrate for white sugar from 124 percent to 180 percent, and for raw sugar it rangedfrom 68 percent to 148percent. However, it should be recognized that hardly any countries inthe world are competitive in sugar when the free market price i s used as the benchmark. The only gaps in this pattern of protection for these crops are preferences negotiated in bilateral agreements for imports of vegetable oils and sugar from Mexico and Chile, and for imports from Asociacidn Latinamericana de Integracidn (ALADI) countries. Imports of coffee are trivial, and imports of sugar are less than 10 percent of sugar exports. For vegetable oils all takentogether, including soybean oil, imports are about four times larger than exports, but Theparafiscalfunds are discussedinsection6.1.2. 79 the latter have been growing much faster since the early 1990s, owing to the dynamism of the oil palm subsector (Table A.2.11). In the case of cacao, exports have not increased but, as mentioned, their compositionhas shiftedtoward processedproducts, and imports have increased rapidly, so the relative trade balance has deteriorated (Table A.2.12). Table A.2.11. RelativeTrade Balanceof Vegetable Oils (trade inmillions of dollars, period averages) IRTB - -0190 -0.90 -0.78 -0:60 H Source: Espinaland SamacB(2003). Table A.2.12. RelativeTradeBalancefor the Cacao Chain (trade inmillions of dollars, periodaverages) Source: EspinalandSamacA(2003). The revealed comparative advantage coefficients for these products are presented in Table A.2.13. The picture that emerges from these coefficients i s mixed. First, obviously some of these products have a proportionately strong presence in export markets, coffee and sugar inparticular. Second, exports of palm oils are rapidly growing and are becoming important in value. Third, most of the products with increasing RCA coefficients have a significant processing component. Even raw sugar passes through an important industrial process before being marketed in that form. Exports of sweets from sugar and chocolate preparations are increasing quickly in importance. In the case of chocolate, they go principally to the Andean Community and the United States. These results tend to confirm the hypotheses that Colombia's principal comparative advantage lies in long-cycle crops and agroindustrial products. Industrial efficiency requires an organizational ability in which Colombia excels compared to most other countries of its income level. A major question that remains i s the extent to which sugar would remain competitive if it did not enjoy the benefits of managed prices on the domestic market. The same question applies to palm oil in light of its high degree of border protectioninrecent years. It i s a beneficiary of the protection policy applied, for example, to soybeanoil. Answers to these questions are sought through the DRC calculations that are presented in Table A.2.14. It can be seen that all four categories of long-cycle crops do possess a considerable comparative advantage. Sugarcane, thanks to its very highyields in the Cauca Valley (among the highest in the world), and oil palmemergeas the most competitive of the four crops by a small margin-in spite of the strength of Malaysia and Indonesia as palmoil producers. However, when the spatial dimension i s taken into account, the most competitive localized crop turns out to be cacao in Antioquia. The only case of doubtful inherent competitiveness is cacao in Huila, which i s one of the larger producers of that crop. For oil palm, Meta is 80 the region with greatest comparative advantage, although Cesar and Bolivar are close behind, andfor coffee it is Tolima. All these crops are highly intensive in labor use and therefore have considerable capacity for creating employment. Data on their costs of production indicate that more than 30 percent of costs are representedby the employment of skilled and unskilled labor. In addition, employment i s relatively stable and well remunerated in most of these crops over the medium term, and therefore it i s employment of high quality. The same is true of bananas and plantains. In the cases of sugarcane, oil palm, and coffee, producers also are well organized. These are crops grown by large numbers of smallholders (especially coffee, cacao, and oil palm). The presenceof smallholders, in combination with the economies of scale in processing and marketing, provides strong incentives for social cooperation and organization in the rural areas where the crops are dominant. Coffee i s especially important in all these respects. In 1999, it created 406,794 directjobs, and 95 percent of coffee producers held less than 5 hectares of land. Therefore it i s of overwhelming importance for rural society as well as for the nationaleconomy. 81 TableA.2.13. Revealed ComparativeAdvantagefor Other Long-CycleCrops and Their Products (export values inmillionUS$) Export Revealed ComparativeAdvantage FA0 Tariff Value, Product Code Code 2001 Product Chain 1997 1998 1999 2000 2001 Raw sugar and 162 1701 150.745 Sugar 9.39 15.30 10.33 14.12 14.10 panela Sweets from sugar 168 1704 117.296 Sugar 7.80 10.84 12.33 10.87 13.82 Glucose syrup, etc. 172 1702 5.729 Sugar 1.89 3.42 4.42 4.83 8.14 Palmoil 257 1511 26.228 Oilseeds, oils, and 2.31 3.13 3.26 3.49 2.92 fats Palmkernel oil 1275 1513 5.860 Oilseeds, oils, and 2.44 2.60 4.91 7.16 6.87 fats Margarine, edible 1242 1517 29.313 Oilseeds, oils, and 1.43 1.07 6.41 17.29 16.41 oil mixtures fats Cardamom, etc. 702 0908 1.534 Others 4.15 1.32 1.83 3.90 3.27 Sugarcane 156 1212 0.604 Sugar 117.7 90.81 141.0 85.41 57.39 0 4 Refined sugar 164 1701 66.245 Sugar 8.75 8.87 7.31 7.47 6.27 Molasses 165 1703 2.970 Sugar `3.47 10.07 5.93 7.20 3.29 Cacao butter and 664 1804 5.519 Cacao 4.06 3.02 3.37 0.90 2.62 oils Green coffee 656 09011 768.573 Coffee 83.86 79.39 66.28 63.20 68.86 Coffee extracts 659 2101 89.587 Coffee 34.39 36.41 26.16 29.86 23.00 Palmkernels 257 12071 0.138 Oilseeds, oils, and 23.47 17.41 7.24 2.84 14.30 fats Glucose, sugar 167 1702 5.729 Sugar 0.02 0.16 0.49 0.63 0.51 syrups n.e.c. Chocolate, other 666 1806 21.582 Cacao 0.46 0.49 0.46 0.57 1.45 Prep. Cacao paste and 665 1805 4.701 Cacao 0.12 0.09 0.08 0.01 0.17 powder Thyme andbay 723 0910 0.324 Others 0.32 0.32 0.44 0.55 0.49 leaves Cacao paste 662 1803 0.410 Cacao 0.50 0.01 0.03 -- 0.42 Raw cacao 661 1801 0.683 Cacao 0.23 0.23 0.07 0.11 0.17 Roastedcoffee 657 09012 0.800 Coffee 0.84 0.34 0.18 0.15 0.31 Note: (1) The first horizontal row section of the table includes products wil an RCA > 1.O (on average) and generally increasing; the second section, products with an RCA > 1.0 and generally decreasing; the third, those with an RCA < 1.0 and generally increasing; and the fourth, those with an RCA < 1.0 and generally decreasing. (2) The tariff codes are those applied inthe Andean Common Market. Source: Adapted from Espinal and SamacB(2003), and official trade statistics. Coffee's comparative advantage has slipped in recent years with the entry of producers such as Vietnam and the consequent decline in international coffee prices. This points to the need to improve productivity (control costs) and place greater emphasis on high-quality coffees, in order to differentiate Colombia's product in world markets. However, while some producers are going out of coffee, those that have been able to control costs remain competitive internationally. They tend to be the smaller producers. 82 Before the coffee crisis smallholders cultivated 445,000 hectares of coffee, representing 45 percent of the total area, and after the crisis set in fully, they were cultivating 670,000 hectares of coffee, which was 77 percent of the One of their strategies for economic survival has been to increase the interplantings of plantain with the coffee. The importance of quality cannot be sufficiently emphasized. Currently, 40 percent of coffee imported into the United States, by value, takes the form of specialty coffees (Comisi6n de Ajuste... 2002:14). Although the quality of Brazilian coffee is improving, neither it nor Vietnam i s as well endowed as Colombia with natural conditions for producing high-quality coffee. However, compared to some Central American countries, Colombia has been slow to improve its coffee grades and develop new, high- quality brands, although historically quality has beena strong selling point for Colombian coffee. Specialty coffees that have gained recognition for their origin, in localities such as Huila, Caldas, and Santa Marta, represent perhaps 10percent of Colombian coffee today, and that i s clearly a worthwhile direction to go in order to enhance competitiveness. In short, the opportunities for Colombia in coffee are still there, but seizing them will require a concerted effort in research, extension, and marketing, all aimed at positioning Colombia at the highend of the evolving market. TableA.2.14. ComparativeAdvantageIndicatorsfor Other Long-Cycle Crops, 2001 Department Cacao Sugarcane PalmOil Coffee Valle del Cauca 0.64 0.55 Santander 0.70 Tolima 0.54 0.49 Magdalena 0.66 Cesar 0.53 Meta 0.49 Narifio 0.59 Bolivar 0.54 Antioquia 0.39 0.64 Huila 1.01 Ouindio 0.68 Minimum value 0.39 0.55 0.49 0.49 Maximum - -0.66 value 1.01 0.55 0.68 - National average 0.66 0.55 0.56 0.60 Note: The national average for each crop is weighted by acreage in correspon ing Departments in that crop. Source: Balcfizar, Orozco, and Samacti (2003). Inspite ofcacao's comparative advantage,its productionandexports have declinedabruptly inthe last three years from levels recorded in the early and mid-l990s, and now chocolate producers import a largest in exports (MADR, Observatorio... 2002). Cacao in many cases i s the only or principal source of share of their cacao requirements. Colombia i s the world's ninth-largest producer of cacao, but only the 4lSt 63Data from the coffee census, cited inForero (2003). 83 income of campesino families that grow it. Their cultivation systems often include several other crops, including coffee, plantain, cassava, fruit, and trees grown for timber. The average area planted in cacao by these smallholders i s about three hectares. Perhaps in part because of the small scale of cacao farms and their geographical dispersion, an effective phytosanitary campaign has not been mountedfor this crop, and plant diseases are a major reason for its lack of dynamism. It is equally weak in the area of agricultural research. In short, technical problems at the farm level need to be resolved before cacao can realize its apparently high potential. Santander is by far the largest producing area for cacao, and therefore it would be a logical place to concentrate an intensifiedresearch and extension effort for the crop, and to the extent that productivity gains could be achieved there, the program could be broadened to other areas. The comparative advantage of cacao would appear to imply that a strengthened technology development effort i s warranted for that crop. A favorable circumstance for palmoil is that world demandis expectedto grow at least inthe range of 3.3 to 3.9 percent annually for the next 20 years. These are projections of the Federucidn Nacionul de Cultivadores de Pulma de Aceite (FEDEPALMA),but it has been pointed out that they are conservative compared to other internationalprojections, especially when the possibilities of the incipient oleochemical industry(substitutingvegetable oils for petroleum products) are taken into account (MisibnPaz 2001: 131). Palm oil i s the tropical counterpart to the temperate soybean oil and can be directed to many of the same uses. Another advantage for the crop inColombia i s that, beyond areas currently planted inpalm, there are literally millions of hectaresof additional lands with appropriate soils for the crop. For calculating the comparative advantage of sugar, the reference price was not the free world market price, which i s acknowledged to be distorted by subsidies in developed countries, but rather the domestic market. Inother words, the reference price was lo$ per pound, rather than the price of 5$ per average price on three markets: the free world market, the international quota export market, and the pound that often prevails on the free market. At the latter price, Colombian sugar would not have a comparative advantage, but neither would sugar from other exporting countries because that price i s the result of surplus production arisingfrom subsidies inthose countries. Inregard to the domestic pricing arrangements and tariff protectionfor sugar, to a degree they are justified so that Colombia could be well positioned to eventually export on a wider world market if international subsidies were reduced. The policy question i s the degree to which the present overdimensioning of the sugar sector should be sustained, at the expense of consumers. For palm oil, the DRC calculations indicate that the crop can be competitive on world markets without the support provided through tariffs andpricing arrangements, but it may be argued that these incentives are conferring important dynamic benefits on Colombia by spurring rapid expansion and technological development in this subsector. The issue is one of balancing those benefits with the interest of domestic consumer welfare. Additionally, given the apparent comparative advantage of some livestock products and processed foods, more competitive domestic palm oil markets could enhance Colombia's export prospects for products for which it i s an input. This set of issues i s discussedfurther in section 6.1.1. 4. Tubers, Tobacco, and Cotton Potato originated in the Andes. Today it i s the 10"-most-important crop in Colombia in area planted, and the sixth most important in value of production. It i s the most important crop in the colder regions of the country. Close to 90 percent of potato producers have less than 5 hectares; together they produce 35 percent of the potato harvest. Medium- and large-scale producers generally have substantially lower productioncosts becausetheir yields are so much higher, although some of the smaller producers are much more productive than others of their scale (Forero 2003). 84 The potato-processing industry is extensive; 70 processing facilities existed in 1997, although 15 medium-sized and large companies control 95 percent of the market for processed potatoes. Exports of potatoes and other tubers earned more than US$20 million in2001. Potatoes and their products have import tariffs of 10and 20 percent, with preferences grantedto a number of Latin American countries. Cassava i s both consumedfresh and used as an animal feed, in milled form, and work i s underway to exploit its potential for making starch and industrial glues. In human consumption it i s an important subsistence crop in some regions of the country, and also i s marketed through the complete range of retail outlets. Cassavai s the tropical alternative to corn, inthe same sense that palmoil i s the tropical counterpart to soybean oil. Physiologically corn yields are favored by the long hours of sunlight at latitudes far from the equator, a factor that would make it difficult for tropical countries ever to compete effectively in world markets for corn. Cassava, however, prospers in the conditions of latitudes closer to the equator, and very highyields of it are obtained incountries likeThailandandeven insouthernAfrica. Two kinds of tobacco are grown in Colombia, light and dark. The tobacco product chain includes extensive processing facilities and a substantial export industry. However, in recent years tobacco imports have increased more rapidly than exports. At the farm level, tobacco i s one of the crops most intensive in labor use, and it often is grown on very small holdings. In Santander, BoyacB, North Santander, and the Atlantic Coast, the averagetobacco farm i s 1.5 hectaresor smaller. Cotton occupied about 262,000 hectares in 1991, but in 2001 was sown on only 53,000 hectares. I t s production fell dramatically in 1994 and then dropped further during 1998-2000. Since then it has recovered very slightly. The nominal rate of protection was negative during 1989-91, then rose to 21 percent and 41 percent in 1992and 1993, respectively, then fell to 5 percent in 1994. It remainedat slightly positive or negative levels until2001, when it rose abruptly to 22 percent, only to drop again to 3 percent in 2002. More than because of low average protection rates, its profitability has been undermined by falling world prices, the risingreal value of the Colombia peso in the 1990s, and risingcosts of plant protection at the farm level. The cotton product chain includes important subsectors producing cotton thread and yarn, fabric, and clothing. Colombia i s a net importer of thread, yarn, and fabric, but also exports them and i s a significant exporter of clothing. Denim and serge fabrics are among the larger exports of fabric. Exports of cotton products have more than compensatedthe fall inraw cotton exports, and total exports of all forms of raw and processedcotton fiber are now in the neighborhood of US$600 million per year. That makes the subsector's role inthe balance of payments comparable to that of flowers. Table A.2.15. RelativeTradeBalancefor the Potato Chain (trade inmillions of dollars, periodaverages) Source: Espinaland Samach (2003). The relative trade balance coefficients for the potato, cotton, andtobacco chains are show inTables A.2.15 through A.2.17. On the surface, Colombia would appear to be losing competitiveness in all three chains but, as indicated, there have been important developments at the industrialend of all of them. This i s 85 another indication of Colombia's relative strength in agroprocessing industries, and it is increasingly shifting in that direction. Table A.2.16. RelativeTrade Balancefor the Cotton Chain (trade inmillions of dollars, period averages) 1991192 1993195 1996198 199912001 425,044 388,516 341,354 331,869 Source: Espinal and Samacfi(2003). Table A.2.17. RelativeTrade Balancefor the Tobacco Chain (trade inmillions of dollars, period averages) Source: Espinal and Samacfi(2003). The RCA coefficients in Table A.2.18 confirm Colombia's growing potential in fresh and processed potatoes, of a variety of types, and in processed tobacco. Leaf tobacco i s a not-insignificant export too, but it appears to be losing ground inrelative importance. This table does not reveal Colombia's strength in cotton textiles and other textiles because the coefficients were not calculated for subsectors considered to be part of industry, but the export values reported in Table A.2.16 confirm the robustness of the textile sector, in spite of the decline of cottoncultivation. It should be noted that cotton is not anespecially employment-intensive crop. It ranksabove grains in employment per hectare, but below sugarcane and well below fruit, vegetables, and tobacco (Table A.2.7). Therefore, the loss in employment owing to the decline of cotton cultivation can be made up through expansion of the area under other kinds of crops, and through growth of the textile industry (based on imports of cotton and other materials). The comparative advantage calculations for these crops are shown in Table A.2.19, in the form of DRC coefficients. These results confirm the comparative advantage, at the farm level, of potatoes and tobaccos in Colombia. Dark tobacco is the most competitive of these crops, especially in Sucre and Santander. The comparative advantage is almost as strong as that of fruits and vegetables, and represents a solid basis for long-run development of these crops. Potatoes also have a comparative advantage when produced by smallholders, in spite of their generally lower yields. A major hurdle for potato growers i s to reduce their dependency on agrochemicals. Chemical residues on the crops are a source of concern from the viewpoints of consumer health, farmworker health, export prospects, and costs of production. It i s widely known that potato growers regularly use excessive quantities of fungicides, herbicides, insecticides, and antibacterial compounds. In the case of the smallholders of the Sabana de Bogoth, these applications of chemicals represent around 31 86 percent of the cost of production. On larger farms, they representabout 27 percent of the cost of production (Le6n 2003:50). In the future, increasing numbers of export markets will be closed to products that do not meet food safety standards. Potato exports also face phytosanitary barriers in the case of Ecuador and Ped, and Venezuela i s the only significant export market for Colombianfresh potatoes (they go to processing plants inVenezuela). Colombia's National Potato Council currently i s working on a project to market french fries to the Andean Community, Panamti, and Trinidad and Tobago. The aim i s to export a high-quality product, since Colombia cannot compete with some other countries in lower-quality potatoes (Forero 2003:Part II:20). Overcoming the phytosanitary barriers to fresh potato exports to Ecuador and Perti will require greater producer organization, and close coordination with researchand extension agencies. Viewed in terms of comparative advantage, cotton presents mixed results. It demonstrates a comparative advantage in four departments (Cbrdoba, Valle del Cauca, Huila, and Meta), although only marginally so in C6rdoba. It has a comparative disadvantage in Cesar, Sucre, Magdalena, and Tolima, although it i s close to being competitive in the first two of these departments. These regional differences spring from differences in cost structures, in particular the costs of pest control. Those costs in tum are related to crop management and rotation practices. In Huila, cotton rotates with sorghum, rice, and corn. Also, the least-competitive regions have experienced degradation of soils over the long term through inadequate soil management. Tolima in particular has suffered compactation and salinization of soils in cotton-growing areas, and Cesar also has suffered degradation of its soils. A key to cotton's success inMeta is the addition of nutrients to the soils, leveling of the fields, and provision of adequate drainage.64 Also, in Meta, only 22 percent of the costs of cotton are devoted to pest control, whereas the corresponding figure in Cesar i s 32 percent. Traditionally, the Atlantic coast zone has experienced the greatest problems with pest infestation in cotton, including Anthonomus grandis, the pink worm Heliothis virescens,and the borer Spodopterufrugiperda. These pests have not tended to appear in the Llanos Orientales because cotton i s grown there in dispersed form, along river floodplains. However, it i s important to be vigilant because in 2001 Anthonomus grandis was detected in Meta. The Ministry of Agriculture and its local agencies have reported significant increases in cotton yields and profitability through the use of organic fertilizers and by plowing with a vibrating "chisel" instead of discs. Cotton is another crop that requires a change in techniques of cultivation to be competitive and sustainable over the long run. Given its susceptibility to pests, it also i s a crop that clearly can benefit from biotechnology, provided the appropriate institutional structures for biotechnology research and controls can be put into place. Worldwide, cotton i s one of the four crops that have received most of the attention in transgenetic research to date (the other three being com, canola, and soybeans). This fact also provides grounds for optimism regarding the potential for that kind of research to improve the competitiveness of Colombian cotton. ~ This paragraphis basedonthereportof Le6n(2003). @ 87 TableA.2.18. RevealedComparativeAdvantagefor Tubers, Tobacco, and Cotton (exportvalues inmillion US$) Otherpotatoaid root products 149 07149 11.963 Vegetables 5.29 8.46 47.68 53.66 68.74 Cottonlint 770 14042 0.149* Cotton 3.23 1.06 1.60 0.20 0.62 Leaftobacco 826 2401 12.608 Tobacco 1.64 1.69 1.84 1.53 1.04 Preparedandpreserved 475 20041 0.351 Potatoes -- 0.01 0.20 0.53 0.79 potatoes Otherprepared, preserved 472 20052 1.704 Potatoes 0.43 0.93 1.22 1.50 0.37 potatoes Frozenuotatoes 118 07101 0.480 Potatoes 0.06 0.01 0.04 0.08 0.20 Cigareties 828 2402 21.695 Tobacco 0.03 0.07 0.27 0.61 0.90 Cottonfiber, uncardedand 767 5201 0.165 Cotton 0.06 0.03 0.02 0.01 0.01 I/cornbed *Exports of cotton fibers, threads, yarn, and weavings (tariffcodes 5201 to 5212) were US$33.010 million in 2001. Note: (1) The first horizontal section of the table includes products with an RCA > 1.0 (on average) and generally increasing; the second section, products with an RCA > 1.0 and generally decreasing; the third, products with an RCA < 1.O and generally increasing; and the fourth, those with an RCA < 1.O and generally decreasing. (2) The tariff codes are those applied inthe Andean Common Market. Source: Adapted from Espinal and SamacA (2003), and official trade statistics. 88 Table A.2.19. ComparativeAdvantage Indicatorsfor Tubers, Tobacco, and Cotton, 2001 Industrial Dark Light Department Potatoes Cassava Tobacco Tobacco Cotton Valle del Cauca 0.55 2.07 0.33 0.67 Santander 0.27 0.44 Tolima 1.os C6rdoba 0.9 1 Magdalena 1.69 Cesar 1.75 1.55 Meta 0.78 0.70 Nariiio 0.49 Sucre 0.21 1.01 Bolivar 0.31 Boyach 0.47 0.34 Antioquia 0.53 2.53 Huila 0.53 0.57 Norte de Santander 0.47 0.33 Quindio 0.43 1.76 P Minimumvalue 0.43 0.78 0.21 0.33 0.57 Maximumvalue 0.55 2.53 0.31 0.53 1.69 0.49 1.78 0.27 0.39 -, 1.02 Note: The national average for each crop is weighted by acreage incorrespondingdepartmentsinthat Cotton's profitability undoubtedly has been affected by subsidies in industrialized nations. To test this hypothesis, the DRCs for cotton were recalculated with a 15 percent increase in international prices. The results (from Balchzar, Orozco, and Samach2003) are shown inTable A.2.20. Table A.2.20. DRCsfor Cotton, Calculatedwith a 15PercentIncreaseinInternationalPrices Valle Tolima C6rdoba Magdalena Cesar Meta Sucre Huila 0.56 0.86 0.72 1.05 1.10 0.58 0.83 0.47 It can be seen that if it were possible to consistently offset the effects of international subsidies on prices (say through tariff policy), cotton would become viable in more regions of the country, essentially everywhere except in Magdalena and Cesar. However, in Sucre and Tolima the comparative advantage i s not very robust, and profitability of the crop would be threatened by fluctuations in the real exchange rate over time. If more border protection were provided to cotton, it would be necessary to ask about the consequences for the textile industry, which would pay a higher price for a basic raw material. Cotton textiles create much greater employment and export earnings than raw cotton does. Cassava has a comparative disadvantage everywhere except in Meta. When sensitivity analysis was done for cassava as well, with an assumed 20 percent increase in international prices, the crop still 89 remained noncompetitive, except again in Meta. Nevertheless, cassava is the one crop with a comparative disadvantage that may have potential over the long run, and warrants closer attention for that reason. The basis for its potential is that its productivity ceiling is very high in relation to actual performance-higher than for any other crop. Current national average yields are about 10 tons per hectare, yet a number of producers, in a wide area from the Department of Atlfintico to the Cauca Valley, are obtaining 35 tons or more per hectare. A group of producers in South Africa i s regularly obtaining yields of about 50 tons per hectare, and in Colombia in the Department of CBrdoba, the clonal variety SM 1433-4 has yielded 84 tons per hectare on a plot of 10 hectares (CLAYUCA n.d.). While the average yields in Thailand are only 15 tons per hectare, it i s thought that the practicalgenetic ceiling i s inthe range of 80 tons per hectare. Although cassava i s an important component of the diet of many families, especially in rural areas alongthe coasts, its most productive uses appear to be inanimal feeds and indiverse industrial applications, especially as starch and glue. This flexibility permits a direct calculation of the parameters necessary for it to be competitive with corn. The technical equivalence between the two crops i s 0.7 a ton of cassava flour i s equivalent to 0.7 a ton of corn. This means that if the reference price of corn were about US$l10per ton, then the equivalent price of fresh cassava would have to be no more than US$34 per ton.65 At that price, a combination of cost and yield that would give cassava a comparative advantage would be a cost of 2.6 million Colombian pesos (at 2002 prices) and a yield of 37 tons per hectare. That combination and even better results have been attained in various regions of the country, according, for example, to reports of the International Center for Tropical Research (CUT) about the Atlantic Coast, and of the CCI about the zone of San Juan de Arama inMeta. Despite cassava's comparative disadvantage, Colombia already i s the world's third-largest producer of it, mainly for supplying the domestic market. The country's favorable conditions for cassavain the long runinclude: (a) CUT has the greatest knowledge of cassava and the largest gene bank for the crop in the world; (b) cassava i s readily adaptableto a variety of soils, including poor and acid soils; and (c) Colombia already has advanced relatively far down the road of developing and testing commercial varieties, and mechanization of some cultivation tasks (MisiBnPaz 2001:144). The adaptability of cassava to acid soils is one reason for the interest inpromotingtrials of it, inrotation with corn and/or sorghum or soybeans, in the altillanura. Cassava's main hurdles include susceptibility to pests (especially the white fly); the high water content of the root (65 percent), which makes it expensive to transport; and the lack of processing facilities that would purchase the crop on a significant scale. The problem of the white fly has been managed successfully in the northern part of the Cauca Valley through techniques of integrated pest management (CLAYUCA n.d.:2). The water content problem i s being addressed through the promotion of small-scale, inexpensive drying plants in rural areas. The question of developing an industrial market on a substantial scale i s more difficult to resolve, because it requires a committed investor who is convinced the farm-level problems can be resolved. However, the banana enterprises are experimenting with a cassava processing plant in the Urabfithat would produce glue for bananaboxes, and interest has been shown in another plant that would produce cassava fries to compete with french fries. A seed-capital fund for related areas could be a useful catalyst. Inthe caseof cassava, trying to change relative prices throughtrade policy does not appear to be the answer, but technological development could be. From a viewpoint of public policy, more rapid exploration and development of cassava's potential has been held back by the lack of commitment of sufficient research resources to the crop. While the outcome of such a commitment cannot be predicted with certainty, all indications are that this i s a promising area. 65 When cassava i s milled, 2.6tons of fresh cassava are requiredto produce a ton of flour. 90 5. Livestock Products Livestock products reviewed in this study include poultry and eggs, pork, and beef and milk. Over the past 20 years poultry has been the fastest-growing product by a wide margin. Poultry production increased by almost two and one half times from 1982 to 2001. Milk production almost tripled from 1979 to 2001, and pork production increased by about 10 percent between 1991 and 2000 (DANE, MADR). Beef production fluctuated around the same level from 1982 to 1997 (of slaughter of 3 million to 3.8 million head per year), and then dropped abruptly during 1998-2001 to about 2 million head. B y carcass weight, the domestic availability of beef was 24 percent lower in 2001 than in 1982. It i s possible that the decline in cattle operations i s attributable to the rural violence; cattle ranchers report a significantly higher frequency of problems of physical insecurity than crop farmers do. Although beef i s no longer a dynamic product, its volume of production remains high. Colombia i s the ninth-largest producer of beef in the world, and the largest inthe Andean region. Corresponding to production trends, beef exports have trended downward in the past 10 years, although with significant fluctuations, but exports of dairy products have grown very rapidly, especially in the last few years (Tables A.2.21 and A.2.22). The small amount of beef exports go to Venezuela; at present they are prohibited from entering the U.S. market because of hoof-and-mouth disease, but there are hopes that parts of Colombia can be declared free of that disease in the foreseeable future. Dairy exports are almost entirely of milk, and their principal markets are Ecuador and Venezuela. Because of the presence of international subsidies on milk and the efficiency of dairy producers such as New Zealand and Denmark, it would not be realistic for Colombia to aspire to dairy exports to the wider international market. However, continuing to export to the Andean region, in the framework of the Andean Community, appears to be a viable route. Inthe poultry subsector, eggs are the principal export, on a relatively small scale. Poultry trade is hampered by the lack of sanitary clearance for poultry meat to enter the United States, and by Colombia's controls over import of chicken parts. Table A.2.21. Relative Trade Balance for Beef andBeef Products (trade in millions of dollars, period averages) Source: Espinaland Samach(2003). 91 Table A.2.22. RelativeTrade Balance for Dairy Products (trade inmillions of dollars, periodaverages) Source: Espinaland Samacii (2003). The study of Misidn Paz speculates, without presenting data, that extensive beef production i s more efficient than fattening in feedlots, because of lower costs and fewer problems for the environment and for animal health. It suggests that a few basic improvements in managing pastures and herds, such as incorporation of legumes in pastures, better attention to animal health, and use of supplemental feeds in the dry season, could significantly increasethe productivity of extensive beef operations. Inthe same spirit, it i s also recommended that installation of cooling tanks for milk could permit double milking each day, increasing milk yields significantly; to date only 35 percent of the country's milk i s produced in that way (MisibnPaz 2001:150-52). Thus the technological frontier for beefand milkproduction couldbe expanded with feasible techniques of management. Milk plays a role in raising incomes of smallholders. It i s estimated that 70 percent of dairy farmers produce less than 70 liters of milkper day (Forero 2003). The domestic resource cost coefficients (Table A.2.23) show a comparative advantage for all livestock products in Colombia. It i s strongest for poultry and eggs, followed by milk. However, when the calculations are made with prevailing rates of protection on imports of corn, sorghum, and soya, pork loses its competitiveness and poultry's competitiveness becomes only marginal. Milk's competitiveness also diminishes somewhat. Thus the protection policy for feedgrains turns out to be an important factor in determiningthe international competitiveness of livestock products, especially pork and poultry. In Colombia, poultry and pork production differ significantly in their scale of operations and the degree of vertical integration of the enterprises. Unit costs decline markedly as scale increases in these industries, becauseof (a) the possibilities of adoption of technologies that are viable only on a large scale, (b) greater efficiency in marketing and purchasing of inputs, and (c) the possibilities offered for vertical integration in the production of feed mixes. The difference in competitiveness between the poultry and pork subsectors may be explained by their differences in scale of operations, poultry being more advanced inthat regard. A characteristic of pork production is that it requires fairly largeexpansesof territory that is lightly populated, because of the need to dispose of large quantities of waste. In principle, Colombia possesses such areas in abundance, but to date they have been little exploited for purposes of large-scale pork operations. 92 Table A.2.23. Comparative Advantage Indicators for Livestock Products vis-&vis the U.S. Market, 2001 Beef Pork Pork, Complete11 Valle del Cauca Note: (1) The national average for each product is weighted by production value in corresponding departments in that crop. (2) These data are preliminary, and there are inconsistencies in the figures for pork.) Source: Balcizar, Orozco, and Samaci (2003). Concerns have been raised about the competitiveness of Colombia's poultry industry in the face of competition from Brazil. However, two factors attenuatethis competition. One i s the highcost of shipping poultry from Brazilian ports to Colombia. The other has been the recent Brazilian currency devaluation. The DRCs reported for poultry in Table A.2.24 refer to comparative advantage vis-&vis the United States market. To address the question about Brazil, the DRCs were recalculated with Brazilian prices as of June 2003. The results of those calculations are shown inTable A.2.24. Table A.2.24. Comparative Advantage Indicatorsfor Poultry v i s - h i s Brazil FromTable A.2.24, it can be seen that poultry has a marginally lower comparative advantage when confronted with competition from Brazil, but it stillretains a comparative advantage. Inthe caseofcattle, analysis ofthe costsofproductionof 17representative livestockoperations in various regions of the country revealed that the comparative advantage i s strongest for the most intensive operations, contrary to the conclusionof Misio'nPaz. Infact, for extensive cattle raising, the DRC indicator i s approximately unity, which indicates that its comparative advantage i s nil. Because many of these ranchers do not take into account the opportunity cost of their land, nor of the financial capital they put into the operation, they perceive a profit in cash-flow terms (financial terms), when infact they may be losing in economic terms, that is, in comparison with alternative forms of investing the resources. Nevertheless, the comparative advantagefor intensive operations, both for beef and for beef plus milk,i s real and substantial. The analysis for all livestock products taken together provides another confirmation of Colombia's comparative advantagein subsectorsthat embody more product processing. Becauseof particular structural barriers in world markets-the issue of sanitary certification for beef and poultry, plus the heavy international subsidies on milk-colombia' s comparative advantage in these products may not have an opportunity to be expressed in the form of more exports in the near future. However, should the sanitary hurdles be overcome, the potential for export growth in these products is considerable. Inthe meantime, 93 until better international arrangements can be negotiated, the distorted nature of international trade in poultry products appears to warrant continuation of a significant import tariff on chicken parts. 6. Forestry The forestry sector includes logs, boards, plywood, and wood manufactures. It i s estimated that in Colombia about 25 million hectares of land are appropriate for forestry (Misibn Paz 2001:120). Currently the country has about 8 millionhectaresof natural forests and forest plantations, of which only a little over 100,000 hectares are plantations. The timber industry i s poorly regulated, with a significant component of illegal, unregistered activity. Frequently the wood felled in Colombia is characterized by low yields and poor quality, and it often i s not delivered ina timely fashion to sawmills. Inspite of these deficiencies, very recently exports of wood products have begun to increase at an acceleratedpace (Table A.2.25). However, in species like pine and eucalyptus the natural growth of wood in cubic meters per year is higher than in exporting countries such as the United States, Argentina, and Chile (ibid.), so Colombia possesses a physical basis for competitiveness inthe sector. Table A.2.25. Relative Trade Balance for Forestry (trade inmillions of dollars, period averages) Item 1991192 1993195 1996198 199912001 11 Imports 7,382 28,219 28,336 21,932 Exports 13,360 9,808 13,470 26,658 Trade balance 5,979 -18,411 -14,867 3,726 Total trade 20,742 38,028 41,806 47,590 RTB 0.20 -0.48 -0.36 0.08 P Source: Espinaland Samaca (2003). In international markets wood products have been increasing at a substantial pace for the past 20 years, and it i s expected that growth will continue to be strong. InSouth America, Brazil and Chile already are large-scale exporters of forest products, and Argentina, Uruguay, and Paraguay are increasing their efforts in this area. Thus the market opportunities exist for an expansion of the forest industry. Among its benefits are considerably greater employment per hectare than cattle raising, and environmental benefits in controlling soil erosion and protecting watersheds. Watershed benefits, normally considered to be economic externalities, can be internalized in financial terms. In Honduras, a municipio on the Caribbean coast pays campesinos in the upper reaches of the watershed for forest protection services, and an internationalproject has been effecting compensation to campesinos for the same purpose in the watershed of the Cajbn dam. InPanad, a pilot project of the Inter-American Development Bank pays campesinos in the Darien US$250to US$300 annually for specified tasks in forest protection, with provision for quarterly monitoring of their fulfillment of duties (Misibn Paz 2001:169). In Costa Rica there are even more examples of developing markets for environmental services. Forests can also provide the setting for the provision of environmental services that have a market value, including sequestration of greenhouse gases and bioprospecting. As of mid-1999, 46 countries had projects for the capture of greenhouse gases, of which 14 are in Latin America, including Colombia. The potential market for the findings of bioprospecting i s also very large; internationalsales of medicinal plants amount to about US$14 billion. Costa Rica has been a leader inboth areas of carbon capture and organized bioprospecting. Therefore, there i s a wide variety of potential economic benefits from adequate forest 94 management; however, Colombia has progressed relatively slowly in this area. Increasingly, success in exporting forest products will require certification of sustainable forest management, and that in turn requires appropriate institutionaldevelopment to support the sector. For the present study, comparative advantage calculations were made for 12 forest species, with the results shown in Table A.2.26. Most of the species analyzed have a clear comparative advantage, more so than in broad-leafed species than in conifers or eucalyptus. Economic analysis thus points to potential strengths in the forestry area. In light of these results, it would be worth rethinking the incentives for reforestation, and also for proper management of natural forests. At the moment the fiscal incentive for reforestation i s tied to the availability of funding to back it, and there have been reports of viable reforestation projects being put on hold in the expectation that eventually sufficient funding will become available to allow the project to access the incentiveprovision. - -DRC TableA.2.26. Comparative Advantage Indicatorsfor Forest Species Specie DRC P Ceiba 0.44 Caribbean pine 0.72 Oak 0.47 Eucalyptus tereticornis 0.82 Gmelina arbo'rea 0.48 Pino oocarpa 0.84 Alder 0.55 Pinopdtula 0.90 ~ I Teak 0.55 Cypress 0.90 0.58 Eucalyptusgrandis 1.39 - 7. Other Products The seafood sector i s an important export earner inColombia, as shown inTable A.2.27. The most dynamic product inthe sector has,beencultivated shrimp, mainly produced along the Atlantic Coast. World demand for fish continues to expand rapidly, andocean and lake fish catches will fall short of demand by an increasing margin, owing to depletion of natural fish stocks. Consequently, there will be no limitation on the demand side to expansion of output. On the supply side, Colombia has natural advantages in shrimp cultivation on the Atlantic Coast. First,that area is less susceptible to diseases of the species than is the Pacific Coast. Second, Colombia's yields, at about 4 tons per hectare of pond, are far greater than those of Ecuador, at 0.85 tons per hectare. Ecuador i s far and away the largest producer of cultivated shrimp in the region, but in recent years it has experienced severe difficulties with viruses of shrimp. Colombia's high yields result from the fact that it i s a world leader in developing shrimp technology. It exports shrimp larvae to other countries. Its use of its own laboratory larvae i s the reason why it has been able to avoid the viruses that have severely affected the cultivated shrimp industry in many countries (Contraloria General... 2002:55). In short, the opportunities for Colombia in this field would appear to be great, and studies on the cost-competitiveness of Colombia's production and the relevant policy options should to be carried out.66 For very optimistic views of Colombia'spotentials inthis area, see Misi6n Paz (2001); andAcero (2000). 95 TableA.2.27. Exportsof PrincipalSeafoodProducts Other tuna, bonito 16.945 19.178 12.700 17.125 7.091 4.346 Frozen cultivated Source: Officialtrade statistics. Organic agriculture i s another area in which world demand i s increasing rapidly. Colombia's exports of organic products were valued at US$4 million in 1998, US$9 million in 2000, and US$11million in 2001. The most important organic products are coffee, palm oil, bananas, mango pulp, and sugar, although several others also are certified for sales as organic products.67 The area under certified organic cultivation i s still small, about 9,000 hectares of crops and 6,500 hectares of cattle and buffalo in 2001, but itis expanding rapidly. Ingeneral, organic farmingis attractiveto many smallfarmers becausethey do not apply extensive amounts of agrochemicals in the first place, so the shift to chemical-free cultivation requires little or no change in production techniques, and to the extent change i s needed, it implies a savings in monetary costs (Damiani 2002). In Latin America, Mexico i s a leader in organic honey, and the Dominican Republic, in organic cacao and bananas. The Dominican Republic now supplies 60 percent of the world market for organic cacao. Its organic bananas represent 70 percent of the country's banana production, and about 2,500 smallholders are involved intheir production(ibid). In some areas in Colombia, smallholders who already apply low dosages of agrochemicals are beginningto make the switch to organic production. These areas include the coffee-growing municipios of Riosucio and Supia in the western part of the country, the Sierra Nevada of Santa Marta, the Valle de San JosC, the Cordillera Oriental in Santander, and other zones. The financial potential for farming with organic methods can be seen in Table A.2.28, which presentsobservedprice differentials in the Bogota market for 28 products cultivated under the two different methods. The price differentials are very substantial, and this i s undoubtedly an area with considerable growth potential, inboth domestic and internationalmarkets. Forero (2003), using data from the Ojicina de Politica Sectorial del Ministerio de Agricultura y Desarrollo Rural, has analyzed the profitability of organic production compared to conventionalproduction inspecific cases. He calculatedthat the profit per kilo of organic mango production is 494 pesos, compared to 361 pesos for conventional production. The corresponding rates of return on total costs are 297 percent and 121percent. The organic production inAtlantic0 (on an 8-hectare farm) had costs that were 15 percent lower than those for conventional production in Tolima (on a 10-hectare farm), but yields that were 53 percent higher. Of course, differences in soil quality may also play a role in these results, but a similar outcome was found for organic panela. For that crop, the rate of return was 54 percent for conventional 67The information reported here on organic production is taken from Ministeno de Agricultura y Desarrollo Rural y Instituto Alexander von Humboldt, with the support of World Bank(2002). 96 production in Santander (10-hectare farm), but 65 percent for organic production in Valle del Cauca (farms of 1to 5 hectares) and 88 percent for organic production in Cundinamarca (farms of 2 to 10hectares). In the case ofpunelu, organic yields were only about half of those obtained under conventionalcultivation, but costs were lessthan a third of those for conventional cultivation. Organic production is expanding, partly with the assistance of nongovernmental organizations, some of which purchase and market the crops. Its future prospects are undoubtedly strong, especially for smallholders. However, continuingdevelopment in this area will require considerable effort in the areas of information and training, producer organization, and market organization. The ECOS project has developed detailed diagnoses of issues in organic agriculture and corresponding recommendations to promote it. Table A.2.28. Bogotii, Comparison of Pricesfor Conventional and Organic Agriculture *In stores inareas of incomelevel 4. PricesobservedinBogoti, June 14,2002. **For home delivery. The delivery charge is separate. Prices observedJune 12,2002. Sources: MADR and the Instituto Alexander von Humboldt, with World Bank support (2002:29). 97 Annex 3 FormulasUsedinthe Analysis of the Competitivenessof ColombianAgriculture The formulas usedinthe quantitative analysesare the following: a) Nominal protectionrates NPRi = pdjpbi,where pd i s the domestic price and pb is the world price equivalent (border price equivalent), adjusted for differences intransport and handlingcosts. b) Relativetrade balance RTB, = (x, -m,)/(x, + m,) where x, and m, refer to exports and imports, respectively, of product chain or group c. This indicator i s calculated for chains or groups of products and thus shows whether a set of related products i s more export or import oriented as a whole. If0.33