Report No. 32763-AR Argentina Agriculture and Rural Development: Selected Issues July 31, 2006 Argentina, Chile, Paraguay & Uruguay Country Management Unit Environmentally and Socially Sustainable Unit Latin America and the Caribbean Region Document of the World Bank TABLE OF CONTENTS ACKNOWLEDGEMENTS ...........................,.............................,...,..........................,....,...................................................v EXECUTIVE SUMMARY vii 1. INTRODUCTION ................................................................................................................................................. ....................................................................................................................................... 2. THE STATE OF AGRICULTURE INARGENTINA ........................................................................................ 31 THEEVOLUTIONOFAGRICULTURAL POLICY.............................................................................................................. OVERVIEW OF THE IMPORTANCE OF AGRICULTURE INARGENTINA'S ECONOMY ....................................................... 3 POTENTIALADDITIONAL PRODUCTIVITY GAINS ...................................................................................................... 4 11 NEW FORMS OF FARM ORGANIZATION ..................................................................................................................... 14 REGIONAL EFFECTSOF POLICY CHANGES ................................................................................................................ SUMMARY ............................................................................................................................................................... 19 17 3. SOME INTERNATIONAL.COMPARISONS .................................................................................................. 20 OVERALLPERFORMANCE ........................................................................................................................................ 20 AGRICULTURALPOPULATION .................................................................................................................................. LAND RESOURCES ................................................................................................................................................... 22 FACTOR Land Productivity ........................................... PRODUCTIVITY.......................................................................................................................................... 24 23 ................................... Labor productivity .......... ......................................... .......................................... ................................... 26 AGRICULTURAL EXPORT POSITIONINGAND MARKET PENETRATION ...................................................................... 29 SUMMARY AND CONCLUSION.................................................................................................................................. 34 4. THE ROLE OF GOVERNMENT ..................................................................................................................... .36 AGRICULTURAL FISCALPOLICY .............................................................................................................................. 36 Agricultural Taxation ....................................... .................................. Public Expenditure i .................................... AGRICULTURAL PUBLIC ..................................................................................................... 46 INTA ................................. ................................ SENASA .............................. .................................... .......................... 48 ANCILLARYSERVICES AND THEROLEOF GOVERNMENT ........................................................................................ 49 Infiastructure and Processing Facilities . ............................... ........50 Rural Finance....................... .................................... ............................... 52 GOVERNMENT A CATALYSTFORCOLLECTIVEACTION...................................................................................... AS SUMMARY AND CONCLUSIONS................................................................................................................................ 54 53 5. LABOR,INCOME AND POVERTYINRURALAREAS ............................................................................. .57 THERURAL LABOR FORCE ...................................................................................................................................... 57 Characteristics ...................... 57 Participation in .............................. ............................. LABOR EARNINGSINRURALAREAS.......................................... RURALPOVERTY ..................................................................................................................................................... 65 Incidence ................................... .............................. 65 Location ........................................ ................................... ................. 65 Characteristics..... ..................................... ........................... 66 Is Land a Poverty Trapfor Small Farmers? ... ............................. 69 SUMMARY AND IMPLICATIONS ................................................................................................................................ 70 6. THE CROP FRONTIER EXPANDS AND THE PAMPASINTENSIFY ...................................................... 72 HowMUCH THE FRONTIEREXPANDED? HAS ......................................................................................................... 72 Salta................ ...................................... ................................... 72 Eastern Santia rn Chaco .................................. .73 THEADVANCE SOYBEANCROPS: WHAT HAVETH 73 FACTORS EXPLAININGTHE LOCATIONSOYBEANEXPANSION............................................................................ OF ...................................................................... OF 75 Regression Resultsfor 1988 .......................................... ...................75 Regression Resultsfor 2002....................................... ...................................... Regression Resultsfor the Change in 1988-2002 ...................................... i Discussion.......................................................................................................................................................... 78 IMPACTON LIVING STANDARDS ............................................................................................................................... 79 INTENSIFICATION INTHE PAMPAS ........................................................................................................................... 80 SUMMARY AND IMPLICATIONS ................................................................................................................................ 83 7.FOOD QUALITY, SAFETYAND PHYTOSANITARYISSUES .................................................................... THEREGULATORY up........................................................................................................................................ 84 84 SET Institutions and Norms PHYTOSANITARY CONTRO ....................................................................... 86 TheNational Fruit Fly Control and Eradication Program (PROCEM) TheFoot and Mouth Disease (FMD) and PHEFA ......... The Carpocapsa Control Program (PLCC) and the Closing of the Brazilian Border to Pomaceous Fruit ......87 Phytosanitay Controls and the Export of Tucumrin Lemons . .............................................................. 90 TheImpending Threat of theAsiatic Soybean Rust ................................................................... Improving Knowledge on Pesticide Management and Occupational Safety Issues QUALITY, INSTITUTIONSAND THE DEVELOPM Diferentiation and Quality in the Wheat Industy Promises and Needs of the Honey Industy ........................................ 95 Economic Governance and the Unequal Regulatoy Situation in the Beef Industy ................... BIOTECHNOLOGY BIOSAFETY AND ................. LESSONS, CONCLUSIONS, AND RECOMMENDATIONS ............................................................................................. 102 Lessons on Sanitay and Phyto-Sanitay Issues........................................................... Lessons on Food Quality and Safety Issues.. Biotechnology ............................................... Final Conclusions and Recommendations.... 8.THE IRRIGATION SUBSECTOR .................................................................................................................... 107 IMPORTANCEOF THE SECTOR ................................................................................................................................ 107 THEPOTENTIALOFIRRIGATION ............................................................................................................................ 108 CONSTRAINTS TO ACHIEVINGDEVELOPMENT POTENTIAL .................................................................................... 109 Incomplete Legal Framework ........ ...................................... 114 Dificulties Related to the Eficiency of Irrigation SUMMARY OF IRRIGATION POTENTIALS AND CONSTRAINTS ................................................................................. 116 Potentials 116 Constraints ........ ......................... 116 POLICY OPTIONS ................................................................................................................................................... 116 9 SUMMARYAND CONCLUSIONS . .................................................................................................................. 120 SUMMARY OF FINDINGS ........................................................................................................................................ GENERALPOLICY CONSIDERATIONSAND SUGGESTIONSFORADDITIONAL WORK.................................................. 120 122 BACKGROUNDPAPERS 124 REFERENCES ........................................................................................................................................................ ..................................................................................................................................... ANNEXI:REGIONAL OVERVIEW .......................................................................................................................... 125 ANNEX11: INDICATIVEANALYSIS OFEXPORTMARKETSTRATEGIESOFTHEREGIONALECONOMIES ...............132 143 ANNEX111:THENEWAGENDAFORFOOD QUALITYAND SAFETY ANNEX IV: INDICATORS OFIRRIGATION DEVELOPMENT INARGENTINA'S PROVINCES 153 MAP OF ARGENTINA .......................................................................................................................................... ..............................................................................151 160 .. 11 ACRONYMS AAPROTRIGO Asociacibn Argentina Protrigo (Argentinean Asociation Pro-Wheat) AFINOA Asociacibn Fitosanitaria del N O A AMBE Agricultural Based Manufacture Exports ANMAT Administracibn Nacional de Medicamentos, Alimentos y Tecnologia MCdica (National Administration o fDrugs, Food andMedical Technology) APHIS Animal and PlantHealth InspectionService C A A Cbdigo Alimentario Argentino (Argentinean Food Code) CASAFE C h a r a de Sanidad Agropecuaria y Fertilizantes (Chamber of Agricultural Health andFertilizers) CNA Censo Nacional Agropecuario (Agricultural Census) CNLA National Foot-and-Mouth Disease Control Commission CONABIA Comisibn Nacional Asesora de Biotecnologia Agropecuaria (National Advisory Comisibn o f Agricultural Biotechnology) CONICET Consejo Nacional de Investigaciones Cientificas y Tkcnicas (National Council o f Research and Technology) COPROSAs Provincial Animal Health Commissions CORENOA ComitC Regional Fitosanitario delNOA EAP Empresaagropecuaria (Agricultural Enterprise) EU EuropeanUnion FA0 Food and Agriculture Organization of the UnitedNations FDI ForeignDirect Investment FUNBAPA FundacionBarreras Patagbnicas (Patagonic Barriers Foundation) FMD Foot andMouth Disease FOB Free On BoardPrice GDP Gross Domestic Product GM Genetically Modified G M O Genetically Modified Organism IASCAV Instituto Argentino de Sanidad y Calidad Vegetal (Argentinean Institute of Health andVegetal Quality) IDB InteramericanDevelopment Bank IIASA International Institute for Applied Systems Analysis INAL Instituto Nacional de Alimentos (National FoodInstitute) INASE Instituto Nacional de Semillas (National Institute o f Seeds) INDEC Instituto Nacional de Estadisticas (National Institute of Statistics) INTA Instituto Nacional de Tecnologia Agropecuaria (National Institute of Agricultural Technology) IP Indicacibn de ProcedencidSource o f origin IRAM Instituto Argentino de Normalizaci6ny Certificacibn ISNAR International Service for National Agricultural Research L A C LatinAmerican Countries LART Laboratorio de Anhlisis Regional y Teledeteccibn (Laboratory o fregional analysis and teledetection) NEA North East o f Argentina N O A North West o f Argentina OECD Organization for Economic Co-operation andDevelopment 111 ... OIE Organization ofAnimal Health PHEFA Plan Hemisfkrico para la Erradicacion de laFiebreAftosa (HemisphericPlan to Eradicate Foot and MouthDisease) PLCC PatagonianProgramto Battle Carpocapsa I PNCCA ProgramaNacional de Certificacion de Calidad de Alimentos PROCEM Programa de Control y Erradicacion de Mosca de 10s Frutos (FruitFly Control and Eradication Program) PRODERNEA Proyecto de Desarrollo Ruralde las Provincias delNoreste Argentino PRODERNOA Proyecto de Desarrollo Rural de las Provincias delNoroeste Argentino PROINDER Proyecto de Desarrollo de PequeiiosProductores Agropecuarios PROSAP Programa de Servicios Agricolas Provinciales R&D ResearchandDevelopment R H S RuralHouseholds Survey RNF RuralNonFarm SAGPyA Secretaria de Agricultura, Ganaderia, Pesca y Alimentos (Secretariat of Agriculture, Livestock, FisheryandFood) SICOFHOR Sistema de Control de ProductosFrutihorticolas Frescos (FreshFruit and Vegetable Control System) SNNCC SistemaNacional de Normas de Calidad y Certificacion TFP Total Factor Productivity UBN Unmet Basic Needs UNESCO UnitedNations Educational, Scientific andCulturalOrganization us UnitedStates ofAmerica USDA UnitedStates Department ofAgriculture WWF World Wide Fund iv ACKNOWLEDGEMENTS This report has beenprepared by a team led by Robert Schneider and comprising Dorte Verner, Jose Maria Caballero, and Marisa Miodosky. Peer reviewers were Derek Byerlee, Lucio Reca, Graciela Ghezhn, Albert0 Valdes, and Dina Umali-Deininger. JosC Maria Caballero took over the finalization o f the report upon the retirement o f Robert Schneider. We would especially like to thank Marisa Miodosky who tirelessly searched down data and information from our Argentinean colleagues, and never ceased to question. We are also most grateful to the team o f Argentinean consultants who contributed to the preparation o f the report. Without their involvement the report would not have beenpossible. Their names and areas o f contribution are given inthe section on References. Duringthe elaborationo fthe report we have received valuable comments from Carter Brandon, Edward Bresnyan, Jorge Caballero, Mark Cackler, Jasmin Chakeri, Estanislao Gacitua, Elsie Garfield, Juan Gaviria, Jesko Hentschel, FranciscoProenza and John Redwood. The team is grateful to the Secretaria de Agricultura, Ganaderia, Pesca y Alimentos for its support during the preparation o f the study. We would like to thank, inparticular, Jorge Neme, NCstor Murgier, Gabriel Parellada and Maria Anchorena Nazar from PROSAP, who provided strong support organizing the December 2004 workshop to discuss the consultants' reports, as well as the trips o f the team to the provinces. Gastcin Bordelois, Susana Aparicio, Susana Soverna, Alejandro Gerardi and M6nica Catania from PROINDER also provided important information concerning rural development programs and rural dwellers' life conditions. Carter Brandon oversaw relations with these andother government counterparts duringthe study. A number o f people deserve special recognition. The support and wise counsel of Raul Fiorentino has been invaluable throughout the process. John Young, on his own initiative, provided the market analysis o f wine and grapes following a brief discussion concerning the relative success o f Argentina and Chile in penetrating high value markets and we would especially like to acknowledge our debt to him. Janice Molina very efficiently translated under difficult circumstances and Nelvia Diaz and Florencia Liporaci provided valuable editorial and logistic support. V vi EXECUTIVESUMMARY Introduction 1. This report examines the performance and trends of the Argentinean rural economy. Its purpose i s to update the World Bank's understanding o f the forces shaping the country's rural economy, and to serve as a vehicle for dialogue with government and civil society on rural development issues. This is especially relevant in view o f the new Country Assistance Strategy finalized by The World Bank and the Government o f Argentina in June 2006. It is hoped that the report will help identify areas where Bank financial and other assistance could be o f value to Argentina over the medium term. The report i s intended as a vehicle for dialogue not a final statement o f Bankpositions. 2. The report i s not comprehensive and does not aim at covering all of the many aspects of agriculture and rural development in Argentina. It reviews selected areas o f major importance, and highlights others where further work might be warranted. In agreement with Government, we chose to emphasize issues relevant to the growing regional economies (i.e., those provinces outside o f the Pampas region). There are two main reasons for this. First, the incidence o f rural poverty i s much higher in the regional economies. Second, the potential for employment and income generation o f agricultural development in the regional economies i s large, and many opportunities exist to develop highvalue products. Some o f the issues that we have chosen to examine are food quality and phytosanitary control, irrigation, and the development o f improved export market participation. Our emphasis on the regional economies has not prevented our reviewing some themes pertaining to Pampean agriculture, such as the outstanding performance o f the grain and oilseeds sector and its contrast with the stagnation of the beef sector; issues related to infrastructure and processing facilities; and the expansion o f the crop frontier. The report i s based on a series o f backgroundpapers by both Argentine consultants and World Bank staff. The list o fpapers and authors' names are listedwith the References. 3. The growth of agriculture in Argentina over the last two decades is cause of satisfaction but should not give rise to complacency. Argentina has enormous agricultural potential and big natural advantages for the production o f many agricultural products. After decades o f slow growth, there has been a big leap forward in the last 15 years. This was the result o f more favorable policies and breakthroughs in farming technology. There i s cause for satisfaction but not for complacency, for several reasons: (i) there is still much to be done inthe grain commodity economy o f the Pampas with respect to infrastructure and processing facilities, quality improvements, environmental surveillance, and phytosanitary control; (ii) the beef sector has remained stagnant; (iii) highvalue crop production in the regional economies has expanded but at a much smaller pace, facing restrictions inthe irrigation andphytosanitary systems, lack o f term financing for farm modernization, and difficulties to organize collective action; and (iv) poverty continues to be important inrural areas. 4. Conditions are favorable for strong agricultural growth in the coming years. Regained macroeconomic stability, favorable international prices and the competitive edge givenbythe 2002 devaluation mark an auspicious opportunity for policy action to vigorously address the above issues. This would make agricultural growth more sustainable and equitable. vii 5. Threelinesof policyactionare highlightedinthis report: 0 There i s evidence o f underinvestment in agricultural public goods in the country. Progressively raising public expenditure in agriculture to levels similar to those o f its closest L A C competitors would do much to increase the competitive edge o f the sector. This has already startedover the last two years, andwe expect it to continue. 0 Mitigating rural poverty would require a combination o f direct and indirect employment generation, particularly in the regional economies, with improved access to and quality o f education and other services. Investing in irrigation, sanitary and phytosanitary systems, and research and extension for small farmers would be an effective way to help improving poverty conditions. 0 Collective action is increasingly critical in modern globalized agriculture, and is particularly significant for the development o f the regional economies. It requires (i) resolving conflicts in the value chains to allow producers, suppliers and buyers to work together to achieve and retain high value markets; (ii) creating and maintaining systems o f vigilance and enforcement o f phytosanitary protection norms; and (iii)effective decentralized local management o f irrigation systems. Summaryof TechnicalFindings Agricultural Performance 6. Agriculture has been critically important throughout Argentina's history. Deep soils, temperate climate, adequate rainfall and good access to sea freight endow Argentina with exceptional potential for agricultural production. This has permitted agriculture to perform well despite over 50 years o f little favorable policies. 7. The contribution of agriculture to the national economy is important. In 2004 the sector contributed 58 percent (US$13.1 billion) o f total goods exports, o f which 39 percent were primary products and 61 percent agricultural manufactures. Agriculture generated 9 percent of GDP and 22 percent o fthe value added o fthe goods sectors. Crops made the largest contribution (63 percent o f the total), followed by livestock (3 1percent). Agriculture is also a major source o f employment. In 1997, direct agricultural employment was about 1.5 million or 11.4 percent o f the total. When the employment generated inthe transport and commerce services related to food and agriculture is considered, employment generation raises to 20 percent (Obschatko, 2002). In 2001, up to 35 percent o f Argentina's population lived in predominantly agricultural zones or residedin small cities linked to agro-industrial activity. This figure increases to 45 percent if a broader definition o frural i s used(de Ferranti et al, 2005). 8. Notwithstandingits favorableconditions,the longterm agriculturalperformancein Argentinawas belowthat of neighboringcountries.The value o foutput increased 158percent in 1961-2005 inArgentina compared to 241 percent inChile and439 percent inBrazil. Over the last 15 years, however, Argentina's agriculture experienced a big leap forward; it grew considerably faster than Brazil's and Chile's during the six-year period from 1993 to 1999, stagnated as Argentina's real exchange rate became uncompetitive during the 1999-2001 pre- devaluation period, and recoveredagain after the devaluation. viii 9. Historically, the Regio'nPampeana has dominated agricultural growth. The Pampas are undergoing a rapid process o f farm consolidation, driven by cost reducing and labor saving technological change, and by new contractual instruments (generically called "planting pools") for combining land, machinery, and high quality management. These pooling arrangements have improved the overall level o f management as well as risk diversification. Primary agricultural exports from the Pampas, especially in grains and oilseeds, grew 46 percent between 2000 and 2004, compared to 29 percent for the rest o f the country. By 2004 79 percent o f total agricultural-based exports came from the Regidn Pampeana. In contrast, the non-Pampean regions have taken the lead inagriculture-based manufactures. 10. The current trend of frontier expa.nsion and farm size increase i s driven by competitive forces and is likely to continue. It is based on technological change, revolution in farm organization and management, and low unit margins in commodity production. Land in annual crops in Argentina grew by 5.5 million hectares in the inter-census period 1988-2002. This expansion took place through intensification (mostly reducing pasture in rotations in the Pampas), and expansion into new frontiers. While prior to 1988 soil characteristics appear to have been the dominant determinant o f frontier conversion, they do not appear to have significant influence in more recent years, presumably due to the spread o f no-till technology which permits cropping on hillier, drier, and shallower soils. 11. Available evidence does not reveal a systematic positive or negative social impact of frontier expansion at the macro level, although the local impacts may be relevant. The environmental impact, however, may be severe as soybeans production expanded mainly on natural vegetation. Thus, 86 percent o f the 1988-2002 frontier expansion in the Chaco and Santiago del Estero occurred in areas previously covered by natural vegetation, and only 13 .percent incropped areas. 12. The beef sector has experienced long-term stagnation in production, exports, and by all available productivity measures, although there has been some recovery in recent years. Between 1961 and 2002, land productivity in livestock grew only 50 percent in Argentina, as compared to 150 percent in Chile and 300 percent in Brazil. Since most farming in the Pampas consists o f mixed crop-livestock operations it i s unlikely that stagnation is a reflection o f management deficiency. It i s more likely that it reflects (i)the very high and competing profitability o f grains and soybeans (especially since the wide-scale adoption o f no-till planting and pooling arrangements), which has moved farmers to concentrate attention on crops and convert good pasture land to cropping; (ii) perception o f greater risk associated with cattle-- the especially since the foot and mouth disease outbreak o f 2001; (iii)the effect o f unstable macroeconomic conditions and high interest rates, which affect beef production more than the crops sector due to the longer term nature o f the economic payoff o f beef; and (iv) some governance problems in the beef chain. The livestock sector i s an area where further research i s recommended inthis report. AgriculturalPolicy and ServiceInstitutions 13. The correction of the urban bias in the nineties produced favorable results in agriculture. The early nineties was a period o f correction o f the urban bias that traditionally ix dominated policy, as well as o f technological development and export market breakthrough. The result was a boom inagricultural exports, mostly from the traditional commodity exporters o f the Pampas but also from high-value exporters o f fruits, vegetable and wine from the regional economies. The period 1991-2001 was tumultuous for Argentinean farmers. It witnessed liberalization, privatization, dollarization, and the coming into force o f MERCOSUR, followed by the emergence o f macroeconomic imbalances that resulted in currency overvaluation and credit unavailability. For most farmers these events dramatically changed their economic environment. Driven by favorable external and domestic markets, agriculture expanded rapidly through 1998. The sector benefited from the elimination o f quantitative restrictions on imports, reductions on import taxes for fertilizers, herbicides, pesticides, machinery and irrigation equipment, the elimination o f taxes on fuels, commercial and financial transactions and exports, the deregulation o f economic activities, and the removal o f inefficiencies and monopoly profits inthe trade channels (elevators, transportation andports). The result was an impressivefive fold increase in fertilizer use, and a three fold rise in the use o f herbicides and pesticides. Duringthe period 1988/1990 to 1996/1998 wheat, corn, soybeans, and sunflower average yields increased 26, 43, 7, and 25 percent, respectively. Land sown with the 31 principal annual crops expanded 25 percent, and annual output growth was 7 percent. 14. However, macroeconomic imbalances towards the end of the decade led to a prolonged recession, and subsequently to capital outflows and a financial crisis. Increasing overvaluation dramatically reduced the margin for Argentinean commodity producers in the Pampas and the value o f exports from the regional economies. At the same time, growing recession and unemployment depressed the domestic demand for income-elastic foods such as dairy, fruits and vegetables. Finally, the devaluation o f the Brazilian Real in 1998 reduced the competitive position o f Argentinean farmers within MERCOSUR. Agricultural profitability was reinstated by the 2002 devaluation upon the abandonment o f the currency peg, and recovery was strong. Agriculture-based exports, which had either fallen or showed insignificant growth throughout 1999-2001, grew by 26 percent in 2002, 27 percent in 2003 and an additional 13 percent in2004. 15. Agriculture has traditionallypaid a considerableamount of taxes in Argentina.Tax pressure on the sector in 2003 was 26 percent, similar to that o f the national economy. Most taxation (97 percent in 2003) comes from national -- as opposed to provincial -- taxes. Tax pressure on agriculture increased much upon the 2001 crisis, mainly as a result o f the introduction (or more exactly reintroduction) in 2002 o f export taxes or retenciones. In 2003, export taxes accounted for 43 percent o f all taxes paidby the sector. The incidence on agriculture o f national taxes is rather different from that on other sectors; proportionally, more i s paid by agriculture in export and income taxes and less in VAT and social security contributions. Various elements in the tax structure are distortionary, chief among which is export taxation. Agricultural taxation i s one o fthe issues flagged inthe report as needing further analysis. 16. Public expenditure in agriculture is very low. There is an imbalance between what agriculture contributes to and takes away from the public purse. In 2005, the main national entities o f the agricultural public sector received only 0.8 percent o f all national government expenditure, equivalent to 1.4 percent o f agriculture GDP. In real terms and as a share o f agriculture GDP government expending in agriculture fell drastically with the crisis. A International comparisons confirm the low level o f spending in the sector. For the average o f 1996-2000 the government expenditure per agricultural worker was substantially less in Argentina than in Brazil or Chile, and less than average in LAC. The "agriculture orientation index'' shows also more discrimination against agriculture inArgentina than inBrazil or Chile or for the L A C average. It is estimated that some 37 percent o f government expenditure in agriculture in 2003 was inprivate goods, which inview of the extreme shortage of public funds going into the sector may result init being deprived o f essential public goods. 17. INTA and SENASA are the largestgovernment institutionsprovidingvital services to the sector--agriculture technology research and extension the former, and phytosanitary protection and food quality and safety services the latter. They are both decentralized public institutions operating under SAGPyA. They jointly accounted in2003 for 46 percent o f national government spending inagriculture and73 percent o fthat estimated for public goods. 18. INTA can claim a number of important successes over the last years related to the introduction o f zero-tillage andbiotechnology, and the improvement o f management practices by small and medium farmers. Many challenges remain, however, to keep pace with the deep changes taking place in agriculture in Argentina in the last 15 years. INTA's strategic planning faces up to these challenges but several areas seem to require more emphasis. They include environmental sustainability, irrigation, bio-energy, quality systems, and the development o f non-traditional and specialty products. 19. SENASA has good technicalcapacity and has been able over the years to establish sound regulatory systems and a number o f valuable regulatory and policing processes. But there are also areas which could benefit from a strategic reassessment o f SENASA's role. SENASA's weaknesses are in strategic planning, communications and information management, and capacity to create and sustain partnerships at the local level. SENASA's management has shown awareness o f these deficiencies, proposing a strategic reassessment o f the organization's role that substantially addresses these weaknesses. The challenge i s also to change the institutional culture towards more focus on "making things happen" rather than "doing them yourself'. 20. Two essential services to ensure competitivenessof farming and the supply chains are infrastructureand processing facilities, and rural finance.Limitations inthe provision o f these services have historically affected the performance o f the sector. Infrastructure and processing facilities and rural risk and credit markets are also areas flagged for further research. 21. There are important needs in infrastructure and processing. They refer to grains, meat, and fruit and vegetables, whose requirements are different. There i s room for improvements in each o f these product groups, which would reduce costs, improve quality and hence prices, expand export markets, and allow the development o f new production areas. A joint public-private strategy would be useful to develop these services, together with a national program to organize the actions o f the public sector. The Ministry o f Planningi s taking welcome steps inthis direction. xi 22. Shortage of bank lending, particularly term lending, is a historic problem in Argentina's agriculture. The pool arrangements that have flourished in the Pampas are to a large extent an answer to this. Lack o f term financing has historically held back the modernization o f regional agricultures. Small farmers are especially deprived o f access to loans. Conditions became especially dramatic upon the crisis, but have slowly started to improve since 2003. RegionalAgricultures 23. The competitiveness of the regional economies is important for the growth of agriculture as a whole. Regional (or non-Pampas) agricultures have a wide range o f agro- ecological conditions, crops, and irrigated and rain-fed farming. On the whole, regional farmers do not enjoy the advantage o f the particularly favorable natural conditions o f their counterparts inthe Pampas. Hence, an exchange rate that allows most Pampa farmers to be competitive may not allow many regional farmers to compete internationally in favorable terms. Enhancing competitive conditions in the regional economies is therefore particularly important. Public policies in areas like technology generation and transfer, irrigation development and management, sanitary and phytosanitary systems, market development, collective action within the production chains, and access to medium-term credit for farm improvements, are particularly relevant to breaking production constraints inthe regional economies. 24. Regional agricultures have a strong potential to directly and indirectly increase rural income and employment. Rural poverty alleviation and the future o f the Argentinean small farming community depend much on the performance o f the regional economies. Employment inthe production chains of fruits, vegetables and wine i s estimated at over 800,000 workers, slightly under a quarter o f workers in primary and agroindustrial employment (Llach, Haniague and O'Connor, 2004). A 30-40 percent increase inemployment inirrigated agriculture appears feasible through a combination of area-expanding and quality (and marketing) improving measures mentionedbelow. 25. Policy reforms and macroeconomic shocks in the post-reformperiod had varying effects on the regions and provinces. Most, but not all provinces exhibited sharp growth in agricultural exports in the 5-7 post reform years, but stagnated or declined as the Peso became overvalued and credit tightened in the late-1990s. In general, the more outward-oriented a province, the more successful it was in responding to new opportunities opened up by the 1991 reforms, and the more successful inrecovering from the post 1997 crisis. 26. Excellent progress was made in exporting regional agriculturalproducts into high value marketsin 1993-97,with the export value per hectare o fArgentina's mainirrigated crops (hit, vegetables and wine) reaching 95 percent o f those o f Chile. The subsequent crisis did serious damage to these exports, however, and in 2002 that value was less than half o f Chile's. Had Argentina's unit export value continued to catch up with Chile's, the value o f Argentina's exports o firrigated crops in2003 would have beennearly US$950 million higher. xii Rural Incomes and PovertyIssues 27. Argentina's rural society is undergoing a process of transformation. Demographic changes in Argentina and changes in the characteristics o f the labor force reflect a rural society in transformation, a view reinforced by the increasing extent of migration. There is a strong feminization o f the labor force, with women showing 48 percent participation. Agriculture i s the dominant source o f employment in dispersed rural areas (72 percent), but the rural non-farm sector is also very important for income generation and poverty reduction. Thus, the estimate from a 2003 survey o f 441 rural households o f the share o f non-farm income, in the income o f poor rural families i s 46 percent, while inthat o f rural non-poor families it is 32 percent. 28. Key factors explaining access to non-farm employment in rural Argentina are education, age, land access, location, and gender. There is strong evidence that educated people have better prospects o f accessing non-farm employment, particularly in the better-paid activities. Being a woman increases the probability o f employment innon-farm activities, and so does age. The opposite i s the case with access to land and farm size. Workers in poor regions are more likely to participate inrural non-farm activities than those inricher regions. 29. Farm size, infrastructure, technology and gender of the farm head are important correlates of farming incomes. At average values o f other variables, income increases with farm size, with larger farmers earning dramatically higher incomes than their smaller counterparts. Gender also matters, although not in an important way; farms runby women yield 7 percent more income than those runby men. Whether the farmland i s rented or owned, on the other hand, shows little measurable effect on income. Not surprisingly, access to infrastructural services (paved roads and electricity) and use o f land-augmenting techniques (fertilizer and irrigation) are important for income generation. 30. There are some 200,000 indigent poor families in rural Argentina. These families tend to be large and young, and they tend to escape from extreme poverty as they mature and children leave the household. They live largely inthe Northeast and Northwest and in dispersed areas where basic service provision is weak and delivery i s difficult and expensive. It i s noteworthy that education attendance beyond 11 years o f age in dispersed areas falls off very rapidly compared to grouped rural or urban areas, reducing the mobility and employability o f youth from dispersed areas. The indigent poor in dispersed areas are significantly more likely to be small landholders thanwage laborers. 31. Out-bound rural migration is significant. Dispersed areas lost 14 percent o f their population over the last decade, reaching 2.6 million in 2001, compared to grouped rural areas, which experienced an 8 percent increase and reached 1.2 million. Around 400,000 people left dispersed rural areas during 1991-2001. Roughly, some 25 percent may have moved to grouped rural areas and the rest to urban areas. 32. Migration is not necessarily an indication of poverty. Inthe dispersed rural areas of the Pampas, high education levels, even among small farmers, combine with an active land market to lure small and mid-size farmers o f f the land, while small farmers without assets remain. In these cases, out-migration should not be taken as prima facie evidence o f decreased xiii well-being. On the contrary, migration may be evidence o f a farm family having achieved a critical level o f assets to successfully make the transition to non-farm employment. We also hypothesize that small farmers left behind tend to be o f below average educational attainment relative to the leavers, living in areas where land values have risen relatively slowly and facing an increasing loss o f public services as neighboring communities depopulate. Inthe Northeast and Northwest, on the other hand, there is evidence (e.g. Chaco) that many farmers leave the farm under conditions o f duress, without human or financial capital to smooth the transition. Thus, inareas o f highagricultural value, especially the Pampas, small farmers capitalize out (or rent out) as technological and institutional change increases the value o f their land beyond the net present value o f their own production. Onthe contrary, inareas o f low agricultural dynamism small farmers have neither buyers nor renters to finance their exit, nor do they have human capital to apply to non-farm employment. Sanitary,Phytosanitaryand Food QualityIssues 33. Argentina has been remarkably active in improving its regulatory framework for food quality and safety over the last decade. The new norms refer both to quality standards voluntarily adopted by producers, and to sanitary norms o f compulsory observance. Contrary to quality regulations, which are rather recent and o f new design (including voluntary ones), phytosanitary regulations have a long history in Argentina. Much work has been done nevertheless inrecent years to update the regulatory framework related to sanitary systems so as to respond to the needs posed by new production technologies, laboratory methods, and consumer demands. 34. Progress has also been made in several aspects of voluntary quality regulation of foods, such as the 2001 law establishing the "Indicacidn de Procedencia" or source indication, and "Denominacidn de Origen" or origin denomination. Other voluntary regulations that have been strengthened regard the certification o f organic products, the promotion o f food quality through the Sistema Integrado de Calidad INTA project, and the establishment o f ajoint Sistema de Normas Agroalimentarias, Insumos Agropecuarios, Produccidn Forestal and Maquinaria Agricola to set standards regarding processes, products andassociated services. 35. In spite of this progress there are still possible improvements in the SPS and quality systems. Some quality and safety issues remain, such as the insufficient differentiation o f standardqualities invarious products, and the different standards applied to export and domestic markets.The need to strengthen phytosanitary controls is underlinedby the cost for the regional economies o f phytosanitary problems. Carpocapsa costs Patagonia and Cuyo an estimate o f US$19 million per year; Fruit Flies cost the Northeast, Northwest, and Cuyo regions some US$21 million; and Canker costs the Northwest region an estimate o fUS$37 million. 36. The Foot and Mouth Disease (FMD) is a major threat to Argentinean livestock production and an area of active government intervention. Argentina's experience with FMD control leaves interesting lessons regardinghow a program based on a regionally concerted effort and strongly supported over a number o f years with adequate legal, technical and financial means finds success, but reversal is still possible if support weakens. Lessons include: (i) the importance o f the involvement o f the private sector through the COPROSAs in achieving xiv success; (ii) importance o f collective action at the regional level epitomized by the PHEFA the program; and (iii) failure to maintainthe conditions leading to success turnedback the clock how for FMD control as shown by the 2001 crisis. The failure o f Argentina to report promptly the 2001 FMD outbreak undermined the credibility of the regulator, and harmed market access not only for beefbut also for other agricultural goods. 37. The use of genetically modified seeds expanded in Argentina at surprising speed. Biotechnology erupted in Argentine agriculture in the second part o f the 199Os, linked to soybean cultivation and the use o f the transgenic Roundup Ready (RR) variety resistant to the herbicide Glyphosate. The fast dissemination o f the GM variety i s a success story in contemporary Argentina agriculture. In 2003 Argentina had 13.9 million hectares planted with GM soybeans, second inthe world to the US, which had 42.8 million, and ahead o f Canada (4.4 million), Brazil (3.0 million), China (2.8 million) and Australia (2.1 million). By 2004, 90 percent o f the area planted with soybeans, 50 percent o f that with maize, and 30 percent o f that with cotton usedGMvarieties. 38. The institutional management of the biotechnologychallenge in Argentina may be considered an international case of best practice. The core o f this response was the establishment o f a regulatory framework which defined clear steps for the release o f GM seeds with clear institutional responsibilities within each step. Risk analyses were systematically conducted to ensure the protection o f public health, the environment, and national commercial interests. The system makes good use o f the comparative advantages o f the relevant entities in the country.- public, private and academic. The use of the genetically modified RR soybean seed has been, however, accompanied by conflict with the owner o f the GM technology, Monsanto, over the payment o f intellectual property rights. 39. Quality andproduct differentiationare importantchallenges.Even though Argentina has a longtradition as an exporter o f agricultural commodities, quality competition inthe form o f market differentiation and niche targeting is relatively recent. There are some important successes, for example in "green beef' and Plata maize, but additional opportunities exist in major commodities like wheat, where the introduction o f graded grain i s slow to materialize in the production chain. In the beef industry, lack o f a standard objective classification of product types and qualities, and the application o f different regulatory systems to export and domestically oriented production, undermines export opportunities. 40. Quality and product differentiation are particularly important in the regional economies.Whereas market differentiation maybe an option for higher profits inthe Pampas, in the regional economies it is closer to a question of economic survival. Unlike the Pampas, the regional economies do not have a clear comparative advantage on either the cost or quality dimension. For fruits and grapes, at least, Argentina has not had in place the kind o f well defined marketing strategy characterizing other world market leaders. While at times Argentina appears to compete on cost inthe commodity markets, at other times it appears to have attempted to move into higher value markets. xv Irrigated Production 41. Irrigated production is an important part of all agricultural production in Argentina.Argentina's irrigated area reached 1.7 million hectares in2001. Some 70 percent o f this area (1.1million has) is located inthe arid or semiarid regions. Fruits and vegetables make up the most important share (28 percent) followed by sugar cane, tobacco and aromatics (15 percent), forage crops and cereals (14 percent), orchard crops (12 percent), citrus, olives, nuts (3.3 percent), cotton (1.9 percent), forest crops (9.1 percent), and others (16 percent). Interms o f economic value, production from irrigated lands represents some 26 percent o f Argentina's total. In several arid provinces such as Mendoza, Rio Negro, and San Juan, agriculture is virtually dependent on irrigation, with irrigated production surpassing 90 percent o f total output value. 42. Productivitylevels and water managementare less than adequatein most irrigated areas. As indicated above, if Argentina received the same export value in h i t s , vegetable and wine per unit o f irrigated land in these crops as Chile, it would gain an additional US$950 million per year (over 70 percent). This would require a combination o f better water control, better phytosanitary control, and better marketing. 43. Additional value and employment from irrigation could be obtained by bringing new land under irrigation.Preliminary estimates indicate that an additional 30-40 percent area could be brought into irrigation with investments in the public irrigation system in the range o f US$lOO-200 per hectare. Additional area in the order o f 20 percent could be irrigated through improved on-farm efficiency from the use o f pressurized irrigation. Complementary benefits from pressurized irrigation are the improved quality o f production due to better water control, andthe eliminationo fwater-logging and salinization. 44. The potentialeffect of increasingly high-valueirrigatedagriculture is exemplifiedby the Cuyo region. In Cuyo, consolidation o f farms has led to much less out-migration than elsewhere, due to the labor intensive nature o f irrigated agriculture, especially inthe production o f highvalue crops. As a result o fthis expanding activity, fanners leaving independent farming inthe Cuyo have a muchhigher probability o f finding wage employment inon- or off-farm occupations. InMendoza, the Cuyoprovincemostsuccessfulindevelopingirrigatedagriculture andpromoting production for highvalue markets, population increased in both dispersed and grouped rural areas despite a 13percent reduction inthe number o f farms. 45. The constraints to the full benefits of irrigation potential are more social than physical.They relate to insufficient decentralization o f authority and responsibility to the water users and to the consequent persistence o f a culture o f dependency. This i s aggravated in some places by low profitability o f agriculture related to small size o f holdings and low human capital. Institutional support to irrigation i s generally weak. xvi PolicyIssues andOptions Genera1 Considerations 46. Public action could be important to encourage further development of the rural economy in Argentina. We see three areas for such action: (i) establishing and upholding rules o f the game that are stable and fair and contain incentives that benefit the wider social good; (ii) investing in public goods, including agricultural research and extension, sanitary and phytosanitary systems, infrastructure, education, and health; and (iii) adopting a clear strategy to support the development o f local and regional collective action, especially in cases where the interests o f individual links inthe marketing chain prevent the competitiveness and development o f the entire chain. Expansion of the Crop Frontier 47. The current rapid expansion of the crop frontier raises social and environmental policy concerns. The social concern is over the potential effect o f frontier expansion displacing more labor intensive activities and eliminating jobs. While there appears to be no imminent social threat at this time, it would be usefbl to monitor the evolution o f social welfare in frontier expansion areas. This would require improving the collection o f rural socio-economic data through household surveys. The environmental concern refers primarily to the conversion o f forest ecosystems to cropland. The design and implementation o f provincial and eco-regional plans to better ensure the conservation o f critical representative ecosystems, particularly in the highly threatened Chaco ecosystem, is an option to address this environmental issue. Sustainabilitv concerns exist inconnection with resource management, whether o f soil, pests, or pesticides themselves. While no-till cropping has made possible the cultivation o fhillier land and shallower soils, threats o f erosion and declining yields due to nutrient depletion remain. In the Pampas, there i s evidence o f a shift to high intensity rotations, especially under annual rental contracts and pool arrangements. The establishment o f a credible monitoring system o f trends in landmanagement and soil conditions from an agricultural sustainability point o f view would be an option to address this issue. Such a monitoring systemwould provide credible landresource information to farmers' organizations, extension workers and landmarket participants. Poverty and Rural Development 48. Poverty reductionstrategiesare different in the Pampas region and in the regional economies. Poverty reduction policies in the Pampas are conditioned by the trend towards increased farm size and "pool" arrangements. The focus could be on the small and medium farmers left behind, inorder to help them make the transition to better paying occupations. Inthe regional economies, the focus could be on agricultural intensification and the development o f highvalue crops. This could be achieved through improved irrigation and phytosanitary systems, technological support, and support to access investment credit. Pursuingrural non-farm growth i s another option for poverty reduction inthe regional rural economies. Inview o f strong synergies inthe development of the farm andnon-farm sectors, this option could be pursuedjointly with agricultural development. xvii 49. Poverty reduction requires actions along several lines. The strategic principles for reducing rural poverty involve strengthening the key assets o f the poor, and taking into account geographic differences inthe poverty situation andpriorities. Four areas could be considered for policy action: (i) targeting indigent households, particularly young large families, and linking income transfers to education through social programs such as Becas and Familia; (ii) increasing access to productive inputs for small firms and farmers through research and extension, more access to land, and rural finance systems tailored to the needs o f small producers; (iii) investing injob creation inthe regional economies to benefit the unemployed and those trapped in low- productivity farms or low paying, low productivity jobs; and (iv) developing and institutionalizing a national framework for provincial and regional rural development and poverty alleviation. In addition, we recommend that a nationwide representative rural household survey be conducted, with the objective o f identifying the family and farm determinants o f poverty and migration inrural areas. Food Safety and Phytosanitary Issues 50. Improving food quality and safety systems is crucial for reaching Argentina's potentialin the productionand export of food.Argentina is a world agricultural power, but it still has potential to increase its participation inworld markets as one o f the leading suppliers o f a number o f agrarian products (soybeans, wheat, maize, beef, temperate fruits, citrus fruits, honey) and a medium supplier o f others (tobacco, sugar, rice, pulses, milk, garlic, onions, groundnuts, cotton). Enhanced quality and safety systems are fundamental to achieving such potential and also maintaining the participation already gained inworld markets. The importance o f this i s underpinned by the increasing importance o f non-tariff barriers. The investments requiredare not particularly highwhile the pay off is large; they are mostly for the fine tuning o f the research, regulatory and surveillance systems already in place. Efforts are also required to strengthen the governance systems o f the productive chains. 51. It would be usefulto defineclear roles and responsibilitiesamongprovincial,public, andprivateagencies, and allocate the budgetaryresources required.Thus, current programs to combat pests, especially the Fruit Fly, Citrus Canker, and Carpocapsa, could be strengthened by: (i) extending their scope to the entire country through national legislation; (ii) defining an institutional structure that includes provinces and production and marketing agents; (iii)a decision on the techniques and procedures to be used in combating pests accessible to all producers and including complementary actions (for example, barriers); and (iv) a suitable budget and identified sources o f financing. A clearer division o f functions and responsibilities between SENASA, provincial authorities and private actors would also be useful. However, a strong national-level surveillance and enforcement o f sanitary and phytosanitary regulations i s a critical, non-delegable attribution o f this government entity. 52. Other areas of SENASA responsibilitythat couldbe strengthenedare: Budgetary processes. A budgetary review and reform could include: (i)transparent budgets by program; and (ii) definition o f tasks for which a service fee will be clear charged. xviii Quality control of commercially marketed fresh products. This control i s currently conducted on a small proportion only o f the total marketed. More support could be given to the efforts o f SENASA to improve the Fresh Fruit and Vegetable Control System (SICOFHOR). First, however, it would be convenient to analyze the alternative systems. Legalframework for pesticides and worker safety. The current legal framework on these issues is cumbersome and at times inconsistent and o f doubtful validity. It would be useful for SENASA to review all o f its current decrees and resolutions, and a similar analysis could be conducted by the Ministryo f Health. Combating Carpocapsa, particularly in abandoned orchards. Stronger incentives could be created for owners to eradicate abandoned orchards, either positive through community-based incentives for eradication or negative through the strengthening o f the police power o f the state. 53. Some of SENASA's functions could be transferred to the SAGPyA. An option to strengthen the capacity o f SENASA to carry out its regulatory and surveillance functions would be to transfer policy and coordination functions to a Plant Protection and Quality Enhancement Unit in SAGPyA. Policy and coordination functions demand different skills and often different mentality than those required by police functions. This streamlining o f functions would allow SENASA to focus on the technical tasks it does best, and be consistent with SENASA's decentralized institutional structure. It would also create a specialized unit in SAGPyA to focus on the promotion o f collective action (for pest control, food safety and quality, and worker safety), removing potential conflicts o f interest. While SENASA could maintain its role in negotiating international agreements, it could work with the SAGPyA unit to ensure that local entities participate inthe definition o f standards andprotocols. Irrigation 54. In spite of irrigationbeing a provincialmatter, the role of the FederalGovernment i s critical to help provinces to plan and execute water strategies. Lack o f action by the Federal Government would perpetuate current provincial weakness. Federal intervention, however, would be more useful if carried out within the context o f a well-articulated strategy to devolve authority and responsibility to the local water associations and to create the national incentive framework to support it. The aims could be to (i) increase the number o f water associations in the provinces; (ii) increase the involvement o f water associations in the areas o f planning and management o f investments; and (iii)strengthen the institutional and financial autonomy o f the provincial water resource andirrigation agencies. 55. In the long run the objective could be envisaged of weaning the water associations from dependence on government. This would require them to develop a business plan and establish a clear "track record" in meeting the plan's goals. Over the medium term the water associations could aspire to qualify for commercial credit. Government investment and assistance could take place within the context o f these business plans, with a clear strategy for the water associations to develop independence and autonomy. 56. The cost of water should be calculatedfollowing a standard methodologyin all the provinces,with due adaptationsfor specific circumstances.The calculation should be made xix even inthose schemes where water is highor fully subsidized, and irrespectively o f whether this is likely to continue. The existence o f water subsidies (that would accrue to certain groups inthe form o f water prices below the fillwater cost) could be supported by a detailed analysis, with fillinvolvementoftheprovinciallegislature. 57. The rehabilitation of water works is needed but it has risks. The Undersecretary of Water Resources i s currently preparing a list o f priority rehabilitation works for each province and a list has also been prepared by PROSAP. This information could be used to prepare a program of investments. It is worth noting the danger, however, o f continuing the pattern o f periodically financing deferredmaintenance through emergency rehabilitation projects. 58. More research on irrigation related issues is needed. INTA and the provincial universities could help provincial governments and water users to prepare programs o f research that attend to the crucial aspects o f irrigation development, including (i)briefbut comprehensive provincial diagnoses; (ii) improvements in technical efficiency; (iii) management systems; and (iv) the social impact o firrigation. Possible instruments to bring about the above aims are as follows: A program o f activities from the national level intendedto orient and support provincial and local governments inthe definition o f strategies, targets, and objectives. Strengthening provincial capacities to manage associative and collective organizations. Detailed master plans for irrigation development in each province, specifying areas to develop, populations to attend, and commercial and institutional strategies to increase productivity. A national program to establish information management systems oriented at improving irrigation management. Water cost calculations for all irrigation systems. The water price could be fixed below cost only as an exception related to important reasons (rural poverty, emergencies, etc.). A clear definition and implementation of a system o fpremios and castigos directed at promoting filland timely payment o fwater charges. Establishing quality control programs for the irrigation systems ineach province. Market Development and CollectiveAction 60. Supporting collective action in the supply chains is one of the most strategic roles of government, impelled by globalizations, which internationalizes procurement standards not only for exports but also for domestic supermarkets. The need o f good economic governance o f the supply chains through collective action is underscored by the enhanced importance o f co- dependency in the traditionally antagonistic relationship among the actors o f the chains. Decentralization o f service provision could focus on promoting incentive-compatible collective actions. Public sector institutions like INTA and SENASA could be at the forefront o fthis. xx Suggestions for FurtherWork 61. Inthe course ofpreparing this review, we have come across several important topics on which there i s lack o f information and analysis. This gives them priority for study. Four of these topics are: Additional work on agricultural taxation and non-fiscal transfers. Although this study contains a review o f fiscal transfers to and from agricultural (Background Paper 2), we recommend a more fully-elaborated treatment o f both fiscal and implicit transfers to and from agriculture, including a more complete analysis o f the issues o f export taxes, tax evasion inthe sector, and local and provincial tax effort. This analysis needs to be carried out within the context o f Argentina's overall tax system and fiscal equilibrium, and address the issue o f regional inequalities and the large regional variation ineconomic rent to land. 0 Risk and credit markets. Lack o f credit is a major barrier for adoption o f improved technology inthe small and medium farm sector. Markets for hedging price and exchange rate risk and crop insurance, appear to be readily available to large farmers but not to the small ones. These markets need to be better understood, especially in the context o f improving the competitiveness o f small andmedium size farms. 0 The livestock sector. The potential to increase beef production in the Pampas and outside the Pampas is high. This i s because much o f the sector operates at medium and low technology levels, and the output gap between productivity levels in cattle production is high--much higher than in crops. Raising productivity is important because the sector i s large, and a moderate raise in productivity would result therefore in a large increase in aggregate output, Also, it i s inthe smaller farms where the productivity gaps are largest. Further study o f the reasons for this sector's relative stagnation, compared to the grain sectors, is warranted. 0 Rural growth and employment creation. Recent Argentine agricultural growth, especially in the Pampas, has not led to significant employment generation due to its relatively high capital intensive and land extensive nature. The potential for higher rates o f employment creation in the regional economies i s higher. Additional analysis o f the possible impact on employment o f the agricultural growth under various scenarios would help inform the debate on rural poverty reduction, the role o f agriculture, and the importance o f developing either non-farm employment or near-by growth centers in towns and cities. 62. This study will help guide future World Bank dialogue and financingof agriculture and rural development in Argentina, particularly with regard to areas where the role o f government may be strengthened or the provision o f public goods increased. The potential for agricultural development in Argentina i s impressive. A strategic support from government to complement the capacity and dynamism o f Argentina's farmers could help much to realize that potential and make agricultural growth more equitable. With the present study, conceived as an instrumentfor dialogue, the World Bank hopes to contribute to that objective. xxi xxii 1. INTRODUCTION 1.1 This report offers an overview o f the performance and trends o f the Argentine rural economy. Its purpose i s to update the World Bank's understanding o f the forces currently shaping the rural economy in Argentina, and to strengthen our dialogue with government and civil society on rural development issues. This i s especially relevant inview o f the new Country Assistance Strategy recently agreedbetween the Government o f Argentina andThe World Bank. 1.2 The report reviews selected areas o f major importance in the agricultural sector, and highlights areas where further work might be warranted. It is intended as a vehicle for dialogue, not a final statement o f Bank positions. For this reason we highlight throughout the report areas where we feel we have important gaps in our understanding. Undoubtedly readers will identify others. Inthis way, we hope that the document will initiate a process that will help the Bank to better contribute to agriculture, rural development andpoverty reduction inArgentina. 1.3 The report has, necessarily, been selective in the areas o f review. Background papers were undertaken on topics which government and other specialists in agriculture in Argentina considered (i) important for future growth, productivity and employment (like irrigation and phytosanitary control); (ii)areas o f potential vulnerabilities (like market access and the environmental effects o f frontier expansion); and (iii) areas which are shaping the future inways which need to be better understood (like plantingpools and rural poverty). These areas were also chosen because the role o f government, either as provider o f public goods or as regulator o f the rules of the game, i s critical, andbecause o f their importance for the regional economies. 1.4 Of course, important topics remain to be addressed. We make recommendations with regardto additional priority studiesinthe final chapter o fthis report. 1.5 This report is largely a synthesis o f background papers prepared by country consultants and World Bank staff. Complete background papers andtheir authors are listed in the section on References, and are available at www.bancomundia1.org.ar.The background papers were also discussed in a series o f workshops in December, 2004, and comments made at that time have been incorporated. The preparation o f the final report required strategic choices to keep the length manageable and the story line intact. As a result, much good material has not been included in this report. This i s not a reflection o f the quality o f the papers. In some cases we decided to wait until a more comprehensive report could be prepared. In others we decided that the Bank would have little to add to local understanding o f the issues. In these cases the backgroundpapers have performed the important function o f upgrading the Bank's knowledge o f Argentina's agricultural economy despite their content not being fully reflected in the final report. 1.6 The report i s divided into nine chapters. Chapter 2 provides an overview o f Argentina's agricultural sector. Chapter 3 makes comparisons with other countries, especially its regional competitors Brazil and Chile. Chapter 4 discusses the role o f government in the agricultural economy, including fiscal issues, main institutions, the provision o f infrastructure and financial services, and government role inpromoting collective action. Chapter 5 examines the situation o f the labor force, incomes and poverty in rural areas. Chapter 6 explores the social and 1 environmental effects o f soybeans expansion. Chapter 7 surveys sanitary, phytosanitary and food quality issues. Chapter 8 discusses irrigation issues. Chapter 9 briefly summarizes the findings andprovides conclusions andrecommendations for further action. 2 2. THE STATE OF AGRICULTURE INARGENTINA 2.1 The purpose o f this chapter i s to provide an overview o f the role o f agriculture in Argentina's economy, and the evolution o f agricultural policy and its impacts. We are particularly interested in exploring the effects o f (i) the policy reforms o f 1991 and (ii)the subsequent macroeconomic imbalances leading to the crisis o f 2001-2002. The chapter examines also the new forms o f agricultural production organization that have emerged over the last two decades, and calls attention to the heterogeneity o f Argentina's provinces with regard to their endowments, production orientation, indicators o f the quality o f rural life, and their capacity to adjust to the changing economic endowments. For reasons o f space, regional and provincial indicators and the description o f agricultural production inthe regions are presented inAnnex 1. OVERVIEWOF THE IMPORTANCEOF AGRICULTUREINARGENTINA'SECONOMY 2.2 Agriculture has been o f critical importance throughout Argentina's history. Most recently, agricultural has led the export-driven recovery o f the Argentinean economy following the crisis o f 2001. In 2004, the sector contributed 58 percent ($13.1 billion) o f total goods exports, o f which 39 percent were primaryproducts and 61 percent were manufactured products I of agricultural and livestock origin. In 2004, the agricultural and livestock sector generated 9 percent o f Gross Domestic Product (GDP) and 22 percent o f the value added o f the goods- producing sectors.' Crops made the largest contribution (63 percent o f the total), followed by livestock (3 1percent). 2.3 Agriculture i s also a major source of employment. In 1997, direct agricultural employment (inprimaryagriculture and agro-industries) was about 1.5 million or 11.4 percent o f total employment. When the employment generated in the transport and commerce services related to food and agricultural distribution i s considered, another 1.1 million jobs are added, with the percentage o f all employment increasingto 20 (Obschatko, 2002). For 2003, Llach et a1 (2004) estimated in 2.75 million jobs the direct employment generated by the food and agriculture sector (including transport and commerce), to which they add another 2.84 million jobs o f indirect employment resulting from backward and forward linkages. 2 2.4 In2001, 35 percent of Argentina's population lived inpredominantly agricultural zones or resided in cities linked to agro-industrial activity. Furthermore, if a definition o f rural as that proposed in de Ferranti et a1(2005), which combines population density with distance to cities i s used, the proportion o f rural population inArgentina is above the average for Latin America, of the order o f 45 percent (see Figure 2.1). Subsectors: agriculture, livestock, hunting, and silviculture-Preliminary estimates in millions of AR$, at current prices, by the Bureau o f NationalAccounts, INDEC. Llach et aZ(2004) figures have been challenged by Rodriguez (2005), who disputestheir methodology for updating results based on the1997 input-output table, and for the calculation of indirect employment. Rodriguez's own figure for the participation o f direct agricultural employment (including transport and commerce) in all employment in 1997 is 18.1percent, similar to that o f Obschatko. 3 2.5 Historically, the Pampean region has dominated agricultural growth in Argentina. PrimaryAgricultural Ex orts from the Pampas, especially ingrains and oilseeds grew 46 percent between 2000 and 2004, compared to 29 percent for the rest o f the country. By 2004, 79 percent r3 o f total agricultural-based exports came from the Pampean region. Incontrast, the non-Pampean regions have taken the leadinagriculture-based manufactures. Figure2.1: How Ruralare LatinAmerican Countries? 0.7 Proportion rural 0.6 I I i n 0.5 00.3 -4 Is, 0.2 0.1 ote: The World Development Index (WDI)uses nationaldefinitions of "rural population." The alternativemeasure proposedby le World Bank uses ameasuredefinedas areas both (a) less than 150 personskmzpopulationdensity and (b) more than 1 hr ravelTime to the nearest large city of>100,000 people. mrce: K. Chomitz,P. Buys and T.S Thomas, World Bank, 2005, Table 6. THEEVOLUTIONOFAGRICULTURAL POLICY 2.6 Deep soils, temperate climate, adequate rainfall and good access to sea freight endow Argentina with exceptional potential for agricultural production. This endowment has permitted Argentina's agricultural sector to perform well despite over 50 years o f policies directed largely at taxing agriculture to promote industrial-based growth (although with varying levels o f transfers over the period). The historical policy regimes are reviewed briefly in the paragraphs that follow, with emphasis on the past 15 years. Prior to 1991 2.7 Up to 1930, Argentina enjoyed a long period o f agricultural expansion that benefited from European immigration and heavy investment, both on-farm and in infrastructure, such as railways, slaughterhouses, and ports. This period was characterized by a strong development o f This figure includes Entre Nos, Santa Fe, Buenos Aires, Cordoba, L a Pampa and SanLuis. 4 private markets for land and c~mmodities.~However, this long agricultural boom ended with the collapse o f world agricultural prices during the Great Depression o f the 1930's. This collapse was followed by the Second World War, duringwhich, due to interdicted naval trade, Argentina builtuphuge stockpiles o fun-saleable grain. 2.8 These disappointing experiences with foreign agricultural markets led to a growing pessimism concerning the agricultural sector as the motor o f economic dynamism. In addition, the depression and the war opened new domestic marketopportunities for industry. These forces combined to usher in a sustained policy promoting industrial growth through import substitution andtaxation o fthe agricultural sector. Agriculture was taxed through overvalued exchange rates, public marketingboards, and export duties on grains andbeef. 2.9 There are numerous indicators, both direct and indirect, o f the negative effect o f these distortionary policies on the agricultural sector. Reca and Parellada (2001) and Mundlak and Regunaga (2001) show that prior to the mid 1930s Argentina had higher agricultural yields than the US, but that U S yields grew much faster subsequently. Comparing the averages for the periods 1913-1930 and 1975-1984, US yields tripled, while Argentina's failed to double. Whereas U S public policies stimulated a revolution in agricultural technology over this period (including substantial public investment in research and extension through the Land Grant College system), Argentina lagged in adopting publicly-available technologies. The most dramatic evidence o f the results o f this policy bias occurred in 1953, when the combined effect o f policy -induced stagnation and drought forced Argentina to import grain to meet domestic needs -- for the first and only time in the 20th Century. Indeed, the average annual production over the three years 1950-52 was 20 percent below that o f 1940-42 (Reca and Parellada, 2001). 2.10 Although this experience led to a substantial reevaluation o f the role o f agriculture in Argentina, and to the creation o f INTA in 1956, the underlying policy bias continued. According to Sturzenneger et. al. (1990) by the early 1980s government policies transferred over 60 percent o f agricultural GDP to other sectors, including consumers and govemment. The cost in both agricultural growth and overall economic growth was severe. Farm gate prices o f cereals and oilseeds are estimated to have been nearly halved by these policies and production was reduced from a potential o f 60 million tons per year to 34 millions tons over the 1980-1985 period. In terms o f the overall effect on Argentina's growth, Mundlak, Cavallo, and Domenech (1989) calculate that had Argentina not adopted an inward growth strategy, on average for the 55 years between1930 and 1984 it could have attained levels o f income, consumption, and investment that were 63, 70, and 112 percent higher, respectively. Growth would have been similar to that experienced by Australia and Canada, countries with similar resource endowments. 1991-2001 2.11 The 1991-2001 period was tumultuous for Argentinean farmers. This period witnessed liberalization, privatization, dollarization, and the coming into force o f MERCOSUR, followed by the emergence o f macroeconomic imbalances that resulted in currency overvaluation and credit unavailability. For most farmers these events dramatically changed their economic environment. Drivenby favorable external and domestic markets, agriculture expanded rapidly 4 BasedonMundlak andRegunaga(2001). 5 through 1998. The sector benefited from the elimination o f quantitative restrictions on imports, reductions on import taxes for fertilizers, herbicides, pesticides, machinery, and imgation equipment, the elimination o f distorting taxes on fuels, commercial and financial transactions, the elimination o f export taxes, the deregulation o f economic activities, and the removal o f inefficiencies and monopoly profits in the trade channels (elevators, transportation and ports). The result was a five fold increase in fertilizer use and a three fold rise in the use o f herbicides and pesticides. Duringthe period 1988/1990 to 1996/1998, wheat, corn, soybean, and sunflower yields increased 26, 43, 7, and 25 percent, respectively. Land sown with the 31 principal annual crops expanded 24.6 percent. Together, these yielded an impressive 7 percent annual growth in production over the same period, as illustratedinFigure 2.2. I Figure 2.2: Evolution of Agriculture inArgentina, 1961-2005. PIN, 1961=100 400.0 I i 350.0 300.0 250.0 200.0 150.0 1 100.0 50.0 I ~ PIN: ProductionIndex. Source: FA0 statistics(2005). 2.12 The explanation for the low growth inlivestock is mostly due to the poor performance of beef production and ina minor way to the decline inthe production o f sheep and pig meat. Milk production feared quite well, particularly in the 1990s, while the increase in poultry production during that decade was spectacular, increasing by a factor o f 2.7 between 1990 and 2000 (Figures 2.3 and 2.4). Yet, beef production explains most o f the evolution o f livestock output since, according to FAOSTAT data, beefmeat i s 71 percent o f all meat output for the average of 1990-2004. 6 Figure2.3: Evolutionofthe ProductionofMilk andBeef, Sheep andPigMeatinArgentina, 1961-2004 Source: FAOSTAT. Source: FAOSTAT 2.13 The long-term stagnation o f beef production seems to derive from a combination o f cattle's losing competition with crops, and the greater uncertainty in cattle production. In much o f the land traditionally under pasture, cattle can simply no longer compete with crops. This has pushed cattle onto less productive and more distant pastures, and cattle-raising has had a hard time competing with crops for management attenti~n.~ Risk in cattle production has also been high, as exemplifiedby droughts and closure o f export markets due to foot and mouth disease (aftosa). ' InArgentina most cattle are raisedinmixedcrop-cattle enterprises. 7 2.14 Beefproduction, however, has experienced considerable growth since 2002. It remains to be seen whether this i s something transitory or a structural phenomenon. The potential to increase beef production in the Pampas and outside the Pampas i s high. This i s because, as shown below, much o f the sector operates at medium and low technology levels, and the output gap between productivity levels in cattle production is high--much higher than in crops. Raising productivity i s important because the sector i s large, and a moderate raise inproductivity would result therefore in a large increase in aggregate output. Also, it is inthe smaller farms where the productivity gaps are largest. Further study o f the reasons for this sector's relative stagnation, compared to the grain sectors, i s recommended. 2.15 National GDP began to stagnate in 1998. Spillovers from the Brazilian crisis and restrictive policies put in place to defend the dollarized peso combined to depress GDP throughout 1999-2001, until the crisis in capital outflows at the end o f 2001. A combination o f factors conspired against the Argentinean farmer during this period: (i)the devaluation o f the Real at the end o f 1998 made Argentina significantly less competitive in newly-won MERCOSUR markets; (ii) rising prices o fnon-tradable combined with record low grain prices to depress the terms o f trade for Pampean producers6, and (iii) collapsed domestic demand affected most regional producers relying on domestic markets. Indeed, the poverty rate rose from 19.9 percent in 1992 to 28.5 percent in 1998, and 38.8 percent in 2001. Unemployment increased from 6.3 percent in 1992, to 12.3 percent in 1998, to 18.3 percent in2001. 2002-present 2.16 Overall, the agricultural recovery since the devaluation in 2002 has been strong. Agricultural-based exports, which had either fallen or showed insignificant growth throughout 1999-2001, grew by 26 percent in2002,27 percent in2003 and an additional 13 percent in2004. National GDP growth from the first quarter o f 2002 to the fourth quarter o f 2004 was 25 percent, nearly sufficient to reestablish GDP losses (not growth losses) incurred over the previous three- and-a-half-year slide. Unemployment, which had peaked at some 21 percent in 2002, fell to 12 percent by the fourth quarter o f 2004. Taken together, the GDP and employment numbers suggest a significant resurgence indomestic demand for income elastic food products. 2.17 Argentina i s able to produce most field crop at top technical levels. It surpasses the average o f the L A C region inall crops included inTable 2.1 except wheat and soybeans which i s surprising. However, not a single field crop surpasses the US yield. L A C averages can be considered a modest standard for Argentina, a middle income country with very favorable natural conditions and a long farming tradition. The future 2.18 The history reviewedabove has created numerouswinners and losers. Many farmers sold out and moved to the city, while, as discussed below, many others turned farm management over to specialists. Farmers inthe Pampas will continue to be extremely competitive in commodities, due to the low price o f land relative to their main competitors and to the current rapid rate o f 6Reca and Parellada (2001), report that the ratio o f Pampeana producer prices to services and labor fell steadily, reaching in2000 nearly 60 percent of their 1990 level. 8 adoption o f new technology. New niches are likely to emerge, both in MERCOSUR and in the markets requires becoming competitive with globalized producers -- on a quality, timeliness, and larger world market for production from the regional economies. However, entry into these cost basis. Experience shows that success requires making strategic linkages with others in the supply chain, jointly identifying and removing bottlenecks and weak links, and entering differentiated, highvalue markets. Table2.1:CropYields inSelectedCountries,Average 2000-2002, (todhectare) Mexico Argentina Chile Brazil LAC USA EU India China Cereals 2.8 3.4 4.9 2.9 2.9 5.8 5.6 2.3 4.8 Maize 2.6 5.8 9.8 3.0 3.O 8.5 9.1 1.9 4.7 Wheat 4.9 2.3 4.1 1.6 2.4 2.6 5.7 2.7 3.8 Rice 4.3 5.3 5.1 3.2 3.8 7.2 6.4 2.9 6.3 Sugar Cane 74.1 65.4 _ _ 69.6 64.9 77.2 -- 67.3 61.3 Citrus 12.4 20.1 15.4 22.0 17.0 34.7 18.3 17.8 8.2 Pulses 0.8 1.1 1.6 0.7 0.8 1.9 2.7 0.6 1.4 Vegetables 16.5 17.2 25.6 17.9 14.9 27.1 26.7 12.9 19.2 Soybeans 1.6 2.5 0.0 2.6 2.5 2.6 3.3 0.8 1.7 Source: Caballero (2005)--calculations based on FAO's AGROSTAT. 2.19 Two important issues for the future o f Argentina's agricultural exports are the long term reliability on the Chinese market, which absorbs now most o f Argentina's soybean exports, and the results o f the negotiations between MERCOSUR and the EU, as well as the possible signing o f a FTAA and the results o f the WTO Doha round. The first o f these issues is explored inBox 2.1 and the second inBox 2.2. 2.20 The differentiation between quality standards required by external and domestic markets will increasingly disappear as supermarkets further penetrate domestic retail marketing. Reardon and Berdeguk (2002) report that fifty-seven percent o f Argentina's retail trade went through supermarkets by 2000, up from 35 percent a decade earlier. Supermarkets compete vigorously for consumers, on the basis o f price, quality, and product differentiation and innovation. They also procure globally. 2.21 Successful farmers will be those who can incorporate themselves into dynamic supply chains that continuously innovate. As we shall see below, in numerous regions farmers are successfully making the transition to high value crops. Indeed, as a result o f the reforms o f the nineties and the currency devaluation o f 2002 a number o f farmers in the regional economies -- inmany cases inalliancewith foreign investors and domestic investors from outside the region-- are investing heavily in production for the export market. Many others, decapitalized as a result o f the events described above, and without access to credit, have no possibility to make the investments necessary to upgrade. The number o f farms is diminishing throughout the country, with the more successful farmers buyingout those who are struggling. 9 Box2.1: Sustainability of SoybeanExports:How secureis the Chinesemarket? Zun-ently, approximately 60 percent o f Argentina's exports o f soybeans and 18 percent o f soy oil are destined f o ~ the Chinese market. Due to the export surge in soy products, China has become Argentina's fourth largest trading partner after MERCOSUR, EU, and US. Thus, there i s a strong reliance on the Chinese market for soy products. Ir determining whether this export market poses risks, close attention needs to be paid to China's biotechnologj policies, since almost 100 percent o f Argentinean soybeans are GMO. Export destination for Argentinean Soybeans 1 2 0 ~ Italy 100 .......Spain --- - --Netherlands / Thailand / .... I -----"c- 0 -.-.a 00 -.-.-..T ... . - - - , 1996 1997 1998 1999 2000 2001 2002 2003 Source: UNCOMTRADE. In 2001, China announced its first set of biotechnology and food safety regulations requiring each shipment o biotech soybeans to obtain an individual safety certificate, This put a temporary halt to imports, bringing d o w import volume from 1.1 million tons in April 2002 to close to zero inJune 2002. In March 2004, the Ministry o Agriculture issued a permanent safety certificate for importing herbicide tolerant GM soybeans. However thc situation remains precarious as import permits are still requiredfor each shipmentand the permanentsafety permiti! subject to review after 5 years. On the Chinese domestic policy side, in 2002 and 2003 the government issued a national policy for boosting GN free soybean production in the northeast region, thus signaling its intention o f substituting soybean imports wit1 domestically produced soy. However, this year the national agricultural policy reversed this policy reemphasizini its commitments to long-term food security by encouraging food grain production. Thus, there is still uncertainty ir national policies on domestic soybean production. It is clear that China will expand its demand for soy products due to its strong demand for meat and vegetable oi and the existence o f new extensive investment in the crushing industry in the coastal areas. For now, thc government seems to have abandoned the promotion o f domestic production in lieu o f imports. Factors that ma: explain this include domestic transportation bottlenecks from the soybean producing provinces inthe northeast, an( benefits associated with the high oil content, consistency o f quality and reliability o f delivery dates o f importec beans. Nevertheless, the possibility o f increased domestic supply competition cannot be ignored. China's soy yield are still 35 percent lower than Argentina's, and they may improve with improved production technology. 11 addition, profits from soybeans (222 Yuan per mu) are double that from corn (110 Yuan per mu) in the northeas region (China Ministryo fAgriculture, 2004 as quoted inUSDA-ERS, 2004). Source: Tuan et a1(2004). 10 POTENTIAL ADDITIONAL PRODUCTIVITY GAINS 2.22 Inthis section we look at two projections of future gains inproduction andproductivity, one by INTA based on field surveys o f existing technologies and technology gaps, and the other by SAGPyA. Chapter 3 discusses productivity further, ina comparative context. The InstitutoNacional de TecnologiaAgropecuaria (INTA) technology adoption study 2.23 In cooperation with its 12 regional centers, INTA has undertaken a long-term study of technology adoption and productivity inArgentinean agriculture. This study is based on detailed production data in 48 categories of crop or livestock activities (disaggregated by homogeneous agro-ecological zones and levels o f technology), and an analysis o f the factors restricting the adoption o f higher levels o f technology (Cap and Gonzblez, 2004). These data were used to project the additional value o f agricultural production that could be obtained under various assumptions concerningrates o f adoption o f existing technological packages. 2.24 Technology adoption was modeled through two parameters, the first related to the rate o f adoption, and the second to a "ceiling" on the amount o f area to which the technology could profitably be applied. The first parameter i s a function o f the characteristics o f the technology and its ease o f adoption (biological, chemical, agronomic, or mechanical; see Byerlee and Hesse, 1982), while the second i s determined by factors external to the technology itself, such as the provision o f public goods (roads, phytosanitary control, etc.), investment by other actors in the production chain (for example in the creation o f a cold storage) and the characteristics o f the producers (access to credit, risk bearingcapacity). 2.25 Two basic scenarios were explored: one assumes that the existing maximum area remains unchanged, but all producers reach the "ceiling" for their area; and the other assumes that all areas currently under production of a crop adopt the highest technology package. The first case assumes that all farmers "catch upyy to the highest technological package that is profitable intheir region. The second scenario further assumes that all necessary public andprivate investments are made to make the highest technology package attractive on all land currently cropped under each o f the 38 items.' 2.26 The results o f the first scenario (full adoption under current external conditions) yield an additional gross value o f annual agricultural production o f US$6.6 billion (at 2000/2002 prices) by 2014. The second scenario-assuming all necessary complementary investment and complete adoption-would raise gross production value by US$11 billion. This suggests a potential increase in agricultural production value (relative to 2002 and at constant 2000/2002 prices) o f 44 percent by 2014 inthe less optimist scenario, and o f 76 percent inthe upper limit scenario. 2.27 Seventy percent of the additional production was explained by four items: livestock, soybeans, wheat, and corn. The potential for productivity increases in cattle is high (Table 2.2), and represents 46 to 47 percent o f the total. Cap and Gonzblez (2004) suggest the following reasons for the large unrealized potential o f livestock: 'Ten o f the i t e m had to be left out of this estimate for lack of all of the necessary data to carry out of the simulation. 11 Box 2.2: Trade Liberalization Scenarios: Will "Sensitive farm products" be included? [tis likely that certain "sensitive farm products" will not be included inthe list o f liberalized goods inthe final trade igreements. EU has not lowered tariffs for this category o f goods in its Association Agreements with Central and Eastern European countries, custom union agreement with Turkey and in its free trade agreements with Mediterranean countries such as Morocco and Tunisia, all areas which have geo-political significance for the EU. In the EU-MERCOSUR agreement, sensitive farm products will probably be dealt with by the "single pocket principle", which means that quotas for sensitive farm products will be split, mostly 50-50, into an immediate share for the EU-MERCOSUR agreement and a later share for the WTO Doha round. Since 75 percent o f MERCOSUR exports to EU is agriculture, including many sensitive farm products, such as rice, cereals, meat, dairy products and sugar, the benefits to MERCOSUR will depend o n the outcomes o f the negotiations. The chart presents simulation results for FTAA, EU-MERCOSUR and WTO trade negotiations. Each model differs inthe assumptions employed, and hence a simple comparison cannot be made but the chart summarizes the result for a full economic liberalization for Argentina under the different trade negotiations.8 An EU-MERCOSUR agreement would yield benefits for Argentina ranging from US$0.6 billion to US$12.61 billion; and benefits from joining the FTAA are believed to be lower ranging from -US$0.5 to US$10.41 billion. The combined benefits o f both could be as high as US$20 billion. However, some sources (Van der Mensbrugghe, 2002, and World Bank Brazil CMU, 2004) estimate that Argentina could have a negative benefit if it joined the FTAA only (and not the EU), as it would lose preferential treatments it currently enjoys with Brazil. The World Bank Brazil CMU model compares full liberalizationwith the case where certain sensitive farm products are omitted from liberalization. Benefits shrink dramatically once it is assumed that certain farm products will not be included. For example, the EU-MERCOSUR agreement i s expected to have a benefit o f US$5.9 billion for Argentina under full liberalization but this shrinks to US$0.5 billion if market access is withheld for the seven sensitive farm products which currently enjoy the highest level o f tariff rates in the EU (paddy rice, cereal grains, processed rice, bovine meat products, dairy products, other meat products and sugar). This indicates the importance o f the details o f the agricultural negotiation indetermining the outcomes. Benefits for Argentina according to trade liberalization scenarios EU-MERCOSUR AR+NAFTA Source: Taminichi(2005). 8 Exceptionis the IFPRI model which assume 0 percent tariff inall goods except for cereals and meat products. 12 The sector operates, ingeneral, at amedium and low technology levels; and 0 The range inproductivity levels i s wide. In livestock, low technology production is only 34% as productive as high technology production, while for soybeans, corn, and wheat the yield gaps between the low and high technology levels are 39, 47 and 48 percent points, respectively (Table 2.2). Table 2.2: Technology Technology level PercentArea Percent %HighTechnology Production Yield Low 38 23 34 Cattle Medium 42 44 60 Low 17 13 61 Soybeans Medium 53 51 81 High 30 36 100 Source: Cap andGonzilez (2004). SAGPyA projections 2.28 SAGPyA has estimated that between 2003 and 2010, land area planted with grains, legumes, and industrial crops will grow 16 percent, and fiom 2010 and 2016 an additional 9 percent. With these projections, 100 million tons would be reached in 2010 and nearly 116 million tons in 2016. The area needed to reach 100 million tons in 2010 was estimated at 32.7 million hectares comparedto 25.4 millionhectares in2003.' 2.29 The projected value o f ago-industrial exports, at constant prices, foresees a 45 percent increase for 2010 and 80 percent for 2016, the year in which a total o f US$30 billion could be reached, at average prices o f 2001-2003 (Table 2.3). Table 2.3: Prqjected Exports: 2010 and 2016 EXPORT VALUE Export complex (in millionsof US$) 2003 2010 2016 Increase, 2003116 SOY 7.191 10.070 11.279 4.088 Wheat 1.086 2.107 2.547 1.461 Sunflower 727 1.347 2.035 1.307 Fruits and horticulture 976 1.669 2.089 1.113 Cotton 7 125 222 215 Maize 1.288 1.126 1.366 78 Tobacco 163 200 240 77 Other cereals 64 76 93 29 Others 338 532 680 342 Source: SAGPyA (2004). Estimatesfiomunpublishedwork by SAGPyA (2004). 13 NEW FORMSOF FARMORGANIZATION 2.30 The 1990s witnessed a plethora o f new forms o f organization o f agricultural production, mostly in the Pampas, but also spreading into other regions. The driving force behind this outburst of institutional creativity has been the large economic gains available from the adoption o f "soft technology" and the economies o f size that this involves. During the first years of the 1990's credit availability from the National Bank improved, rates fell, and longer term credit became available--which in turn facilitated a greater participation o f credit in the financing o f investments in the agriculture sector. As credit tightened, new ways o f financing emerged, including various barter arrangements exchanging goods and marketing for crop products. Warrant systems," leasing, trust funds, and reciprocal guarantee societies developed as alternatives to traditional financing, and also as a means o f bringingtogether lands from different owners, technical and managerial know-how, machinery, and inputs. The availability o f alternative financing was stimulated inpart, by a favorable evolution o f agricultural prices. 2.31 The agreements among producers and various actors that associate in the productive process through the provision o f production factors (sometimes in kind, sometimes in labor or money), commonly called planting "pools" or `>ooZs de siembru ",have increased in quantity and variety. Normally management o f the farm remains inthe producer's hands (often assisted by professional technical personnel) with the agreements being a form o f financing and risk diversification, and in many cases allowing increases in scale. With the incorporation o f fertilization technology and direct cultivation these arrangements have also generally involved the conversion o f rotationalpasture to permanent agriculture (see Box 2.3). 2.32 Looking at agricultural establishments (EAP) engaged in predominantly agricultural (as opposed to livestock) activities, brings out the importance o f pooling arrangements in Pampean agriculture. The proportion o f agricultural EAPs that work third party lands (whether or not they also own land) i s shown inFigure2.3. Figure2.5: PercentageofAgriculturalEAPswith Landunder ContractinPampeanProvinces 70.0% 60.0% 50.0% 40.0% 30.0% 20.0% 10.0% 0.0% B u e n o s Aires Cordoba Entre Rios L a P a m p a Santa F e Source: Bertolassi (2004). Basedon the 2002 AgriculturalCensus. 10 The warrant systemprovides the farmer a way to fix the price of his grain while holding in storage in country without havingto rushit to the port to sell. 14 2.33 Effect on management. Data from the 2002 agricultural census provides evidence that farm enterprises operated under "pool" arrangements achieved higher management standards than others. The census includes various questions related to enterprise management, including mechanisms to manage risk and improve efficiency, use o f agriculture insurance, and use o f mechanisms to provide price coverage. Figures reveal substantially higher rates o f adoption o f risk management practices on farms operated under "pooling" arrangements. For example, in Cordoba 71 percent o f farms operating under pooling arrangements have hail insurance, compared to 11 percent for farms without participation in pools. Similarly for the adoption o f agricultural technology: 27 percent o fpool farms inEntre N o s performed soil analysis compared to 14 percent o f other farms, and 55 percent monitored pests compared to 22 percent innon-pool farms. Although these differences are not as high for all provinces, adoption rates are higher for pool than for non-pool farms inall Pampeanprovinces for virtually all modem practices. 2.34 Effect on farm size. The 1988 agricultural census registered about 85,000 agricultural enterprises (EAPs) with fewer than 100 hectares for the Pampean region. This number fell to 50,500 EAPsby 2002, a drop o f over 40 percent. The number o f EAPslarger than 1,000 hectares increased from 14,000 to nearly 15,000, an increase o f 7 percent. Overall inthe Pampas region, there were about 54,000 fewer EAPs registered in 2002 than in 1988. This shows the clear consolidation o f farms (Table 2.4). Table 2.4: Variationinthe number of PampeanEms,by stratum Scale (has) EAPs88 EAPs2002 Difference Percentage Diff. up to 5 8720 4484 -4236 -48.6% 5.1- 10 7159 3692 -3467 -48.4% 10.1-25 15925 8858 -7067 -44.4% 25.1-50 21740 13397 -8343 -38.4% 50.1-100 31528 20099 -11429 -36.3% 100.1-200 35846 24294 -11552 -32.2% 200.1-500 37666 29352 -8314 -22.1% 500.1-1000 15544 14978 -566 -3.6% Subtotal, up to 1000 174128 119154 -54974 -31.6"LO 1000.1-2500 9735 10294 559 5.7% 2500.1-5000 2900 3107 207 7.1% 5000.1-10000 1081 1155 74 6.8% 10000.1-20000 255 318 63 24.7% over 20000 72 84 12 16.7% Subtotal, over 1000 14043 14958 915 6.5% Total 188190 134112 -54078 -28.7% Source: 1988 and 2002 Agricultural Censuses. Bertolassi(2004). 15 Box 2.3: Development and Spread of Zero Tillage: Spontaneous and Collective Action The development o f Zero Tillage (ZT) in Argentina shows positive, and often unplanned results o f virtuous public-private interactions. ZT is the most important agricultural technology introduced inthe last 50 years inArgentina. It is the planting o f crops inpreviously unprepared soil by opening a narrow slot or trench o f the smallest width and depth needed to obtain proper seed coverage. ZT evolved over 30 years o f formal and informal interactions o f public and private agents. Three different stages can be identified. The first stage was in the INTA station o f Marcos Juarez, Cbrdoba, in 1970, when a group of researchers decided to test ZT in farmer's fields. Researchers sought help from local manufacturers to develop adequate planters. Interactions between the researchers, machinery manufacturers, and farmers, were informal and very active. Manufacturers explored a new line o f machine with small investments in research and the researchers developed the tools they needed. As a result, a package based o n the herbicide paraquat was developed and tried by many farmers. It was abandoned, however, because o f costs and difficulties with weed control and machinery. At the same time, the Pergamino INTA station started an experimental project with tillage and realized that higher yields were obtained with ZT. However, managers at the experimental station did not want to commit land to a technology whose potential was uncertain. The trials were established in a nearby agricultural school, and innovative farmers followed it closely. The second stage lasted through the 1980s as three groups worked simultaneously. On the one hand, INTA researchers from Marcos Juarez and Pergamino studied weed control, with slowly but steadily improvements in the efficiency o f herbicides. On the other hand, pioneer farmers continued to work on the technology package. The same year, INTA Pergamino launched the Program for Conservation Agriculture (PAC), which promoted several conservation practices including ZT. Soon after, INTA and Banco Nacidn launched a subsidized credit program for the purchase and maintenance o f tillage machinery, although it did not work well because the technology was still uncertain. Inthe mid 1980s about 15 pioneer farmers and INTA researchers began meeting to discuss ZT. A Monsanto researcher, who had previously participated as a trainee in Pergamino's trails, pressed them to form an association to promote ZT. In 1988, The Argentine Association o f Farmers for Direct Planning (APRESID) was created, after Monsanto paid all start-up costs. With the creation o fAPRESID, the third stage started. Through APRESID, use o fZT surged from 300,000 ha in 1991to 9,250,000 ha in2000. ZT i s superior to alternative technologies in every aspect considered. APRESID coordinates most o f the activities o f an innovation system formed by thousand o f farmers, input suppliers, research finders, public research institutions, individual researchers and governmental agencies. The experiments, conducted by farmers, deal with rotation and soil management. Lessons learned. This experience shows how the interaction and the leadership o f different actors at each stage resulted in the most important agricultural technology recently introduced in Argentina. The state, through INTA, started the process and interacted positively with the private sector, who contributed with inputs to the trials. However, INTA did not want to expand the project outside their experimental station. At that stage, the private sector through APRESID took the leadership o f expanding in expanding ZT. APRESID has played two key roles: it has reduced the costs o f information generation and transfer by exploiting economies o f scale, and has filled the gap left by the formal researchsystem, which has been slow to respond to farmers' needs. Source: EkboirandParellada (2002). 2.35 Effect on migration. Importantly, there is no clear relationship between farm size concentration and out-migration into more urban areas (see Annex 1).Furthermore, as discussed in Chapter 5, out-migration and farm consolidation should not be necessarily taken as an indication o f reduced social well-being in Argentina. On the contrary, they suggest that farmers who have the assets necessary to leave the farm and take advantage of opportunities off the farm are doing so. Most important among such assets are education, which provides the opportunity 16 to maintain income in non-farm employment. Second i s good quality land (even if in a small parcel) that i s under demand in an expanding agricultural economy. Good quality land can be either sold to provide the capital to finance non-farm activity, or it can be rented to supplement non-farm income. Data inAnnex 1suggest that in areas o f highagricultural value, especially the Pampas, small farmers capitalize out (or rent out) as technological and institutional change increases the value o f their land beyond the net present value o f their own production. In contrast, in areas o f low agricultural dynamism small farmers have neither buyers nor renters to finance their exit, nor do they have human capital to apply to non-farm employment. This, for example, appears to be the case o f the Northwest, which has the highest poverty and un-met basic needs rates o f all the regions, but nevertheless has the lowest rate o f out-migration from dispersed areas andthe lowest rate o f reduction infarm numbers. 2.36 The Cuyo region appears to illustrate the potential employment effect o f increasingly high-value irrigated agricultural. In the Cuyo, consolidation o f farms has led to much less out- migration than elsewhere. This i s undoubtedly due to the labor intensive nature o f irrigated agriculture, especially in the production o f high value crops. As a result of this expanding activity, farmers leaving farming in the Cuyo have a much higher probability o f finding wage employment in the agricultural or non-agricultural area. In Mendoza, the province most successful in developing irrigated agricultural and promoting high value products, population increased inboth dispersed and grouped rural areas despite a 13 percent reduction inthe number o f farms. REGIONAL EFFECTSOF POLICY CHANGES 2.37 Macroeconomic performance in the post-reform period had varying effects on the different regions and provinces. Most, but not all exhibited sharp growth in agricultural exports inthe 5-7 post reform years, followed by stagnation or decline as the peso became overvalued and credit tightened.'' The following paragraphs show the wide range in outcomes o f these macroeconomic swings on individual provinces, and attempt to shed some light on the determinants o f these outcomes. In order to do so, we have looked at growth in exports o f manufactures o f agricultural origin over two periods: (a) the earl post-reform period, and (b) from the peak o fpre-crises growth (usually around 1997) to 2003.1Y These two measures capture how successfully a province took advantage o f trade liberalization to expand value-added agricultural exports, and how vulnerable the province was to overvaluation and credit rationing during the pre crises and the early post-crises period. The analysis .indicates important differences among the performance o f the regional economies. 2.38 Note that 13 provinces had rapid growth (over 10 percent a year) following the immediate post-reform period (Table 2.5). Of these provinces, 5 also showed net non-negative growth from their pre-crises peak to 2003. These 5 are Chubut, Catamarca, Cbrdoba, L a Rioja, ~ ~ ~ Details on agricultural production inArgentina`s various regions are presented inAnnex 1. l2 We chose to look at exports o f manufactures o f agricultural origin because (1) unlike primary exports they do not contain mineral exports, and (2) they measure the success o f a province in adding value beyond the primary agricultural stage. I t would had been interesting to perform further analysis along these lines on primary exports as well, but we were not able to obtain a clean series for this period for primary exports. 17 and Mendoza. This group contains provinces with a long history o f competitive agricultural exports (Mendoza, Cbrdoba, and to a lesser extent Chubut), and provinces which have successfully introduced "new" export products, namely olives, olive oil, wine, and industrial crops, at least partially with the aid o f fiscal incentives (Catamarca and L a Rioja). 2.39 Three other provinces, Santa FeyMisiones, and Tucumhn, also grew duringboth the pre- crises and post crises period, albeit at a slower rate than the previous group. These provinces are also generally competitive inagroindustrial products. Santa Fe is highly competitive ingrain, oil and dairy products, while Misiones is highly competitive in wood products and tea, and Tucumhn incitrus, mostly lemons. 2.40 Reflecting strong investment, all o f the Patagonian provinces exhibited strong export growth during the early years o f the reform. Interestingly, nearly all Patagonian provinces experienced significant declines, following their peaks (in 1995-1997). Only Chubut (as noted above) had recovered its (1996) peak by 2003. The rest were still exporting 6-8 percent less o f agricultural-based manufactures in2003 than inthe pre-crises period. Table 2.5: Growth Rate of Exports of Manufactures of Agricultural Origin, 1990- Rapid decline Slow decline Growth La Pampa -2.5;-9.3 (1998) EntreNos -9.6;-0.8 (1995) Corrientes-3.3; 5.4 (1996) Jujuy -12.7; -0.2 (1996) 2 Decline Sgo. del Estero-1.2; -10.1 (1995) 0 Y 0 Formosa 1.4; -19.3 (1992 ") Buenos Aires 9.0; -3.9 (1997) Santa Fe 8.4;5.2 (1998) o\ Show growth Salta 0.8; -0.9 (1996) Misiones 6.9;s. 1(1997) Tucudn 8.9;l.g (1996) Chaco 11.O; -6.8 (1994) San Luis 30.7;-6.1 (1998) Chubut 29.8; 0.0 (1996) II Neuqukn 11.6;-6.1 ((1996) Catamarca 34.5; 4.2 (1998) Rapid Santa Cruz 21.8; -5.8 (1997) C6rdoba 16.4; 6.0 (1997) growth 30Negro 11.6;-20.1 (1996) San Juan 17.0; -4.3 (1997) T. Fuego25.0; -7.9 (1995) L a Rioja 34.0; 0.8 (1997) Mendoza 10.2; 1.0 (1997) II Notes: I r each province, the two growth rates are given, and the peak year is inparenthesis. For Formosa, 1992 was selected a 2.41 Santiago del Estero, Jujuy, L a Pampa, Entre Nos, and Corrientes all suffered sharp reductions in agriculture based manufacturing exports (ABME) immediately following the 1991 reforms. Entre Rios, Jujuy and L a Pampa all saw ABME fall over 70 percent in the years immediately following the reform. The trough was reached in 1993 inthe case o f Entre N o s and Jujuy, and stretched over the period 1994-1995 for L a Pampa. ABME grew slightly in Formosa inthe immediate post reform year but experienced an 80 percent drop between 1992 and 1993, and never recovered significantly within the period under review. These provinces represent a wide range o f situations. Santiago del Estero and Jujuy are both relatively uncompetitive in external markets, and consequently have few export-oriented agroindustrial activities. Beef from Santiago del Estero can, in small proportion, be exported, but only after processing in Santa Fe or Cbrdoba. Jujuy is competitive in sugar cane, citrus and high quality paper, but only the citrus 18 i s exported. L a Pampa i s very competitive inprimaryproducts, but agricultural processing tends to take place in Santa Fe and Buenos Aires prior to e~porting.'~Entre N o s is fairly competitive in poultry and grain production, but suffered a decline in exports of rice to Brazil. Of these provinces, only Corrientes has recoveredto or beyond its pre crises export peak, based on growth inwood products, beef, and some citrus. SUMMARY 2.42 In 1991 Argentina emerged from over 50 years of policy bias against agriculture. The result was a boom in agricultural exports, especially from the traditional commodity exporters o f the Pampas, but also from emerging high-value exporters o f fruits, vegetable and wine from the regional economies. Land sown with the 31 principal annual crops expanded 24.6 percent and yielded an impressive 7 percent annual growth inproduction over the period 1988/90 to 1996/98. The more outward-oriented a province, the more successful it was in responding to new opportunities opened up by the 1991 reforms, and the more successful it was inrecovering from the post 1997 crisis. 2.43 Emerging macroeconomic imbalances inthe middle o f the 1990s led, inthe first instance, to a prolonged recession intended to defend the peso, and finally to capital outflow and a financial crisis. Increasing overvaluation dramatically reduced the margin for Argentine commodity producers in the Pampas, and the value o f exports from the regional econ~mies.'~ At the same time, growing recession and unemployment depressed domestic demand for income- elastic foods such as dairy, fruits and vegetables. Finally, the devaluation o f the Brazilian Real in 1998 further exposed Argentine farmers to competition within MERCOSUR, markets, including from Brazilian imports. 2.44 The sector has enjoyed inthe past years strong international comparative advantages and prices inboth primary and processed exports, and has experienced a significant consolidation o f farms and only modest employment creation, at least in the primary crop production in the Pampas. Evengreater growth and employment potential lies inthe regional economies inhigher value added production for both domestic use and export. l3 Since a higher proportion of Pampas' exports are inprimary form they escape this analysis. 14 Inthe following chapter we present evidence suggesting that within the regional economies it was the productsin whichArgentina hadthe highestmarket share that sufferedmost. 19 3. SOME INTERNATIONAL COMPARISONS 3.1 It is helphl when diagnosing performance to have comparators, either with similar characteristics or with variation on characteristics that can help explain difference in outcomes. Inthis chapter we compare Argentina's performance with Brazil and Chile, along a number of dimensions. We use Brazil primarily as a comparator with Argentina's Pampeana region, and Chile as a comparator with Argentina's regional economies, occasionally making also comparisons with other southern cone countries. 3.2 With regard to agricultural performance in Brazil and the Pampas there are a number o f important similarities:l5 e Since 1990, soybean production has tripled inArgentina, while inBrazil it has more than doubled. Argentina's wheat and corn production is up 75 percent and 105 percent, respectively, andBrazil's corn production i s up 40 percent. a Argentina and Brazil have become numbers 3 and 2 soybean exporters, respectively. Together, the two countries have some 50 percent o f world trade in combined soybean- and-product exports, easily surpassing the 35 percent share o f the US. e Economic and political reforms undertaken inthe early and mid-1990 underpinthe surge inagriculturaloutput inbothcountries. e Both countries have abundant land and good soils, and are therefore natural low-cost producers. 3.3 Argentina's regional economies have a number o f characteristics incommon with Chile: 0 They have roughly comparable irrigated area, producing inpredominantly arid and semi- aridclimates. a They are bothmajor producers o fpomaceous fruits, grapes, and citrus. a Both have made significant progress in moving upscale in the export market through quality improvement, product differentiation, andbranding, especially inwines, although, as shown below, Chile has done so more successfully. OVERALLPERFORMANCE 3.4 Not surprisinglyinview o f the history o f policy bias reviewed inChapter 2, agriculture in Argentina has had a long term performance below that o f its neighbors. The value o f agricultural production over 1961-2005 increased 158 percent inArgentina compared to 241 percent inChile and 439 inBrazil (Figure 3.1). Figure3.2 focuses on the post reform period. It highlights the fact that Argentina's agriculture grew considerably faster than Brazil's and Chile's during the early years o f the reform (the six years from 1993 to 1999), and stagnated as Argentina's real exchange rate became uncompetitive duringthe 1999-2001 pre-devaluation period, with a strong recovery since 2002. l5See Randallet a1(2001) for further details. 20 Figure3.1: AgriculturePerformance: Argentina, Brazil, and Chile 1961-2005 - (FA0 Productionindex-PIN) 600 --eArgentina 500 +Brasil Chile 400 300 200 100 l o 1 1 I I Source: FAOSTAT data (2006). Figure3.2: AgriculturalPerformance: Argentina, Braziland Chile 1990-2005 - (FA0 ProductionIndex-PIN) --Argentina Source: FAOSTAT data (2006). 21 LAND RESOURCES 3.5 Argentina's neighbors show contrasting development paths, largely determinedby their land endowments. Figure 3.3 examines land availability for crops in Argentina, Brazil, and Chile. Between 1990 and 2000, Argentina andChile cleared new landby less than one percent o f the 2000 cropland area. On the other hand, the new lands opened inBrazil were equal to nearly 40 percent o fthe 2000 cropland area. Figure 3.3: LandAvailability for Crops: Brazil, Argentina, and Chile 4 5 4 3 5 a .E 3 1 . b 5 1 0.5 0 Chlie Argentina Brazil Source: Schneider (2004). 3.6 The evolution o f agricultural land and land under crops reflects availability (Figure 3.4). From 1961 to 2002 Brazil's total agricultural land grew by 75 percent and its crop land by 134 percent. Because Argentina's overall land frontier (including crop and pasture land) was already virtually closed, crop growth took place mostly through the switch o f pastures to cropland, with cropland growing by 23 percent with little change in total agricultural land. Chile expanded its agricultural land significantly, but with a fall o f 40 percent o f the land in crops. Since the1980s, landinagriculture andincrops bothfell inChile. Figure3.4: Agricultural Land and Landin Crops: Argentina, Brazil, and Chile 1961-2002 - A. Index of TotalAgriculturalland: Argentina, B.Index of Land inCrops: Argentina, Brazil, Brazil.and Chile - 1961-1002 and Chile - 1961-2002 ---`Argentina -Brazil zoo - Chile 22 AGRICULTURAL POPULATION 3.7 Of the three countries, only Chile ended the century with a larger agricultural popu1ationl6 than in 1961 (Figure 3.5). Argentina's and Brazil's agricultural populations were, respectively, 17 percent and 35 percent lower in 2003 than in 1961. All three countries lost agricultural population in 1990-2003 but at different rates: Chile 4 percent, Argentina 11percent, and Brazil 23 percent. As a reflecting o f its roots as a livestock economy, Argentina has a high land-agricultural population ratio: while Chile and Brazil have less than 10 hectares per unit o f population in agriculture, Argentina has nearly 50 hectare (Figure 3.6). This ratio has also been rising much faster inArgentina than inBrazil inthe last decade. Figure 3.5: Index of Agricultural Population: Argentina, Brazil, and Chile - 19 2003 120 110 ' 100 90 80 70 1961 1964 1967 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 20 Source: FAOSTAT data (2006). Figure 6: Agricultural Land Per Ag Population: Argentina, Brazil, and Chile - 161-2002 fl 6 o I O - c I 0 , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . , . , , , , , , , > ~ 16This definition of agricultural population follows FAO, which includes agricultural workers and their dependents. 23 FACTOR PRODUCTIVITY LandProductivity 3.8 Figure 3.7 compares the productivity o f cropland in Argentina and its neighbor^.'^ The increase by a factor o f 4.5 in the productivity o f cropland in Chile since 1983 i s remarkable. During the same period, the productivity o f Argentina's cropland increased 70 percent, and Brazil's 50 percent. Chile's strong performance partly reflects the abandonment o f unproductive cropland, while the incorporation o f new cropland in Brazil and Argentina has probably put downward pressure on land productivity. Chile's results also reflect a remarkably successful shift into highvalue crops. 3.9 Productivity o f pasture land inArgentina has laggedbehindthat o f its neighbors (Figure 3.8), reflecting low efficiency in the cow-calf sector, low levels o f technology adoption, reproductive diseases, and inefficient farm size (ERS, 1998). Due largely to the longer term nature o f the economic payoff, unstable policies and high interest rates have also hurt the livestock more than the crops sector. Over 1961-2002 land productivity in livestock grew only 50 percent in Argentina, compared to 150 percent in Chile and 300 percent in Brazil. Productivity gains in Brazil are particularly impressive since the pasture area grew 61 percent in this period. Chile's productivity growth was also remarkable, with an increase o f 145 percent, while the pasture area expanded 35 percent. Despite a slight decrease in area, the productivity of Argentina's pastures increasedby only a third o fthat inChile and a sixth of that inBrazil. Figure3.7: Evolutionof LandProductivityinCrops: Argentina,Brazil,and Chile(1961-2002) 500 450 400 350 250 200 150 100 Source:FAOSTATdata (2006). 17Figures are based respectively on FAO's crop PIN divided by arable and permanent crops and livestock PIN dividedbypermanentpasture. 24 Figure 3.8: Evolutionof Land Productivity inLivestock:Argentina, Brazil, and Chile(1961-2002) 4505 400 350 300 250 200 1so 100 I I 50 i , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , / , , , , ( Labor productivity 3.10 Labor productivity in agriculture grew at similar rates in Argentina and Chile in 1961- 2002, more than doubling (Figure 3.9). Growth in Brazil was much higher, more than 450 percent. As mentioned above, Brazil and Argentina's labor productivity figures are accompanied by a 17 percent and 45 percent decrease in agricultural population, respectively, while Chile's agricultural population remained relatively constant. Argentina's labor productivity benefits from a highlandlabor ratio, whereas that of Chile benefits from a highoutput/land ratio. Figure 3.9: Index of Agricultural Labor Productivity (1961-2003) Source: FAOSTATSdata (2006). 25 Mechanization 3.11 Figure 3.10 compares tractorization in the three countries with the United States and . Canada. Chile mechanized first and continued to mechanize longer, reaching a degree of mechanization comparable to that of the U S and greater thanthat o f Canada. Argentina beganto mechanize before Brazil, but Brazil mechanized more rapidly throughout the 1960s and 1970s, catching up with Argentina in 1975.' * Figure 3.10: Arable plusPermanent Crops per Tractor: Argentina and Comparators(1961-2002) Argentina -Brazil - - Chile 350 - United States 300 2501 200 150 100 50 3.12 The indicators we have been looking at are partial productivity measures, which measure the average contribution to output o f individual factors. They have the limitation o f being affectedby elements other than the factor inthe index. Total factor productivity (TFP) examines instead the productivity o f the aggregate collection o f inputs. It is measured as the ratio o f the production index over an index of aggregate inputs. 3.13 Table 3.1 shows TFP estimates from two recent sources. Both use FAOSTAT data and include Argentina and other countries within a common methodological framework. Avila and Evenson (2004) use basic accounting relationships. Bravo-Ortega and Lederman (2004) estimate translog production functions. Taking the entire 40 year period 1961-2001 and estimating total '*It should be noted that (i) data do not distinguish between tractors o f different size or power, (ii) FA0 tractor needs have been reduced in both Brazil and Argentina since the late 1980s as minimum and zero tillage have become more widespread, and (iii) farm size structure inArgentina probably favors fewer and larger tractors than i s the case inthe comparators. 26 agricultural production (agriculture plus livestock), Bravo-Ortega and Lederman found annual TFP growth to have beenhighest inBrazil (1.93 percent), second inArgentina (1.84), and lowest inChile (1.20). These values compare to 2.12, 2.11and 1.23 for Australia, the US, and Canada, respectively. Avila and Evenson, on the other hand, break the four decades into two 20 year periods (Table 3.2). For the earlier period they estimate Argentina's TFP growth to be highest (1.83), followed by Chile (0.69), and Brazil (0.49). Table3.1: TotalFactor ProductivityGrowthEstimates Authors Method Period Estimated total factor productivity growth(YOannual) Argentina Brazil Chile Australia US Canada Bravo- Estimated Ortego, trans-log 1961- 1.84 1.93 1.20 2.12 2.11 1.23 Lederman production 2000 (2004) function Evenson -- -- -- Accounting 1961- 1980 1.83 0.49 0.69 (2003) and relationship 1981- 2,35 _ _ _ _ 2001 3.22 2.05 -- 3.14 Consistent with observed technological improvement over the past 20 years, estimates for the period 1981-2001 indicate a sharp increase inTFP growth. Inthis period, Brazil leads overall TFP growth with an annual average of 3.22, followed by Argentina (2.35) and Chile (2.05). Note that the simple average of two 20 year estimates from A d a and Evenson i s broadly consistent with the 40 year estimates of Bravo-Ortega and Lederman. Table 3.2: Output GrowthRatesandTotalFactor ProductionGrowthRates LAC SouthernCone Countries 1961/81and 1981/2001 1961-1981 Countries Crops Livestock Crops & Livestock TFP Output TFP output TFP output Argentina 3.08 2.86 0.90 1.24 1.83 1.86 Brazil 0.38 3.20 0.71 4.28 0.49 3.72 Chile 1.08 1.40 0.24 1.92 0.69 1.53 Paraguay 3.97 5.35 0.36 1.26 2.63 3.53 Uruguay 1.29 1.16 0.32 0.00 0.01 0.18 1981-2001 Argentina 3.93 4.43 0.43 0.92 2.35 2.18 Brazil 3.00 3.60 3.61 4.58 3.22 3.41 Chile 2.22 2.99 1.87 3.92 2.05 3.67 Paraguay -1.01 1.31 1.29 4.17 -0.30 3.27 Uruguay 2.02 2.58 0.53 1.16 0.87 1.48 Source:A. Avila and R. Everson (2004). 3.15 From Table 3.2 we can notice that aggregate output growth follows closely growth in TFP and was significantly higher in 1981-2001than inthe previous 20 years. Output growth was mostly driven by increased efficiency and technological improvement, rather by increased input use. Brazil's high growth rates reflect predominantly input growth inthe earlier period, and both 27 significant input and productivity growth in the latter period. Chile's strong growth during the most recent period reflects both solid technology growth and strong input growth. 3.16 TFP growth in crops in Argentina has been consistently high since 1961, rising from 3 percent per year during the earlier period to nearly 4 percent per year in the latter period. N o other country in the Southern Cone exhibits this consistently high performance in efficiency growth in crops. The contrast with the livestock sector i s remarkable; Argentina shows the lowest productivity gain inthe Southern Cone in 1981-2001, havingbeen the first in 1961-1981. 3.17 Growth in TFP results from the adoption o f new technologies such as zero tillage and new varieties like genetically modified RR soybeans and BT corn that permit higher yields or lower production costs. A country's adoption o f new technologies depends on its capacity to innovate and imitate. 3.18 Avila and Evenson classified countries according to their stren h in these dimensions through an "innovation capital" index and an "imitation capital" index'? Figure 3.11 illustrates the rankingo f L A C Southern Cone countries interms o f innovation and imitation, and the 1981- 2001 TFP growth. Note that Brazil, with the highest TFP growth, had already reached the highest innovation level (according to this index) in the early period and gained two levels in imitation capacity. Argentina was tied with Chile in the earlier period but failed to improve either its innovation or its imitation capacity. Chile improved its innovation capacity to Brazil's level. With the exception o f the ordering o f Argentina and Chile, a country's ranking on the innovationlimitation space corresponds to its rankinginthe TFP space. 19The innovation capital index is based on ISNAR data for agricultural scientistkropland, and UNESCO data on R&D/GDP. The imitation capital index is based onthe years of schooling ofthe working population (males) and the number ofextension workerslcropland. 28 Figure3.11: TotalFactor ProductivityGrowth(1981-2001) and Capacity to Innovate andImitate:LAC SouthernCone Countries I 3.22 Brazil n 6 - I* n 0 Chile 2.05 (no change) Argentina 0 2.35 v 5 - a ti 0.87 Uruguay 3 4 5 6 Imitation Source: Basedon Avila and Evenson, (2004). Note: The base of the arrow indicates a countries ranking on these indices in 1981. The point of the arrow indicatedits ranking in 2000. Argentina made no improvement over the period. The numbers refer to each country's estimated annual TFP growth over the period. AGRICULTURAL EXPORT POSITIONING AND MARKET PENETRATION 3.19 Table 3.3 compares the position in world markets o f Argentina, Brazil and Chile. Argentina has maintained its position inthe cereals market, as the 4th or 5th larger exporter, and has improved dramatically its share inthe soybeans market. It i s now the third largest exporter of bulk soybeans (Brazil is second, andthe US first), and the first exporter o f soybean cake and oil, with Brazil in the second position. With respect to meat and beef, Argentina has been loosing market share since 1970. It i s now the 10th exporter o f fresh bovine meat and the 17th exporter of meat and prepared meat. Brazil, on the other hand, has rapidly expanded its share inthe world market o f livestock products, ranking now from 1st to 4th exporter in all categories. Regarding exports from the regional economies, Argentina has significantly improved its performance in oranges, lemons, pears, and wine since 1970, and incherries and grapes since 1980, and has lost significant positions inapples. Argentina i s now the 2nd largest world exporter o f lemons and the 3rd o fpears. Chile has improved its position inall h i t s considered inTable 3.3 and inwine. It i s now the world's first exporter o f grapes and the 4th or 5th o f apples, cherries andwine. 3.20 Among Argentina, Brazil and Chile, efforts to increase investments in value added downstream production are inversely related to the scope for additional primary production (Brooks and Locatelli, 2004). Brazil and Argentina remain heavily focused on primaryproducts (although Argentina relatively less so), while Chile, with no scope to incorporate additional land, has been very successful inincreasing the unit export values. 29 3.21 It is interesting to compare the exports o f fruits, vegetables and wine in Argentina and Chile. These products have major importance in Chile and are typical o f Argentina's regional economies. Figure 3.12 suggests that the reforms o f 1991 had a significant effect in helping Argentina narrow the value gap with Chile, and that the imbalances o f the period 1997-2001 dealt a severe blow to Argentina's penetration o f world markets. Argentina made substantial progress during the six years following liberalization, reaching export values per hectare 95 percent o f those o f Chile in 1997). With the loss o f high value markets associated with overvaluation and lack o f credit, per hectare export values fell to less than half o f Chile's in 2002. Had Argentina's unit export value continued to catch up with Chile's, the value o f Argentina's exports o f irrigated crops in2003 would have been some US$950 million higher. 3.22 Figure 3.13 looks at the relative success o f Argentina and Chile in expanding into new markets inthe three periods discussed above. We divide exports inthree categories according to market penetration in 1990: (i) high(greater than2 percent ofworld market share); (ii)medium, (between 2 and 0.5 world market share); and (iii) (less than 0.5 percent o f world market low share). In Chapter 2, we found that provinces with larger outward orientation were more successful in increasing exports o f agriculture-based manufactures during the early years o f the reform period, and also more successhl in maintaining exports through the crises period. We expected that markets where Argentina had a low market share would be new, less well established markets, and would therefore be less likely to resist the difficult 1998-2002 period. Hence, we expected the growth profiles o f Chile and Argentina to be similar where market penetrationwas high, and for Chile to have been more effective inincreasing penetration innew markets.The result however is the opposite. Argentina and Chile show similar (modest) growth profiles for low and medium penetration markets. For markets with highmarket share, however, export values o f Argentina fell by 35 percent in 1997-2002, while those o f Chile expanded by 45 percent. Apparently, Argentina's more traditional export industries suffered more than the newly established niche markets, and Chile did historically better inbuildingon those marketswhere it had clearly established strengths. 30 Figure 3.12: Value of exportsof fruits, vegetable andwine per hectare of irrigated land: Argentina and Chile 1600 I Source: FAOSTATdata (2006). Figure 3.13: Growth of export value by 1990 market penetration 1800000 1600000 1400000 1200000 .Medium marketpenetration 1000000 800000 600000 400000 200000 0 1990 1997 2002 1990 1997 2002 Chile Argentina Source: FAOSTATdata (2006). 32 Box 3.1: The contrastingdevelopment ofwine industriesin Argentina and Chile Chile's initial export strategy was to penetrate international markets with average quality wines sold at a low price. To this aim, different strategies were applied. Insome areas small wineries sold out to large ones and concentration permitted scale economies. Other local producers created associations, which reduced transaction costs. The small wineries grouped together in companies (Sociedades Andnimas) which were able to finance investments through capital markets. Thls enabled the sector to attract large FDI inflows. Joint ventures with foreign firms also transferred technology to local producers, a process which was further supported by a number o f governmental projects that funded R&D expendituresto help smaller producers to improve their grape and wine varieties. Chile i s now moving into higher quality segments o fthe wine market, with greater emphasis on quality andbranding (which requires consistent investments inR&D and advertising). Large producers have investedinpromoting brand reputation, while small wineries' associations have been applying quality standards and promotion strategies. Producer associations and governmental bodies (e.g. the government export promotion agency PROCHILE) have succeeded in creating a country image: Vino de Chile. FDI and improvements in the quality o f infrastructure (transportation, logistics and communication systems), ledto Chile becoming by 2000 the fifthworld wine exporter. Argentina has a natural comparative advantage inwine production due to a temperate climate and abundant natural resources. I t is the largest wine producer in Latin America, the world's fifth largest producer and the 9' largest exporter in 2000. In contrast with Chile, the wine industry in Argentina was traditionally fully oriented to the domestic market (Argentina has one o f the world's highest rates o f wine consumption per capita). But domestic demand favored lower priced wines and wine production was dominated by old and low quality grape varieties. The industry started changing at the end of the 1980s when domestic demand of quality wines increased while total domestic consumption fell. Also, export opportunities for quality wines became more apparent and FDI investment started flowing in. However, despite huge potential and on-going changes in the industry, total wine exports accounted for less than 1percent o f all exports inthe late 199Os,with only around 6 percent o f total wine production exported. Moreover, Argentine wine exports are not highly differentiated, as more than half o f total wine exports consist o f bulk or tetra-packed wine, and the composition o f exports interms o f common and quality wine has been very uneven since 1990. The transition from a traditional sector to a modem and differentiated one has been constrained by the quality of grapes. Many farmers have been unable to invest in this area and the lack of coordination between wine industries (bodegas) and farmers has hampered the transformation. There was also little public policy effort to helpthe conversion from a traditional to a modem sector. FDIinArgentina's wine industryhasbeenquite limited. While many foreign groups acquiredwineries inArgentina over the last ten years, the technology transfer didnot materialize at the farm level. The latter would have requireda learning process supported by policy initiatives (training courses; improvement o f information channels between different levels o f the food chain; initiatives from producers' associations). Nevertheless, a stronger presence o f foreign companies and an increase inFDIi s helping the sector to produce higher quality wines. Inaddition, there is a recognition on the part o f the public and private sectors o f the benefits o f creating a national wine brand, as was done inChile. 3.23 Annex 2 analyzes the export marketing strategies o f the regional economies with particular reference to wine (see Box 3.1). The analysis i s consistent with observations by OECD emphasizing the difference in the role o f Foreign Direct Investment (FDI) in the food chain in Argentina and Chile. In Argentina, the first wave o f FDI in the food industry used the country's abundant natural resources to produce commodities. More recently, FDIhas been associated with "supermarketization" and procurement for local and regional markets. Chile's more successful penetration o f value markets has been associated with macroeconomic policies and a stable regulatory environment that encouraged FDI. These policies include improvements in infrastructure through road concessions to the private sector, and measures designed to integrate small farmers into commercial networks, improve access to credit, and extend the benefits o f R&D'(Brooks and Locatelli, 2004). 33 3.24 Brooks and Locatelli (2004) find that both the role o f FDI and that o f govemment has been limited in the case o f Argentina, especially with regard to transferringmodem technology to the farm level. This appears to reflect a lack o f effective partnerships between government and producers associations, as well as lack o f cooperation among different levels inthe supply chain (e.g. bodegas and farmers). InChile, govemment-fimded R&D was an important help to smaller producers to improve grape andwine varieties and to promote a national name. SUMMARY AND CONCLUSION 3.25 Productivity Growth. Argentina's output growth over the past 40 years has taken place inthe crops sector, and has come predominantly from productivity growth. This is the result of an exceptional progress inthe development and adoption o f new crop technology, inthe context o f traditionally highly land extensive agricultural practices with relatively scarce new landbeing available. 3.26 Growth in Argentina's livestock sector has been elusive, with only Uruguay exhibiting lower rates o f growth in output and total factor productivity. We suspect that total factor productivity growth in the livestock sector in Argentina has been somewhat underestimated, however, as high-quality pasture has been converted into cropland and existing TFP measures do not take into account differences in land quality. Nevertheless, INTA technology surveys confirm that adoption o f advanced technology in livestock has lagged, and that the payoff for adoption seems high. 3.27 Investment Climate. It i s noteworthy that Chile had the highest rate o f aggregate growth inagriculture inthe Southern Cone in 1981-2001, despite having the third highest TFP growth figure (Table 3.3). This illustrates the point that while productivity growth is very important, so i s the quantity o f investment. Chile's somewhat lower TFP growth was more than compensated byits stronginvestment and governance climate. 3.28 Research and Extension. Argentina holds a strong position in the Southern Cone in terms o f capacity to innovate and imitate new technology, but has lost ground to Brazil and Chile in 1981-2001. This lag is consistent with Argentina's low unit export values of irrigated crops compared to Chile, and the slow uptake o f advanced technology in cattle, both areas where additional technological efforts may have a highpayoff. 3.29 High value markets. The regional economies made excellent progress in exporting into high value markets during the post reform years. During the six years following liberalization export values per hectare in highvalue irrigated crops reached 95 percent o f those o f Chile. The subsequent crises did serious darnage to the exports from these economies into high value markets, and in 2002 the value per hectare o f fruits, vegetables and wine exports o f Argentina was less than half o f Chile's. Had Argentina's unit export value continued to catch up with Chile's, the value o f Argentina's exports o f irrigated crops in 2003 would have been some US$950 million higher. Surprisingly, it appears that the exports most affected were those where Argentina had large market shares. This i s an area for hrtherresearch. 34 3.30 Marketing strategies for the regional economies. The behavior of Argentinean exporters of wine and table grapes reviewed above and in Annex 2 are suggestive of a trading mentality rather than a marketing strategy. A marketing strategy would target a specific market niche with a product designed or selected to appeal most to a particular segment of the market, and then adopt a long-term market penetration plan that `fits' or `aligns' all aspects of the business to serving the target customer. The data suggests that Argentina competes neither on price, nor cost, but rather hesitates between the two. A transition to higher quality wines started at the end of the 1980s, but it has been slow in showing export results in spite of Argentina's large naturalpotential. 35 4. THE ROLE OF GOVERNMENT 4.1 Government can play a critical role in improving the competitiveness o f agriculture and influencing the well being o f rural households, but can also distort incentives and hence negatively affect competitiveness. Government can help expand the resources for agriculture through the creation o f a favorable investment climate for the private sector; it can invest directly in rural infrastructure and other public goods and services; and it can help improve the productivity o f resources through research and development, extension, and strong and consistent rules o f the game. Government can also promote collective action among private and public sector actors-an important ftnction to enhance the competitiveness o fproduction chains. We address these issues in the present chapter, examining agricultural fiscal policy, the main government institutions related to agriculture, and the role o f government in enhancing ancillary services such as infrastructure and rural finance, and in promoting collective action. Our discussion refers to Argentina but some international comparisons are also made. AGRICULTURAL FISCALPOLICY 4.2 There are several aspects in the analysis o f agricultural fiscal flows. First, we want to know how strong agricultural taxation is and ifthe amount spent by government in the sector i s sufficient to sustain competitiveness and equity-an issue related to the level o f the flows. A connected issue refers to fiscal neutrality, i.e. the balance between the taxation and expenditure flows. Finally, we want to know whether the way in which fiscal resources are extracted from agriculture through taxation and ploughed back into it through public expending i s consistent with competitiveness and equity, i.e. what are the efficiency and equity implications o f the structure o f fiscal flows. 4.3 Several studies throw light on the above issues in Argentina in a historical perspective. Thus, it has been shown that: (i) the best part o fthe 20th century agricultural taxation was during considerably higher than public expenditure inthe sector (Cavallo and Mundlak, 1982; Reca and Parellada, 2001); (ii) tax structure traditionally penalized the sector by decreasing the relative the price between its outputs and inputs, thus reducing investment levels and slowing sector growth (Sourrouille and Mallon, 1973; Ragunaga and Mundlak, 2002); and (iii) public expenditure in agriculture has traditionally been comparatively small in Argentina, insufficient for the needs o f public goods and services. We examine below in more detail some o f these issues in a contemporary framework. Agricultural Taxation 4.4 Tax pressure on agriculture i s estimated at around 26 percent in 2003. Hence, 26 cents out o f each Peso o f value added generated in agriculture inArgentina goes to government. This is very close to the 25 percent general tax pressure in the country (Table 4.1) in that year. Agriculture, thus, contributes to taxation ina proportion similar to the rest o f the economy. 36 Table 4.1: Argentina Tax Structure andPressureon the Agricultural Sector 1998and 2003 (million2003 AR$) 1998 2003 AR%(m) YO AR$ (m) YO Agricultural Taxation A. NationalTaxes Corporate Income tax 588 15.6 711 6.7 Personal income tax 218 5.8 1378 12.9 Minimumincome tax -- 499 4.7 Export tax -- _-__ 4617 43.3 Value added tax 1682 44.6 1348 12.7 Social Security Contribution 394 10.4 285 2.7 Tax onbank transactions -- _ _ 509 4.8 Dieseltax 193 5.1 401 3.8 Personal property tax 74 2.0 153 1.4 Sub-total National Taxes 3149 83.5 9902 93.0 B.Provincial Taxes Sales tax -- -- 202 1.9 Ruralproperty tax - - . 624 16.5 549 5.2 Sub-total Provincial Taxes 624 16.5 750 7.0 Total Agricultural Taxation 3773 100.0 10652 100.0 AgriculturalGDP 23037 41515 Tax Pressureon Agriculture (YO) 16.4 25.7 Source: O'Connor (2004). 4.5 Tax pressure increased between 1998 and 2003. The introduction in2002 o f export taxes (retenciones),20which account for 43 percent o f all taxes paid by the sector in2003, i s the main reason for the 1998-2003 increase; without retenciones, tax pressure on agriculture in 2003 would have been 14.5 percent, less than in 1998 (Table 4.2). Other changes were also made in the tax system in this period. In particular, two more taxes were introduced with a relevant impact on agriculture: the minimum income tax (impuesto a la ganancia minimapresunta), and the tax on bank transactions (impuesto a Zos cheques). 4.6 Provincial taxation falling on agriculture is small; it amounted to only 7 percent in 2003. It consists o f a sales tax (which is different from the national VAT) and a personal property tax (on rural property in the case o f agriculture). There are other provincial taxes like the stamps tax (impuesto a Zos seZZos) and the vehicles tax (impuesto automotor), but they have little incidence on agriculture. The importance of provincial taxation and the incidence o f individual provincial taxes change much across provinces. The dispersion o f provincial tax incidence i s big, incidence beinggenerally higher inricher provinces like Buenos Aires, C6rdoba and Santa Fe. 2oExport taxation of agricultural products has beena recurrent event inArgentinean history. 37 Table 4.2: Argentina: Tax Pressure on Agriculture and the National Economywith and without Export Taxes, 1998 and 2003 (percentages) 1998 2003 %Change Agriculture only 16.4 25.7 56.7 Agriculture less export taxes 16.4 14.5 -11.6 NationalEconomy (all sectors) 18.2 25.2 38.5 NationalEconomy less export taxes 18.2 22.6 24.2 Source: O'Connor (2004). 4.7 There are considerable differences betweenthe tax structure o f agriculture andthat o f the national economy (Table 4.3). In agriculture, the importance o f export taxation is much larger than in the rest of the economy, while that o f the value added tax is much smaller. Not surprisinglyinview o f the highincidence o f self-employment inagriculture, the contributions to social security are also much smaller inthe agricultural sector. Table 4.3: Argentina: Structure of National Taxation, MainTaxes, 2003 (millionAR$) Taxes All Sectors Agriculture AR$(m), YO AR$ (m) YO Value added tax 20,948 29.0 1,348 13.6 Income taxes 14,751 20.4 2,588 26.1 SS Contributions 9,668 13.4 285 2.9 Export taxes 9,212 12.7 4,617 46.6 Other taxes 17,696 24.5 1,064 10.7 Total Taxation 72,275 100.0 9,902 100.0 Source; O'Connor (2004). 4.8 There are potential problems o f distortions with the current system o f agricultural taxation in Argentina. One case is export taxation (retenciones). They were introduced in 2002 with the double purpose o f (i) off windfall gains to agricultural exporters resulting from wiping devaluation and rising world prices, and (ii)ameliorating the impact o f the recession by decreasing the domestic rice o f exportable food staples and also by raising public revenue to finance social programs.2P The introduction o f export taxation was an emergency response to the crisis to achieve these worthwhile objectives, but maintaining it i s not necessarily an efficient way o f meeting them in a longer perspective. First, retenciones discriminate against exports vis- a-vis importables and non-tradable. Second, agricultural rents could be dealt with using other instruments like income and land taxation, and social programs are better financed from the ordinary budget than from earmarked taxation. Keeping low the domestic price o f exportables (which in the case o f Argentina coincides with basic staples) through export taxes implies a forceful transfer o f income from producers to consumers, with economic effects that should be considered. During times o f high export revenue, farming can be profitable in spite o f the transfer, but as soon as conditions normalize the transfer i s at the expense o f negative incentives to farm production and investment. Also, it i s usually during booming periods when delayed agricultural investments are made and many farm loans are paid back. Cutting profits in these 21 Much in fact o f the proceedings o f the export tax goes to finance the program Jefas y Jefes, a conditional cash transfer to the poor. 38 periods may slow investment and the repayment o f loans. The impact o f taxation on farm profitability is illustratedinBox 4.1. Box4.1: Profitabilityof aMixedAgriculturalCompanyinthe Pampasin2004 We examine in this box the impact o f taxation on the profitability o f a mixed crops-livestock company in the Pampas. Figures are from an actual farm located inthe western part o f Buenos Aires Province. The farm has 1,500 hectare of which 25% planted with sunflower, 32% with maize, 10% with wheat and 33% devoted to beef fattening. Agricultural technology is average for the zone and the farm uses its own machinery. Yields are: 7.0 tonshectare of maize, 2.8 tonshectare of wheat, and 2.0 tonshectare o f sunflower. The livestock yield i s 384 kghectare. Prices are those prevailing in the area in 2004. The farm i s valued in AR$12.8 million. Enterprise results are shown in the following table. Pesos Return on Assets (YO) VALUE OF ASSETS 12,782,875 GrossIncome 2,209,392 minus Livestock purchases 349,578 Inputs& interest payments 652,109 Labor costs 158,595 Marketing costs 339,497 Indirect costs 77,964 Various taxes 83,964 GrossProfit 547,685 4.3 minus Depreciation allowance 111,510 0.9 Net Profit 436,175 3.4 minus Income tax (35%) 149,662 Distributable Profit 286,513 2.2 Total taxation 233,626 1.8 Net profit Without taxation 520,139 4.1 The return on assets is low, only 2.2% ofthe distributable profit. Total taxation is AR$233,626 and amounts to 45% o f the net profit without taxation. Two taxes are not included in the table because they are recoverable from thc income tax: the minimum income tax, and the diesel tax. These taxes amounted respectively to AR$76,697 anc 8,570, with the latter being paid inadvance by the owners. Had the net profit been less than the minimum tax, i.e AR$76,697, the distributable profit would have beennegative. That was infact the case inthis farm in 1999. Source:O'Connor (2004). 4.9 Because o f the importance o f retenciones, agricultural taxation falls heavily on export crops, especially on grains and oilseeds. Grains and oil seeds pay 23.5 percent and h i t s , horticultural products, honey, rice and other minor crop exports are taxed at 10 percent. Regionally, this means that the Pampas bear a heavy burden o f agricultural taxation relative to the other regions. This i s so for three reasons: (i) a much higher proportion o f the Pampa's agricultural production is exported than for the other regions; (ii) the tax rate for Pampa,products i s higher; and (iii) agricultural incomes and the value o f production i s also higher. Despite the higher export tax, the natural advantage o f the Pampas in grains and oilseeds is so strong that there is no evidence o f substitution to other crops. 39 4.10 Lack o f consistent rules o f the game represents a serious constraint to investment in the agricultural sector in Argentina. Unlike a tax on income, an export tax comes "off the top", leaving producers' incomes much more exposed to commodity price variability. To an unhealthy measure the current system causes agricultural profitability to depend on year-to-year policy decisions. 4.11 The main purpose o f the minimum income tax (impuesto sobre ganancias minimas presuntas) i s to prevent evasion inincome taxation by taxing imputedrather than actual income, which i s often difficult to measure. The drawback is that it penalizes producers who because o f beingat the start o f an investment project or for other reasons, e.g. climatic or pest hazards, have low or no profits. Vulnerable producers, more exposed to economic shocks, can be particularly hit. Farming, where profits-to-assets ratios are normally lower than in other sectors, is also especially penalized. Furthermore, like retenciones, the impuesto sobre ganacias minimas presuntas is not neutral with respect to the economic cycle; its negative impacts are felt mostly during the downturn. Evasion could perhaps be better fought through other means, such as simplifyingthe tax system and improving information andcollection. 4.12 Evasion i s also a problem in other areas o f agricultural taxation. Thus, an AFIP (1999) study estimates VAT evasion in 1997 and 1998 at around 26 percent for the national economy but 67 percent for agriculture. Another study by Libonatti (2000) also shows VAT evasion in agriculture to be well above the average. A different area o f large evasion in the agricultural sector i s in the contributions to social security, following the high incidence o f informal employment in agriculture. From the 1997 input-output table, informal labor use in agriculture, and hence the evasion o f SS contributions, has been estimated by O'Connor (2004) at 76 percent in horticulture and fruit production, 63 percent in industrial crops, 61 percent in livestock production, and 46 percent ingrains. 4.13 Agricultural taxation does not seem to have a significant direct impact on poverty, since poor producers generally do not pay income tax, retenciones (most o f which fall on Pampean grain production) or VAT. They may pay the municipal impuesto immobiliario rural, but this is low. Indirect effects could be more significant to the extent, for instance, that retenciones lowers the price o f exportable products for all producers, not only for exporters. But this would depend on whether poor rural producers are net producers or net consumers o f the relevant foods. 4.14 Although this study contains a review o f fiscal transfers to and from agricultural (Background Paper 11) we recommend a more fully-elaborated treatment o f both fiscal and implicit transfers to and from agriculture, including a more complete analysis o f the issues o f export taxes, tax evasion in the sector, and local and provincial tax effort,` and the impact o f agricultural taxation on relevant variables like prices and wage rates. The analysis needs to be carried out within the context o f Argentina's overall tax system and fiscal equilibriumneeds, and address the issue o f regional inequalities and the large regional variation in economic rent to land. 40 PublicExpenditureinAgriculture 4.15 Contrary to taxation, which i s overwhelmingly national, public expenditure in Argentina i s more or less equally shared between the national government and the provincial plus municipal governments. There is, thus, a high degree o f vertical fiscal inequality in Argentina (Table 4.4). 4.16 National public expenditure in agriculture is extremely low by international standards. Expenditure in2005 was AR$613 million out o f a national total o f ~ ~ $ 7 7 , 9 1million, which is8 only 0.8 percent. This i s less than in 1998 (Table 4.4), andmuch less than the share o f agriculture inGDP.We do nothave aggregate figures for provincialgovernment spending inagriculture, but from a fiscal analysis o f four provinces (Buenos Aires, Catamarca, Mendoza, and San Juan) it seems that provincial spending in agriculture i s larger than national spending but is also low (Figure 4.1). Agriculture expenditure inthese four provinces is also small as a proportion o f their total and agricultural GDP, ranging from 5.4 percent o f provincial agricultural GDP inMendoza to 2.9 percent in Buenos Aires. Comparing with the taxation figures reported above, it is clear that there i s a strong imbalance between the fiscal flows coming out o f and going into agriculture. Table 4.4: Argentina: Composition ofPublicExpenditure, 1998 and 2005 (millioncurrentAR$) National Agriculture" % of national AR$(m) AR$ (m) expenditure 1998 47,531 516 1.09% 1999 50,046 473 0.95% 2000 49,720 449 0.90% 2001 48,903 384 0.79% 2002 46,980 254 0.54% 2003 58,867 348 0.59% 2004 64,828 465 0.72% 2005 77,978 613 0.79% Source: MECON (2006). * Includes SAGPyA, INTA, INV, INIDEP and INASE. SAGPyA expenditure includes the following programs: Formulacih de politicas del sector primario, Programa social agropecuario - PROINDER, Promocion de comercio y produccih de semillas (years 2001-2003) ,and Atenci6n del estadode emergenciapor inundaciones. 4.17 Figure4.2 shows the evolution o f an index o f total and agricultural expenditure from the national government in real terms from 1998 to 2005, and Figure 4.3 shows the evolution o f national expenditure in agriculture as a share o f agricultural GDP during the same period. National expenditure, which grew between 1998 and 1999, fell dramatic in 2002 as a consequence o f the crisis. Contrariwise, national expenditure in agriculture fell from 1998 to 2002 but grew in the following years. As a share o f agricultural GDP, national public expenditure in agriculture decreased strongly after the crisis. This was due to a combination of factors: (i)a change in relative prices in favor o f agriculture due to the devaluation; (ii) the increase in international prices o f Argentina's export products; and (iii) the comparatively strong performance o f agriculture duringthe crisis. 41 Figure4.1: Public ExpenditureinAgriculturein 2003 by Four Argentinean Provincesas a Proportionof all ProvincialExpenditure, ProvincialGDP and ProvincialAgriculturalGDP (percentages) Bs. As Mendoza Catamarca San Juan Source: Basedon O'Connor (2004). Figure4.2: Indexof Public NationalExpenditureTotaland in Agriculture inRealTerms from 1998to 2005,1998=100 * +Agriculture +National 120.0 , I 20.0 42 Figure4.3: NationalPublicExpenditureinAgriculture as a Share of Agricultural GDP, 1998to 2005 (percentages) 14.5% 3.9% 4.0% 3.5% 3.0% 2.5% 2.0% 1.5% 1.O% 0.5% I0.0% 1998 1999 2000 2001 2002 2003 2004 2005 Source: Table 4.4 for public expenditure and INDECfor agriculturalGDP. 4.18 Government support to agriculture inArgentina has been historically low relative to other L A C countries. Table 4.5 compares Argentina with selected comparator countries. We look at government expenditure on agriculture as a proportion o f agricultural value added, and as a proportion of total government expenditure. The first measures the support to agriculture relative to agriculture's importance in the economy. The second is an indicator of its priority relative to other government expenditures. Table 4.5: Government Expenditureon Agriculture Percent AgricultureValue Added 1997 1998 1999 2000 2001 2002 2003 Argentina -- -- -- 5.7 6.1 1.9 -- Australia -- _ _ ---- -- -_ 14.1 13.9 13.9 Brazil 8.7 9.3 -- -- _ _ -- Canada _- -- -- 28.7 11.1 12.6 Chile -- -- _ _ -- -- _- -- 3.4 Mexico _ _ -- 10.3 10.4 _- -- Uruguay -- -- 8.6 6.2 6 -- -- Percent Total Government Expenditure Argentina -- -- -_ 1.o 1.o 0.7 -- Australia _ _ -- 1.5 1.5 1.5 Brazil 2.6 2.5 -- -- -- -- -- -- -- Canada -- -- -- 1.7 0.7 0.8 Chile -- _--_ -- -- -- _ _ 1.4 Mexico -- 4.1 3.1 2.7 -- -_ -- Uruguay -_ -- 1.5 1.2 1.2 -- _ _ Source: 2004 GovernmentFinance StatisticsYearbook, IMF. 43 4.19 Although lack o f consistent data coverage prevents firm conclusions, the overall impression is that Argentine farmers receive substantially less than their competitors by both measures.22 Looking at pre-devaluation numbers, Argentina's support to agriculture appears to have been around 6 percentage points o f agricultural GDP while that o f Brazil was 9 percentage points. Looking at post devaluation numbers, Chile's support i s 3 percent o f GDP and Mexico's 10 percent-compared to 2 percent for Argentina, As a percent o f total government expenditure, only Canada in 2002 falls to the same low level (0.7 percent) as Argentina.23 By this measure, Uruguay appears to spend some 20-50 percent more than Argentina, and Brazil more than twice. Note that in developed countries the agricultural sector i s small relative to overall government expenditure. Thus, for Canada and Australia large contributions relative to the size o f the sector take up a small proportion o f the budget. 4.20 The FA0 study o f public expenditure in agriculture in L A C countries offers another set o f comparator estimates. Figure 4.4 shows data on public expenditure in agriculture per agricultural worker for the average o f 1996-2000, and Figure 4.5 shows an Orientation Index, defined as the share o f agriculture innational expenditure divided by the share o f agriculture in GDP.24 The dispersion o f the figures o f expenditure per worker i s big, with some countries spending twenty times more per agricultural worker than others. Chile i s first, expending per worker close to US$400, followed by Mexico with US$303. Argentina is below the L A C average and also below its most direct competitors, Chile and Brazil. This is remarkable inview o f the highland-labor ratio characteristic o f Argentinean agriculture. 4.21 The agricultural orientation index in Figure 4.5 measures the intensityo f the fiscal effort in agriculture relative to the economic importance of the sector. A value of one indicates that expenditure in agriculture i s in accordance with its economic importance, less than one means "discrimination" against agriculture, and the opposite if it i s more than one. Here again the dispersion among countries is large. The low value o f the index for Argentina, well below the L A C average, confirms the scarce attention that public expenditure in agriculture has traditionally received in Argentina, compared to other countries in L A C among which its direct competitors, Brazil and Chile. 22Note that when comparing expenditure as a percent o f agricultural value-added it i s important to keep inmindthe effect o f currency overvaluation o n the agricultural value-added. As indicated in the text, the dramatic drop in Argentina's ag expenditure/ag GDP in 2002 i s largely a result o f the increase in ag GDP from devaluation. In comparing Southern Cone expenditure by this measure, note that all Southern Cone expenditure with the exception ofArgentina 2002 is prior to the major devaluations that each country experienced over the 1999-2002 period. 23Note that due to the much smaller size o f Canada's agricultural sector relative to government expenditure, this 0.7 percent o f total government expenditure translates into 11percent o f ag GDP in Canada, and only 2 percent of ag GDP inArgentina. 24Figures refer to expenditure inproduction-related programs only, not to all rural development expenditures. They are incurrent U S dollars. For some countries the average is for 1996-1999 due to lack o f data for 2000. Not all rural development spending is included, only that devoted to productive development. 44 Figure4.4: PublicExpenditureinAgricultureper AgriculturalWorker 450 , inLACCountries, Average 1996-2000 - I 400 350 300 250 200 150 100 50 0 Source: World Bank (2005), based on Kerrigan (2003). Figure4.5: AgricultureOrientationIndexinLAC Countries, Average 1996-2000 Source: World Bank (2005), based on Kerrigan(2003). 4.22 With its low expenditure on agriculture, Argentina cannot afford to divert resources from public to private goods. Yet, in2003, 37 percent of Argentina's agricultural expenditure appears to be so directed (Table 4.6). Expenditure on public goods includes agricultural research and extension (INTA, 26 percent; INDEP, 1 percent; INV, 2 percent), sanitary and phytosanitary control (SENASA, 20 percent), and seed quality (INASE, 0.5 percent). Additional expenditure o f a public good nature passed as transfers from the Secretariat o f Agriculture (SAGPyA) to the provinces, including externally financed projects directed to support small and medium producers (PROSAP, PROINDER, PRODERNOA, PRODERNEA), and emergency for farmers inzones affected byflooding. 45 Table 4.6: Public vs. Private Goods: Classification of Government Expenditures inRural Sector, 2003 Millionsof AR$ Percentage Public Private Public Private SENASA 128 -_-__ 20 ----- INTA 168 ----- 26 __----_ _ _ _ INIDEP 9 ----- 1 INASE 3 __-__ 0 ----- INV 11 _---- 2 ----- SAGPyA 85 52 8 FET ----- 187 _ _13_ _ _ 29 TOTAL 404 239 63 37 Source; O'Connor (2004). 4.23 The largest expenditure o f a private good type is the Fondo Especial del Tabaco (FET), constituting some 30 percent o f total national agricultural expenditure, and financed by an earmarked tax on cigarette purchases. This tax provides price supports and other incentives to the tobacco industry. In addition, nearly 40 percent o f SAGPyA expenditure goes to subsidies for forestry, sheep production, and subsidized credit, largely o f a private good nature. 4.24 As shown recently by L b p e ~ the , ~ ~cost to agricultural growth o f subsidizing private rather than public goods i s high. For the nine countries in his study L6pez calculated that reallocating 10 percent o f subsidy expenditures to supplyingpublic goods may cause an increase in per capita agricultural incomes of about 2.3 percent, without increasing total government expenditures. At 37 percent, Argentina's agricultural expenditure on private goods i s somewhat under the 44 percent average for the nine LAC countries analyzed in L6pez's study.26While Argentina's private goods portion i s not high by Latin American standards,27 coupling low overall agricultural expenditure with a 37 percent proportion on private goods leads to the risk o f a substantial under-expenditure on vital public goods. AGRICULTURAL PUBLIC SERVICE INSTITUTIONS 4.25 We examine briefly in this section the two main decentralized public institutions providing vital services to the agricultural sector: the Instituo Nacional de Tecnologia Agropecuaria (INTA) and the Sewicio Nacional de Sanidad y Calidad Alimentaria (SENASA).~~ 25 Ldpez, Ramh, 2004, Why Governments Should Stop Non-Social Subsidies: Measuring the Consequencesfor Rural Latin America. 26These countries are Costa Rica, Dominican Republic, Honduras, Panama, Paraguay, Peru, Venezuela, Ecuador, and Uruguay. 27InLopez's study containing 9 Latin American countries, 4 spent a lower percent on private goods and 5 spent a larger percent. 28 For reasons of space we do not deal with other, less important decentralized public sector institutions operating in the agricultural sector such as the Instituto Nacional de Semillas (INASE), the Instituto Nacional de Vitivinicultura (INV)or theInstituto Nacional deInvestigacidny Desarrollo Pesquero (INIDEP). 46 INTA 4.26 INTA, a decentralized institute o f the Secretariat o f Agriculture, created in 1956, is SAGPyA's arm for the development and transfer o f food and agriculture technology. It has its headquarters in Buenos Aires, and regional operational offices throughout the country. Its main functions are: (i) the formulation o f strategic and operational policies for research, innovation and extension in food and agriculture; (ii) the design and implementation o f work programs in the area o f food and agriculture research and extension; (iii)liaison function with enterprises a and other institutions to implementjoint programs inits area o f specialization; and (iv) to design and implement communication and institutional strengthening strategies. INTA participates in various inter-institutional entities, and has agreements with many agricultural technology firms and research bodies inand outside Argentina. 4.27 Overall guidance is ensured by a Governing Council integrated by a President, a Vice- president, and eight members representing different institutions (university faculties o f agriculture and veterinary medicine, Federacidn Agraria, Sociedad Rural, CONINAGRO, CRA, and AACREA). Day-to-day operations are under the responsibility o f a National Director and four Assistant Directors (for Scientific Research and Development Planning, Operations, Human Resources, and Administration). 4.28 The territory is divided by INTA into 16 regions, each with a Centro Regional and a Direccidn Regional, There are 42 Experimental Stations in charge o f basic and applied research and extension activities, and 92 Rural Extension Agencies devoted to extension andparticipatory research. 4.29 A number o f important successes can be claimed by INTA over the last 10 or 15 years. Major among them: (i) its contribution to the development and dissemination o f the direct planting, zero tillage technology (see Box 2.3 inChapter 2); (ii) the development (or contribution to development) o f a number o f specialized grain and fodder varieties suitable for particular agro-ecological conditions; and (iii) the dissemination among small and medium farmers o f modern management techniques through the Cambio Rural program. Many o f INTA's contributions were frustrated by the extreme shortage o f budgetary resources during the 1990s and at the beginning o f the 2000s, which resulted in the loss and insufficient renovation o f qualified staff, little staff training, and strong shortage o f operational funds. Fortunately, the budgetaryconstraint has beensubstantially reduced inthe last years. 4.30 INTA faces a number o f difficult but important challenges. Major among them is the settingup o fplanningmechanisms and operational programs consistent with the bigchanges that have been taking place inArgentina's food and agriculture sector over the last 15 years. We can include here the development o f biotechnology, the intensified participation of the private sector inagricultural research, new environmental threats, andthe difficult competitiveness conditions faced by the small farm sector. The new Plan Estratkgico o f INTA has faced up to these challenges, and defined appropriate areas o fwork, methods and activities. 4.3 1 There are, however, several strategic areas for technology development and disseminationwhere action inthe short and medium term seems particularly important: 47 0 Environmental sustainability.Technological and other developments in the sector are posing new or exacerbating old environmental hazards, which demand technology and extension answers. Among them are deforestation and loss o f biodiversity due to the expansion o f the crop frontier (see Chapter 6); soil and aquifer contamination from increased pesticide, herbicide, and fertilizer use; loss o f soil fertility because o f reduced mixed farming and crop rotations; and soil deterioration from inadequate irrigation practices. Irrigation.Irrigation is not included among INTA's priorities, yet research and extension are urgently needed in relation to irrigation efficiency, organization and governance o f water schemes, water cost and charges, and others (see Chapter 8). 0 Bioenergy. Bioenergy is expected to be instrong demand inthe mediumand longrun.It i s important to carry out the technology research that could permit Argentina to compete inthis field, where it is already well behindBrazil. Important areas for research are the selection o f primary materials, industrial plant characteristics and scale, logistics, and market aspects. 0 Quality systems. INTA launched in 2001 the Programa Nacional de Calidad to disseminate production protocols and the certification o f good practices, with training as the main instrument. This i s an important program because o f the need to enhance and standardize food quality inimportant food chains (see Chapter 7). The program could be enhanced by building stronger alliances with provincial governments and the private sector. 0 Developmentof non-traditionaland specialtyproducts. Concentration o f exports on a few products poses evident risks and signals an inadequate use o f Argentina's high potential for diversification. There i s hence the challenge o f developing technologies for products with highvalue added indifferent agro-ecologicalniches. SENASA 4.32 SENASA, the government's entity responsible for phytosanitary protection and food quality and safety, is a decentralized agency. It operates under SAGPyA, and has 25 regional and 319 local offices. Its Government Council consists o f 15 membersrepresenting producers, agro- industry, consumers, the provinces, and SENASA staff. It has eight Departments: (1) Animal Health, (2) Plant Protection, (3) Food Inspection, (4) Laboratory and Technical Control, (5) Agrochemical, Pharmacological and Veterinary Products, (6) Technical, Legal, and Administrative Coordination, (7) Operational Vigilance, and (8) International Coordination. SENASA employs some 3,800 staff, o f which slightly over one third have completed university studies. The average age i s over 50 years, and hence there will be a highrate o f retirement over the coming decade. SENASA's mandate has expanded rapidly over the past 15 years. 4.33 SENASA has broad powers to plan for, diagnose, and control or eradicate animal and plant pests and diseases throughout the country. This includes certifying export quality, establishing phytosanitary and epidemiological zones and frontiers, and applying whatever measures are necessary to safeguard the health o f Argentina's plant and animal assets. Also included are all measures related to food safety, including slaughter, warehousing, transport, use o f additives, biotechnology, and veterinary medicine. In addition to its traditional role in ensuringanimal health (inparticular controlling FMD), it has assumed primary responsibility for 48 creating and maintaining access to export markets. The complexity o f international food and biosafety regulations imposes the need o f continuous staff training, including not only SENASA's own staff, but also that o f other national, provincial and municipal institutions, public andprivate. 4.34 SENASA's great strength is in its technical capacity and processes (see Chapter 7). Its weaknesses, some o f which are self-ackn~wledged,~~are in strategic planning, the continued presence o f double standards between domestic and export markets, communications and information management, and insufficient capacity to create and sustain partnerships at the local level. SENASA's management proposes a strategic reassessment o f the institution's role with emphasis on the following: a Convergence of standards. A process o f convergence o f quality standards i s proposed for the domestic and export markets, under a system o f "co-responsibility" o fmembers o f the product chains and sanitary authorities. a Clarification of roles and responsibilities of different jurisdictions and the private sector. SENASA proposes clarifying roles in areas o f overlapping federal and provincial jurisdictions, and, where appropriate, delegating certain responsibilities to private certification agencies or to producers. a Decentralization to the regional level. The central level would retain the national functions o f planning, strategizing, establishing norms, providing orientation, strengthen- ing capacity, and auditing performance. Operational functions would be decentralized with the objective o f creating better opportunities for coordination with other public and private entities and bringing services closer to clients inthe interior o f the country. a Ensuring financial resources. The new model o f cooperation would require institutional agreements that guarantee the financing o f agreed activities. Achieving a large degree o f financial autarchy would be important for SENASA. 4.35 SENASA's new strategic plan shows a clear resolve to devolve and delegate responsibility within a clear framework o f shared responsibility. The challenge is to change its culture as an inward looking institution which resists giving up historic tasks that may not be the most strategic under the new circumstances. Its focus would be best kept on its role as the most strategic player inits area aiming at "making things happen" rather than "doing things itself'. ANCILLARY SERVICESAND THE ROLEOF GOVERNMENT 4.36 We examine in this section two fundamental ancillary services to farm production on which agricultural competitiveness strongly depends: agricultural infrastructure and processing facilities, and rural finance. Shortcomings in the provision o f these services are affecting the profitability and competitiveness o f the agricultural sector. We provide a brief diagnosis o f the situation o f these services in Argentina today and give some suggestions as to possible government interventions to improve them. 29See "Change and Strengthening of SENASA", SENASA, 2004. 49 Infrastructure and Processing Facilities 4.37 Production o f agro-based products requires support o f four infrastructure pillars: roads, navigable ways and ports, storage centers, and facilities for the preservation o f fresh produce. Lack o f these facilities negatively impacts costs, timeliness o f transport, quality o f final products, and the use o f potential production areas for lack the necessary investments. It has thus a major impact on competitiveness. Most affected are non-perishable bulky products and high value perishable ones. 4.38 InArgentina, infrastructure needs relate mainly to three product groups: grains, which require good storage and shipping infrastructure; meat, which demands quality processing facilities; and h i t s and vegetables, which need good conservation facilities. The combined output o f these three groups amounted to over 95 million tons in 2003 (more than 2.4 ton per inhabitant), and i s expected to grow at about 2.5 percent per year, reaching over 125 million tons inadecade. 4.39 The importance o f improved roads can be inferred from the following: transport o f 70 percent o f the grain crop (approximately 50 million tons) requires some 2 million truck trips for a mean o f 400 kilometers per trip during 6 months, concentrated during harvest time. Grain classification requires some 35,000 storage centers with an individual capacity o f 2,000 ton each. These facilities are not available inArgentina nowadays. 4.40 The importance of improving slaughtering and meat processing facilities can be gauged from the fact that equipment complying with international standards can serve at present only 25 percent o f output, estimated at 13.5 million tons per year. Uniformity and quality i s a requisite to improve beef production and increase exports. Quality has an effect on price. Poor meat processing results in price losses in processed products o f approximately 15 percent. In an industry that processes more than 13 million tons for approximately AR$54 billion, lack o f quality can result inlosses o f over AR$8 billion per year. 4.41 Cold storage o f vegetables and h i t i s essential to small and medium size producers. Major hit-producing areas like central Tucumh (lemon), Rio Negro valley (apples, pears, grapes, vegetables), northeast Entre Rios and southern Corrientes (citrus) harbor small fanners, but their production is linkedto that o f large ones, and hence their lack o f cold storage facilities is less acute. New production areas like the mountainous regions o f Rio Negro, Santa Cruz, and Neuquen, where production i s in the hands o f small farmers, show greater deficiencies. These deficiencies are linked to products with a gate value o f some AR$6,000 per ton, with highexport potential (especially cherries andbilberries). 4.42 Lack o f adequate facilities results in a remarkable loss o f opportunities to develop good producing areas. This was shown by Peri et a1 (1998) with respect to fine h i t s in Chubut. Del Pino et a1 (2004), show that the availability o f roads is one o f the main causes o f soybean and other grain expansion inthe north o f the country. 4.43 Work by Pesce (2003) and U.N.del Sur (2001) assessing the infrastructure and logistics needs for grain production inArgentina show the following hurdles inmajor Pampean areas: 50 The condition o f storage plants ranges from average to bad in Entre Nos and from average to good inthe South o f Buenos Aires province. Inboth cases the storage, drying and reception capacities are compromised. Regarding equipment maintenance, quality o f management and product losses, most plants inboth areas range from partially adequate to deficient. Investments to upgrade these plants are feasible and would save some US$2.85 per stored ton. Extrapolating this information we obtain a national saving figure o f over US$50millionsper year. Bagged grain i s more than 20 percent o f total output inArgentina (some 15 million tons). The storage cost per ton and per season in 2003 was AR$2.46, 1.8 times the cost o f storage in "fixed" storage centers (AR$1.36 per ton and per year). The difference incost for a volume o f 20 million tons i s US$20 million, to which the classification and storage advantages o f "fixed" centers should be added. 0 The increase o f processing costs due to drying facilities using expensive energy sources (wood, gas-oil) instead o f cheaper ones (gas) is significant. Highdrying costs causes loss o f competitiveness equivalent to 0,72 percent o f the value o f grain production, equivalent to some AR$230 million. 0 The proportion o froad infrastructure ingood repair is low (not higher than 15 percent) in the districts o f the areas analyzed, However, the budgets o f Provincial Road Departments have fallen during 1992- 2002 (but there is a remarkable rise in the period 2003-2004). Meanwhile output has doubled and the use o f roads increased notably. Cost increases derived from bad road quality are estimated at 0.25 percent o f grain value equivalent to AR$700 million. 0 Transport by river could increase tenfold in volume from its current level o f 2.8 million tons per year by taking advantage of multimode transport strategies. This would have important economic advantages since the energy cost o f transport by barges is five times lower than by train and 11 times lower than by truck. There is, however, insufficient availability o fbarges and lack o f competition inthe provision o f this service. 4.44 A strategy for the development o f agroindustrial infrastructure (roads, navigable ways, storage centers, ports, equipment for processing) involves the public and private sectors. Public sector participation i s required at different levels, because the orderly planning o f roads and storage facilities involves national, inter-provincial, provincial and municipal jurisdictions. A national program for the development o f infrastructure seems to be required. It could include actions regarding diagnosis, training, planning, and institutional strengthening, and sources o f financing. The Ministry o f Planning has already taken some steps inthis direction. 4.45 Local participation could also be fostered. Road consortia with more than 10 years o f experience (albeit with some frustrations) could be evaluated and perhaps revived. The Chilean experience in this regard could be analyzed. In Chile, the expansion o f ports, roads and airport facilities through a system o f concessions to private firms, which resulted in large saving o f fiscal resources (Valdes, 2005), has been extended to the management o f secondary roads (with a government subsidybecause o f less traffic volume). 51 RuralFinance 4.46 Monetary stability during 1992-98 influenced positively sector financing, but this was reversed by the 2001-02 crisis, which led to the virtual disappearance o f bank lending to agriculture. Slow recovery started in 2003. In 2004, bank lending was below 12 percent o f the value o f output o f primary agriculture. This contrasts with the U S (around 45 percent) and Chile (around 40 percent). The deposits/loans ratio i s also high and the availability o f term financing very exiguous. 4.47 Bank lendingto agriculture comes mainly from government banks like the Banco Nacidn andthe Banco de la Provincia de Buenos Aires, with more than 70 percent o f lendingbenefiting medium and big producers. In2000, the Secretaria de la Pequefia y Mediana Empresa together with Banco Nacidn and other banks launched an important lending program for small and medium enterprises for which rural producers are eligible. Few loans, however (no more than 2,000), have yet gone to the sector. Banco Nacidn has also a program for small and medium producers jointly with the Consejo Federal de Inversiones, which has been on-going since 1994, butthe number o floans i s also small andthe approvalprocedures are complex. 4.48 There are other sources o f funding for small producers from special programs and international financial NGOs, such as the Grameen Foundation, Promujer and others. The work o f these institutions, however, is recent in Argentina and they have not yet granted many loans. The special programs lacked financing capacity during2002-2003 but there seems to have been a recovery in2004 and 2005, although no information on their recent lendingwas available. There is policy interest to increase the number andvolume o f activities o f financial institutions working with small producers but no important results have come out yet. 4.49 An important source o f agricultural financing is that provided by input companies. Client lending by five o f the larger agricultural supply companies in 2001-2002 was US$250 million going to 30,000 producers. There i s a wide dispersion in these loans, which range from US$l,OOO for a modest producer to US$l,OOO,OOO for a big input retailer. Loans mostly finance seeds and agrochemicals and are usually based on a relation o f trust between the producer and the local retailer who acts as intermediary. Payment i s in accordance to the value the merchandise in international prices at that time o f repayment. This form o f financing was particularly useful duringthe banking crash, but i s not generally accessible to small producers. 4.50 Insufficient and expensive term financing has traditionally been a major obstacle for the modernization o f agriculture in Argentina. Inadequate term financing can be explained by poor institutional arrangements for this type o f lending andmacroeconomic imbalances which made it extremely risky. In the case o f grain production, pool arrangements o f different types were successfully used to overcome this constraint (see Chapter 2). Also, term financing i s not us much needed to innovate in grain production as to improve, for instance, irrigation systems or the root stock o f vineyards and fruit trees. The regional economies have suffered particularly from the lack o f adequate term financing. 4.51 Two main strategies could be envisaged to improve access to operation and term financing in the rural sector. First, the present economic situation o f commercial and fiscal 52 surplus and reasonable monetary stability should be taken advantage o f to enhance rural banking and improve the deposits/loans ratio. Ordinary monetary policy instruments can be used, but there i s need to assess the particular requirements o f the sector and use also persuasion and incentive mechanism. To this effect, a task force could be created, possibly coordinated by the Central Bank and including inter alia FINAGRO and SAGPyA, to assess the situation and needs o f operation and term lending inthe rural sector and design an appropriate strategy. Actions such as stimulating the h c t i o n i n g o f investment funds and usingnew forms o f rural financing could be considered. The experience o f the varied financing arrangements under the pools could serve as inspiration for new policy proposals. 4.52 Second, special attention could be given to the needs o f small producers. The financing side o f special programs could be expanded, operational procedures streamlined and the loan amounts increased. It would be important to simultaneously introduce provisions to assist inthe preparation o f investment projects by small producers, and to improve loan control. Also, microfinance institutions could be attracted to work in rural areas, and technical support and incentives provided for bankinginstitutions to open microfinance divisions. GOVERNMENT A CATALYST FORCOLLECTIVEACTION AS 4.53 Perhaps no power o f government is as strategically important as its power to support or undermine efforts by individuals and groups to get together to solve problems-that is to support or undermine collective action. 4.54 This role o f government takes increasingimportance ina global economy because global procurement redefines the role o f the supply chain. First, competition in a global economy requires efficiency and quality in every level o f the supply chain. It is not enough to produce a cheap product o f good quality ifthere are delays intransport, concerns over phytosanitary issues, or poor market identification and promotion. Second, global competition is not simply for exports. Global procurement by supermarkets in an increasingly liberalized world implies that producers for the domestic market must meet the same quality and sanitary conditions as exporters. Supermarkets compete vigorously for consumers, on the basis o f price, quality, and product differentiation and innovation, bringing about substantial restructuring o f supply chains (see Chapter 7 and Annex 3). Collective action to ensure good economic governance in the supplychains becomes thus essential ina globalizedworld. 4.55 Also, firm level efficiency i s not sufficient. Critically, each supply chain i s only as strong as its weakest link. In global competition, whether for exports or for a place on the shelves of domestic supermarkets, every member o f the supply chain loses if a link is weak. This adds an important dimension o f co-dependency to a naturally competitive and antagonistic relationship. Historically, the relationship among members o f the marketing chain is one o f competition for economic rent--resulting in antagonism and distrust. Increasingly, global competition dictates that this antagonism be tempered with a recognition o f the collective interest. In this environment, government catalysis for the required collective action is critical. While good government policy foments collective action, clientelistic practices undermine it. We examine examples o fthis insubsequent chapters. 53 4.56 We focus inthis report on the following roles o f government: 0 Defend rules o f the game that are stable, fair, and contain incentives that benefit the wider social good; 0 Invest in public goods, including inter alia agricultural research and extension, the quality agenda, sanitary and phytosanitary standards, infrastructure, and education and health; 0 Promote equity, protecting the weakest and more vulnerable sectors o f society; and 0 Support the development o f local and regional collective action, especially incases where the interests o f individual links inthe supply chain may be holding back competitiveness inthe entire chain. 4.57 The dialogue about decentralization could become more focused on promoting an incentive-compatible collective action. Nothing undermines collective action more than unclear rules about who finances what-especially when the government sporadically finances semi- public goods that require clear rules o f cost recovery. Two examples where clientelistic financing undermines local incentives for collective cost recovery are rural roads and irrigation maintenance (see Chapter 8). Government i s increasingly aware o f the need for a new institutional vision; one which gives it an important role as catalyst for coordination, provider o f key public goods, and negotiator and moderator of interests which threaten to block the advancement o f the larger common good. The role of promoter o f collection action is extremely important for public sector institutions like INTA and SENASA. SUMMARY AND CONCLUSIONS 4.58 Agriculture has traditionally paid a considerable amount o f taxes in Argentina. Tax pressure on the sector in 2003 was 26 percent, similar to that o f the national economy. Most taxation (97 percent in 2003) comes from national -- as opposed to provincial -- taxes. Tax pressure on agriculture increased much upon the 2001 crisis, mostly as a result o f the introduction (or more exactly reintroduction) in 2002 o f export taxes or retenciones. In 2003, export taxes accounted for 43 percent o f all taxes paid by the sector, The incidence on agriculture o f national taxes is rather different from that on other sectors; proportionally more is paid by agriculture inexport and income taxes and less inVAT and social security contributions. 4.59 Various elements in the tax structure are distortionary. One o f them i s export taxation. For various reasons: (i) it discriminates against exportables in relation to importables and non tradable; (ii)it i s not neutral with respect to the economic cycle and interferes with the farm investment and loan repayment cycle; (iii) it discriminates against some regions, especially the Pampas; and (iv) coming "off the top" leaves producers' incomes exposed to price variability, oandretenciones was understandable under the very special conditions created by the economic causes agricultural profitability to depend on year-to-year policy decisions. The introduction ' f crisis and subsequent devaluation, but there i s need to investigate more its effects and reflect on whether it could be substituted by other forms o f taxation on a more continued basis. The minimumtax on presumedincome, similarly introduced in2002, is also not neutral with respect to the economic cycle. It penalizes vulnerable producers and producers starting investment 54 projects, more likely to have abnormally low profits, as well as the farm sector as a whole because o f its characteristically low profits-to-assets ratio. The main purpose o f this tax i s to prevent evasion, but there are other means to achieve this goal. 4.60 Public expenditure in agriculture is extremely low. There is a high imbalance between what agriculture contributes to and takes away from the public purse. In2003, agriculture only received 1.1 percent o f all national government expenditure, equivalent to 1.7 percent o f agricultural GDP. In real terms and as a share o f agricultural GDP government expending in agriculture fell drastically with the crisis. International comparisons confirm the discriminatory treatment to the sector. Government spends per agricultural worker substantially less in Argentina than inBrazil or Chile, and less than average inLAC. The "orientation index" shows also more discrimination against agriculture inArgentina than in Brazil or Chile or for the L A C average. 4.61 Some 37 percent o f government expenditure in agriculture i s estimated to be in private goods. This i s less than the average for nine L A C countries studiedby Lopez (2004), but inview o f the extreme shortage o f public finds going into the sector this may result init being deprived o f essential public goods. 4.62 INTA and SENASA are the largest government institutionsproviding vital services to the sector--agricultural technology research and extension the former, and phytosanitary protection and food quality and safety services the latter. They are both decentralized public institutions operating under SAGPyA. They jointly accounted in2003 for 46 percent of national government spending inagriculture and 73 percent o fthat estimated for public goods. 4.63 INTA can claim a number o f important successes over the last years related to the introduction o f zero-tillage andbiotechnology, and the improvement o f management practices by small farmers. Many challenges remain, however, to keep pace with the deep changes taking place in agriculture inArgentina inthe last 15 years. INTA's strategic planning faces up to these challenges but there are several important areas in need o f fresh technological research and extension efforts which seem to require more emphasis. They include environmental sustainability, irrigation, bioenergy, quality systems, and the development o f non-traditional and specialty products. 4.64 SENASA has good technical capacity and has been able over the years to establish regulatory systems and a number o f valuable regulatory and policing processes. But there are also weaknesses which require a strategic strengthening o f the institution's role. The emphasis could be on the convergence o f quality standards for domestic and export markets, the clarification o f roles and responsibilities with other jurisdictions and the private sector, decentralization to the regional level, and insuringfinancial resources to be able to operate hlly. The challenge is also to change the institutional culture towards more focus on "making things happen" rather than "doing them itself'. 4.65 Two essential services to ensure competitiveness o f farming and o f the supply chains are infrastructure (including processing facilities), and rural finance. Shortcomings in the provision o f these services have historically underminedthe performance o f the sector. 55 4.66 andvegetables -- whose requirements are different. There is large room for improvement ineach Infrastructure and processing needs refer to three product groups -- grains, meat, and fruit o f these groups. Improvements would reduce costs, improve quality and hence prices, expand export markets, and allow the development o f new production areas. A joint public-private strategy would be useful to develop these services, together with a national program to organize the actions o fthe public sector. The Ministryo fPlanningi s taking steps inthis direction. 4.67 Shortage o f bank lending, particularly term lending, is a historic problem in Argentina's agriculture. The pool arrangements that have flourished in the Pampas are to a large extent an answer to this. Lack o f term financing has hold back considerably the modernization o f regional agricultures. Small farmers are particularly deprived o f access to loans. Conditions became especially dramatic upon the crisis, but have slowly started to improve since 2003. Two main strategies are recommended to deal with this situation. The first one is to take advantage o f the present situation o f reasonable monetary stability to enhance rural banking and improve the deposits/loan ration. To that effect, the creation of a task force i s suggested to design an appropriate strategy, which could include new policies and instruments alongside traditional ones. Second, particular attention could be given to the needs o f small rural producers. Expanding the financing side o f special programs, attracting microfinance institutions to work in rural areas, and promoting the opening o f microfinance division by banking institutions are possible options. 4.68 Supporting collective action in the supply chains i s one o f the most strategic roles o f government, impelled by globalizations, which internationalizes procurement standards not only for exports but also for domestic supermarkets. Good economic governance o f the supply chains through collective action is underscored by the enhanced importance of co-dependency in the traditionally antagonistic relationship among the actors o f supply chains. Decentralization o f service provision could become ever more focused on promoting incentive-compatible collective actions. Public sector institutions like I N T A and SENASA could be at the forefront o f this. 56 5. LABOR, INCOME AND POVERTY INRURAL AREAS 5.1 There seem to be three main types o f livelihood strategies o f the rural population in Argentina: (i)on-farm, agriculture-based livelihood, where some 15 percent o f the rural population are engaged; (ii) off-farm, agriculture and non-agriculture employment and transfers, which comprises some 21 percent o f population; and (iii) a combination o f the former two, which i s the livelihood strategy o f the majority, some 64 percent o f the rural population. This chapter throws light`onthese livelihood strategies examining household and labor characteristics inrural areas as well as the characteristics andincidence ofruralpoverty. We refer mainlyto the population resident in dispersed rural areas, as defined by INDEC, i.e. areas in the open countryside where households are not linked through urban streets or enjoy urban services. The remaining rural areas, up to settlements with 2,000 residents, are defined as "grouped" rural areas. 5.2 Most o f the information presented in the chapter is based on a rural household survey carried out in 2003 (RHS or RHS 2003) in dispersed rural areas o f four provinces, Chaco, Mendoza, Santa Fe and Santiago del Estero. These provinces contain one third o f the rural population in Argentina. Altogether, 441 households were surveyed. It should be kept in mind that results from the survey are representative for the aggregate o f the four provinces without crossing o f variables. As soon as variables are crossed, as in most tables in this chapter, results can only be referred to the sample. It i s unfortunate that no other large survey information was available, but household surveying inArgentina is only carried out inurban areas. THERURAL LABOR FORCE CharacteristicsandEmployment 5.3 The Argentinean rural labor force i s highly feminized. Women are strongly engaged in rural labor markets; in2003,48 percent o frural workers were women (Table 5.1). Linkedto this, the participation o f family and unpaid workers is low compared to other countries inthe region. In 2003, unpaid family workers accounted only for 4 percent o f all workers in dispersed rural areas. Education levels o f rural workers are generally low, with an average o f 6.7 years o f schooling. Male and female workers with completed primary education were 64.2 percent (Table 5.1), while only 8.5 percent had completed secondary schooling. 5.4 Argentina is well known for its relatively higheducation levels in comparison with other L A C countries. Yet, disparities are large between rural and urban areas and across regions. Youngsters 12-14 and 15-17 old in dispersed rural areas register school attendance rates 12 and 20 percent lower than those o f their urban peers. School attendance in rural areas in poor provinces i s considerable less than inrich provinces. Thus, 28 percent o f youngsters inthe 15-17 age group in Santiago del Estero attend school, compared to 72 inthe Province o fBuenos Aires. 5.5 The agricultural sector i s the main rural employer, even iffarm employment has fallen by roughly 34 percent from 1991 to 2001 according to the population censuses. According to RHS 57 2003, 72 percent o f workers were engaged in agriculture (Table 5.2).30 Males work proportionally more in agriculture than females: 77 percent versus 50 percent. Of the rural working population primarily engaged in non-agricultural activities, most work in services, followed by industryand government. Table 5.1: Characteristicsof the Labor Force inDispersedRuralAreas inArgentina, 2003 (percent) Gender -- Male 52.4 Female 47.6 Labor Status ---- Family Salaried worker 41.5 Self-employed 46.3 Employer 8.4 &UnpaidWorkers 3.7 Education -- No Education & Primary Incomplete 35.7 -- Higher Primary Complete 52.7 Secondary Complete 7.6 Education Complete 3.9 Source: Verner (2004), basedon RHS 2003. Table 5.2: Distributionof Workers amongSectorsDispersed RuralAreas inArgentina, 2003 (percent) Male Female Total Sample Industry 4.2 5.8 4.4 Services and Commerce 9.8 36.0 15.3 Agriculture and Livestock 77.1 50.4 71.6 Public Administration 3.4 6.3 4.0 Other Sectors 5.6 1.5 4.7 Source:Verner (2004). Basedon RHS 2003. 5.6 We show inFigure 5.1 the distribution o f household heads by labor status distinguishing farm and non-farm household^.^^ While the majority o f heads o f non-fam households are salaried workers the majority o f heads o f farm households are self-employed. Most jobs in dispersed rural areas are informal in nature. In 2003, only 28 percent o f Argentina's heads o f household indispersed rural areas were engaged inthe formal labor market. 30Unfortunately, there are no data on rural sector employment inArgentina with which to compare the results from the RHS survey. 31The distinction is based on whether the majority o fhouseholdincome comes from farm or non-farmsources. 58 Figure 5.1: DistributionofHouseholdHeadsby Labor Statusin DispersedRuralAreas, 2003, FarmandNon-FarmHouseholds(percent) Percentageof HHheads 80 t 7 I I 50 40 Salaried Worker Self-Employed Employe Family &Unpaic r Worker Source: Vemer (2004). Based on R H S 2003. Participationin Non-FarmOccupations 5.7 The rural non-farm sector is important in a number o f ways in rural areas.32It helps absorbing a growing rural labor force, slows down rural-urban migration, contributes to national growth, and can promote a more equitable distribution of rural incomes (Lanjouw and Lanjouw, 2001). What determines participation in the non-farm sector? We present an exercise based on the RHS 2003, using a probit model to determine the probability o f individual involvement in non-farm activities as primary occupation, conditional on some personal, household and geographic characteristic^.^^ Because o f limitations inthe RHS survey, some important variables likely to influence employment options cannot be considered, including ethnicity and social networks. 5.8 Recent research has pointed out that the non-farm sector can be seen as a source o f high- return employment or as a "last resort" occupational option (Ferreira and Lanjouw, 2001; Verner, 2004b). To capture this, we estimate two additional models, one for high-return and the other for low-return non-farm activities.34 Table 5.3 shows the results for the three models. The first model indicates, for all activities, the probability that the principal occupation o f the worker be in the non-farm sector. The second and third models split activities into two groups, o f low- and high-return.We review the results inthe following paragraphs. 32See Lanjouw and Lanjouw (2001) and Reardon, BerdeguC and Escobar (2001) for two recent surveys. 33Rather than reporting parameter estimates, difficult to interpret, we present the marginal effects associated with each explanatory variable. They can be interpreted as the effect o f a percentage change in the explanatory variable o n the probability o f involvement in non-farm activities, taking all other variables at their means. For dummy variables, the marginal effect i s the change inthe dependent variable associated with a change from zero to one, holding all other variables constant at mean values. 34Activities are designated as high- or low-return depending on average monthly earnings. If the earning i s below the poverty line, the activity is considered low-return. It i s considered high-return ifit i s above the line. 59 5.9 Gender. Women have considerable higher probability than men to participate in both low- and high-return activities, but men are more likely than women to find high-return employment. The difference between male and female participation rates tends to level out in high-return activities. We conclude that women do not have less access than men to high-return activities inArgentina, although the probability o f men participating innon-farm occupations to beinhigh-returnactivities is larger thanthat o fwomen. 5.10 Age. The probability o f non-fann employment increases with age, which may be taken as a proxy for skill level. Age i s positively associated with non-farm employment in general and with high-return activities, but not with low-return ones. There is no evidence that participation begins to decline at a certain age35.This finding contrasts with a similar one from Brazil where older workers have smaller probability o fbeing employed off-farm (Ferreira and Lanjouw 2001). Table 5.3: Probability of beingEmployedinthe Nonagricultural Sector, RuralDispersedAreas inArgentina, 2003 Nonagricultural Low-return Nonagricultural High-return Employment Nonagricultural Employment Employment Worker Characteristics: dF/dx P>lz dF/dx P>lz/ DF/dx P>lzl Education , Primary complete 0.069 0.000 -0.075 0.000 0.186 0.000 Secondary Complete 0.189 0.000 -0.076 0.000 0.389 0.000 University Complete 0.434 0.000 -0.041 0.000 0.524 0.000 Skills Age 0.010 0.000 -0.003 0.000 0.002 0.000 Gender Male -0.202 0.000 -0.246 0.000 -0.017 0.000 Land Landper capita -0.001 0.000 -0.001 0.000 0.000 0.660 Family characteristics Family size -0.013 0.000 -0.013 0.000 -0.003 0.000 Region Mendoza 0.055 0.000 -0.033 0.000 0.102 0.000 Santiago del Estero 0.196 0.000 0.135 0.000 0.108 0.000 Chaco 0.288 0.000 0.172 0.000 0.187 0.000 PseudosR2: 0.118 0.160 0.074 Source: Vemer (2004b). Basedon RHS 2003. Notes: Excludedcategories:No education or primaryincompletefor Education,and SantaFe for Region. 5.11 Education.Participation in the non-farm sector is significantly correlated to education. As education attainment rises, so does the probability o f being employed inthe non-farm sector ingeneral, and also inhigh-returnoccupations. At average values of other variables, completing primary and secondary education raises the probability of employment inhigh-returnjobs by 19 35An "Age Squared" variable was tested but was not significant. 60 and 39 percent, respectively. Contrarily, education decreases the probability o f participation in low-return activities. 5.12 Land. Access to land slightly reduces the probability o f employment in rural non-farm activities. At average values o f other variables, increasing the land holding by one hectare reduces the probability o f employment innon-farm low-return jobs by 0.1 percent. Participation inhigh-returnactivities doesnot appearto be affectedbylandsize. 5.13 Region. Workers in poorer regions are more likely to participate in rural non-farm activities in general and also in high-return ones, Relative to those in Santa Fe, workers in Chaco, Mendoza, and Santiago del Estero are more likely to be employed in high-returnnon- farmjobs and innon-farmjobs ingeneral, controlling for individual characteristics. LABOR EARNINGS RURAL IN AREAS 5.14 Wage levels in rural Argentina have remained low, except for high skilled, well- educated workers, even in times o f relatively high economic growth. This seems to be a consequence o f population growth in rural areas in previous decades, which has resulted in an abundant supply o f unskilled labor, and the type o f technical and crop mix changes that have been takingplace inArgentina, especially inthe Pampas, which are labor displacing. 5.15 Rural earnings are related to labor status. Formal sector workers, i.e. those contributing to the pension system, earn more than their peers in the informal sector, and permanent workers earn more than temporary workers (Table 5.4). Employers earn roughly double than the self- employed and four times more than wage-workers. Self-employed workers are generally better o f f than wage-workers, unless when they are temporarily employed. Table5.4: Average AnnualEarningsfor Permanent andTemporary Workers inDisperseRuralAreas of Argentinain2003 (AR%) Permanent Temporary Formal Informal Self-employed 4,325.2 1,44 1.7 7,895.1 2,602.5 Wage-worker 3,811.4 1,507.7 4,122.8 2,061.8 Source; Vemer (2004). Basedon R H S 2003. 5.16 Child labor still exists in disperse rural areas in Argentina, but to a much lower extent than in other countries inthe region. RHS data reveal that 4 percent o f children inMendoza and Chaco worked in 2003. Hence, child labor does not appear to be a serious problem in rural Argentina, andworking children may be both working and studying. 5.17 We investigate factors related to the level o f earnings using quantile regression analysis.36 Results are shown in Table 5.5. All variables inthe model are significantly different 36 Quantile regression makes it possible to examine the relation between wages and the explanatory variables in different parts o f the distribution. Incomes are modeled using log annual labor incomes as the dependent variable. 61 from zero for all income quantiles. Findings indicate that earnings are not related in the same way to the explanatory variables at the different levels of the earnings distribution. For example, females earn much less than males inthe low end o f the distribution relative to their peers inthe highend, andreturns to lower levels o f education are smaller inthe lower income quantiles than inthe higherones. Wereview the results below. Table 5.5: DeterminantsofLaborIncomeinDisperseRuralAreas ofArgentina, OLS andQuantileRegressions, 2003 OLS 25th 50th 75th goth Return Return Return Return Return Yo P>ltl Yo P>ltl Yo P>ltl Yo P>ltl Yo P'ltl Age 0.30 0.00 0.30 0.00 1.21 0.00 0.80 0.00 1.11 0.00 Female -53.51 0.00 -43.33 0.00 -35.21 0.00 -34.56 0.00 -17.06 0.00 Education Primary education 7.79 0.00 9.75 0.00 27.12 0.00 25.99 0.00 15.60 0.00 Secondary education 144.49 0.00 58.57 0.00 72.12 0.00 55.89 0.00 88.89 0.00 University education complete 353.13 0.00 192.41 0.00 135.84 0.00 92.13 0.00 52.50 0.00 Labor status Has a permanentjob 121.00 0.00 138.69 0.00 42.19 0.00 34.72 0.00 54.19 0.00 Has a formal job 149.93 0.00 66.36 0.00 59.84 0.00 40.07 0.00 16.53 0.00 Self-employed -23.43 0.00 -35.60 0.00 -26.36 0.00 4.50 0.00 24.23 0.00 Wage-worker -16.14 0.00 -5.45 0.00 -3.34 0.00 12.30 0.00 -7.96 0.00 Sector Commerce and 5Services -72.11 0.00 -43.62 0.00 -22.59 0.00 -21.96 0.00 -9.06 0.00 Agriculture-Livestock -71.75 0.00 -35.92 0.00 -27.67 0.00 -21.26 0.00 -13.24 0.00 Other sector -13.93 0.00 2.74 0.00 -15.72 0.00 -15.21 0.00 -9.43 0.00 Public Administration -81.33 0.00 -36.87 0.00 -31.55 0.00 -27.82 0.00 -12.89 0.00 Province Santiago del Estero -33.44 0.00 -18.37 0.00 -25.40 0.00 -32.23 0.00 -46.74 0.00 Chaco -68.75 0.00 -83.29 0.00 -43.62 0.00 -36.43 0.00 -55.34 0.00 Mendoza -20.23 0.00 1.82 0.00 -26.36 0.00 -33.77 0.00 -48.93 0.00 Constant 222286 0.00 103177 0.00 156612 0.00 262705 0.00 436673 0.00 Adjusted R2 (OLS) and Pseudo R2 0.13 0.085 0.097 0.I03 0.136 Source:Verner (2004). Basedon RHS 2003. Notes: Excludedcategories: no educationor primary incompletefor Education, piece-workerfor Labor Status, industrial sector for Sector, and SantaFeprovincefor Province.The percentagereturnis calculatedas (exp (coefficient estimate) - 1) * 100. 5.18 Education. The education earningpremium increases rapidly with the level o f education and is positive for all quantiles. Compared to the earnings o f non- educated workers and those with incomplete primary education, median earnings of workers with complete primary education were 27 percent higher, with complete secondary schooling 72 percent higher, and The general model contains explanatory variables in levels and allows for non linearities inthe data. For example, the log labor income equation i s found to be nonlinear ineducation. 62 with tertiary education 136 percent higher.37Workers with complete secondary education face increasing returns across the earnings distribution. The poorest (25th quantile) receive a wage premium for complete secondary education o f 59 percent, while the richest (90th quantile) receive 89 percent. Inthe case o f tertiary education the education premium clearly decreases as we move up the income distribution, Ingeneral, results would seem to indicate that: (i) is there wide heterogeneity in the quality o f education in rural areas; (ii) capacity o f workers to the convert their educational capital into higher earnings through labor market networks is variable; and (iii) opportunities vary by locality. 5.19 Labor Market Status. Labor status is an important correlate o f earnings. Coefficients for all the included occupational groups are statistically significant and different from zero. Looking at the median o f the distribution, wage and autonomous workers earn less than piece- workers, controlling for other factors such as education. For the 75th and 90th quantile, the premium-gap changes infavor o fthe self-employed andthe autonomous workers. 5.20 Age. Age increases earnings. Two questions are addressed: (i) age important in the is earnings determination process?; and (ii) returns to age homogeneous across income levels? are The answer is yes to the first question and no to the second one. Age is statistically significant for all reported quantiles, controlling for other individual characteristics. However, returns to age are low: only 0.3 percent inthe 25th quintile, 1.2percent inthe 50th, and 1.1percent inthe 90th quantile. 5.21 Formal vs. Informal Workers. The positive effect on rural earnings o f formal sector employment decreases across the distribution: a worker in the 25th quantile obtains an income premium o f 66 percent whereas one in the 90th quantile and above receives a 17 percent premium. Workers in the informal sector are disadvantaged in at least two ways: they do not have access to social security, and they obtain lower incomes with possibly greater seasonal variation in earnings. Since the formal sector generally consists o f higher productivity jobs, and since higher productivity may require more skills, the informal sector variable may be partly capturing skill differences not signaled by other variables inthe model. 5.22 Gender. Large measurable inequalities persist in rural areas between men and women. Female earnings are significantly lower than male wages in all quintiles. Moreover, the gender gap decreases as we move up inthe distribution. The largest earning gap appears at 25th quantile where women receive around 43 percent lower earnings than men, narrowing to 17 percent at the 90th quantile. The gender-earning gap may, to some degree, be explained by choice o f jobs by women. Women are more likely than men to select more flexible jobs; they may choose, for example, part timejobs or jobs with lower working hours. 5.23 Location. Location matters. Thus, Santa Fe workers enjoy a premium with respect to other provinces as they have earnings significantly larger than those in Chaco, Santiago del Estero, and Mendoza. Controlling for other characteristics, the poorest workers earn inChaco 83 percent less that their peers in Santa Fe. At the top end o f the distribution, workers in Chaco, Mendoza, and Santiago del Estero earn between 47 and 55 percent less than inSanta Fe. 37 Recent research shows that returns to education and skills inurban areas inArgentina have increased for all three levels of education inthe last decade, but more pronouncedly for tertiary education (World Bank 2004). 63 5.24 We complete this analysis o f rural earnings examining in more detail farming incomes applying an augmented earnings function. Log annual income from farming activities i s the dependent variable. The model contains explanatory variables in levels and allows for nonlinearities in data. Findings are presented in Table 5.6. All coefficients o f the explanatory variables included have the expected signs and they all are statistically significantly different from zero. 5.25 As inthe case o f rural earnings in general, education is strongly correlated with income levels. Controlling for other characteristics, returns to primary, secondary, and tertiary education were statistically significantly different from zero and positive,38 showing a rapidly increasing premium to educational attainment. More-educated farmers earn significantly higher incomes thantheir less educated peers. Table 5.6: Determinantsof FarmIncomesinArgentina in2003 Marginal impact (YO) P>ltl Gender Male -7.13 0.00 Education Primary educationcomplete 10.96 0.00 Secondary complete 124.34 0.00 University studies 185.48 0.00 Land 2-10 hectares 100.77 0.00 11-35 hectares 283.44 0.00 36-100 hectares 312.06 0.00 101-250 hectares 877.67 0.00 More than 250 hectares 1880.63 0.00 %rentedhectares/ total hectares 0.10 0.00 % owned hectares/ total hectares 0.00 0.01 % sharedhectares/ total hectares -1.09 0.00 % occupiedhectares/ total hectares 0.40 0.00 Infrastructure andproduction inputs Access to a paved road 29.30 0.00 Access to electricity 43.48 0.00 Use fertilizer 25.61 0.00 Access to irrigation 28.27 0.00 Constant 63170.23 0.00 AdjustedR2: 0.35 Source: Verner (2004) Annex I11.Owncalculationbasedon R H S 2003. Notes: Excludedvariables: No educationor primary incompletefor Education, and<2 hectares for Land. The marginalimpadpercentagereturnis calculatedas (exp (coefficient estimate) - 1) * 100. Numberof observations: 124. 38Rates o f return are calculatedbythe earnings function method due to Mincer (1974). 64 5.26 Farm size is important. All farm size variables included are statistically significant and positive. At average values o f other variables, income increases with farm size, in a nonlinear fashion. Compared to farmers with less than two hectares, farmers with farms o f 2-10, 11-35, 36- 100, 101-250, and 250 or more hectares have incomes 101, 283, 312, 878 and 1881 percent higher, respectively. Larger farms earn dramatically higher incomes than smaller ones, but whether the farmland i s rented or owned has little measurable effect on incomes. Gender also matters although not in an important way. Farms run by women are 7 percent more profitable thanthose runbymen.Access to infrastructureincreases the profitability offarms, andso does the access to paved roads and electricity, which increase income by 29 and 44 percent respectively. The use o f production enhancing techniques i s also important. Fertilizer and irrigation are both significantly positive determinants o f farm income, although there may be endogeneity at play. Farms using irrigation and fertilizer experience 28 and 26 percent higher incomes, respectively, than farms not usingthem. RURAL POVERTY Incidence 5.27 Measured by income, there were 1.2 million extreme poor in rural Argentina in 2003, accounting for some 15 percent of the extreme poor inthe country. Incidence o f extreme poverty i s greater in rural than in urban areas. By the income measure, nearly 40 percent o f rural households are in extreme poverty, compared to just over 30 percent in urban areas.39 With an average household size just under six, this means that there are more than 200,000 rural households inextreme poverty inArgentina, most o f them indispersed areas.40 5.28 Highpoverty incidence inrural areas and wide rural-urban differences are confirmed by other poverty measures. Thus, in2001, 33 percent o f the rural population had unmet basic needs (UBN)--a significantly larger share than in urban areas where the percentage was 14. A rural UBNof 33 percent is highfor a middle-income country like Argentina. Measuredby UBN,the rural poor account for 19 percent o f the poor, while rural residents only account for 11percent o f the population. Location 5.29 There are large differences inpoverty among regions. In 2003, the extreme poverty rate indispersed rural areas in Santa Feyin the Pampean region, reached 7.6 percent; one fourth o f that in Santiago del Estero, in the Northwest region, where the poverty headcount was 29.1 percent. The Pampean poverty figure also contrasts with that o f Chaco, inthe Northeast region, 39 This poverty comparison refers to income poverty because consumption poverty estimates are not available for urban areas. However, since consumption poverty rates give a more accurate picture o f household poverty we will use them for rural areas in the rest o f the report unless stated otherwise. In 2003, the poverty line was set at AR$118,61per month (some US$40) and the extreme poverty or indigence line at AR$69.65 (some US$21) per adult equivalent (Gerardi 2003). 40 We should mention that due to the effect o f the 2002 economic crisis rural poverty in 2003 was probably well above the historical trend. Lack o f data prevents knowing the evolution o f ruralpoverty over time. 65 where 20.7 percent were poor, and o f Mendoza, with 26.6 percents4*The variation in UBN across provinces i s also large. Data from 2001 reveal that the Northeast and Northwest regions have the largest share o f the rural population with UBN. In Salta and Formosa, more than 50 percent o f the rural population has UBN,while only around 15 percent o f the rural population in Buenos Aires andthe Pampa provinces face this situation. Characteristics 5.30 Table 5.7 gives a profile o fthe characteristics o fpoor households indispersed rural areas. Poverty is concentrated in younger households and tends to be transitory. Thus, the strong incidence o f poverty is inhouseholds headed by a person under 25, and it dramatically drops off as age increases. The probit regression analysis presented inTable 5.8 shows that the probability o f being extremely poor falls by 0.2 percent for every year o f age o f the household head. The Decline inpoverty with increasing age o f the household head i s strongly related to the average number o f children in the household (Figure 5.2). Probit results indicate that indigent poverty falls by nearly 0.4 percentage points when the dependency ratio falls by one percent (Table 5.8). 41Mendoza illustrates the difficulty o f interpreting these numbers. While headcount poverty is high, indicators of basic needs o f dispersed rural populations in Mendoza are much better than those o f the other provinces sampled. Labor earnings are also highrelative to other provinces -especially inthe lowest quintile- and Mendoza is one o f the few provinces where rural population grew over the last decade. All this suggests that high headcount poverty in Mendoza i s likely to be due to an influx o f yet-unabsorbed unskilled workers. Inshort, poverty is more llkely to be a temporary condition than a structural problem. 66 Table 5.7: PovertyProfileof HouseholdsinDispersedRuralAreas of Argentina in2003. ConsumptionPoverty(Percentof RuralHouseholdsinthe Sample) Consumption Consumption Poor Indigent Gender Male 45.4 20.4 Female 64.4 27.0 <25 100.0 100.0 25-44 46.5 24.3 45-65 54.4 22.9 >65 37.1 8.9 Literacy Literate 47.2 20.3 Illiterate 60.8 33.8 Yearsof schooling None or less than 1 66.0 31.5 1-4 56.6 25.2 5-8 45.8 20.5 9-12 30.3 10.7 More than 12 0.0 0.0 Workposition Wage-worker 53.3 23.2 Self-employed 43.1 18.7 Piece-worker 50.9 26.0 Employer 14.8 4.4 Work Sector Agriculture & Livestock 43.3 24.5 Industry 51.6 5.2 Commerce & Services 54.9 25.7 Other sectors 57.6 0.0 Work condition Formal 38.6 21.2 Informal 50.1 14.6 Family Size 1-3 members 23.2 4.1 4-5 members 44.2 14.8 More than 5 members 80.5 47.4 Land Tenure No land 45.0 12.9 0-1 ha 74.9 23.8 1.1-10 ha 62.8 18.4 10.1-35 ha 26.4 30.8 35.1-100 ha 8.9 30.5 100.1-250 ha 16.4 13.3 More than 250 ha 11.4 2.6 Source:Vemer (2004). Based on RHS. 5.31 Other characteristics o f rural poverty emerging from Tables 5.7 and 5.8, which mirror those regarding earnings discussed before, can be summarized as follows.42 First, education i s strongly related to poverty: as the years o f schooling go up the incidence of poverty goes down. Second, poverty incidence i s larger among female- than among male-headed households. Third, poverty seems to be more prevalent among wage-workers than among the self-employed. Fourth, households involved in agriculture seem to be less prone to poverty than those in other work sectors. Fifth, poverty incidence i s less among formal than informal workers (although this result does not hold for the extreme poor). Sixth, households without land seem to be less prone to poverty than those with small farms, but more than those with large farms. 42 Itshould be emphasized that the results here commented refer only to the 441 households inthe sample. Inview of the small sample size, results cannot be expandedto the entire population ofhouseholds indispersed rural areas. 67 5.32 Remittances and transfers are a significant source o f income in general in rural areas, accounting for 19percent o f the income o f the non-poor and 27 percent o f that o f the poor (Table 5.9). This suggests that as children leave the household they continue to contribute significantly to their parents' earnings. These transfers are mostly private, as only some public programs are actually available for the rural poor. For example, the Familia program is only implemented in urbanarea. TheJefas andBecas programs are implemented inbothrural andurbanareas, but no disaggregate data were available to evaluate rural participation in these programs. Some observers contend that Jefas encourages rural-urban migration, as some household heads traveling to the city to collect their allowances end up staying. Table5.8: ProbabilityofBeinganExtremePoor Household inDispersedRuralAreas inArgentina,2003 dF/dx P>lzl x-bar Skill characteristics Age -0.002 0.000 48.87 Education -0.022 0.000 5.84 Gender Male -0..58 0.000 0.87 Family characteristics Dependencyration (children < 15/householdsize) 0.386 0.000 0.25 Land Holdings * 0.1-100 ha. 0.057 0.000 0.633 100.1-250 ha. -0.014 0.003 0.089 >250 ha. -0.025 0.000 0.059 PseudoR2: 0.1I73 ObservedP: 0.189 PredictedP: 0.160 Source: Verner (2004). Basedon RHS 2003. . . Note: Excludedcategories,No land 5.33 Table 5.9 shows that the rural poor are less dependent on agriculture (54 percent) than the non-poor (68 percent), which i s consistent with the observed consolidation o f smaller farms into bigger ones. Inany case, the pattern o f large, young families, highrate o f departure o f maturing children from the household, relatively lower agricultural dependence, and significant remittances, go a long way towards explaining the observed fall inpoverty with household head age. 68 Figure5.2: Average Numberof Childrenand Childrenunder 15 Residentin - RuralHouseholdsby Age of HouseholdHeadin Argentina,2003 I 3 5 3 0 1 2 5 1 2 0 B l5 a 1 0 0 5 0 0 18- 25-30 31-35 36-40 41-45 46-50 51-55 56-60 61-65 66-70 71-75 76-80 81-85 86-88 23 Head of householdage cohort Source Vemer (2004). Based on RHS 2003. Table 5.9: HouseholdIncomeSharesinDispersedRuralAreas Argentina,2003 (percent) Poor Non-poor Independentfarming 28.7 50.4 Agricultural wage labor 24.9 17.8 Total agricultural income 53.6 68.2 Non-agricultural income 19.8 12.8 Other Sources (transfers andremittances) 26.6 19.0 Total non-agricultural Income 46.4 31.8 Source: Vemer (2004). Based on RHS 2003. I s Land a PovertyTrap for Small Farmers? 5.34 Poverty rates are higher among households with little land than among those with no land.According to the cross tabulations o fTable 5.7, consumption poverty among landless rural- dwellers i s less than among landholding households with less than 100 hectares. Thus, extreme poverty for landless households i s 13 percent, compared to 24 percent for those with c 1hectare, 18 percent for 1-10 hectares, and 30 percent for 10-100 hectares. There seems to be a "poverty drag" o f small-scale land ownership. Small farm ownership may be closely correlated with other variables that might increase poverty like low education or high dependency ratio. If so, Table 5.7 would overstate the poverty drag. 5.35 The results o f the probit analysis shown in Table 5.8 give a more reliable indication o f the independent effect o f farm size on poverty. Controlling for age, household dependency ratio, education, and gender, the conclusion holds. Thus, Table 5.8 suggests a small but statistically 69 significant impact o f land ownership on extreme poverty, with farmers with holdings < 100 hectares having a probability approximately 6 percent higher o f experiencing extreme poverty than landless laborers. On the other hand, having > 100 hectares decreases the probability o f beingpoor. SUMMARY AND IMPLICATIONS 5.36 This chapter has shown a number o f features o f the labor force, poverty and livelihoods inthe dispersed ruralareas ofArgentina. We synthesize thembelow: Participation o f women in the labor force as paid or-self employed workers (not as non- paid family members) i s important inrural Argentina. Agriculture is the main employment sector, with the poor deriving less o f their income from agriculture than the non-poor. Informal laborers constitute the vast majority o f the rural labor force. Women have higher probability than men to participate in rural non-farm activities, and are not confined to low-return employment. Involvement in the non-farm sector is positively associated with education attainment. Workers in poorer regions are more likely to be employed inthe non-farm sector. There i s an earning gap betweenmen andwomen, which is wider at low income levels. Earnings are strongly correlated with education but the education premium decreases as we move up inthe earnings ladder. Non-farm income and employment are highly correlated with gender, age, access to land, and education. Farm incomes increase with land size, education levels, road access, and the use of electricity, fertilizer, and irrigation. Poverty incidence is higher inrural than inurbanareas; 40 vs. 30 percent in 2003. Fifteen percent o f the extreme poor are inrural areas compared to 11percent o f the population. There are some 200,000 extreme poor rural families. They tend to be large and young, andtend to escape from extreme poverty as they mature and children leave the household (and often continue contributing to it). They live largely inthe Northeast and Northwest and in dispersed areas where basic service provision is often weak and delivery is difficult. They are more likely to be small landholders than landless laborers. Remittances and other transfers are an important source o fincome (27 percent) for these families. A comprehensive vision o f rural development, beyond the sectoral approach, emerges from the above findings. The rural world i s not only an a&cultural world -- although it i s more so inArgentina than, for example, inMexico -- and the rural economy i s not only an agricultural economy; the combination of economic activities is the dominant feature o f rural Argentina. Demographic and labor force changes inrural areas reflect a rural society intransformation, with education offering increasingly better prospects in the farm and non-farm sectors. Some differences, however, can be observed when non-farm occupations are divided into low- return and high-returnactivities, since education i s a particularly important determinant o f employment in the better-paid ones. The complexity o f the income determination process in rural areas is illustrated by our analysis o f earnings, which shows a heterogeneous impact on earnings o f 70 individual characteristics across the income distribution. The magnitude o f effects varies depending on workers being comparatively rich, poor or placedinthe median o fthe distribution. 5.38 The importance o f the rural non-farm economy is a major conclusion o f our analysis in this chapter. Pursuing rural non-farm growth should not be seen as an impediment or an alternative to agricultural development; there are strong synergies between the farm and non- farm sectors and no contradiction in supporting the development o f the two, as shown by Lanjouw and Lanjouw (2001). What is needed is a comprehensive rural development policy where the contributions o fboth sector are acknowledged. 5.39 Our findings suggest concentrating the fight o f rural poverty on young and small farming families indispersed rural areas, particularly inthe Northeast and Northwest. This would require a targeted focus on education for poor families and on the acquisition o f skills needed to compete inan increasingly competitive world. Increased access to infrastructure, markets, andproductive inputs for poor rural farmers would also be important. Poor small farmers need assistance to improve productivity, which crucially depends on access to research and extension, land, and rural finance. A rural development strategy could usefully include specific recommendations for the rural poor. In this regard, PROINDER (a government small-farmer support program) has shown that: (i) it is desirable to tailor strategies to the regional and local characteristics o f the rural poor, and that (ii) interventions aimed at increasing the productivity and sustainability o f poor small farmers are viable provided that there is institutional support available to them. 5.40 Options to reduce poverty inrural areas could consider four dimensions: 0 Targeting extreme poor households and linking income transfers to education. Young large families could be targeted with transfers linked to education through the secondary level, bymeans o f social programs such as Becas andFamilias. Increasingaccess to productiveinputsfor smallfarms and ruralfirms. Inaddition to education, other mechanisms could be explored to facilitate small farmers and rural firms moving up the productivity ladder. They might range from work to improve social linkages between households in dispersed and grouped areas to programs to increase access to productive inputs and to ensure titling o f land. 0 Creatingjobs through investment in the regional economies. Many households are poor because they are trapped on low-productivity land or are in low paying, low productivity jobs in the informal sector. We argue elsewhere in this report that high productivity jobs can be created in the regional economies by improving, inter alia, the provision of public goods and the environment for collective action in irrigated agriculture. Developing a rural development strategy. The design and implementation o f a rural development strategy would be a useful tool to address the issues o f Argentinean rural poverty. 71 6. THE CROPFRONTIEREXPANDSAND THE PAMPASINTENSIFY 6.1 Land in annual crops in Argentina grew by 5.5 million hectares (40 percent) between 1988 and 2002. This increase in cropland derives from both expansion o f the crop frontier and intensification through shortening o f rotations and elimination o f pastures. These changes have raised environmental and social concerns. As indicated in Chapter 2, much o f the expansion o f the crop frontier is related to the technical advances and expanding market opportunities registered for soybeans. Inthis chapter we look first at expansion o f the frontier inthe Northeast and Northwest, and second, at some evidence concerning the intensification o f land use in the Pampas. The purpose o f i s to shed light on environmental and social outcomes and to assess whether or not a stronger role for government i sjustified. 6.2 To better understand the social and environmental implications o f frontier expansion, the Laboratorio deAntilisis Regional y Teledeteccidn (LART) undertook a detailed study o f landuse inthe six Northern provinces (LART/FAUBA, 2005). This study was carried out at two levels. First, a department-level analysis based on census data for each o f the provinces (96 departments). Second, a detailed analysis inthe northeastern part o f Salta and the westernpart o f Santiago del Estero based on interpretation o f LANDSAT 5 TM images from 1998-1999 and MODIS-TERRA images for 2002-2003. Comparison o f these images, together with data on the transportation network, location o f urban areas, quality o f life indicators, and other information permitted a detailed characterization o f the nature and effects o f frontier expansion. HOW MUCH THE FRONTIER HAS EXPANDED? 6.3 Data from the 1988 and 2002 National Agricultural and Livestock Censuses show that in the study area,43 agricultural land increased 70 percent from 2.5 to 4.3 million hectares, or 3.9 percent annually. Nearly 120,000 hectares o f the study area were incorporated into agriculture each year, accounting for approximately 14 percent o f the total increase in agricultural land in Argentina duringthis period. Land expansionconcentrated infour well-definedareas: 0 Northeastern Salta (Tartagal) 0 Southeastern Salta (Las Lajitas) 0 Northeastern Santiago del Estero and southwestern Chaco (Charata) 0 Southeastern Santiago del Estero and northwestern Santa Fe (Bandera) Salta 6.4 Inthe extreme north of Salta the increase in area occurs toward the east of agricultural lots existing in 1988. Soybeans account for most o f the nearly 200,000 hectare o f crop area 43The study area encompasses 96 departments in 6 provinces of northern Argentina: Formosa, Chaco, Santiago del Estero, Salta, Santa Fe, and Corrientes. This area includes a large part of the Argentine Republic's Chaco Region, with the exception of a portion of the province of Corrientes, which is located outside the region sensu stricto. The study area consists of 568,664 km2,which represents approximately 15 percent of the country's territory. 72 expansion; from about 114,000 hectares in 1988 to almost 311,000 in 2002. New lots have reduced the areas o f natural habitat between the lots. Nearly all the area previously occupied by natural vegetation has thus been covered. One can observe evidence o f abandonment o f previously farmed areas. Some relatively large patches that had been cultivated in 1989 do not appear with crops in 2003. They are located on the western edge o f the agricultural zone, in the transition area between Chaco forests and the piedmont jungles o f the Yungas. Grau et al. (in publication) indicate that inmountainous areas agricultural lots are being abandoned, as a result o fmigration to urban centers. Eastern Santiago del Estero and Western Chaco 6.5 Inthis zone, the estimated total crop area for the 1988-89 agricultural year (based on a LANDSAT 5 TM image) was 142,158 hectares (9 percent o f the total area analyzed). In the 2002-03 agricultural year (based on MODIS satellite images) it was 454,964 hectares (28 percent o f the area). Sixty-six percent o f the increase in agricultural area in the region studied was explained by soybean cropping, 24 percent by wheat, and 7.2 percent by corn. The area planted with other crops like beans, cotton, sorghum, and sunflower decreased duringthis period. THEADVANCE OFSOYBEANCROPS: WHAT HAVETHEY REPLACED? 6.6 Crops replacedby soybeans varied by department. InSalta, only 25 percent o f the current soybean area i s on lands which were planted with soybeans in 1988-89. Another 24 percent i s in areas with different agricultural use in 1988-89, about half o f which in beans. In the Chaco- Santiago del Estero zone, 13 percent o f the area plantedwith soybeans was inareas already used for agriculture, especially cotton. There i s little evidence o f competition with cattle. In departments where the agricultural area increased, there generally was little significant change in the cattle population (Figure 6.1). 6.7 New crops mostly replaced forest innorthern Salta. In2002-03, 50.6 percent o f soybean cultivation, approximately 157,409 hectares, was in areas previously covered by natural vegetation. Eighty-nine percent o f the natural vegetation replaced by soybeans was dry Chaco vegetation (quebracho, palosanto, duraznillo, and other species), 5 percent rainforests, and 6 percent Chaco Serrano, according to the vegetation unit classification proposed by Zapater de del Castillo (1985). Table 6.1 shows the loss o f hectares o f each type o f natural vegetation due to the advance o f agriculture (principally soybeans) in the zone, with the most severely affected vegetation type decreasing by nearly 13 percent. Areas that are less fertile, flood-prone, or have excessive salinity were not convertedto agriculture. 73 5007 .-alB n - r +400 Ln 2 c m- .E ; a- al + + - 5n 200 " .- ** al + 74 6.8 In Eastern Santiago del Estero-Western Chaco, 86 percent of the new agriculture activity (with soybeans as the principal crop) was carried out in areas with natural vegetation in 1988-89. The advance o f soybeans on natural vegetation occurred mainly on open forest (quebracho, palosanto, duraznillo, and other species, 43.2 percent), grazing land and savannas (33 percent), and areas with exposed soils (16.4 percent). See Table 6.2. Table 6.2: Encroachmentof Agricultureon NaturalVegetation InEastern Santiago delEstero-Western Chaco Area (hectares) Loss Vegetation type 1988-89 2002-03 Hectares Percent Open forest 770,002 612,969 157,034 20.4 Pastures and savannas 383,949 264,148 119,801 31.2 Exposedsoils 93,216 33,422 59,793 64.1 Closedforest 76,456 58,194 18,262 23.9 Paramo(moorland) 36,240 29,800 6,440 17.8 Flood-prone areas 96,73 1 94,922 1,808 1.9 Total 1,456,594 1,093,455 363,138 24.9 Source: LARTFAUBA (2005). FACTORS EXPLAINING LOCATION SOYBEANEXPANSION THE OF 6.9 The factors associated with expansion o f the soybean frontier were analyzed ~tatistically.~~ This analysis consisted o f stepwise regression to select the variables most strongly associated with the percent o f soybeans in each o f the 74 departments in the study area. Explanatory variables include size structure o f farms, type o f tenancy (rented, private, corporate), irrigation, population density, drainage and soil characteristics, climate, and transportation access. Models were selected that best explained the variation in percent o f land in soybeans among departments in 1988 and 2003, as well as the change between 1988 and 2003. Regression Resultsfor 1988 6.10 Table 6.3 presents the results for model that best explains the cross section spatial variation inthe area insoybeans in 1988. The model is the following: % soybeans =-0.013 +0.00019 PROF+ 13.72 Kmroads/ha -0.06s~-0.01ALKA-0.22 Riego 6.11 Approximately 44 percent o f the variation is explained by soil characteristics and 7 percent by road density. The percentage o f land in small farms and the percentage in irrigated land explained an additional 10percent, with both variables carrying a negative coefficient. 44 For details see LARTEAUBA (2005). 75 Table 6.3: SpatialVariation inthe Area of Soybeansin 1988 Variable PartialR2 ModelRz F P Soil density (PROF) 0.398 0.398 47.71 <0.0001 Road density (kmroadsha) 0.072 0.471 9.74 0.0026 Proportion o f department's 0.055 0.526 8.25 0.0054 landarea occupied by agricultural enterprises with fewer than 100has (Sc) Alkalinity Index (ALKA) 0.036 0.562 5.68 0.0199 Proportion o f department's 0.041 0.604 7.17 0.0093 landareaunder irrigation (RIEGO) Regression Results for 2002 6.12 Table 6.4 presents the results for model that best explains the cross section spatial variation inthe area insoybeans in2002. The model i s the following: % Soybeans= -0.09 +0.44 Scm+ 0.001 IP-0.77 Sc +0.09 DREN 6.13 Inthis year, approximately 48 percent of the variation was explained by farm size, with the proportion o f land under soybeans increasing with the dominance of large farms and decreasing with that o f small farms. Soils-related variables explained the remaining 16.5 percent. IP, a synthetic variable that summarizes climatic and soils-related dimensions explained 12.4 percent, and DREN, a variable related to soil texture and topography, explained the remaining 4.1 percent o f variance. 6.14 At the end of the 1980s the area planted with soybeans was mainly associated with soil variables while by the beginningo f the present decade it was associated with variables that relate to farm size. Both the statistical model obtained to explain the change in the area planted with soybeans and analyses based on teledetection showed conclusive results: the increase inthe area planted with soybeans was carried out mainly at the expense of natural vegetation and independent o f environmental variables (climate and soil). The replacement o f other crops by soybeans was less important in explaining soybean expansion but very important in explaining the decrease inthe area o f the crops replaced. 76 Table 6.4: SpatialVariation inthe Area of Soybeansin 2002 Variable PartialRz Model Rz F P Prouortion o f department's area 0.403 0.403 58.08 <0.0001 occupiedby agricultural enterprises with over 100 and fewer than 1,000 has (Scm) Indexof cartographc productivity 0.124 0.527 22.34 <0.0001 (IP) Proportion o f department's area 0.083 0.61 17.93 <0.0001 occupiedby agricultural enterprises with fewer than 100ha (Sc) Drainage Index (DREN) 0.041 0.65 9.92 0.0023 6.15 The change in the relative importance o f environmental variables (Le., soils) and socioeconomic variables (ie., scales o f farm size) may reflect the relaxing o f environmental constraints, stemming from the use o f new technologies. Access to new genotypes and agronomic practices requires capital andknowledge that is not equally available to all producers. The positive association between the area planted with soybeans and the proportion o f large- scale producers suggests that the latter, which generally use more technology and have greater access to information, were able to incorporate more quickly soybean crops in their productive schemes. Inaddition, the cost o f obtaining information, which makes it possible to determine the best planting date or the fertilizer dose to be applied, does not vary according to the number o f hectares on which it will be used. Thus the benefit obtained from this information will be greater when the cultivated area is greater. 6.16 Road accessibility i s a factor. Although a large part o f the increase in agriculture in the period analyzed (1988-2003) occurred inareas o f natural vegetation, the start o f this expansion i s clearly associated with accessibility (roads and urban centers). Once the replacement process began, the agricultural area expanded "contagiously". This phenomenon generated large agricultural patches, with small "islands" o f natural vegetation-pattern particularly clear in Salta. 6.17 Inbothregions examinedthe growth ofagricultural areastook place onArgiustoll soils. The ustic soil regime indicates that these soils were formed under conditions o f less water availability. This suggests that the conditions o f favorable water balance currently experienced may well not persist in the future. Contrary to what one might expect, average annual precipitation did not contribute to explaining the variability o f the area used for soybeans, probably because o f the limitedvariation o f rainfall inthe study area. RegressionResults for the Change in 1988-2002 6.18 The following model explains the variation among departments inthe change inthe area insoybeansbetween 1988 and2002: % Difference inSoybeans=-0.13 -0.8 DSc + 0.0007 PROF+0.14 DREN+0.36 DSArr +0.06 ALKA 77 6.19 Table 6.5 shows the statistical results. A change in farm size variable is the most important correlate with soybean expansion, with the proportion o f land in small farms explaining 14 percent o f the variation (negative coefficient). Together with the percent o f land being rented (positive coefficient) socioeconomic variables explained 23 percent o f the variation. Soils-related variables explained the remaining 27 percent o f the variation. Discussion 6.20 The statistical analysis reported above suggests a change in the relative importance of edaphic and socioeconomic variables indetermining the expansion o f the soybean frontier. Inthe earlier period edaphic variables and transportation explained 51 percent o f the variation and socioeconomic variables (percentage land in small farms) 4.5 percent. In 2002, on the other hand, farm size variables accounted for 48 percent o f the variation, with edaphic variables explaining only 16.5 percent. To explain the change in area devoted to soybeans, the importance o f edaphic and socioeconomic variables was about equal, although the single most important variable was the change inthe number o f small farms. Table 6.5: Changeinthe Area of Soybeansbetween 1988 and 2002 Variable Partial R2 Model Rz F P Difference inproportion o f department's land area occupied by agricultural 0.144 0.144 13.68 0.0004 enterprises with fewer than 100 has (DSc) Soil depth (PROF) 0.135 0.279 15.02 0.0002 Drainage Index (DREN) 0.086 0.365 10.74 0.0016 Difference in the proportion o f department's land area rented 0.082 0.448 11.68 0.001 (DSArr) Alkalinity Index (ALKA) 0.046 0.495 7.09 0.009 6.21 These results are consistent with the observation that zero tillage and fertilization practices have strongly widened the range o f edaphic conditions under which soybeans can be profitably cultivated. Similarly, the results in farm size and tenancy are consistent with the bias o f modemsoybean technology toward large management units. 6.22 The positive association between the area planted with soybeans and the proportion o f large-scale producers at departmental level suggests that the latter, generally with greater technical improvements and greater access to information, were more quickly able to incorporate soybeans intheir productive systems. 78 IMPACT ON LIVINGSTANDARDS 6.23 Adequate data i s not available to properly assess the impact o f this agricultural expansion on living standards, including those o f the poor. The only indicator available i s the index o f Unsatisfied Basic Needs (UBN).Analyzing departmental level data, we did found no systematic relationship between change inlanduse and change inthe UBNinthe inter-census period. 6.24 Over the period analyzed the population inthe study area increased 18 percent-from 3.4 million to almost 4 million, while that with UBN decreased by 3.6 percent. The change in a department's area used for annual crops showed no relation to the proportion o f the population with unsatisfiedbasic needs (Figure 6.2). Figure6.2: RelationshipbetweenChangesinUBNand inLandArea used by Agriculture at DepartmentalLevel(1988 -2002) Changes in the proportion of land area used by agriculture Source: LART (2002). 6.25 Independence between social well-being and increase in agricultural area, principally with soybeans, couldbe related to the following: (i)Theincomegeneratedbythiseconomicactivitywasfbndamentallyofaprivate, highly concentrated nature, and in many cases was repatriated to other areas. Added to this would be the over 50 percent increase inthe proportion o f leased areas between 1988 and 2002 in the area under study. Rental contracts often involve non-local enterprises who allocate income to their investors many residingoutside the region. (ii) maybeadelaybetweentheincreaseofeconomic activityanditsinfluenceonan There index such as the UBN. 79 (iii)The effect o f changes in agricultural area may be masked by the scale utilized for this analysis (there may be effects at municipal scale that are not reflected at the departmental scale o f the analysis). (iv)Local tax collection i s very poor, generating little inthe way o f increased expenditure on public goods. 6.26 There is no evidence from either the census or the detailed LANDSAT analysis that small farmers are being significantly displaced by the expansion o f the frontier mainly because expansion took mostly place on unoccupied lands. Consistent with this picture o f occupation o f previously unused land, analysis o f population change in the 6 departments in the Chaco ecosystem experiencing rapid soybean expansion, revealed a 14 percent rate o f population increase in these departments over the 1991-2001 period, suggesting a modest net positive employment effect o f the soybeans expansion inthis area. 6.27 Surprisingly, the expansion o f soybeans area exhibited no clear influence on average farm size. Over the 96 departments inthe expansion area the average farm size fell by 32 percent in 1988-2002. The effect of increasing soy cultivation showed no clear influence on average farm size, however: o fthe 21 departments with greatest proportional increase insoybeans, 4 had a significant decrease in average farm size, 11had no significant change, and 6 had a significant increase. 6.28 Over the full study area (96 departments) the area under rental contracts increased 50 percent. Of the 21 departments with the most rapid growth o f soybeans, 15 evidenced a significant increase inthe land under rental contracts. These data suggest that despite the finding that increases in soybean hectares i s preferentially related to areas with larger farms units, small farmers maybe benefiting from landtransactions with the new soybean entrants. 6.29 Our analysis suggests that the expansion o f soybeans i s not likely to have been associated with a significant net loss o f agricultural jobs in the region, but is, on the other hand, having a major impact on biodiversity. INTENSIFICATIONINTHE PAMPAS 6.30 New technology and management techniques not only helped spread soybeans into frontiers areas where it was not previously planted, but resulted also inthe intensification o f land use in the Pampas. Here we look at some evidence o f this intensification. Given the clear trend towards larger management units discussed in Chapter 2, we look at environmental and social dimensions by tenure category. 6.31 Soil Management. The technical and institutional changes that occurred in the Pampas over the past decade have drastically reduced pasture in rotation cycles. Pasture has been the traditional form o f recovering organic material and soil fertility following intensive cropping. Zero-tillage technology i s different however. Under zero tillage, it i s necessary to rotate soybeans, which leave little stubble, with crops such as wheat and corn that contribute a good quantityo f'stubble (Michelena, 2004), andbecause their root system contributes to maintain soil 80 f soil pathogens and increases the of land dctcrio on ~ K ~ b a ~ 2004). ~Correctly a s ~ i wed. The incentivcs for conect utili ers conduct sail analysis, which wc t i ~ cfzcrc: as a iny to ccx2sus data soil lysis i s niost widcsprcad by 66rdoba (25 percent), SanEa s greater use of soil analysis in ough the overall lly associated tvith both scale ccono tor o f pests to to his practicc is al protection and e c a s ~ ~~ ta~i ~~i~ ~ ~aridaisn closely z c ~ , Eevet, A s in the o f soil arialysis, pest ~ ~ ~ ois ~ ~ ~ ~ the oteral1 adoption o f thc practice remains tow 81 SUMMARY AND IMPLICATIONS 6.35 The current rapid expansion o f the crops frontier and the intensification o f land use inthe Pampas raise three concerns o f potential public interest: social, environmental and sustainability concerns, 6.36 Social concerns exist over whether the expanding agricultural frontier displaces more labor intensive activities and eliminates jobs. In general, the picture that emerges is one o f relatively little effect o f the expansion o f the frontier, either positive or negative. The frontier has advanced primarily onto natural ecosystems in areas o f low population density. While it would be useful to maintain a continued review o f rural employment generation as an element o f provincial regional strategies, we identify no serious social threat at this time. 6.37 Environmental concerns arise from soybean cultivation expanding mainly at the expense o f natural vegetation. As mentioned above, cropland conversion i s taking place largely inforest ecosystems. The development and implementation ofprovincial and eco-regional plans to better ensure the conservation o f critical representative ecosystems, particularly in the highly threatened Chaco ecosystem would be an option to address this concern. In addition, there remains the concern that a significant portion o f frontier expansion has been made possible by recent shifts inthe rainfall regime o f uncertain duration. These doubts strengthen the justification for creating reserves o f representative ecosystems, as the economic sustainability o f commercial agriculture inthese marginal ecosystems i s yet to be proven. 6.38 Sustainability concerns exist due to the relatively low level o f scientific resource management, whether o f soil, pests, or pesticides themselves. While no-till cropping has made possible cultivation o fhillier landand shallower soils, it would be important to monitor the threat o f soil deterioration and declining yields due to nutrient depletion. Inthe Pampas, there is clear evidence o f a shift to ever higher intensity rotations and monocultivation, especially under annual rental contracts, and pool arrangements. In general, professional management associated with pooling arrangements improves adoption o f soil analysis, andpest monitoring. Farmers who wish to buyor rent land probably have no difficulty identifying the effects o fprevious husbandry on the land. There is undoubtedly a short time horizon, however, especially in light o f the current difficulties in financial intermediation in Argentina. The establishment o f a credible monitoring system o f trends in land management and soil conditions from an agricultural sustainability point o f view would be an option to address this concern. The system would provide credible land resource information to farmers' organizations, extension workers and participants inthe landmarket. 83 7. FOOD QUALITY, SAFETY AND PHYTOSANITARYISSUES 7.1 A new agenda for quality and safety has emerged over the last decades not only in Argentina but throughout the world - driven mostly by consumer pressures (see Annex I11for details). Government and producers have reacted swiftly in Argentina to this pressure. Institutions have been created or reformed, new regulations have been issued, special programs have been launched, and new preventive systems have been established. The reaction, however, has beenunevenand has had unequal success. 7.2 This chapter deals with the progress made inArgentina to meet the challenges mentioned above. It briefly explains the institutional and regulatory set up, and then presents and discusses several areas where phytosanitarycontrol issues and experiences have been important. These areas are: 0 The fruit fly eradicationprogram; 0 The carpocapsa control program; 0 The problems experienced with the foot and mouthdisease (FMD); 0 The phytosanitary restrictions to the export of lemons to the US andEUmarkets; and 0 The threat posedbythe soybean rust. 7.3 Issues related to the importance o f quality and stronginstitutionsfor the development of food industries are then examined andillustratedwith the cases of: 0 The differentiation o fwheat qualities; 0 The exploitation o fthe exportpotentialof the honeyindustry;and 0 The unequal standardsinthe beefindustry. 7.4 Finally, the issues posed by biotechnology and the use o f genetically modified (GM) crop varieties are examined, and the experience with the establishment o f risk assessment procedures for GM seeds i s reviewed. The chapter concludes with some reflections and policy options based on the analysis o fthe above issues and experiences. THEREGULATORY SETUP 7.5 Animal andplant health, and food quality issues are institutionally complex. While strong technical expertise is needed at all levels, it would also be important to clearly define responsibilities between national and provincial governments and between them and the private sector. Within the private sector, there i s a strong need for effective collective action. InstitutionsandNorms 7.6 Argentina has been remarkably active in improving its regulatory framework for food quality and safety over the last decade. New international norms refer both to (i)quality standards voluntarily adopted by producers, and (ii) sanitary norms of compulsory observance. 84 The former require establishing quality standards, developing procedures for identifying product types and qualities, and setting up systems for the accreditation o f producers, the certification o f processes and products, and the necessary monitoring and auditing (Gutman, 2003). The latter demandestablishing sanitary norms and standards, andcreating expost control systems to ensure compliance with the norms as well as ex ante systems to promote good sanitary practices throughout the production chain. 7.7 Contrary to quality regulations, particularly voluntary ones, which are rather recent and o f new design, sanitary and phytosanitary (SPS) norms have a long history in Argentina. Much work has been done nevertheless in recent years to update the regulatory framework related to sanitary systems so as to respond to the needs posed by new production technologies, new laboratory methods, andnew consumer demands. 7.8 The legal basis o f the food quality and sanitary system is the Cddigo Alimentario Argentino (CAA). To ensure the application o f the C A A in a more modem form, the Sistema Nacional de Control de Alimentos (SNCA) was created in 1999. The SNCA designates the Sewicio Nacional de Sanidad y Calidad Agroalimentaria (SENASA) and the Administracidn Nacional de Medicamentos, Alimentos y Tecnologias Me'dicas (ANMAT) as the responsible entities for the safety and quality aspects o f animal and vegetable foods. SENASA focuses on fresh and frozen foods whereas ANMAT is mandated to focus on industrially processed foods. A crucial issue is the articulation o f these entities with provincial SPS authorities. SPS control o f export products is carried out by SENASA, and i s done more rigorously than the control o f domestic market products, which i s fragmented among different entities. 7.9 Inthe case ofhorticulturalproducts,special norms were issued in2001 to complement the SNCA by means o f a Sistema Nacional de Control de Productos Frutihorticolas Frescos (SICOFHOR). This system operates through a Comite' Te'cnicoAsesor Frutihorticola, under the jurisdiction o f SENASA. SICOFHOR establishes norms and procedures to ensure the safety and traceability o f products, determining their quality, and detecting the presence o f pesticide residues and contaminating microorganisms. Four stages are envisaged to establish the SICOFHOR, o f which the first one, related to the identification o f product types and qualities, is now underway. 7.10 The global regulatory framework for quality standards voluntarily adopted by individual producers and firms is the Sistema Nacional de Normas de Calidad y Certzjkacidn (SNNCC). This system, which applies to all kind o f products and notjust to foods, i s directed by a national council, the Consejo Nacional de Normas de Calidad y CertiJicacidn (CNNCC), and has two operational bodies, the Instituto Argentino de Normalizacidn y Certzjkacidn (IRAM) and the Organismo Argentino de Acreditacibn (OAA). IRAMis responsible for establishing and disseminating quality standards and norms, while O A A i s responsible for regulating laboratories, accreditation and certification entities, and auditors. 7.11 In the case of foods, quality certification operates through SENASA's Programa Nacional de CertiJicacidnde Calidad de Alimentos (PNCCA), introduced in2001. This program provides a regulatory framework for quality standards voluntarily applied byprivate producers or processors. The standards can be set by the producers themselves, who may want to have them 85 independently certified and thus acknowledged, or derive from international good practices like I S 0 norms or HACCP, or reflect particular standards requested by foreign importers. The system lays down the accreditation and certification procedures for food quality, regulating individual producers or firms willing to certificate the quality o f their products, and also the authorization to operate o f independent certification firms. 7.12 There i s superposition o f regulatory norms between the PNCCA system and the SNNCC, and there i s superposition o f functions between SENASA as the entity designated by PNCCA to establish norms, carry out the accreditation o f firms, and certificate processes and products inthe area o f food quality, on the one hand, and IRAM and OAA, which have similar responsibilities with respect to products in general, on the other hand. Another issue is that SENASA has both accreditation and certification functions, which in principle should not be held by the same entity. 7.13 Progress has also been made in other aspects o f voluntary quality regulation of foods. Thus, a law o f 2001 regulates the so-called "Indicacibn de Procedencia" (IP) or source indication, and "Denominacibn de Origen" (DO) or origin denomination. SAGPyA is responsible for registering the IPSand DOs. Other voluntary regulations refer to the certification o f organic products, introduced in 1992 and 1993. SENASA i s the responsible entity for supervising organic production and applying the organic certification system. Also the project Sistema Integrado de Calidad INTA (SIC-INTA) promotes actions to improve food quality, and has developed good practice manuals for agriculture production, handling and packaging, and h i t s and vegetables. Finally, in2000 an agreement was signed by INTA, IRAMandFundacibn ArgenINTA to establish a joint Sistema de Normas Agroalimentarias, Insumos Agropecuarios, Produccidn Forestal and Maquinaria Agricola to set standards regarding processes, products and associated services. As a consequence of this agreement, joint norms have been developed for several agricultural products. PHYTOSANITARY CONTROLPROGRAMS: ISSUES AND EXPERIENCES 7.14 We illustrate in this part some o f the main issues and programs in phytosanitary promotion and control in Argentina and glance at the strengths and weaknesses of the institutional SPS system and especially SENASA. The National FruitFly Control and Eradication Program(PROCEM) 7.15 The fruit fly inArgentina causes significant losses to h i t s andvegetables and constitutes the principal non-tariff barrier hindering international trade o f these agricultural products. Argentina had to restrict exports to the EUand various Asian countries due to the h i t fly.45To 45Argentina has two species of h i t flies: Ceratitis capitata (Wiedemann) (Diptera: Tepphritidae) known as the Mediterranean h i t fly, originally from Africa and introduced to the country in the early 20" century, and Anastrephafiuterculus (Wiedemann) (Diptera: Tephritidae), the South American h i t fly, a native species that is widespread inthe Neotropical Region. 86 manage these restrictions, PROCEM46was created to (i) fruit fly-free zones inArgentina certify for national and international recognition, and (ii)protect the zones free from fruit flies. PROCEM i s executed through the Patagonia Barrier Foundation (FUNBAPA), a joint agency that includes SENASA, the Provinces, and producers' associations, and uses the following technical instruments to combat the fruit fly: (i) integrated Pest Management (IPM),(ii) detection systems for pest monitoring, (iii) chemical crop control efforts, (iv) direct pest control by means o f the Sterile Insect Technique (SIT), and (v) a quarantine protection system through the isolation o fregions under the Program. 7.16 PROCEM's principal achievements may be summarized as follows: 0 National and Chilean recognition o f the Andean Patagonian Valleys as Argentina's first economically important area free o f fruit flies; 0 Maintaining the status o f areas with little prevalence o f fruit flies inthe PatagoniaRegion and inMendoza; 0 Decreasing the pest population in the remaining PROCEM areas (Provinces o f San Juan and LaRioja); and 0 Implementing a protocol with Chile for free transit o f fruit fly-hosting fruits, for export from ports located on the Pacific Ocean. 7.17 PROCEM faces significant challenges, however, chief among which that o f controlling fruit flies inN O A andNEA (including those provinces without commercial crops), consolidating the achievements reached in L a Rioja and San Juan, reversing the setbacks observed in recent years, and maintaining the fruit-fly free status already achieved. The Carpocapsa Control Program (PLCC) and the Closing of the Brazilian Border to PomaceousFruit 7.18 Apple and pear exports to Brazil, Argentina's main market, have been affected by Carpocapsa, a moth. To combat it, the Patagonia Region Carpocapsa Control Program (PLCC) was created in 1994, in order to: (i) improve the health and quality o f pomaceous fruit (mostly apples and pears) by reducing the levels o f damage caused by Carpocapsa; (ii)intensify phytosanitary surveillance of the production o f fresh pomaceous fruit and by-products throughout the entire productive chain; (iii)increase pest control through the eradication o f abandoned orchards and/or those with phytosanitary risks; (iv) disseminate the adoption of non- contaminating, alternative techniques for pest management by means o f a communications campaign aimed at the fruit producers o f the region; (v) coordinate the operation o f Comisiones Fitosanitarias Locales (CFL) for pest control, granting producers greater participation in the PLCC's design and the use o f resources; and (vi) support the generation and adaptation o f alternative techniques for pest management that are environmentally friendly and favor the production o f fruit with low levels o f residues. 7.19 The achievements under PLCC since 1994 have been modest, and include: (i)the formation o f fourteen CFLs in Rio Negro and NeuquCn; (ii) training and technical assistance the 46PROCEM, Programa Nacional de Controly Erradicacidn de la Mosca de la Fruta, or National Program for the Control and Eradication ofthe Fruit Fly. 87 for fruit sector producers and technicians inintegrated orchard management techniques; (iii) the surveying and eradication o f abandoned orchards and those with phytosanitary risks; (iv) the control o f the pest monitoring network; and (v) operational assistance to the Programa de Agroinsumos Regional (PAR) aimed at financing pesticides for producers through a trust fund system. 7.20 Inmid-May2002, Brazilcloseditsborderstopomaceousfruit fromArgentina, duetothe magnitude o f Carpocapsa infestation. Although the closure lasted 45 days, the decision represented a serious economic problem for the Province o f Rio Negro. The measure was lifted through an agreement between both parties to create in the Rio Negro Valley-the principal apple- and pear-producing area-a bi-national technical commission to inspect and certify future fruit shipments to Brazil. Thus, the potential damage caused by Carpocapsa remains an obstacle to these exports, which represent an average value o f US$62 million per year. The importance o f the ban on exports to Brazil lies partly in the fact that fruits not exported must be placed on the internalmarket, provoking a significant drop inprice. 7.21 The current and potential negative effects o f this phytosanitary problem have been estimated as follows: (i) million per year for the control system established at Brazil's US$1.5 request in2002; (ii) US$34.3 million per year for a possible total closure o f the Brazilian market; and(iii) US$17.0 millionper year inlight o f a possible total embargo by the US market. 7.22 The 2002 border closing for pomaceous fruit reflects the combination o f factors that has made Carpocapsa difficult to control. Principal among these has been the lack o f profitability of the so-called "traditional h i t orchards", in turn a consequence o f the lack o f rootstock and technological improvements among many small- and medium-scale producers. The costs o f converting to new techniques and changing to higher yielding more marketable varieties or from apples to pears are not affordable for small- and medium-scale producers. Inaddition, when an orchard i s found to be "abandoned," and a decision is made to eradicate it, strong legal impediments may be encountered which make SPS regulations difficult to apply. To get around this, any plant health policy aimed at reducing the prevalence o f Carpocupsu could consider the strengthening of current legislation. 7.23 A serious limitation o f PLCC's lies in its financing, collected by means o f the Canon Contributivo -similar to an excise tax on producers-, the resources from which are insufficient to reach the program's objectives. Financial restrictions adversely impact the dissemination o f available new pest control technologies (e.g. sexual confhion pheromones) and hinder adequate pest monitoring and surveillance. As is the case with the fruit-fly under PROCEM, ensuring the proper functioning o f the quarantine protection system and overall compliance with legal regulations on plant health is crucial to attaining suppression o f Carpocapsa. Unfortunately, no fruit production development planor program is currently beingcarried out inthe Upper Valley of the Rio Negro to regain the growth that characterized the region until the 1960s. This illustrates the difficulty o f developing a regional program that works with all stakeholders, including small and medium-scale producers. 88 The Foot and Mouth Disease (FMD) and PHEFA 7.24 In1987, the countries ofSouthAmerica signed a HemisphericPlanfor the Eradicationof the Foot-and-Mouth Disease (PHEFA). Under PHEFA, the region applied an average o f 250 million FMD vaccinations annually, resulting in a 94 percent vaccination rate o f the combined South American cattle herdby 1995. By 1999, clinical signs o f FMDwere absent in 60 percent o f all geographic area o f the continent. Largely because o f PHEFA, the average number of reportedFMDoutbreaks in South America decreased from 955 in 1990 to 130 in 1999 (PHEFA, 1999). PHEFA also contributed substantially to the strengthening o f national veterinary systems throughout the continent, and promoted private sector cooperation in the administration and execution o f FMD control and eradication activities, yielding an overall improvement in the effectiveness o f national animal health programs and services in nearly all South American countries. Argentina, through its efforts under PHEFA, was recognized by the World Organization o f Animal Health (OIE) as "FMD-free with vaccination" in 1997 and "FMD-free without vaccination" in May 2000. Once the "FMD-free with vaccination" status was granted, the U S and Canadian export markets were opened to Argentine fresh beef. Exporters also led trade missions to Japan and other Asian countries in anticipation o f a sustained "FMD-free without vaccination" status. 7.25 The successes o f the MFD control program under PHEFA during the 1990's generated favorable recognition o f SENASA, both nationally and internationally. That success was due to a combination o f factors, including: (i)a firm political decision to control FMD, which in turn translated into (ii) solid legal and institutional framework involving the provinces and sectoral a agencies; and (iii)adequate technology and technical assistance to the producers. Thus, PHEFA was able to manage the competing and conflicting interests within the commercial beef sector by creating a public good for society at large inthe efforts to eradicate FMDinArgentina. 7.26 In 2000, new outbreaks of FMD occurred, which were made worse by the fact that vaccination had ceased in April 1998. Cattle in three provinces in the northeast part o f the country were involved and the situation was linked to illegal imports o f cattle. The disease spread rapidly to Brazil and Uruguay. To combat the epidemic, the three countries used a combination o f slaughter and ring vaccination, which was effective in quickly controlling the outbreak. By October 2000, based on findings from an in-country assessment team, the OIE announced that Argentina would remain on the list o f countries recognized as "FMD-free without vaccination." Fresh beef exports to the United States, which had been suspended in August 2000, were restarted inDecember 2000, with an additional requirement for certification o f origin. Then, in February 2001, the first cases o f what was to become a massive FMD outbreak inArgentina, Brazil and Uruguay were reported to SENASA. However, these incidents were kept from the Argentine public, importers o f Argentine beef and the international authorities until March 2001. The OIE suspended Argentina's "FMD-free without vaccination" status and, in June 2001, the U S issued a ban on imports o f Argentine fresh and frozen beef, retroactive to February 19, 2001. The outbreak peaked in M a y 2001, with 605 reported cases during that month, and more than 2,400 cases were reported during 2001 before FMD was brought under control (Smitsaar, et a12002) (Figure 7.1). 89 7.27 The subsequent closure o f foreign markets reduced export volume by 53 percent and export value by 65 percent within the first eight months o f 2001 compared to the previous year. The overall losses to Argentina's beefindustry have been estimated at over US$450 million from lost exports, with additional losses resulting from lower livestock prices (Reca, 2002). The dairy industrywas also affectedbecauseo freducedproductivity per cow andrestrictedmarket access. I Figure7.1: Foot andMouthDisease Outbreaks inArgentina I 700 7 1 7.28 With this historical record, the reputationo f SENASA was severely tarnished. The failure o f SENASA to provide prompt and official notification o f the disease outbreak created an issue o f trust with foreign regulatory agencies o f lasting consequences. This loss o f trust i s a likely contributing factor to the reticence o f the US to reopen its market. Eventhe region o f Argentina south o f the 42nd parallel, which has retained its "FMD-free without vaccination" status, remains excluded from the U S market, at a combined cost o f approximately US$l50 million per year. PhytosanitaryControls andthe Exportof TucumitnLemons 7.29 Citrus production in Argentina, which accounted for 2.5 percent o f the world total in 2003, is found largely in the Northwest (NOA) and the Northeast (NEA) regions o f the country. O f this, 90 percent o f lemon production comes from the province o f Tucumh. The high acid content o f Tucumtin lemons makes them more appropriate for use as juice, essential oil and dry skin than for whole h i t consumption. Tucuman exports some 350,000 tons o f fresh lemons per year and processes a further 800,000 tons, accounting for 40 percent o f the world's market for processed lemon products (FAS, 2004). In 2001, lemons and lemon by-products accounted for 41 percent o f all exports from T u c u m h with a value o f US$169 million. The main countries o f destination are Greece, Italy, the Netherlands, Spain, Russian Federation and the US. Sustaining and expanding Tucumh's lemon export industry depends on ensuring that lemons satisfy the SPS requirements o f importing countries. 7.30 ExportingLemons to the EU. The markets o f Southern EUmember states were long closed to citrus imports from countries where citrus diseases such as citrus canker, citrus black spot, and citrus leaf spot were known to occur. In 1998, the EUintroduced a system o fprotective 90 measures that allowed imports even if the exporting country was not free o f the disease o f concern if a set o f criteria was met ensuring that the product as well as the soil where it was grown was disease-free. In 1998, the European Commission recognized the citrus producing provinces o f the NOA, namely Catamarca, Jujuy, Salta, and Tucumhn, as free from Xanthamonas campestris (causal agent o f citrus canker).47 This was not extended to the rest o f Argentina including the NEA, which remains affected. This same Commission decision recognized Argentina as free from the fungus Cercospora angolensis (causal agent o f citrus leaf spot). In 1999,48 the Commission did not recognize any part o f Argentina as being free from Guignardia citricarpa (causal agent o f citrus black spot). 7.3 1 In2003, Spain reported the adoption o f emergency measures prohibiting the introduction o f citrus fruits originating from Argentina and Brazil. Spain indicated that it had received consignments o f citrus h i t s from both countries infected with the prohibited pathogens Guignardia citricarpa, Xanthomonas campestris pv. citri and Elsinoe spp. in 2000, 2001, 2002 and 2003, with significant increases in 2003. Similar infestations o f citrus h i t s with G. citricarpa were also reported by the Netherlands and the United Kingdom in 2003.49 Imports were stopped inNovember o f that year. These EUemergency measures, as applied by SENASA, continue to exclude some o fthe largest citrus producers and exporters inArgentina. 7.32 Exporting Lemons to the US. The U S has the potential to become the world's largest importer o f fresh lemons as domestic lemon production in the states of California and Arizona has peaked but countrywide demand has not. Projections (Spreen, T.H. 2001) estimate that the U S could account for 20 percent o f the world's lemon imports by 2010. However, Argentina i s currently shut out o f the U S citrus market because o fphytosanitary restrictions. 7.33 In2000, following years ofnegotiation, the US allowed imports ofArgentine lemons for the first time, and Argentina quickly became the dominant supplier from outside the US. Argentine imports were also permitted in 2001. However, by September, 2001, imports were discontinued when, as a result o f a law suit brought by US growers, the U S agreement was suspended. U S citrus producers, anxious to prevent the entry o f Argentine citrus, particularly lemons, successfully used SENASA's corporate history to undermine its reputation as a trustworthy regulator. In the case o f Harlan Land Co. v. USDA, brought by U S citrus growers, SENASA's failure to disclose a serious outbreak o f FMD in Argentina became an issue in the court's deliberations: "...the court i s concerned about whether SENASA can be entrusted to enforce the mitigation measures used by the systems approach. Although the President o f SENASA and the Argentine Minister o f Agriculture have been replaced, the court i s not convinced that other SENASA officials who were involved in the cover-up have been removed from office.yy50 47 Commission Decision 98/83/EC (OJ NoL 15, 21.1.1998, p.41) and amendment 2001/440/EC (OJ No155, 12.6.2001, p.13). 48Commission Decision 1999/104/EC(OJ NoL 33, 6.2.1999, p.27). 49 Commission Decision of 29 April 2004 on temporary emergency measures in respect of certain citrus h i t s originating inArgentina or Brazil. Official Journal ofthe EuropeanUnion (L208/68-70). 50https://web01.aphis.usda.govlPRkStatusWeb2.nsf 91 7.34 Since then, U S technical teams have visited Argentina to carry out a pest risk assessments on citrus plantations in NOA, but there has been no new phytosanitary agreement between the U S and Argentina. The ImpendingThreat of the Asiatic SoybeanRust 7.35 Inrecent years a new pest, the Asiatic soybean rust (Phakosporapachyrizi), a fungus which attacks soybean plants, has penetrated Argentina seriously threatening soybean production. This form o f rust i s more virulent than the American soybean rust (Phakospora meiboniae) already present in South America.". The Asiatic soybean rust is already present in Brazil, Paraguay, northern Argentina (in 2001/2) and Bolivia (in 2003). It i s hence likely that it will soon attack the soy crops o f the Pampas. Inthe 2003/4 season, it attacked commercial crops inlocalizedpartsofMisiones, Corrientes, northernSantaFeyEntreRios andElChaco. 7.36 The losses from this pest canbe enormous. Inthe 2001/02 season the Asiatic rust affected 90 percent o f Brazil's soybean area causing major losses. Because o f the prevention and control measures introduced, at a cost estimated in US$576 million, losses were much reduced during the 2002/03 season. 7.37 To prevent and combat the Asiatic rust, SAGPyA created in2003 the Programa Nacional de Roya de la Soja with the participation o f SENASA, INTA, the experimental station Obispo Colombres o f TucumAn, provincial administrations and private entities. The program has three components: prevention and monitoring, training and dissemination o f control practices, and research. So far there are no varieties resistant to the American rust, and hence the importance o f the research component. Control is carried out through agronomic practices and chemical means, which are to be promoted through the training and dissemination component. INTA i s working on the identification o f resistant genes. There i s a sufficient stock o f fungicide inthe country, so there i s no need o f quarantine measures. However, the cost o f control measures could be high- as it has been inBrazil. ImprovingKnowledgeon PesticideManagementand OccupationalSafetyIssues 7.38 Ingeneral, the ArgentineanGovernmenthasinsufficient informationto movebeyondthe current, scattered actions in order to implement a more integrated policy towards pesticide handling and worker safety. The current legal framework on these issues is cumbersome and o f doubtful validity in some cases. Analysts and authorities agree that the large number o f current regulations imposes a serious problem for effective interventions. It would be useful the carrying out o f a study to provide a comprehensive diagnostic o f such issues. The study would need to consider pest management protocols as well. The diagnostic could contain an analysis o f shortcomings inthe institutional capacity o f agencies that are operating or should operate, and an evaluation o f the most relevant institutional overlaps and omissions. Other policy options to support worker safety include: (i)strengthening the Occupational Safety Law, with special 51 Iti s a biotrophic parasite that affects the number o f pods and the formation o f beans and therefore the potential soybean yield. Inserious situations it may kill the plant. It i s not transmitted by seed or stubble but it spreads easily through its windborne spores. I t is thought that the spores arrived in America from Africa through the Atlantic Ocean carried by the winds. 92 attention to pesticide-induced chronic illnesses; (ii)tightening controls on unregistered employment (which reduces employers' incentives to promote rural labor safety);52 and (iii) strengthening the surveillance and enforcement capacity o f SENASA and the provincial authorities, and repealing ad hoc regulations o f questionable legality. QUALITY, INSTITUTIONSAND THE DEVELOPMENTTHE FOOD OF INDUSTRY 7.39 Food quality and safety issues are as important as SPS measures for improving market access and the development o f a healthy agro-export industry. As indicated earlier in this chapter, SPS issues are not new to Argentinean regulators, but regulating food quality i s mostly new. Much has been done in this regard in the last decade, in particular on the normative framework for voluntary regulations related to the quality o f food production processes and products. Many quality and safety issues remain, however, such as (a) the lack or insufficient differentiation o f standard qualities in various products, and (b) the double standards applied to production oriented to the export and domestic markets. These issues are illustrated here with reference to three very different industries: the wheat industry, the honey industry, and the beef industry. Differentiation and Quality in the Wheat Industry 7.40 Argentina i s a renowned producer and exporter o f wheat. It accounted for 25 percent o f world exports in the 1930s. Today it i s the world's 5th exporter with a 7 percent share of the world market. Domestically, wheat i s the 3rd largest crop in area planted, after soybeans and maize, but it i s a main staple inthe Argentinean diet. Since domestic demand i s inelastic, it can only grow at the pace o f population. Total domestic utilization i s around 5 million tons per year, or approximately one third o f total output. Production increased in the mid-90s as a result o f technical change, but i s flat now. Further growth depends on the possibility o f expanding exports on advantageous terms. Product differentiation and quality issues are important for this. 7.41 There have been important changes inthe internationalmarket for wheat over the last two decades. Thus, there i s a tendency for wheat to become "decommoditized", with demand being more differentiated and more focused on the characteristics o f the grain such as flour extraction, cleanness, and functional properties (gluten content, elasticity, baking strength, stability, protein content, and other parameters) (Gutman and Lavarello, 2003). Also, the market increasingly demands uniform characteristics o f the different quality groups. Three main factors have influenced these changes. The first i s increasing consumer sophistication in the demand for wheat products, transmitted to the grain market through the milling and secondary processing industries. The second i s the privatization o f the import marketing o f grains by the main importers, since private importers deal in smaller quantities and are more interested in quality than government import companies operating through large contracts o f mixed quality grains. Finally, the industrialization o f wheat products demands a system o f clearly identifiable quality groups o f uniform characteristics, There have beenchanges on the supply side also, with Eastern 52The Occupational Safety Law is currently being analyzed andrevisedinaccordance with an important decisionby the Supreme Court of Justice on the scope o f a worker's lawsuit against a company, a matter that unleashed a technical and legal debate on this Law and the authority to implementit. 93 European, ex-Soviet Union countries and India coming in as significant world exporters o f mixed grains o f generally modest quality, thus puttingpressure on the less differentiated part o f the market. 7.42 Argentina exports around 8 to 10million tons per year, o fwhich 85 to 90 percent as grain andthe rest as flour. Brazil is the mainmarket due to low transport costs andthe Mercosur tariff preference. Argentina has also traditionally exported to Chile and other Andean countries, and to some Asian and African countries (Indonesia, Iran, Kenya and South Africa), although exports to these markets have decreased under the pressure o fthe new export incomers. 7.43 Argentina exports mixed, non differentiated grain, mainly "Argentina trigo-pan". In contrast, Argentina's main competitors -US, Canada and Australia- have sophisticated wheat classification systems and export different standard varieties o f uniform quality. There are several reasons for Argentina's not having been able to catch up with its competitors in this respect. One i s that contrary to what happens inother countries, wheat varieties inArgentina are not associated with geographical production zones, and classification i s hence more difficult. Another reason i s that Brazil, which i s Argentina's mainbuyer, i s not very demanding interms o f quality. Third, storage plants do not normally have classification facilities. Fourth, producers have little information as to the agricultural requirements o f different varieties, price advantages, and possible trade-offs between grain quality and yield. Finally, although there has been awareness o f the wheat quality issue for a number o f years, there was little public action to promote quality untilrecently. This may be due to the absence o f effective institutions "bringing together actors from the different parts o f the chain to design production, technology and marketing strategies, coordinate actions, and negotiate sectoral agreements as well as joint action with government authorities" (Gutman andLavarello, 2003). 7.44 As a consequence o f this lack o f differentiation and quality control, Argentinean exports receive a lower price than those o f its major competitors. Between 1993 and 2003 the average unitvalue inUS$ per ton for wheat exports was 154.1for Canada, 151.8 for Australia, 147.7 for the US, and 132.2 for Argentina.53 The average difference between the unit value received by these three countries and Argentina during the 11 year period was US$18.3 per ton exported. Since Argentina exported a total o f 91.0 million tons over the whole period, the aggregate implicit loss was US$1.7 billion or US$151 million per year. The evolution o f unit values i s shown in Figure 7.2. It can be observed that (i) differences in unit values have increased since 1996, and (ii) differences tend to move inversely with unit value levels. This is logical since in years o f excess supply quality differences become more important as price determinants than in years of excess demand. 7.45 Action has been taken in recent years to improve the quality situation. Thus, since 1995 protein content analysis has been made compulsory for commercial consignments. The Asociacibn Argentina Protrigo (AAPROTRIGO) and INTA have proposed a classification system o f wheat varieties based on quality groups and protein bands, to be adopted on a voluntary basis. In 2003 SAGPyA launched a Programa Nacional de Calidad de Trig0 to promote the competitiveness of Argentinean wheat through improved quality, the promotion o f 53Calculatedfrom FAO's AGROSTAT. Includesthe wheat equivalentof flow exports. 94 wheat classification, the identification o f quality requirements from importers, and measures regarding wheat seeds to facilitate the differentiation or grouping o f cultivars according to quality and type o f use. Also in 2003, SAGPyA approved the Red de Ensayos Comparativos de Variedades de Trig0 to be applied to all registered wheat cultivars as a means to create a quality index to serve as the basis for a Sistema de Clasificacidn de Variedadespor Grupo de Calidad (Cuniberti, 2003). -3 .* c, $ 100.0 5 50.0 - 0.0 I I I I I I I I I I I _ _ _ _us - --Canada Australia mma.-.-Argentina I Source: FAOSTATSdata (2005). 7.46 The actions underway from INTA, SAGPyA and AAPROTRIGO will continue and may eventually result in the generalized application o f a system o f segregation o f wheat varieties. Other measures that would contribute to that end are the dissemination o f information among producers as to the markets, prices, and production and post-harvest technologies for quality wheat, incentives for the introduction o f segregation technologies in storage plants, support to the association o f wheat farmers and AAPROTRIGO, and the enhancement o f the regulatory system with the introduction o f an official classification o f group qualities to facilitate more generalized segregation. Differences in unit values for quality wheat fully justify investments along the above lines. They would allow Argentina to make full use o f the advantages offered by its comparatively low labor costs, its favorable agro-ecological conditions, and its good genetic materials with ecological adaptation. Promises andNeeds of the Honey Industry 7.47 With an output o f some 80 thousand tons per year, Argentina is one o f the world's largest producers o fbee honey, together with China and the US. Honey production represents an annual value o f some US$120 million, and has expanded rapidly in recent years due to good export opportunities following strong demand, the exclusion o f China from the EUmarket (due to the 95 presence o f nitrofurane residues in Chinese honey), and climatic problems affecting production inAustralia, the US andEurope. 7.48 There are some 25,000 honey producers in Argentina with around 2.5 million beehives, and it is estimated that the number o fbeehives could be expanded to 4.5 millions without major changes. Argentina has very favorable conditions for the production o f quality honey capable o f satisfying high demand standards. It can produce honeys o f different types and botanical origins as well as other quality apicultural products, including live queens. Domestic consumption i s small, and more than 90 percent o f output i s exported. 7.49 There have been important advances in the regulatory and support system for honey production in recent years. A traceability system was introduced in December 2003. SAGPyA opened a registry o f honey producers, which has already registered more than 17,000 producers. Inaddition, aqualityprotocolto certifyArgentineanpremiumhoneyisunderway.SENASA has put in practice a Plan de Control de Residuos for the sector, and many enterprises have introduced their own quality norms and control systems. The establishment o f a country-of- origin trade mark for Argentine honey, linked to specific quality parameters, i s being considered. 7.50 Notwithstanding the above measures, Argentina's exports experienced two setbacks in recent years. One was the imposition o f a compensatory tariff for exports to the U S market based on alleged subsidies and the application o f antidumping dues to a number o f Argentinean exporters. The second setback was the presence o f nitrofurane residues inhoney shipments to the United Kingdom and Canada, apparently due to the illegal production and use o f a beehive disinfectant. The illegality compromised some 40 percent o f Argentina's exports. Urgent measures were taken by SENASA and this situation is now under control. 7.51 Argentina's honey industry enjoys a significant competitive advantage. Its main weaknesses are the inability so far o f adding value by exporting honey fiactioned rather than in bulk form, andthe vulnerability o fthe sector to sanitary issues andcommercialretaliation. These weaknesses are to a large extent the result o f deficient governance inthe production chain, with atomized and poorly organized producers facing a concentrated industrial and export sector. There is no strong value chain organization. Collective action to ensure sanitary standards, regulate commercial practices, and adding value to honey exports i s thus encumbered by the low level o f economic governance inthe chain. EconomicGovernanceandthe UnequalRegulatorySituationin the BeefIndustry 7.52 Argentina's comparative advantage for the production and export o f beef has a long tradition. There is a large potential to increase the production and export o fbeefbut, as examined inChapter 3, bothproduction and stock have remained stagnant duringthe last two decades at a level o f 2.0 to 2.5 million tons per year (from around 50 million heads), although with increasing trends inthe last three years. Exports, which oscillate between 5 and 15 percent o f output, rose to nearly 20 percent in 2004 and may continue strong in 2005 due to favorable market conditions. In view of stagnant output, however, this is at the expense of domestic consumption and domestic price rises. 96 7.53 The reduction o f the pasture area over the last 10 to 15 years, particularly due to soybean expansion, i s a factor affecting the beef industry, and further cropland expansion would certainly put a downward pressure on beef output. Even so, there is space for considerable increase in output and exports if a technological revolution parallel to that experienced in the grain sector took place in the beef industry. This would involve intensified field production, mainly through improved fodder varieties, higher fertilization, enhanced use o f supplementary feeding and o f mixedfeed lot-grass systems, and improved livestock management. 7.54 For such an intensification to take place, beef producers evidently need strong market good -- since domestic consumption only grows at the slow pace o f the Argentine population. incentives. Incremental demand should come from the external market -- where prospects are The processing sector would also have to invest in bridging the differences in sanitary and quality aspects that currently exist between domestic and export markets. 7.55 It has been convincingly argued that an important reason why beef production in Argentina does not awaken i s the economic governance situation o f the beef chain, which is not conducive to growth. At issue here i s the difficulty o f chain actors to cooperate in ways that would increase the competitiveness o f the entire industry, with increased returns to its various components. Poor economic governance "results in disincentives for (i) an efficient division o f activities, (ii)adequate coordination o f the productive stages; (iii) the generation o f synergies that would result inhigherjoint competitiveness; and (iv) the adequate distribution o f the income generated" (Bisang, 2003). 7.56 The beefchain is structured inthe following way. At the farmers' end there are two main actors: those breeding the steers and those fattening them. Fattening in Argentina starts at a younger age than in competing countries, and slaughtering also takes place at a younger age and lower weight (350 to 400 kg.)--except when the animals are raised for the export market since foreign markets usually demand cuts from bigger animals (around 500 kg.). Fattening i s mostly done with grass with eventual ration supplements. Feed lot fattening i s not common, although is increasing. 7.57 Slaughtering i s carried out infour types o f abattoirs. Type A plants satisfy all the sanitary requirements and quality regulations and are able to export to the most demanding markets. Type B plants satisfy lower sanitary and quality standards, usually have lower technology but still can sell at the national level and to the less demanding export markets. Plants type A and B are supervised by SENASA. Type C plants are regulated at the provincial level and can sell their meat within the boundaries of the province only. They normally have lower technology and safety standards than those o f A and B. Finally, there are small slaughterhouses regulated by municipalities. 7.58 The weak monitoring capacity o f provincial and municipal authorities introduces differences in the application o f sanitary and quality regulations. The current system o f cuenta zinica, whereby the resources coming from sanitary inspections and fines revert to the provincial treasury and not to the provincial sanitary authority, undermines incentives and withdraws resources for good monitoring. The unequal application o f rules to processors extends also to taxes (inparticular the VAT and company profit taxes) and to labor legislation. 97 7.59 Much o f the economic governance problem i s related to different regulatory systems being applied to producers and processors according to the type o f market. Inthe case o f exports, different standards apply for the Hilton Quota54, FMD-free, and FMD markets. Inthe current system, rents are created in favor o f processors subject to lower standards and less exacting monitoring, which typically occupy the less formal end o f the market. Current incentives, thus, discriminate against exports, favor informality, and foster divergence o f interests inthe industry, obstructing economic governance o f the value chain. 7.60 There are other reasons for inadequate economic governance. An important one is the lack o f a standard objective classification o f product types and qualities, without which it i s difficult to pass on incentives to steer producers in the form o f price differentials for different qualities. Steers are sold according to live weight without regard to the quality o f the final product. Because o f the involvement o f different authorities in sanitary and quality regulation, the diversity o f norms, and the absence o f an objective classification o f types and qualities o f final products, domestic standards are largely determined unilaterally by the actors themselves. They usually follow the norms that, because o f marketing image or other reasons, large slaughtering plants and supermarkets decide to apply to themselves or impose on the up-stream links inthe chain. 7.61 A contributing factor to poor economic governance is that only one product is marketed domestically at the intermediate stage: the half carcass, which i s the only output that abattoirs usually sell to retailers. This introduces unnecessary rigidities and inefficiencies in the down- market system. 7.62 Improving the competitive ability o f Argentina's beef depends on meeting required standards for safety and quality, which i s linked to rebuildingconfidence inSENASA. There are some signs o f positive change. As o f August 2003, SENASA required that all ranches and feedlots registered for exports identify all animals with a tag for traceability purposes. This was the result o f increased sanitary demands from importing countries, particularly the EU. In addition, there i s recent regional recognition o f the need to revitalize a concerted and sustained regional eradication program. Because o f the porous nature o f national boundaries, maintaining disease-free status without vaccination in some countries while neighboring countries remain infected i s nearly impossible. This will require a reorientation o f some national priorities infavor o f regional objectives, and stronger alliances with the livestock industry so that even small producers become interested in, and supportive o f disease control and eradicationprograms. BIOTECHNOLOGY AND BIOSAFETY 7.63 Biotechnology erupted in agricultural production in Argentina in the second part o f the 1 9 9 0 ~linked to the use o f the transgenic Roundup Ready (RR) variety o f soybeans. It was ~ introduced by Nidera and released in 1996, less than two years after genetically modified (GM) 54The Hilton Quota, a favorable tariff rate quota o f 20 percent (compared to the full tariff o f 104 percent), was established by the EU for high quality beef imports. It is currently o f around 58,000 tons and it i s shared with six other countries: Australia, Brazil, New Zealand, the US, Canada, and Uruguay. Argentina i s the main beneficiary with 28,000 tons, some 47 percent o f the total. 98 soybean seeds started being planted inthe US. As shown inFigure 7.3 the area planted with the RR soybean seed grew rapidly. The fast disseminationo fthe GMvariety is a major success story incontemporaryArgentina agriculture. After soybeans, came GMmaize and cotton although not with the same dissemination speed. In 2003 Argentina had 13.9 million hectares planted with GM soybeans, second inthe world to the US, which had 42.8 million, and ahead of Canada (4.4 million), Brazil (3.0 million), China (2.8 million) and Australia (2.1 million).55 By 2004, 90 percent of planted soybeans, 50 percent o fmaize and 30 percent o f cotton used GMvarieties. I Figure7.3: Area Cultivatedwith GMO (millionhectares) 7.64 The extremely rapid expansion o f the planting o f RR soybean seed (see Figure 7.4) can only be explained because o f its role as an ideal complement to the new agronomic package beingadopted by soybean farmers, consisting o f direct plantingplus double cropping (inrotation with maize and wheat) plus intensified fertilizer use. By allowing Glysophate to substitute for a combination o f various herbicides, the modified seed simplifies cultivation, reduces labor inputs and production costs, and increases yields although modestly so. The production cycle i s also shortened, thus allowing cultivation to be expanded to some areas with climatic restrictions (Bisang, 2003). ~~ 55Figuresfrom ISAAA (2004). 99 Figure 7.4: Evolutionofthe SurfacePlantedwith GMOs inArgentina 90% -- Cotton I C o m OSoy 70% 0 60% 50% 40% 30% 20% 10% 0% 1996 -1997 1997-1998 1998 -1999 1999-2000 2000 -2001 2001 -2002 2002-2003 2003 -2004 2004-2005 Source: A S A (2005). 7.65 The utilization o f GM seeds in Argentina was facilitated by the introduction o f institutional innovations, which allowed the opportune release to the market o f GM seeds while systematically conducting the risk analyses required to ensure the protection o f public health, the environment, and national commercial interests. Institutional innovations consisted o f the creation o fnew organizations and the introduction o f new regulations. 7.66 The institutional response to the biotechnology challenge can be considered a best practice case o f public policy infood quality and safety aspects inArgentina inrecent years. The core o f this response was the establishment o f a regulatory framework which defines clear steps for the release o f GM seeds with clear institutional responsibilities within each step. Keywas the Agropecuaria (CONABIA) -- with representatives from the public, private and academic sectors, creation o f a specialized committee -- the Comisih Nacional Asesora de Biotecnologia and the entrusting to this committee o f the main technical and scientific responsibilities. The system makes good use o f the comparative advantages o f the relevant entities in the country-- public, private and academic.56 56The members of CONABIA include the Instituto Nacional de Tecnologia Agropecuaria (INTA), the national universities of Buenos Aires, Mar del Plata, Quilmes, and Comahue, the Foro Argentino de Biotecnologia, the Comitd de Biotecnologia de la Asociacidn de Semilleros Argentinos (ASA); the Consejo Nacional de Investigaciones Cientificas y Tdcnicas (CONICET), the Instituto Nacional de Semillas (INASE), the Sewicio Nacional de Sanidad y Calidad Agroalimentaria (SENASA), the Secretaria de Ambiente y Desarrollo Sustentable and the Secretaria de Politicas, Regu1acid.n y Relaciones Sanitarias o f the Ministerio de Salud, the Sociedad Argentina de Ecologia, the Cdmara de Sanidad Agropecuaria y Fertilizantes (CASAFE), private sector representatives of crop, animal and fish production, the Instituto Nacional de Investigacidn y Desarrollo Pesquero, 100 7.67 Risk analyses are based on the differences between the GMO and the non-GM equivalent. It i s by examining these differences that the environmental impact and the safety or not o f the new organism are appraised. The regulatory system includes risk analyses related to (i) the environment, under the responsibility o f CONABIA; (ii) consumption by humans or the animals o f foods produced with the GMO, under the responsibility of SENASA; and (iii) the potentially negative impacts on markets and international trade, assessedby the trade department o f SAGPyA. 7.68 For a new GMO to be released the following steps must be followed: 0 First, the interested party should present a request to CONABIA accompanied by the relevant technical documentation. 0 The request can be to carry out laboratory, greenhouse or field tests o f new organisms or to release to the market GMOs already tested. 0 In the first case CONABIA gives a recommendation to SAGPyA regarding the authorization to carry out the testing, indicating the biosafety conditions under which it should be conducted. SENASA and INS are responsible for monitoring the tests by means o f direct inspections. 0 For the release to the market o f the GMO three independent reports are needed: (i) from CONABIA, regarding impacts on the environment; (ii) from SENASA, regarding the safety o f the new organism; and (iii) the trade department o f SAGPyA, regarding from market impact . 0 The authorization to release the GMO is given by the Secretary o f SAGPyA based on the recommendations made in the above reports, which, however, are not binding for the Secretary. 7.69 Altogether ten GM seed varieties have been released so far in Argentina, one for soybeans, seven for maize, and two for cotton. Tolerance to herbicides and resistance to insects were the central characteristics o f the genes introduced. These varieties are shown inTable 7.1. 7.70 As indicated by Trigo et a1 (2002), the biotechnology-biosafety system has worked effectively since 1991, and has shown its capacity to evolve according to needs. The International Service for National Agricultural Research (ISNAR) has indicated that "the regulatory system for biotechnology in Argentina is a good model for other countries facing the challenge o f ensuring the safe and responsible use o f agricultural biotechnology" (Burachik. and Trynor, 2002). and the Coordinador General o f the Oficina de Biotecnologia of the Secretaria de Agricultura, Ganaderia, Pescay Alimentos (SAGPyA), who i s the Executive Secretary and Technical Coordinator of Biosecurity o f CONABIA. 101 Table 7.1: GMCropVarietiesReleasedinArgentina Seed Identification Company Characteristics Year Soybeans "40-3 -2"" Nidera-Monsanto Tolerance to glyphosate 1996 Maize "176" Ciba-Geigy Resistance to lepidopters 1998 Maize "T25" AgrEvo Tolerance to ammonium 1998 gluphosinate Maize "Mon 810" Monsanto Resistance to lepidopterists 1998 Maize "Bt 11" Novartis-Agrosem Resistance to lepidopters 2001 Maize "NK603" Monsanto Tolerance to glyphosate 2004 Maize "TC 1507" Dow Agro- Tolerance to ammonium 2005 Sciences & Pioneer gluphosinate andResistence to Argentina lepidopters Maize "GA 21" Syngenta Seeds Tolerance to gluphosate 2005 Cotton "Mon 531" Monsanto Resistance to lepidopters 1998 Cotton "Mon 1445" Monsanto Tolerance to glyphosate 2001 Source: Oficina de Biotecnologia, SAGPyA. 7.71 The introduction, however, o f GM seeds, especially o fthe RR soybean seed, has not been without conflict. For a number o f years Monsanto, the owner o f the RR technology, has clashed with Argentinean producers and the Argentinean Government inconnection with the payment o f intellectual property rights for the commercial use o f the RR seed. The characteristics o f this conflict are summarized inBox 7.1. LESSONS, CONCLUSIONS, AND RECOMMENDATIONS Lessons on Sanitary and Phyto-Sanitary Issues 7.72 Economic Losses. Table 7.2 characterizes the direct annual economic losses caused by pests in pear, apple, and citrus crops in different regions o f the country. Annual losses are conservatively estimated at US$75 million. These direct losses are historically between 15 and 20 percent o f p r o d ~ c t i o n .There are also indirect costs caused by quarantine restrictions-both ~ ~ for exports and among regions within the country. These costs may be manifested as: (i) rejection or rerouting o f shipments and destinations, with a corresponding price penalty; and (ii) quarantine requirements that generate additional marketing costs. These requirements affect all types o f shipments o f fresh h i t or vegetables considered to be hosts for quarantine pests (e.g., methylbromide or cold treatments for h i t flies). Most important however, is the opportunity cost o f lost export markets and loss o fmarket confidence. 57Values o f losses caused by pests and diseases were drawn from (i)Phytosanitary Program for Northwest Argentina, Tucumin Citrus HealthProgram. Provincial Agricultural Services Program (PROSAP/SAGPyA), August 2002; (ii)Phytosanitary Project "Reduction of the Prevalence of Fruit Flies in the Northeast Region." PROSAP/SAGPyA, July 2002; (iii) Technical-Economic Evaluation Report, "Impact o f Carpocapsa on Pomaceous Fruit in Northern Patagonia, Argentina." Fundacidn Barreras Patagdnicas (FUNBAPA), February 2003; and (iv) Phytosanitary Project "Suppression of Carpocapsa in Fruit Orchards o f the Rio Negro Valley. PROSAP/SAGPyA, July 2002. 102 Box 7.1: IntellectualProperty Conflict over the RR SoybeanSeed With more than 2 million tons of grain seeds marketed each year, Argentina is the second largest Latin American market for grain seeds. Moreover, Argentina i s a world leading market for soybean seeds, more than 90 percent o f which is the genetically modified RRvariety (some 14 millionhectare planted with RR seed inthe 2004-05 season). There is a standing conflict between Monsanto, the San Luis based agribusiness company, developer and owner o f the RRtechnology, andproducers, exporters and Government inArgentina. There are two aspects to ths conflict. First, Monsanto does not produce and sell directly RR seeds inArgentina; it licensedthe right to use the technology to seed companies operating in the country. Hence, Monsanto cannot charge directly to farmers for the use o f RR seeds. These seeds have not been patented in Argentina, where the country's legislation to protect intellectual property over living organisms i s based on a breeder's rights approach, not onpatent registration. The second aspect concerns the freedom of farmers under Argentina's seed law (art. 27 of Law 20,247) to reuse their own seed without paying any intellectual property fee. From a technical perspective, reuse o f the RR seed is entirely possible. Farmers do not only reuse it, but an active informal market o f second generation seed has developed which government finds difficult to stop--the so called bolsa blanca. Monsanto complains that through the bolsa blanca producers avoidpaying any fee for the use o f the RRtechnology. Monsanto negotiated with government and producers' representatives to try to reach a global agreement for the payment o f a technology fee for the use o f the RR seed. This had been done inBrazil where a compensation fee i s paid to the company per ton o f soybeans produced. Negotiations, however, did not prosper, and in 2003 Monsanto discontinued its soybean genetic improvement program inArgentina. Part o f the problem is that, initially, the major business for Monsanto was not the seed itself but selling the associated glyphosate pesticide o f which Monsanto had the patent. This patent, however, has now expired and glyphosate i s being imported from China at a lower price. InMarch 2005 Monsanto made a strong move by deciding to charge no fees to the seed companies producing and selling RR seeds in Argentina, and seeking compensation directly from exporters o f GM soybeans to Europe. Monsanto has patented the RR technology in EU importing countries and i s counting on legal action in these countries against Argentina's exporters to obtain payment for the use o f seeds where the RR technology i s incorporated. This decision has met strong opposition from Argentina's producers and government, but inthe end it may force a negotiation. Going from the experience of the agreements reached by Monsanto with Paraguay and Brazil, a payment o f between 3 and 5 U S dollars per ton seems possible. 7.73 Reputational Issues. On two occasions, lack o f confidence inthe regulator has impacted rulings in importing countries against Argentina's interests. One was the ability o f U S citrus growers to overturn an agreement with SENASA inorder to block citrus imports, andthe other i s the continuing poor access o f Argentine beef exports to the U S market. Not only did the 2001 FMDcrisis have a big economic impact on the Argentineanbeef sector, with losses estimated at over US$450 million, but it also created a crisis o f confidence in SENASA. Confidence in the integrity of regulatory bodies is fundamental to maintaining good working relationships with foreign agencies. Table 7.2: Estimate of Annual Losses to Fruit Crops CausedbyPestsandDiseases Pest or disease Crop Regions affected Values of losses (US$ million) Carpocapsa or Pears, apples, walnuts, and Patagonia and Cuyo 19.0 Pear Moth quince Fruit fly Citrus and stone h i t s Northwest, Northeast 21.2 and Cuyo Cankers Citrus, especially lemon Northwest 35.6 Source:Huergaand San Juan (2004). Basedon FUNBAPA (2003). Note: Direct losses causedby Carpocapsaalludeonly to those occurringinthe UpperRioNegro Valley. 103 7.74 The Importanceof Collective Action. The FMD experience leaves interesting lessons regarding (i) how a program based on a regionally concerted effort and strongly supported over a number o f years with adequate legal, technical and financial means meets success; (ii) the importance o f the involvement o f the private sector through the COPROSAs in achieving that success; (iii)the importance o f collective action at the regional level epitomized by the PHEFA program; and (iv) how failure to maintain the conditions leading to success turned back the clock for FMDcontrol inArgentina as shown by the 2001 crisis. 7.75 Monitoring. A major contributor to the reemergence o f FMD in Argentina was the weakeningo fthe highlevels o fprevention and surveillance requiredto maintain the disease-free status that had been achieved. Once the economic and political pressure to reach this milestone had subsided, there was deterioration in the veterinary infrastructure, reduction in cooperative efforts with the private sector, and decrease inFMD awareness-raising activities with producers. This was not unique to Argentina, and data from FMDprograms in South America indicate that once a country declared the disease eradicated, there was a shift in government resources and support to other high-priority needs (Correa Melo et al, 2002). 7.76 Pesticide Handling and Worker Safety. The current legal framework on pesticide handling and worker safety is cumbersome and o f doubtful validity in some cases. A study would be useful to provide a comprehensive diagnostic o fthe issues involved. Lessons on FoodQuality and SafetyIssues 7.77 Wheat Industry. This industry exemplifies the economic opportunities missed from the lack o f product differentiation according to quality, which constitutes a dynamic disadvantage vis-a-vis a progressively "decommoditized" wheat market. Little government action in the past and weak chain organization are to be blamed for this. Institutional actions would be required to stimulate changes across the entire chain. 7.78 Honey Industry. This case exemplifies the economic opportunities missed because o f not adding value to the exported product. The strong presence o f Argentina in the honey market and the quality and variety o f Argentinean honey should allow for more high-value honey exports. Again, the weakness o f chain organization i s at least partly to blame for this. 7.79 Beef Industry. The economic opportunities missed due to insufficient quality differentiation are also the central theme in this case. The beef industry also exemplifies the discrimination against the export sector due to unequal industry standards according to market destination. The absence o f an objective standard classification o f the final product and the unequal regulatory system are partly responsible for this. Biotechnology 7.80 Opportune institutional and regulatory action was crucial to the successful introduction o f GM seeds in Argentina, especially soybean GM seeds. Part o f the explanation for the good results rest on the fact that innovative institutions were created which were capable o fmobilizing within a flexible administrative framework and with a minimum o f legal change much o f the 104 knowledge existing or being developed in the country on biotechnology. The concurrence o f public, private and academic experts injoint committees with the clear purpose o f giving advice to the Secretary o f SAGPyA was the right approach. There was also transparency inthe process o f releasing GMOs, and respect to the autonomy o f CONABLA. This contributed to its legitimacy as a responsible scientific regulatorybody with advisory functions. Conflict, however, emerged over the payment of intellectual property rights for the new seeds, which has yet to be resolved. FinalConclusionsandRecommendations 7.81 We have seen in this chapter, through various illustrations, the progress being made in Argentina in some o f the critical SPS and food quality issues. On the whole, Argentina has reacted swiftly to the challenges posedby the new SPS and food quality agenda. Nowhere i s this more apparent than inthe institutional systemput inplace to adapt to the opportunities and risks posed by biotechnology. Some challenges remain, however, and continuing efforts to strengthen the food safety and quality system deserve priority in the policy agenda. The next section indicates some o f the systemic areas where Government could act to accelerate the on-going process o f improving food quality, safety and phytosanitary standards. 7.82 Clearer definition of roles. There seems to be needed to define clear roles and responsibilities among national and provincial, public and private agencies, and the corresponding budget resources. Thus, for PROCEM (fruit flies), for example, it would be useful to define specific actions for NEA and NOA, and to identify the corresponding institutional roles, including for example the provision o f sterile flies for the investment program being prepared. A particularly serious problem for combating Cavpocapsa i s that o f abandoned orchards. Similar institutional issues o f coordination and delegation remain in the beef (both FMDandquality issues), wheat, andhoney examples cited above. 7.83 Strengthening SENASA's Police and Quality Control Functions. A strong national- level surveillance and enforcement power o f SENASA i s a critical, non-delegable attribution o f this government entity. It is important that SENASA's enforcement measures are upheld inthe courts o f law, and that there are adequate budget allocations to support the police function. 7.84 Another area of clear SENASA responsibility i s the quality control o f commercially marketed fresh products. This control is currently conducted on only a small proportion o f the total produce marketed. Support could be given to the efforts o f SENASA's current management to obtain financing to improve the Fresh Fruit and Vegetable Control System (SICOFHOR). First, however, it would be important to analyze the alternative systems, especially to compare procedures based on the analysis o f market samples, with alternatives based on traceability and supported by fruit and vegetable producers. 7.85 Separating SENASA's Police Function from Promotion and Policy Coordination. Strengthening the capacity o f SENASA to carry out SPS, and food and worker safety surveillance and enforcement would be facilitated by transferring policy and coordination functions out o f SENASA to a plant protection and quality enhancement unit in SAGPyA. This division o f responsibilities would have several advantages. First, it would focus SENASA's 105 capacity on the technical aspects where it i s most qualified. Second, it would create a unit specialized in SAGPyA focused on the promotion o f collective action (for pest control, food safety and quality, and worker safety). This demands a very different skill mix, and often a different mentality than that required and desired for the police functions o f SENASA, with an emphasis on negotiation and conflict resolution and the economic analysis o f alternatives. Third, it would eliminate corporatist incentives to centralize activities. Despite SENASA's current expressed willingness to do so, past experience is that internal corporatist tendencies have interfered with efforts to decentralize and delegate. Finally, as discussed above, a strong police function in SENASA would create the incentive framework for the promotion o f collective action by SAGPyA. Placing these functions in separate institutions removes also potential conflicts of interest and focuses efforts. 7.86 To conclude, we offer several other specific suggestions to strengthen the country's SPS and food quality agencies: 0 Strengthening the international negotiation role o f SENASA. SENASA's role in negotiating international agreements i s important, but it requires (i) strengthening human resources to more effectively carry out this work, and (ii) working with SAGPyA to ensure that local entities participate inthe definition o f standards and protocols. 0 Seeking the convergence o f regulatory standards. More pro-activity in seeking the convergence o f double standards between domestic and export markets would be welcome. This could be accomplished not by reducing export standards but by improving those for the domestic market. Market forces led by supermarkets and consumer demand are bringingthis about inany case. 0 Establishing objective standards and norms for different food qualities. Quality improvements generally depend on producers receiving signals from the market in the form o f premium prices. Setting up objective quality standards and disseminating them promotes market transparency and facilitates transferring to primary producers the premiums paid for final products o fbetter quality. Strengthening financing and transparency. Budgetary processes could be reviewed to provide greater clarity inthe rules o f the game. This could include (i) transparent budgets by program to strengthen social control, (ii) definition o f tasks for which a service clear fee could be charged, and o f those for which their public goods nature suggests that they be paid from the general budget. Generating a forward-looking capacity. SAGPyA could help catalyze and lead regional institutions with strong private sector participation to anticipate changes in market requirements and conditions (e.g. variety changes and sanitary standards). These institutions could also take the lead, inpartnership with INTA, in research and extension inthemes relatedto sanitary control andmarket demand. Finally, these institutions could serve as observatories, monitoring the impact o f on-going activities to identify sanitary andother threats to local andregionalproduction. 7.87 Overall, the investments required to carry out the above suggestions do not seem particularly expensive andthe pay-offi s potentially very large. 106 8. THEIRRIGATIONSUBSECTOR IMPORTANCE OF THE SECTOR 8.1 This chapter reviews the importance o f the irrigation subsector, especially inthe Regional economies. It looks at indicators o fperformance, and examines policy options for reform. 8.2 Argentina's irrigated area reached 1.7 million hectares in2001, o fwhich some 70 percent (1.1 million has) is located in the arid or semiarid regions. Fruits and vegetables make up the most important part (28 percent) followed by sugar cane, tobacco and aromatics (15 percent), forage crops and cereals (14 percent), orchard crops (12 percent), forest crops (9.1 percent), citrus, olives, andnuts (3.3 percent), cotton (1.9 percent), and others (16 per~ent).~' 8.3 Interms of economic value, production from irrigated lands represents some 26 percent o f Argentina's total. In several arid provinces such as Mendoza, San Juan, and N o Negro, however, agriculture i s virtually dependent on irrigation, with irrigated production surpassing 90 percent o f total agricultural output value. In these provinces, irrigated agriculture represents between 22 percent and 45 percent o f the provincial value added, and is a major source o f employment. 8.4 Approximately 80,000 farms use irrigation out o f a national total o f 300,000 in 2001 (CNA, 2002). Irrigated agriculture includes many farm types. Thus, 75,000 farmers (nearly 92 percent o f the total) have farms o f less than 25 hectares, while 30,000 farmers (more than 36 percent o f the total) have farms o f less than 9 hectares (World Bank, 2000). 8.5 The irrigated area increased from slightly over 500,000 hectares in 1958 to some 1 million hectares in 1988 and 1.7 million hectares today. Irrigation expanded substantially following the 1950s with the use o f tube-wells for groundwater irrigation, which increased both the area andthe efficiency o f irrigation. The quality o f groundwater, however, is betterinhumid than in arid areas. Inthe latter, the overexploitation of aquifers and obsolescence of the wells have contaminated many underground sources. 8.6 Growth took place largely in two distinct periods: 1979-1982, and 1988-2003. The first period corresponds to the carrying out o f irrigation infrastructure development projects financed by international donors in N o Negro, San Juan, Santiago del Estero, and other provinces. The second and more recent period was private-sector led but with a strong influence o f provincial expenditures in several provinces, and with irrigation based on groundwater. This period saw an active incorporation o f innovative technologies for products directed to export markets, in particular table olives, olive oil, andtable grapes. 8.7 Inanumber of areas, recent years have seen substantial growth inthe value ofoutput per irrigated hectare. For example, in Mendoza, where irrigation is oriented to wine grapes, the replacement o f low quality grapes with grapes appropriate for fine wines has nearly doubled the Data for 2000. 107 value o f irrigated production. The Valle Centra2 in the province o f San Juan has also replaced low quality grapes with high quality wine grapes, but has specialized in quality table grapes which command highprices ininternational markets. THEPOTENTIALOFIRRIGATION 8.8 There i s good additional potential for irrigation in Argentina. The country's water regulatory capacity consists o f 124 h c t i o n i n g water storage structures. Water stored in these structures i s destined for various uses, but 116 systems are used for irrigation. The total capacity o f water storage is o f the order o f 160 km3(World Bank, 2000). Similarly, the country has 125 public and private irrigation zones. The potential irrigated area i s 6.3 million hectares, o f which 2.5 million could rely on irrigation all year around. With 1.75 million hectares currently under full irrigation, this implies that with favorable economic conditions some 700,000 hectares could be added. 8.9 Much o f current irrigation infrastructure i s underutilized, however. InSan Juan, Santiago delEstero, Tucumin, Chubut, and Formosairrigated agriculture utilizes only about 50 percent o f areas with irrigation facilities. This figure is particularly strikingwhen it i s noted that more than 80 percent o f the crops under irrigation have had favorable conditions in international markets since mid-2001. As discussed below, underutilization i s largely due to institutional constraints. 8.10 There is a serious problem of technical efficiency. Average efficiency o f irrigation schemes -- including water conveyance and distribution -- does not exceed 30 percent. This figure i s low or very low when compared to irrigation leaders in Europe, but also in relation to neighboring countries in South America. The area with highly efficient technology (tube conveyance, localized distribution) does not exceed 3 percent o f the irrigated surface. 8.11 Inefficient water use leads to water shortages. These are not a generalized problem in Argentina, but local shortages are important. We give two examples: 0 Inefficiency in water intake and distribution makes it impossible to adequately irrigate highproductivity plantations andhighreturncherries inLas Golondrinas and ElHoyo in the province o f Chubut; 0 Water use inefficiency limits also irrigation in Coronel Moldes, Salta, to low value tobacco andhorticultural crops. Examples o f this nature can be repeated throughout the length and width o f the country. L o w efficiency inwater use on existing scheme prevents the expansion o f the irrigated area. 8.12 Water scarcity at the scheme level i s also related to the availability o f infrastructure. First, there i s lack o f critical works. Thus, for example, lack o f adequate structures for intake and distribution creates water shortages in Chubut's Valle de Octubre system. Second, there i s lack o f maintenance. For instance, the poor state o f storage infrastructure inChicoana, Salta, prevents the irrigationo f about 40 percent o fthe available area. 108 . 8.13 Finally, some 20 to 30 percent o f irrigated areas have high salinity problems due to deficient drainage. This results inlarge economic losses due to foregone production. 8.14 The main features of contemporary irrigation development in Argentina can be synthesized as follows: Large increase inirrigated surface, production and value o f output over the last decade. Expansion o f the production in irrigated lands o f export crops with sustained prices in internationalmarkets. Complementary irrigation grows more than full irrigation; the benefits, hence, o f . incremental irrigation go proportionally more to middle and large scale farms in the Pampas, who are the main users o f complementary irrigation. Irrigated areas are incorporating technical improvements in the genetic material, crop management, post-harvest practices and processing, but there i s little technical improvement inirrigation techniques and water management. Low technical efficiency in irrigation continues and i s mostly determined by limited technical capacity o f water users and poor state o f irrigation infrastructures. In several provinces where irrigated production is relatively stagnant, irrigation mostly benefits small poor farmers. Production stagnation has hence a social impact. CONSTRAINTSTO ACHIEVING DEVELOPMENT POTENTIAL 8.15 Achieving the development potential o f the regional economies will require considerable improvements in water institutions and policies as well as in SPS and quality improvements to gain market access. The major impediments to better water use are: (i) an incomplete and imperfect legal framework; (ii) weak and poorly articulated water institutions; and (iii) unclear rules o fthe game for financial independence o f water users associations. IncompleteLegalFramework 8.16 Argentina does not have a general water law and a master plan for the use o f water resources. A national legal framework and a long term national water management strategy would be important in order to (i)establish national level criteria for the allocation o f water resources and national guidelines for water management; (ii) prevent and solve conflicts related to waters shared by different jurisdictions; (iii)establish national water standards; and (iv) establish uniform norms for the protection o f water sources. 8.17 Provinces have their own legislation for water management, which i s complex and heterogeneous (World Bank, 2000; FAO, 1996; Alvarez, 1997). The main problems in the provincial legal frameworks are that they do not give security to users, and the overlapping o f management responsibilities and competences. Also, regulatory frameworks diverge much across provinces. Development and streamlining o f water legislation could be done in conjunction with the revision o f environmental and soil use norms. This could be donejointly by COHIFE and COFEMA. 109 Weak and Poorly Articulated Water Institutions 8.18 Weak water institutions affect not just technical efficiency but, more fundamentally, the capacity to mobilize resources, both financial and human. There is overall weakness in the institutional framework at the federal, provincial and scheme levels. Reasons for the weaknesses, which vary at each level, are analyzed separately below. Federal Level 8.19 For various reasons the coordination and development o f Argentina's water resources requires strategic direction from the national level. First, the scale o f investment i s large relative to the financial means o f individual users. Second, efficient and equitable rules o f the game must be established for federal assistance to provincial (and local) resource development. 8.20 The coordination and strategizingroles are compromised by lack of planning resources. Thus, the top agency responsible for water resource policies (the Under-Secretariat for Water Resources in the Ministry o f Infrastructure, Planningand Public Works) has (in2005) only one full-time professional dedicated to irrigation5'. INTA does not include irrigation among the strategic areas included in the new Strategic Plan 2005-2015. The Secretariat o f Health and the Environment has no projects for the environmental monitoring o f natural resources in irrigated areas.60 In SAGPyA, irrigation depends o f a management unit (PROSAP) that does not appear inthe institution's organizational chart and has no clearly definedresponsibilities related to the planning and management o f irrigation. PROSAP has been very active and useful, however, in capacity building, technical assistance, andthe formulation o f investment projects. 8.21 In addition to material scarcity there is scarce coordination among the relevant agencies (the three agencies mentioned above). It i s critical inthis regard that all the institutions related to irrigation participate in the recently created research network for the integrated management o f water resources. Current participants include the Secretariat o f Science, Innovation and Technology (SECyT), the Under-Secretariat o f Water Resources, the Universities o f Santa FeyCuyo, and others. 8.22 The third factor i s the weak system of information to support irrigation development. This weakness crosses all levels and actors, including public sector institutions and local irrigation consortia. This situation could be remedied with the rapid development o f a complete information system, such as the one proposed at the meetings o f the Federal Irrigation Council in 2003 and 2004. Provincial and Local Levels 8.23 Annex IV summarizes the status o f irrigation agencies in the provinces. The first factor affecting their performance i s lack of autonomy, stemming from their weak position in the organizational structure o f provincial governments. Only two agencies are autonomous and can ~~ 59 Other professionals inthe institution collaborate part time. 6o I t should be mentioned, however, that this Secretariat recently formulated a program to evaluate the contamination o f agricultural soils. 110 administer their own budgets-in Mendoza and N o Negro. In many provinces, irrigation authorities are Direcciones Generales located in the third or fourth level o f the administrative hierarchy. Their resources are so limitedthat insome cases (like inJujuy and Formosa) they lack vehicles and fuel to carry out basic surveillance. 8.24 A second factor is the scarce endowment of technically qualified human resources. Without an increase in the minimum endowment, efforts at capacity building are futile. Staff numbers are large in some provinces but that o f professionals i s minimal. In Santiago del Estero andJujuy, for example, there are insufficient engineers specializing inproject works, andcurrent legislation and budgetary restrictions prevent additional hiring. The excessive number o f unqualified workers suggests that the public irrigation agencies act in some cases as a refuge against open unemployment. Of additional concern is the fact that many professional staff are reaching a relatively advanced age and will retire soon. Also, the training o f technical staff in provincial agencies i s weak, especially in management techniques. The active participation o f PROSAP has brought about substantial improvement in various provinces (like Tucumhn, Chubut, Neuquh) but the challenge remains. 8.25 Other factors are the scarcity of heavy equipment for irrigation work maintenance, and the weakness o f the information system for irrigation development. Poor information prevents adequate planning for capturing resources because there is no updated list o f irrigators--not even in Mendoza and N o Negro where management standards are highest. Nor does available information allow the precise calculations o fwater demand. Insome systems even the number of farms with water rights is not known. DecentralizationIssues 8.26 All provinces have promoted the formation and fimctioning of water users associations (WUA), especially since the beginningo fthe 1990s.The objective has beento increase irrigation efficiency by moving decision making closer to the users, and to reduce management costs and the budgetary burden on the provinces. In virtually all cases the first step was to give this decentralized structure a legal basis, either directly within the provincial water code (cddigo de aguas) or through administrative regulations. 8.27 Decentralization effectiveness varies across and within provinces. Provinces which have progressed the most in decentralization are Mendoza (in practically all local schemes) and Rio Negro (inmost o f the systems, although several are still managed by the provincial authority). In the other provinces a minority o f decentralized, locally managed systems coexist with others which have not yet formed associations or whose associations are not functioning. In general, local management i s more effective in associations o f large- and medium-scale holdings. Schemes mostly formed by small fanners, with low economic and technical level tend to depend still on the provincial agency. 8.28 The rate o f formation o f W A S decreased after 1997, due to the economic and institutional impacts o f the economic crisis. There are signs, however, o f recovery, and various provinces are once again promoting the creation and functioning o f WUAs. 111 8.29 Notwithstanding some failures, decentralized management has brought important benefits. It has lowered costs, increased flexibility, improved the participation o f local farmers, and increased the capacity to mobilize local resources. These results lead to the conclusion that establishing local management mechanisms, together with improving the pricing of water, are the most critical components in a strategy of irrigation development over the nextyears. 8.30 Experience in recent years also shows the importance o f developing a greater commitment on the part o f the associations to medium-term planningand financial management. The Upper and Middle Valleys o f Rio Negro, the N o Tunuyb watershed inMendoza, and the Valle 12 de Octubre o f Chubut all are examples o f the benefits o f greater financial involvement andcommitment. DifficultiesAssociatedwith Water PricingandWater Charges 8.31 Problems relatedto water pricing and charges can be summarized as follows: 0 Difficulties intransferring water between users or even among uses for the same user; 0 Low pricing o fwater that results inlow water fees (canon) inmost provinces; 0 Highrates o f default, even on low water charges. 8.32 A problem for more efficient water use is the lack of transferability of water due to provincial water norms that prevent selling or otherwise transferring water even among members o f the same association. Worse yet, the law even prevents an owner o f water rights from using the water to irrigate those parts o f hisher own landholding without water rights. Lack of transferability reduces the potential productivity o f the available water. Users may simply not use the water, through lack o f authorization to transfer it to others, or may use it inexcess for a given use, through lack o f incentives to manage it efficiently. Both outcomes worsen the technical efficiency and productivity o f water (for instance in Mendoza, and L a Rioja). Not surprisingly, legal constraints have not managed to totally prevent water exchanges that are mutually beneficial to buyers and sellers, It i s well known that covert markets operate despite legal restrictions (for example in L a Rioja), but under conditions that affect the system's transparency and efficacy. 8.33 Low water charges is another problem which, fortunately, has been somewhat ameliorated recently in a number o f provinces. Existing water charges, however, filly cover the costs o f providing water in one province only--Mendoza. Inmany others, water charges are not sufficient to cover operating costs, much less to recover investment costs in infrastructure. In a number o f provinces administrative costs are inflated by the employment o f a large number o f unqualified personnel (for instance in Santiago del Estero, Jujuy, and L a Rioja). In these provinces, however, water charges would continue to be insufficient, even in the hypothesis o f substantial improvements inorganizationalmanagement. 8.34 The difficulties o f cost recovery worsen with the high rates of default on water charges in most provinces. This problem has two roots. The first is the poor financial health of many farmers, especially as a result o f the economic crisis. With devaluation and good prices for 112 exports, however, many fanners are now beginning to work their way out o f debt. The second root is the weak governance o f irrigation institutions. Weak governance can be traced to lack o f a credible commitment on the part of water institutions to provide a good service, and to the lack of a system to reward good payers and punish defaulters. In the absence o f these incentives "free-riding" becomes endemic. "Cheating practices" can only be corrected through solid improvements in the capacity o f federal, provincial, and W A S to enter into credible commitments to provide strategic and technical assistance, but not to bail out failing water associations. W A Sshould provide services efficiently inreturn for water charges, and enforce sanctions against those who attempt to free-ride. DifficultiesRelatedto Infrastructure 8.35 This type o f difficulties canbe summarizedas follows: 0 The advanced state o f deterioration o f much o f the irrigation infrastructure, constructed from the 1940sthrough the 1980s; The absence o f complementary investments which would make it possible to modernize the functioning o f irrigation systems - for example computerized gate valves and flow meters; Slow incorporation o f modern, mechanized irrigation methods, such as drip or pulse irrigation. 8.36 The degree of infrastructure deterioration varies substantially across provinces. The most deteriorated infrastructures are found in Santiago del Estero, Salta, Catamarca, L a Rioja, and Formosa. The highest quality infiastructures are found, not surprisingly, in the provinces with the highest quality management and the most advanced institutional development - Mendoza and Rio Negro. Infrastructure deterioration i s also less serious in"rich" provinces like Santa Cruz, Chubut, and Buenos Aires. Nevertheless, problems o f infrastructure maintenance are relevant in all provinces, even in richer ones. In a number o f provinces, including San Juan, Santiago del Estero, Tucumin, Chubut, and Formosa, the installed irrigation area which i s not irrigated i s almost as large as the area effectively irrigated.61 The absence o f modem irrigation equipment at both the system and the farm level i s generalized in all provinces. The potential impact o f improving,deterioratedworks, such as intakes and canal sections, i s very high. DifficultiesRelatedto Financing 8.37 Shortage and unpredictability o f financing has fundamentally affected the development o f irrigation infrastructure, and has prevented its active and orderly recovery. These difficulties have three important aspects: The macroeconomic situation o f the country; 0 The low level and rate o f collection o f water charges; 0 The insufficient decentralization o f the system and the incomplete internalization o f new responsibilities withinthe W A S . ~~ For details see Fiorentino, 2004 and provincial case studies. 113 8.38 Duringthe past years the macroeconomic situation prevented the normal functioning of the financial system, including commercial or institutional credit for irrigated agriculture. Even inthe late 1980's and early 1990's term financing was scarce and expensive. The availability of lines o f credit in the provinces for irrigation. investments varies much across provinces and has typically been very erratic6* Scarce and unpredictable financing, particularly o f term financing, has been inimical to investment in large- and medium-scale irrigation works. The lack o f creditworthiness o f the W A Si s as important as finance availability. The problem o f credit availability combines with weak financial discipline o f the W A S to greatly weaken their capacity to upgrade their infrastructure. Difficulties Related to the Technical Capacity of Water Users 8.39 A large number of water users in virtually all the provinces do not employ known techniques to improve productivity. Inaddition to the problems o f credit availability mentioned above, this difficulty i s linkedto: 0 A shortage o f training opportunities (short courses and workshops) for strengthening water users' capacities; and 0 A shortage o f basic diagnostics concerning the levels o f efficiency o f irrigation and its causes. This inturn is related to the lack o f resources applied to diagnostic research inthe various irrigation areas and systems. Insufficient technical capacity o f water users i s an important determinant o f the low level o f technical efficiency o f irrigation in Argentina. It affects the capacity o f irrigation to create wealth, reduce rural poverty, andbecome environmentally sustainable. Difficulties Related to Research 8.40 Research related to the improvement o f irrigated agriculture has widely divergent outcomes. On the one hand, research and extension activities related to crop technology have yielded excellent results. Yields o f irrigated crops have grown considerably, most noteworthy in sugar cane, tomato, potatoes, vegetables, and irrigated fodder crops. Similarly, quality has improved for wine grapes and tobacco. 8.41 On the other hand, the adoption o fnew methods and systems o f irrigation has been slow. Still more serious, the incorporation o f good practices intraditional technologies, such as furrow irrigation, has been particularly weak. Good practices which have been known for decades are not being used. Excessive gravity irrigation, poor drainage, and low efficiency inthe conveyance and application o fwater are all persistent problems. This is not surprisinginview o f the lack o f 62 For example, since 2003 Neuquen and Santa Cruz are enjoying new rural credit lines which rely o n provincial funds, whereas Formosa and Jujuy have no credit resources. In some provinces, during 2002 and 2003 there was little credit to medium and large farms but yet some funds for small farmers, coming from Programa Social Agropecuario (PSA). Credit availability changed through time. In 1989 bank deposits were around US$4 billion, and there was hardly any credit for farmers. In 1994 deposits rose to US$48 billion and there was commercial farm credit at reasonable rates. Following the crisis o f 1998 interest rates nearly doubled insix months, and again nocredit was available to the rural sector. 114 incentives; water i s cheap or free, the regulatory framework i s poor, and self-reliance and collective action are weak. 8.42 Research applied to irrigation issues i s not receiving sufficient official attention. Both I INTA and the provincial research and development institutions have few professional experts in irrigation. INTA currently has national projects for specific irrigated crops but not for irrigation research, extension, and development per se. Difficulties Related to the Efficiency of Irrigation 8.43 Practically all the systems have low levels o f technical efficiency o f intake, storage, conveyance, and application o f water. According to provincial specialists, water use efficiency (including in-take, conveyance and application) is o f the order o f 15-30 percent in many provinces. It i s noteworthy that provinces do not generally have studies to measure and evaluate with precision their irrigation efficiency. The estimate above is hence only an educated guess o f provincial experts. The main exceptions are the irrigation systems with a highproportion o f lined canals, or those which have developed pressurized irrigation. The irrigation system o f San Juan's Central Valley, where most o f the primary and secondary canals are lined, has a combined efficiency o f conveyance and gravity application o f the order o f 25-40 percent. Efficiency for pressurized irrigation exceeds 85 percent in all the cases analyzed.63 The area, however, under pressurized irrigation, estimated inless than 100,000 hectares, does not exceed 5 percent o f total irrigated area inArgentina. 8.44 Low efficiency o f irrigation i s relatedto the following difficulties: 0 Lack o f economic incentive to maximize the return to water; 0 Poor state o f irrigation infrastructure; 0 Deficiencies inthe technical capacity o f water users, who tend to use excessive quantities o f water, and fail to use techniques o f gravity irrigation proven superior to more traditional methods; 0 Inadequate planning o f the irrigation cycle, which often takes place at excessively long intervals for proper fruit development and maturation. 8.45 The main effects o f low technical efficiency are: 0 Reduction o fthe irrigated surface below its potential: 30-40 percent lower inmany cases; 0 Inadequate applicationo f water to the plant, both involume andinfrequency; 0 Environmental deterioration, linked to soil salinization and uncontrollable increases in groundwater levels (Upper Rio Negro Valley, irrigated areas o f PiranC in Formosa, and many rice production areas inCorrientes). 8.46 Reduction o f the irrigated area and low quality o f irrigation decreases productivity. To the extent that these problems are generalized in systems with predominantly small producers (Catamarca, L a Rioja, Santiago delEstero, Formosa), they tend to increase rural poverty. 63They were irrigation systems of large and medium farms that profited from programs o f fiscal incentives for producers inSan Juan, Catamarca, and La Rioja. 115 SUMMARY OF IRRIGATIONPOTENTIALSAND CONSTRAINTS Potentials 8.47 Additional value and employment from irrigation could be obtained three ways: (i) upgrading PSP practices, product quality, yields and crop mix; (ii) bringing new land under production; and (iii) improving irrigation technology and efficiency. 8.48 An indication of the potential benefits of upgrading product quality is the difference in value o f h i t s and vegetables exports per hectare o f irrigated land in Argentina and Chile discussed inChapter 3. 8.49 In terms of bringing into irrigation additional areas, current estimates indicate that an additional 30-40 percent o f irrigated area could be brought into production with public investments inthe range o f US$lOO-200 per hectare. Additional area on the order o f 20 percent could be irrigated through improved on-farm efficiency from the use o f pressurize irrigation. The main benefit from pressurized irrigation, however, i s not an increase in area as much as (i) improved product quality due to better water control, and (ii) elimination o f water-logging and salinization. Constraints 8.50 The constraints to the full benefits o f irrigation potential are more social than physical. They relate fundamentally to insufficient decentralization o f authority and responsibility to the irrigators, and to the persistence o f a culture o f dependency. This i s aggravated in a relatively few places by low profitability o f agriculture related to small size o f holding and low human capital. Institutional support to irrigation i s weak at all levels. POLICY OPTIONS Overall principles and objectives 8.51 We do not recommend interveninginthe sector except within the context o f a clear, well- articulated strategy. Such a strategy could be designed to devolve authority and responsibility to the W A Sandto create the national incentive framework to support it. 8.52 The long run objective o f such a strategy could be to wean W A Sfrom dependence on government. This would require that they develop a business plan and establish a clear "track record" in meeting the plan's goals. Over the medium term W A Scould aspire to qualify for commercial credit. Any government investment and assistance should take place within the context o f this business plan, with a clear strategy for the water association to develop independence and autonomy. 8.53 Supporting policies could be put inplace at the national level. Inaddition to conditioning transfers and investment on a business plan it would be important to ensure that other 116 institutional impediments to efficient water use are eliminated. In particular, investment in infrastructure development or rehabilitation could be forthcoming for provinces which dismantle impediments to efficient water use, like legislation preventingtransferring water. The advantages oftrading water are discussed inBox 8.1. Elementsfor a strategy 8.54 The central elements o f an irrigation strategy would fall along the following lines: 0 Institutional upgrading o f national, provincial and local institutions; Decentralization o f financial and operationalplanning; 0 Improved management quality; Improved determination water pricing; 0 An emergency programto recuperate critical infrastructure; 0 Increased applied research inirrigation; and Review o fthe legal framework of irrigation andenvironmental control. These elements are discussed inturn, below. 8.55 Upgradingof institutions.This could include: (i) humanresources, better trained, more favoring recruitment o f young people and the hiring of new, qualified personnel; (ii) more coordination among agencies at the national level, and among national, provincial and local agencies; and (iii) involvingmunicipal governments andincluding them intraining programs. 8.56 Decentralizing irrigation planning and management. This could have the following objectives: (i)increasing the number of W A S in all the provinces; (ii)increasing the involvement o f W A S in the areas o f planning and management o f investments; (iii) strengthening the institutional and financial autonomy o f provincial water agencies, and (iv) increasingthe involvement o fmunicipal governments. 8.57 To strengthenthe financial capacity o f WUAs and their capacity to borrow over the short and long term, they would need to gain experience in financial activities; for example, through the formation o f trusts, local investment funds, and the creation o f reserves for the expansion o f irrigated areas. 8.58 Improving irrigation management. This would include the following activities: (i) establishing information management systems at the national, provincial and local levels, (ii) strengthening the management capacity o f W A S ;and (iii) establishing quality control programs at the irrigation system level ineachprovince. 117 Box 8.1: TheRightto Trade Water Water markets are a contentious issue in Argentina. I t is clear that this i s a topic which needs more time to mature. We do not recommendanational policy ofmoving to water markets at this time. We do recommend a national policy to remove impediments to any WUA which might, through a democratic decision process, decide to permit water transfers, sales, or purchases among its associates. We see the following advantages o f this freedom to choose. 1. Leaving current water rights constant (which we would recommend) no irrigator can be made worse off through a water transfer system. He/she could either (1) do not trade water and continue as he/she i s or (2) trade water (ifthe value o f water to hidher neighbor i s greater than to hidher) if he/she feels he/she will be better off by so doing. 2. Without change in the current legislation, many irrigators cannot achieve the full benefits from upgrading to pressurize irrigation technology. Lowering the benefits o f upgrading prevents technological upgrading and the string o f associated improvements-higher value production, greater employment, etc. 3. Decisions regarding the rules for water allocation should be made by the water users, and reflect local conditions. I 4. Water associations are inthe best position to establish the rules o fthe game for water sharing. I 8.59 Determining the cost of water and the water charge. The cost o f water could be calculated, even in those areas which are today completely subsidized and might continue to be subsidized. The calculation should cover all the costs of supplying water, including those related to the improvement and expansion of the irrigation infrastructure. The required calculation o f full water costs would provide the transparency necessary to establish appropriate values for the potential subsidies and to carry out an informed dialogue between the provinces andthe W A S . 8.60 It would be desirable that the above calculation o f irrigation costs be carried out following a standard methodology in all provinces, with due adaptations for specific circumstances. The information obtained could be freely circulated. It i s also important that the establishment of subsidies (that would accrue to certain social groups inthe form o f a water price below the full water cost) be supported by a detailed analytical process. 64 The subsidy scheme could be adequately selective, avoiding subsidies to farmers capable o fpaying, and could decline through time, without failing to accommodate the poorest farmers. 8.61 Emergency program for irrigation infrastructure rehabilitation. The Under Secretariat o f Water Resources and PROSAP are preparing lists o f priority works for each Province. This information could be used to prepare a program o f investments. We note the danger, however o f continuing the pattern o f periodically financing deferred maintenance through emergency rehabilitation projects. As mentioned above, ideally, investment should be made after the institutional framework i s inplace to ensure its future maintenance. 64 Itwould be desirable that inthe long run all water subsidies be removed. The requiredassistance to poor farmers should take a form that increases their economic productivity, not one that simply lowers their costs below the true value o fthe resources employed. 118 8.62 Researchand development.INTA and the provincial universities could help provincial governments and water users to prepare a program o f research that attends to the crucial aspects o f irrigation development. Inparticular, it could include (i) brief but comprehensive provincial diagnoses; (ii) improvements in technical efficiency; (iii) management; (iv) irrigation system development strategies; and (v) the social impact o f irrigation. 8.63 Improvementof the legalframework.This could entail (i) formulation o fa national the legal framework for water resources management, which could establish allocation criteria, operational norms for water management, water quality standards, and protection norms for water sources; and (ii)streamlining and harmonizing provincial norms o f water resources managementjointly with land use environmental norms. Instruments 8.64 Possible instruments to bringabout this strategy are the following: 1. A program o f activities from the national level intended to orient and support provincial and local governments inthe definition of strategies, targets, and objectives. 2. Detailed master plans for irrigation development in each province, specifying areas to develop, populations to attend, and commercial and institutional strategies to increase productivity. 3. A national program to establish information systems oriented at improving irrigation management. This could include activities related to information management for both the provincial andW A Slevels. It could be viewed as a subprogram o f (1) above. 4. Calculation o f water costs for all irrigation schemes. The water price could be fixed below cost only as an exception related to important reasons (rural poverty, emergencies, etc.). Exceptions could be more general inthe beginningto ensure political sustainability. 5. A clear definition and implementation o f a system o f "premies" and "sanciones directed " at promoting full andtimelypayment o f water charges. 119 9. SUMMARY AND CONCLUSIONS 9.1 This chapter summarizes the observations o f the previous chapters and highlights the major implications. It i s organized in two sections. The first summarizes the findings of the individual chapters, and the second synthesizes general policy considerations and makes suggestions for additional work. SUMMARY OF FINDINGS 9.2 Chapters 2 and 3 reviewed the agriculturalperformanceo f Argentina and compared it with other countries-in most cases with Brazil and Chile. They examined the evolution o f agricultural policy and the potential for productivity gains, and compared productivity measures, export market penetration, and ability to move into higher quality markets. The major conclusion from this section was that while Argentina has done well in increasing crop output, its overall agricultural growth has lagged despite evidence o f hightotal factor productivity growth incrops. The point was made that poor factor productivity growth in cattle and slow growth o f agricultural inputs more than offset Argentina's high productivity growth in crops. It was also argued that the loss o f competitiveness due to growing overvaluation in the 1990s was more damaging to the emerging regional economies than to the established commodity economies o f the Pampas, but that both regions were hurt by phytosanitary problems. It was shown that Argentina's irrigated agricultural sector was on track to catch up with Chile in terms o f penetrationo fhighvalue markets, but fell far behindduringthe 1998-2003 crisis. 9.3 Chapter 4 examined the participationof government in the agricultural economy in Argentina. The revision included the analysis o f fiscal flows, o f the roles, strengths and weaknesses o f the two major government agencies for the sector, INTA and SENASA, and o f the situation and possible government action to improve the supply o f infrastructure and rural finance, and to promote collective action in the supply chains. A major conclusion from this chapter i s that government expenditure in agriculture in Argentina is extremely low, proportionally much smaller than that o f other South American countries, including Brazil and Chile, insufficient to meet the sector's requirement o f public goods. Government expenditure bears no relation to the fiscal pressure on the sector, which is high, slightly above the average for other sectors. From each AR$lO that leave the sector as taxes, between 1 and 2 AR$ come back as government expenditure. Agricultural taxation has some distortionary elements, in particular the retenciones and the minimum tax on presumed income. Another major conclusion refers to the inadequacy o f the infrastructure, processing and finance services available to the farm sector, andthe importanceo f government actionto improve the situation. 9.4 The chapter on ruralincomes, povertyandthe labor force addressed the characteristics o f the rural labor force and rural employment, including correlates o f participation in rural non- farm employment, and the dynamics o f earnings inrural areas. It also presented a poverty profile o fthe rural poor, and looked at the statistical evidence concerning possible routes o f escape from poverty. It showed that women participation inthe rural labor force is strong. It also showed that the extremely poor depend more on agricultural incomes than the non-poor, and that the vast 120 majority o f rural labor i s informal. Labor income analysis revealed a large gender gap that increases at the lower quintiles o f the earnings distribution. Employment analysis showed that non-farm employment i s highly correlated with gender, age, access to land, location, and education. Rural women tend to participate inhigh-return and low-return non-farm employment more than men, but men participate proportionally more in high-return occupations. Farm income analysis revealed that farmers' incomes increase monotonically with land size, and are positively correlated with education, road access, and the use o f electricity, fertilizer and irrigation. The chapter also showed that poor rural families tend to be young and large, and are more likely to be small landholders than landless workers. Remittances and other transfers are an important part o f their incomes. Statistical evidence is based on a limited household survey o f dispersed rural households which makes conclusions tentative. 9.5 The chapter on frontier expansion and land use intensification sought to better understand the expansion o f soybeans into the Northwest and Northeast, driven by both new technology and highprofitability, as well as cropping intensification inthe Pampas. The chapter tried to assess the environmental and social consequences o f these recent phenomena. Main points were the following: Soybean intensification in the Pampas generates concern about nutrient depletion andpest buildup.Prior to 1988, the location o fcropland expansion was largely determined by soil class, but between 1988 and 2002 zero tillage made expansion much less sensitive to soil and slope conditions. Transaction costs o f obtaining economic-sized units seem to dominate the current location o f cropland expansion. Therefore, the expansion has occurred primarily in the north, where new cropland has largely replaced natural vegetation, with relatively slight displacement o fpasture and cattle. Although existing data do not provide a clear picture o f the social impact, statistical analysis revealed no impact on m e t basic needs. However, the environmental consequences o f this expansion are worrying. 9.6 The chapter on sanitary and phytosanitary issues and food quality reviewed the progress made in Argentina to meet new challenges o f phytosanitary and quality standards. It briefly described the institutional and regulatory set up, and then presented and discussed several areas where phytosanitary control issues and experiences have been important and left significant lessons. These areas are the fruit fly eradication program, the Curpocupsu control program, the problems experienced with the foot and mouth disease, the phytosanitary restrictions to the export o f lemons to the U S and EU markets, and the threat posed by the soybeans rust. Issues related to the importance o f quality and appropriate institutions for the development o f food industries were then examined and illustrated with the cases o f the differentiation o f wheat qualities, the exploitation o f the export potential o f the honey industry, and the unequal standards inthe beefindustry,The issues posed bybiotechnology andthe use ofgenetically modified crop varieties were also examined, and the positive experience with the establishment o f risk assessment procedures reviewed. The chapter concluded with some reflections and policy options based on the examination o f the above issues and experiences. Some o f the aspects important for policy action highlightedwere: the convenience o f an integrated approach to food quality and safety; the high potential returns o f investments on food quality and safety; the importance o f establishing effective partnerships and a clear definition o f responsibilities between national and provincial governments and between them and the private sector; and the usefulness o fpromoting effective collective action. 121 9.7 The chapter on irrigation discussed the performance o f the irrigation subsector. It presented evidence o f week performance--underutilization o f irrigation systems, low technical efficiency, poor irrigation techniques, and inadequate water charges. Along with a diagnosis o f sector problems and potentials, the main points were: (i) institution weakness at the national, provincial and local level i s a major reason for poor performance; (ii) provincial water laws are often an impediment to the adoption o f improved technology and efficient water use; (iii) greater decentralization o f water management and decision making to water users associations i s a top priority; and (iv) soft budget constraints prevent water associations from assuming (management andfinancial) responsibility. GENERAL POLICYCONSIDERATIONS AND SUGGESTIONSFORADDITIONAL WORK 9.8 Public action i s essential to encourage further development o f the rural economy in Argentina. Some general policy considerations are as follows: 0 There is evidence o f underinvestment inagricultural public goods inArgentina including agricultural research and extension, sanitary and phytosanitary systems, investments to improve export market participation, infrastructure, and promotion o f financial services. Progressively raising public expenditure in agriculture and rural development to levels similar to those of its closest L A C competitors would do muchto increase the competitive edge o f the sector. This has already started over the last two years, and we expect it to continue. 0 Mitigating rural poverty would require a combination o f direct and indirect employment generation, particularly inthe regional economies, with improved access to and quality o f education and other services. Investing in irrigation, sanitary and phytosanitary systems, and research and extension for small farmers would be an effective way to help improving poverty conditions. 0 Collective action i s increasingly critical in modem globalized agriculture, and i s particularly significant for the development o f the regional economies. It requires (i) resolving conflicts in the value chains to allow producers, suppliers, and buyers to work together to achieve and retain high value markets; (ii) creating and maintaining systems of vigilance and enforcement o f phytosanitary protection norms; and (iii) decentralized local management o f irrigation systems. 9.9 Inthe course of preparing this review, we have come across several important topics on which there is lack o f information and analysis. This gives thempriority for study. Four priorities areas are: Additional work on agricultural taxation and non-fiscal transfers. Although this study contains a review o f fiscal transfers to and from agricultural (Background Paper 2), we recommend a more fully-elaborated treatment o f both fiscal and implicit transfers to andfrom agriculture, including a more complete analysis o fthe issues o f export taxes, tax evasion inthe sector, and local andprovincial tax effort. This analysis needs to be carried out within the context o f Argentina's overall tax system and fiscal equilibrium, and 122 address the issue of regional inequalities and the large regional variation ineconomic rent to land. 0 Risk and credit markets. Lack o f credit is a major barrier for adoption o f improved technology inthe small and mediumfarm sector. Markets for hedging price and exchange rate risk and crop insurance, appear to be readily available to large farmers but not to the small ones. These markets need to be better understood, especially in the context o f improving the competitiveness o f small and medium size farms. 0 The livestock sector. The potential to increase beef production in the Pampas and outside the Pampas i s high. This i s because much o f the sector operates at medium and low technology levels, and the output gap between productivity levels in cattle production i s high-much higher than in crops. Raising productivity i s important because the sector is large, and a moderate raise in productivity would result therefore in a large increase in aggregate output. Also, it i s inthe smaller farms where the productivity gaps are largest. Further study o f the reasons for this sector's relative stagnation, compared to the grain sectors, is warranted. 0 Rural growth and employment creation. Recent Argentine agricultural growth, especially in the Pampas, has not led to significant employment generation due to its relatively high capital intensive and land extensive nature. The potential for higher rates o f employment creation in the regional economies is higher. Additional analysis o f the possible impact on employment o f the agricultural growth under various scenarios would help inform the debate on rural poverty reduction, the role o f agriculture, and the importance o f developing either non-farm employment or near-by growth centers in towns and cities. 9.10 In conclusion, this study will help guide future World Bank dialogue and financing of agriculture and rural development in Argentina, particularly with regard to areas where the role o f government may be strengthened or the provision o f public goods increased. The potential for agricultural development in Argentina i s impressive. A strategic support from government to complement the capacity and dynamism o f Argentina's farmers could help much to realize that potential and make agricultural growth more equitable. With the present study, conceived as an instrument for dialogue, the World Bank hopes to contribute to that objective. 123 BACKGROUNDPAPERS This report is largely a synthesis o f background papers prepared by country consultants and World Bank staff. Complete background papers are available on the website (www.bancomundiaLorg.ar). Interested readers are referred to these much longer papers. As shown below, background papers were commissioned in 7 areas: (i)sanitary and phytosanitary control (including worker and environmental safety inpesticide use); (ii) irrigation and its contribution to the regional economies; (iii) new mechanisms for combining land, the labor, capital, and management in the Pampas (planting "pools"); (iv) the environmental and social effects o f the expansion o f the crops frontier; (v) issues related to market access for Argentina's agricultural exports; (vi) the role o f taxation and government expenditure in the agricultural sector; and (vii) rural poverty. World Bank-CommissionedBackgroundPapers Topic Author Cases studies Changingpatterns o f agricultural Roxana Bertolassi (1) Junin, Buenos Aires organization (2) Las Lajitas, Salta Taxes and government expenditure in Emesto O'Connor (1) Buenos Aires agriculture Jorge Cabellero (2) Mendoza (3) SanJuan (4) Catamarca Rural Labor Force, Incomes and Dorte Verner -_ Poverty Spatial patterns o f frontier expansion LARTFAUBA (1) Northeast Salta (Tartagal), (Federico delPino, (2) Southeast o f Salta (Las Lajitas), coordinator) (3) East Santiago del Estero/West Chaco (Charata) (4) Southeast o f Santiago del Estero andNortheast o f Santa Fe (Bandera) Irrigated agriculture and its Raul Fiorentino (1) Mendoza contribution to the regional (2j Catamarca economies (3) Rio Negro (4) Jujv (5) San Juan Market access Yurie Tanimichi Phytosanitary institutions, Pesticide Miguel Huerga (1) Carpocapsa inRioNegro safety Sebastiin San (2) Asiatic Soybean rust, Juan (3) Horti-fruiticulture inMendoza (4) Horti-hiticulture inLules and Tafi del Valle (5) Tobacco inMisiones Foodquality and sanitary and CarmenVincien (1) Honey phytosanitary institutions. (2) Wheat market differentiation (3) BeefinBuenos Aires (4) Biosafety Impact o f international phytosanitary MorvenMcLean (1) Beef requirements (2) Lemon (Tucudn) 124 REFERENCES Araujo, C., A.de Janvry and E. Sadoulet (2003). Measuring the role o f social networks on behavior with an application to rural off-farm employment. Department o f Agricultural and ResourceEconomics Universityof California Berkeley.Mimeo. Avila D. A. and R. Evenson (2004). Total Factor Productivity Growth inAgriculture: The Role of Technological Capital. Paperpublishedby the Economic GrowthCenter, Yale University. Bertolassi (2004) Contract farming inArgentina. Consultancy for this report. Annex I. Bisang R. (2003). Eficiencia y redes productivas: la industria de las carnes en Argentina. Instituto de Industria. UniversidadNacional de General Sarmiento, 2003. Bravo-Ortego, Lederman (2004). Agricultural productivity and its determinants: revisting international experiences. InEstudios de Economia Vol. 31N"2. Brooks J. and Lucatelli S. (2004) International competitiveness o f the A-B-C agro-food sector, Chapter 4 in"Trade and competitivenessinArgentina, Brazil, and Chile: Not as easy as A-B- CyOECD. Burachik,M.and Trynor, P (2002).Analysis of aNationalBiosafety System:Regulatory Policies andProcedures inArgentina" ISNARCountry Report 63. ISNAR.The Hague. Byerlee D. and Hesse de Polanco (1982) La tasa y la secuencia de adopci6n de tecnologias cerealeras mejoradas: el cas0 de la cebada de secano en el altiplano mexicano. Working paper no. 82/6. CIMMYT. Caballero, J. C. and O'Conner (2004) Estudio Fiscal - AnSlisis comparativo de la tributacibn y el gasto p~blicoagropecuario. Annex 11. Part 1. FA0 Investment Center consultancy for this report. Caballero J. M.(2005) "Rural poverty inMexico" .WorldBank, mimeo. Cap E. J. and GonzSlez P. (2004) La adopcion de tecnologia y la optimizacih de sy gestion como fuente de crecimiento de la economia. INTA. Cassman, K.G., 1999. Ecological intensification of cereal production systems: Yield potential, soil quality, andprecision agriculture. Proc. Nat. Acad. Sci. 96, 5952-5959. Cavallo, D. y Y. Mundlak. "Agriculture and economic growth in an open economy; the case of Argentina"; IFPRI and IIERAL; Washington D.C. and Cordoba, Argentina. November 1982. Chomitz K.M., Buys P., and T.S. Thomas (2004) Quantifying the Rural-Urban Gradient inLatin America and the Caribbean. Background paper prepared for de Ferranti et a1 (2005). The World Bank. Washington DC. 125 Correa Melo, E., Saraiva, V. & Astudillo, V. (2002). Review of the status of food and mouth disease in countries of South America and approaches to control and eradication. Rev. sci. tech. Off. Int.Epiz. 21(3): 429-436. Cuniberti M. (2003): Propuesta de Clasificacibn del Trigo Argentino y Programa Nacional de Calidad. Available on line at httu://www.elsitioa~cola.com/~acetillas/marcosiuarez/mi2003070/trigo.asD 1 De Ferranti D., Perry G.E, Foster, W., LedermanD., andA. ValdCs (2004) Beyond the City: the Rural Contribution to Development. World bank Latin American and Caribbean Studies, The World Bank, Washington DC. D e l Pino, F. (Coord.). "Spatial patterns of agricultural frontier expansion in Argentina". LART- Facultad de Agronomia -Universidad de Buenos Aires; Buenos Aires. 2005. Fundaci6n Sofia. Analisis del presupuesto en Argentina; annonia y conflicto. Buenos Aires. 2001* Ekboir J. and G. Parellada (2002) Public-private interactions and technology policy in innovation processes for zero tillage inArgentina, in Byrlee D. and R. Echeveda (editors) Agricultural researchinan Era of Privatization, CABI, New York. ERS-USDA (1998) Argentina's Economic Reforms Expand Growth Potential for Agriculture http://www.ers.usda.gov/publications/agoutlooMmar1998Ia0249f~~df FA0 (1996). Proyecto agricola provincial-Resumendel subsector riego. Documento de trabajo No.1;Roma. FA0 (2000) Agricultural production and productivity in developing countries, in The State of Food and Agriculture. Available on line at ttp://www. fao.org/documents/show_cdr.asp?url_file=/docrep/x44OOe/x44OOeOO.htm) FAOSTATS (2005) Available on line at http://faostat.fao.org/ FAS (2004) Argentina Citrus Annual 2004. GAIN Report No. AR3048. Foreign Agricultural Service, Washington DC, 2004 Ferreira, F. and P. Lanjouw (2001). Rural Poverty and Nonfarm Employment in Brazil. World Development, 29/3, pp. 509-528. Finan, F., E. Sadoulet, and A. de Janvry (2002). Measuring the Poverty Reduction Potential of LandinRuralMexico, Universityo f Californiaat Berkeley.Mimeo. Fiorentino R. (2004) La Agricultura Irrigada en Argentina y su Contribucih A1 Desarrollo de las Economias Regionales. Consultancy for this report. 126 Gacitua Mario E., C. Sianes and Q. Wodon (2001). Reproductive Health in Argentina's poor rural area". InMeasuring and measurement: Combining qualitative and quantitative methods for the analysis ofpoverty and social exclusion inLA. Washington, DC: World Bank. Ghezan G., Mateos M.y Viteri L., (2002). Impact of Supermarkets and FastFood Chains on Horticulture Supply Chains in Argentina: Challenges and Opportunities for Development in Development PolicyReview Vol. 20 Issue. Giarraca, N and C. Grass (2000). El Mercado De Trabajo Caiiero En Tucumh. Vinculos Laborales y Caracteristicas Demograficas de la Poblaci6n Asalariada. Asociacibn Latinoamericana de Estudios del Trabajo I11Congreso Latinoamericano de Sociologia del Trabajo. BuenosAires 17 a120 de Mayo de 2000. Giarraca, N.,S. Aparicio and C. Gras (1999) `Los Campesinos Argentinos: Estrategias Sociales en una Agricultura en Transformacih. Multiocupacionen Pequefios Productores de Caiia de Azucar', (Unpublished paper), Grupo de Estudios Rurales (GER). Buenos Aires: UBA / FONCyT. GutmanG. (2003): "Estudio de base para la implementacibn de unsistema de certificacih de la calidad y la sanidadde laproduccion agricola". Manuscript. FAO, 2003. Gutman G. and Lavarello P. (2003). El complejo trigo. Elementos de diagnostic0 para la implementacih de politicas de promoci6nde la calidad en la Provincia de Buenos. Gutman G. (2002) Impact of the rapid rise of supermarkets on dairy products systems in Argentina. InDevelopment Policy review Vol20 N04. Hicks, N. (2000) Poor People in a Rich Country. A Poverty Report for Argentina, World Bank Report No19992AR, Vol 1.The World Bank: Washington DC. Huerga M. y San Juan S. (2004) El Control de las Plagas en la Agricultura Argentina. Consultancy for this report. Annex VII. IICA (2004) Argentina en Cifras. Available on line at www.iica.org;.ar INDEC (1988) Agricultural Census. Available on line indec.gov.ar INDEC (1991) Households andPopulation Census. Available on line indec.gov.ar INDEC (2001) Households andPopulation Census. Available on line indec.gov.ar INDEC (2002) Agricultural Census. Available on line at indec.gov.ar ISAAA. Enhttp:llwww.isaaa.ord 127 Lanjouw, J. 0.and P. Lanjouw (2001). The Rural Nonfarm Sector: Issues And Evidence From Developing Countries. InAgricultural Economics 24, pp.1-23. Lattuada (2000). Ponencia presentada en las X Jornadas de la Asociacion Argentina de Extension Rural, Mendoza 18 - 20 dejunio de 2000. Levy, F. and R. J. Murnane(l992). USA Earnings Levels and Earnings Inequality: A Review of Recent Trends and Proposed Explanations. InJournal o f Economic Literature, Vol. 30, pp. 1333-1381. Llach J.J., H d a g u e M. and O'Connor E. (2004) La generacion de empleo en las cadenas agroindustriales. Fundacion Producir Conservando, 2004. Lbpez, Ramon (2004) Why Governments Should Stop Non-Social Subsidies: Measuring the Consequences for Rural Latin America. Background paper for the World Bank 2005 Flagship Beyond the city: the rural contribution to development, Washington DC: The World Bank. Mallon, R. and Sourrouille, J. "Economic policy making in a conflicting society". Harvard UniversityPress; Cambridge, Mass. 1975. McLean M. (2004) The Impact o f International Phytosanitary Requirements on the Lemon IndustryinTucumh. Annex IX. McLean M. (2004) The Impact of International Sanitary Requirements on the Beef Industry in Buenos Aires. Annex IX. Mincer, J. (1974). Schooling, Experience andEarnings. NBER Working Paper, New York. Morello, J. and Adimoli, J. (1974). Las grandes unidades de vegetacih y ambiente del Chaco argentino. Segunda parte: vegetation y ambiente de la Provincia del Chaco. Secretaria de Estado de Agricultura y Ganaderiade la Nacibn. INTA, Serie Fitogeogrifica No13. Buenos Aires. 130pp. Mundlak and Regunaga (2001) Economic Aspects of Agricultural Growth inArgentina, Center for Agricultural Economic Research, Working PaperNo. 20104. Mundlak, D.; Cavallo, D. and Domenech, R. "Agriculture and economic growth in Argentina; 1913-1984. IFPRI - ResearchReport No.76. Washington D.C. November 1989. Mundlak, Y. andReghaga, M. "Economic aspects of agricultural growth inArgentina". Centre for Agricultural andEconomic Research. Working Paper No. 2104. Washington D.C. MunnisMiguel (1996) Pobrezaruraly ocupacion: revision de algunos datos inkditos. InRevista estudios deltrabajo N12. 128 Mwabu, German0 and T. Paul, Schultz (1996). Education Returns Across Quantiles o f the Wage Function: Alternative Explanations for Returns to Education by Race in South Africa. American Economic Review, vol. 86 (2), pp. 335-339. O'Connor E. (2004) Estudio Fiscal - Analisis Comparativo de la Tributacion y el Gasto Phblico Agropecuario en la Naci6n y en Cuatro Casos Provinciales: Buenos Aires, Mendoza, San Juany Catamarca. Consultancy for this report. Annex 11part 2 Pesce. S. "Analisis y evaluacion de la infraestructura granaria de Entre Rios y su contribution a la competitividad". Tesis de Maestria en Negocios. Universidad de Belgrano; Buenos Aires; Septiembre de 2004. PHEFA (1999).Report on the Situation o f the National FMD Eradication Programs in South America. PanAmerican Food-and-Mouth Disease Center, 40 pp. Randall, D. et a1 (2001). Trade Economics Division, Economic Research Service Agriculture in Brazil andArgentina: Developments and FieldCrops. Re, Agriculture, Agriculture and Trade Report. WRS-01-3. Reardon, T. and Berdeguk A. (2002). The rapid rise o f supermarket in Latin America: Challenges and opportunities for development. InDevelopment PolicyReview Vol20 N"4. Reardon T. J., BerdeguC A. and Escobar G. (2001). "Rural Nonfarm Employment and Incomes inLatinAmerica: Overview andPolicyImplications". WorldDevelopment 29(3). Reca, A. (2002). Foot and Mouth Disease in Argentina: turning a crisis into an opportunity. Industry Note 040-2002. Rabobank Buenos Aires. Reca, L.andParellada, G. "El sector agropecuario argentino". Editorial Facultadde Agronomia. Universidad de Buenos Aires. Buenos Aires. 2001. Roccatagliata J. (1992) Regionalizacion. In L a Argentina. Geografia regional y 10s marcos regionales, Planeta: Buenos Aires: Argentina. Rofinan A. (1999) Modernizacion productiva y exclusion social en las economias regionales. economias regionales. InRealidad Economics. Vol. 162 pp107-136. SAGPyA (2004) Escenarios agricolas para el Bicentenario de la Independencia Nacional ("Agricultural Scenarios for the Bicentennial o f National Independence"). Office o f Economic Analysis, Bureau o fAgricultural Economics. Schnepf D, Erik Dohlman, and Christine Bolling (2001) Agriculture in Brazil and Argentina: Developments and Prospects for Major Field Crops. USDA Economic Research Service Agriculture and Trade Report No. WRS013, December 2001 Available on line at tp:/www.ersda. gov/publications/wrsO 131 129 Schneider R. (2004). Background paper for the World Bank flagship 2005 Beyond the city: the rural contribution to development. Charts available in chapter 3 o f the Flagship. Washington DC: The World Bank. SENASA (2004) "Plan estratkgico para el cambio y fortalecimiento del SENASA 2004". Available on line at www.senasa.gov.ar. SENASA (2004) Cambio y Fortalecimiento del SENASA. Lineamientos estratkgicos. Available on line at www.senasa.gov.ar. ShenggenFan, IFPRI(2005) (pers. communication) updating Fanand Rao, 2003. Slafer, G.A., Kernich, G. (1996). Have changes in yield (1900-1992) been accompanied by a decreased yield stability inAustralian cerealproduction? Aust. J. Agric. Res. 478,323-334. Smitsaar, E., N. Mattion,, G. Mazzuca, B. Robiolo., E. Maradei, J.Filippi., A. Sadir, , A. Falczuk, J. La Torre, A. Pedemonte, R. D'aloia, 0.Periolo, G. Cadenzaai, E. Palma. & R. Bellinzoni (2002). Foot and Mouth Disease in Argentina: development o f vaccines for emergency control and eradication o fthe disease. Spreen, T.H. (2001). Projections o f World Production and Consumption o f Citrus to 2010. Proceedings of the China/FAO Citrus Symposium, FAO, Rome. Sturzennegeret. al. (1990) Trade, ExchangeRate, andAgricultural Pricing Policies inArgentina, Sturzennegere.t al. World Bank Comparative Studies. Tapella E. (2001)Globalization and the agrarian question in Argentina: I s there any chance for the peasants to survive? In Gallina E. (editor) Regional and National Responses To Globalization' Federico Cafe Centre Publisher, Roskilde University - Dept. of Social Sciences. Trigo, E., Chudnovsky, D., Cap, E., L6pez, A. (2002). Los Transgknicos en la agricultura argentina. Unahistoria con final abierto. Libros ElZorzal. Buenos Aires. Tanimichi Y. (2004) Internationalmarketrisks and agricultural trade related issues. Annex VI. Tuan F, C. Fang, Z. Cao ( 2004) China's soybean imports expected to grow despite shirt-term disruptions, Electronic Outlook Report OCS-04J-01.USDA-ERS. Universidad Nacional del Sur. "Diagn6stico econ6mico de seis municipios del sureste de la Provincia de BuenosAires. BahiaBlanca. Octubre de 2002. Valdes, A. "Comentarios sobre el Informe del Banco Mundial Argentina - Rural Development Review (Febrero de 2005). Santiago de Chile, Septiembrede 2005. 130 Van der Mensbrugghe D. (2002) Quantifying Trade Policy Options for Argentina, Washington DC. Vemer D (20041,) Rural Poverty in Mexico During 1992-2002, The World Bank, Washington, DC. Mimeo. Vemer D.( 2004) Ruralpoverty andlabor markets inArgentina. Paper for this report. Annex I11 World Bank (2000). Argentina, gestion de recursos hidricos. Inf.20729 AR, Washington D.C. World Bank 2004 "Rural Infrastructure inArgentina: Its Challenges, Key Issues and Options for its Development" Report No. 26271. World Bank :Washington, DC. World Bank (Brazil Country Management Unit) (2004) Brazil Trade Policies to Improve Efficiency, Increase Growth and Reduce Poverty, Report Number 24285-BRYWashington DC Zapater de del CastilloM. (1985) Esquema fitogeogrifico de la Provincia de Salta. Secretaria de Estado de Asuntos Agrarios, Direction General Agropecuaria. 131 ANNEX I:REGIONALOVERVIEW Rural Argentina i s extremely heterogeneous, in terms o f both resource endowments and the quality o f life o f its rural residents. It i s divided in five economic and geographical regions: the Pampas, North East (NEA), North West (NOA), Cuyo and P a t a g ~ n i a .All~o f these regions but ~ the Pampas are usually called "regional economies". While the Pampas produces mainly commodities, largely for the foreign market, the regional economies have grown in response to internal demand and state efforts to promote agricultural development. Only inthe 1990s did the regional economies begin to significantly develop agricultural exports. Agriculture inthe Pampas is generally modem and dynamic, while the regional economies (with some important exceptions) tend to bemore traditional andto function farther from the technology frontier. As discussed in Chapter 2, a combination o f rapid expansion o f credit and an abrupt collapse o f internal demand left many farmers with serious economic problems. In addition, adoption o f labor-saving technology (for example mechanical cotton-pickers) accelerated rural dwellers' migration to the urban areas (Rofman, 1999). Between 1991 and 2001, the rural population decreased by 8.4 percent. The decrease was higher (14.5 percent) in the dispersed rural areas, where 68 percent o f the total rural people live than in the grouped rural areas66. From 1988 to 2001, the number o f farms decreased by 18 percent in Cuyo, 13 percent inthe NEA, 16 percent inPatagoniaand7 percent inthe NOA. (Table A1.1) The following paragraphs describe ingreater detail the main agricultural activities and feature in each o f the regions, and discuss specific regional issues. Tables A1.2 and A1.3 summarize this discussion. MAIN AGRICULTURAL ACTIVITIES INTHE REGIONAL ECONOMIES AND THE PAMPAS TheNortheast The major traditional agricultural activity o f the Northeast region is the production of yerba mate, cotton and tea, especially in the provinces o f Corrientes and Misiones. Soybeans have made important advances, especially into the province o f Chaco, and have become the major crop, currently representing some 35 percent o f total cropland. 65 There are several criterion to divide Argentina and this political division is one o f the most common one (Rocctagliata, 1992; Rofman, 1999).The Pampeanaregion comprises Buenos Aires, La Pampa, Cordoba, Santa Fe and Entre Nos. NEA i s composed by Corrientes, Chaco, Formosa and Misiones. NOA by Catamarca, Jujuy, Santiago del Estero, Salta, La Rioja and Tucumhn. The provinces o f San Juan, San Luis and Mendoza make up the Cuyo region. And the Patagonia regioncomprises Chubut, Neuqukn, Rio Negro, Santa Cruz and Tierra del Fuego. 66InArgentina, according to the National Institute of Statistics (INDEC), a person is considered to live ina rural area ifhe/she lives indispersedareas inthe open countryside, where households are not linkedthrough urban streets (dispersed rural areas), or in urban settlements with less than 2000 habitants (grouped rural areas). About 68 percent of rural dwellers live indispersed rural areas inthe four provinces included inthe sample o f the WB Rural Household Survey. 132 Table Al.1: Indicators of Quality of Life by Province and Region YOchange in Of Unsatisfied farms 1988- % Population change % School attendance Ag GDP basic needs 2002 1991-2001 15-1 7yrs. old 2001 1991- 2001 2001 Grouped Dispersed Grouped Dispersed Rural Buenos Aires -32 12 -30 80 72 1.6 14 L a Pampa -10 -7 -29 78 56 5.8 15 Cordoba -36 8 -22 69 51 2.3 21 Santa Fe -24 2 -21 76 61 2.3 21 Entre Rios -20 18 -19 67 57 3.5 26 Total Pampeana -32 7 -25 75 61 __ 19 Mendoza -15 32 0 70 57 2.3 27 San Juan -23 -16 -17 67 57 2.3 31 San Luis -29 8 -27 69 49 2.5 31 Total Cuyo -18 9 -5 69 56 -- 30 Catamarca -4 14 -1 78 63 2.2 33 Jujuy -5 20 -14 68 48 2.2 48 Salta 16 32 -13 72 56 3.3 53 Santiago del Estero -6 9 2 57 28 2.1 49 Tucumhn -40 17 1 58 42 0.7 40 L a Rioja 9 -12 -4 76 68 3.6 32 Total Northwest -13 14 -3 68 42 -- 43 Corrientes -34 -1 -8 67 43 1.5 46 Chaco -11 13 -29 62 35 1.2 52 Formosa -6 -10 -17 74 46 2.8 51 Misiones -2 -4 -3 62 38 2.4 36 Total Northeast -13 0 -14 65 39 __ 46 Santa Cruz -14 -57 -33 87 67 4.1 12 Chubut 3 12 -12 75 61 3.5 27 NeuquCn -13 13 -3 73 51 2.8 33 Rio Negro -9 1 -24 69 61 4.7 27 Tierra del Fuego 14 159 8 92 87 3.2 21 Total Patagonia -7 2 -16 73 58 -- 24 Total Argentina -22 8 -15 71 49 2.3 Source: PopulationCensus 1991 and 2001. Note *: Employment rate = employed population/Totalpopulation older than 14years old. AARG = average annualrate of growth 133 Soybeans expansion into Chaco has beenmostly at the expense o f natural vegetation. Analysis o f the expansion o f soybeans into the 50 departments in Chaco and Santiago del Estero with the most active expansion o fthe soybeans frontier, found that 80 percent o f the expansion took place in areas previously covered by natural vegetation, and only 13 percent in areas previously covered by crops. This suggests that the expansion o f soybeans i s not likely to have been associated with a significant net loss o f agriculturaljobs inthe region, but is, on the other hand, having a major impact on biodiversity. About 6.4 percent o f total cropland in the region is irrigated. This reaches a maximum of 20.5 percent inCorrientes and a minimumo f 1.2 percent in Chaco. Rice i s the major irrigated crop in the Northeast, accounting for some 50 percent o f irrigated cropland, followed by horticulture (approximately a third) and citrus (17 percent). Horticulture i s largely practiced by small farmers, growing mainly tomatoes, pepper, onions and lettuce for the domestic markets. InMisiones and Corrientes, the major irrigated crops are rice and citrus, both produced for internal and external markets. Yerba mate is irrigated in Corrientes, where a single firm dominates nearly the entire Mate production (inMisiones, small farmers involved inyerba mate do not irrigate). Citrus and bananas are irrigated by small farmers (less than 10 ha) and medium sized farmers (10-100 ha). In Formosa, bananas are a main crop; even though the area cultivated has been decreasing throughout the nineties, Formosa accounts for 52 percent o f the total area plantedwith bananas inArgentina. Almost all areirrigated. In Chaco, the main crops are soy, wheat, sunflower and cotton, none of which are irrigated. With the exception o f cotton, these crops are produced mainly for export. Wheat and cotton are also important inFormosa. Small farmers constitute the majority o f producers in the Northeast. While 72.5 percent o f all farms are smaller than 100 hectares, they account for only 7.2 percent o f all farm land. Farms between 100 and 1,000 hectares account for 21.6 o f all farms but for only 24 percent o f all farm land. Finally, farms larger than 1,000 hectares represent only 5.8 o ftotal farms, but they account for 68.4 percent o f all farm land. The reduced area in cotton, combined with mechanization o f cotton harvesting, has caused a severe fall o f rural employment opportunities in Chaco and to a lesser extent in Formosa and Corrientes. Inthese provinces small farmers inthe cotton sector have historically combined own production and wage employment. With the introduction o f the mechanical cotton picker, labor demand decreased by 50 percent. As a result, unemployed workers migrated to the urbanareas and/or neighbor cities, where poverty rates have increased dramatically (Rohan, 1999). In 2003, Chaco had the highest rural indigence and poverty rates 46.7 and 65.3, respectively (measured through income). The number o f dispersed rural dwellers in the Chaco decreased by 29.3 percent between 1991 and 2001, while the number of farms decreased by 34 percent (between 1988 and 2001). Population outflow from dispersed rural areas in Chaco i s exceeded only by Buenos Aires and Santa Cruz. However, unlike Buenos Aires and Santa Cruz, where 72 and 67 percent respectively o f 15-17 year olds are in school in dispersed rural areas, in Chaco 136 only 35 percent continue. Clearly the readiness for out-migrants to embark on an urban livelihood i s very different inthe two cases. We suspect that outmigration from rural areas inthe Chaco is caused by loss o f off-farm employment for small farmers who have historically combined own production with wage employment, and represents serious deprivation. Under this hypothesis the growth in grouped rural population is likely to be pushed by loss of rural opportunity in dispersed areas, not a pull from growing non-farm employment in urban and grouped areas.67 We note that the rate o f rural out migration is relatively low inCorrientes and Misiones (7.8 and 3.3 percent respectively), despite highrates o f poverty and unsatisfied basic needs. Corrientes is among the provinces with the highest rate o f farm consolidation, with the number o f EAPs with defined limits falling 34 percent. Both provinces are among the small group o f 7 that exhibited growth in agricultural-based manufactures relative to their pre-crisis peak, suggesting potential growth innon-farm employment. Growth ina very competitive wood products industrymay play a role. However, given the low rate o f growth o f agricultural GDP (especially in Misiones), the low education levels, and the lack o f significant activity inthe land markets, one cannot discard the hypothesis that small farmers are trapped inpoverty bythe lack o f assetsrequired to seek off- farm alternatives. TheNorthwest Inthe Northwest regionthe main crops are sugar cane, soy and citrus. Sugar cane is cultivated in Jujuy, Salta and, most importantly, in Tucuman; total surface with sugar cane accounts for 268,450 hectares with most production consumed internally. Soy i s grown in all the provinces except L a Rioja. Tobacco is important in Salta, Jujuy and Tucumhn (as well as inCorrientes and Misiones inNOA). Irrigated area inthe northwest accounts for 17.8 percent o f total farm land. InJujuy 61 percent o f cropland i s irrigated, followed by L a Rioja, Tucumin and Catamarca with 56,25 and 23 percent irrigated, respectively. The provinces with the least irrigation in the Northwest are Santiago del Estero and Salta, with 8 and 5 percent o f their cropland irrigated, respectively. Sugar cane accounts for over a quarter o f the irrigated land in the region. InJujuy, sugar cane accounts for about 63.4 percent o f total irrigated area and in Tucumhn for about 55 percent. Tobacco is important in Jujuy and Salta, and for the region as a whole represents nearly 8 percent o f irrigated area. InTucuman, there are about 34,600 hectares ofcitrus production, ahighproportion ofwhich are lemons for the export market (see Chapter VII), also produced in Salta and Jujuy. InCatamarca and L a Rioja, olive production is increasing rapidly and grapes are also important. Olives are exported both as olive oil and as table olives. Olive production i s mostly in the hands o f large producers (with a significant boost from fiscal incentive). Small producers are involved in horticulture (Le. tomatoes, peppers, pumpkins and onions) for the domestic markets. In Catamarca, main irrigated crops are olives, nuts and grapes with 16, 11 and 9 percent o f total irrigated farm land. In addition small farmers growinghorticulture crops use about 13 percent o f 67We note that Chaco is among the provinces where agricultural-based manufactures responded well following the 1991 reforms, but which has stagnated since 1994. 137 total irrigated farm land. Similarly, in L a Rioja, olives and grapes account for 56 and 22 percent o f total irrigated farm land. InSalta, tobacco and fodder account for the highest share o f irrigated farm land, with 11 and 9 percent o f irrigated farm land, respectively. The rest o f the irrigated land i s divided between fruits and horticulture; among .irrigated farm land with fruits, citrus accounts for 50 percent. In Santiago del Estero, the main irrigated crops are fodder and cotton with 44 and41percent o ftotal irrigated land, respectively. Major non-irrigated crops are soybeans (50 percent o f non irrigated cropland), wheat, beans (Jujuy and Salta), and sorghum (Santiago del Estero). Soy production, a part o f wheat production and a high share o f tobacco production are exported. Wheat, sorghum and beans receive supplementary ground water irrigation by large farmers. As in the NEA, in the NOA the majority o f farms are small. Farms smaller than 100 hectares account for 76.1 percent o f all farms, and own 3.8 percent o f total farm land. Farms lager than 1,000 hectares account for 6 percent o f total farms, but for 76.3 percent o f total farm land. Small farmers inNEA have had difficulty adjusting to the post-reform environment. Weak demand in domestic markets, lack o f credit, and low management capacity combine to accelerate the loss o f small farms. In the sugar industry, only large farmers have been able to compete with imports from Brazil, and the rest have abandoned the activity. InTucumh, between 1988 and 2001 the number o f farms has decreased by 40 percent (INDEC, 2001). Giarraca et a1 (1999) estimated that inthe period 1988-1996 the number o f sugar-cane minifundistas (caieros) decreasedby 31.5 percent and that the proportion o f temporary off-farm labor greatly increased relative to permanent employment. Giarraca and Grass (2000) report that sugar workers in Tucuman involved in technical, organizational, or managerial roles generally have social benefits and fixed term appointments. Those involved inharvesting or ingeneral collection activities and young people (under 25 years old) have precarious occupations. However, older workers tend to have formal labor status and better social benefits. The authors point out the general poor living conditions o f all these workers, but highlight a difference between those living in their houses in marginal neighborhoods and those who live inencampments, because they have migrated from other parts o fthe province or from other provinces. In Jujuy, where sugar production is based in large farms, the re-orientation of the production process resulted inhighunemployment and accelerated migration to urban areas. Between 1991 and 2001 rural dispersed population decreasedby 13.5 percent. By indicators o f poverty, basic needs, and educational levels the northwest contains some o f Argentina's poorest provinces. Although only 6 percent o f the region's cropland is irrigated, irrigated agricultural is very important in several provinces, with Tucuman being a world leader in lemon exports. The soy frontier has expanded rapidly in Salta and Santiago del Estero. As discussed above, in the Chaco-Santiago expansion, only some 13 percent o f soybeans replaces previously cropped land, indicating a high likelihood o f a net increase in agricultural employment. In Salta, the soybean area replacing other crops increases to 24 percent. Nevertheless, analysis o f population change in the 6 departments in the Chaco ecosystem experiencing the most rapid soybean expansion, revealed a 14percent rate o fpopulation increase 138 in these departments over the 1991-2001 period--suggesting a modest but positive net employment effect o fthe soybeans expansion inthis area. Consolidation o f small farms in sugar cane production has been rapid, due in part to increased price pressure from Brazilian imports, especially in Tucumih, where the number o f farms decreased by 40 percent. Currently available data raise important questions concerning the poverty dynamics o f the Northwest. For example, rural areas in Santiago del Estero have, by a significant margin, the lowest rate o f school attendance (measured by percentage o f 15-17 year olds in school). Satisfaction o f basic needs i s also among the poorest. Yet both dispersed and grouped areas have grown inpopulation. Perhaps this reflects new entrants related to the soybean boom. It may also reflect relatively low mobility o f existing farmers, which has prevented them from out migrating despite their poor conditions. It is noteworthy that the area o f soybeans expansion exhibits no significant change in average farm size, and only a small increase inthe use o f rental contracts, providing little evidence o f small farmers benefiting from land transactions with the new soybean entrants. To provide the basis for a poverty strategy adapted to the economic dynamics o f the region, we need additional data concerning the economics o f poor farm households and their regional contexts. Catamarca and La Rioja have, by a considerable margin, the best indicators on education and basic needs o f all the Northwestern provinces, and considerable recent expansion o f "new" export products such as olives, olive oil, wine, and industrial crops, L a Rioja appears to export a higher share o f its agricultural products in manufactured form than any other province in Argentina (18 percent), whereas Catamarca's exports in value-added form were weak (1 percent). At 3.6 percent per year, L a Rioja's agriculture GDP growth over the 1991-2001period was among the highest in Argentina, while Catamarca's was slightly below average. Both provinces demonstrated good growth in ag-based industrial exports, both pre-crises and post crises, although Catamarca's post-crises ag-based manufacture export growth has been stronger (from a much lower base). They both exhibit low rates o f population loss from dispersed areas. Nevertheless Catamarca's grouped rural population has increased by 14 percent, while that o f L a Rioja has fallen by 12 percent. This i s especially curious given L a Rioja's higher rate o f agricultural GDP growth and stronger export orientation o f ag-based industrial products, which one would expect might generate additional non-farm employment ingrouped rural areas. Again the formulation o f appropriately differentiated regionalpoverty andrural development strategies will require that we better understand the dynamics o fpopulation movements inthese provinces. The Cuyo The Cuyo region is composed o f three provinces, Mendoza, San Juan, and San Luis. Mendoza and San Juan depend almost entirely on irrigation for agriculture, while in San Luis, irrigation i s minimal. Mendoza and San Juan are famous for their vineyards. San Luis, on the other hand, produces grains and oilseeds; notably wheat, sorghum, sunflower and soy, with 51,300, 36,000, 36,000 and 4,000 hectares, respectively. 139 Mendoza alone concentrates 70 percent o f Argentina's land planted with grapes. In Mendoza, fruits account for 82 percent o f total irrigated farm land. Among fruits, the higher shares correspond to grapes, plums, apricot and olives with 54, 7, 7 and 6 percent o f total irrigated land respectively. In addition, inMendoza horticulture accounts for 14 percent o f total irrigated farm land area. In San Juan, fruits account for 85 percent o f total irrigated farm land. Among fruits grapes are noteworthy with 57 percent o f total irrigated area, followed by olives with 20 percent o ftotal irrigated area. Onions and garlic is also important inthe region. InSan Juan andMendoza, all size producers are involvedwith grapes, but exports are mainly limited to medium and large producers. In Mendoza, grapes are exported as wine and also as fresh fruits. In San Juan, exportable fresh grapes predominate (they are also consumed in the domestic markets). Apricots and a portion o f the onion and garlic crop are also exported from both San Juan and Mendoza. In both provinces, the farmers involved in the export business constitute the minority. Few producers have been able to meet the requirements o f the external market, although those who have, have done well. A significant numbero fsmall farmers grow horticulture (i.e. industrial tomatoes, carrots, onions) in both provinces, and in San Juan there is an increasing production of melons for internal consumption. The majority o f small farmers either continued to produce for the domestic market, and seek new non-agricultural activities or sell out and abandon agriculture. Data from the agricultural census shows important decreases in the number o f farms in Cuyo. In San Luis, the number o f farms decreased by 38 percent, in San Juan by 23 percent and in Mendoza by 13 percent. Not surprisingly, in San Juan and San Luis, rural population decreased by 17 and 27 percent, respectively. InMendoza, on the other hand, rural population increased -- by 0.3 percent indispersedruralareas, and32percent ingroupedruralareas. All three o f Cuyo's provinces are among the provinces that exhibited rapid growth of agricultural-based manufactured exports in the immediate post reform era. Of the three, only Mendoza had, in 2003, regained the level achieved before the crisis. Overall, Mendoza has been much more successful than San Juan in achieving high value export markets. Mendoza's agricultural-based manufactures are equal to 6 percent o f agricultural GDP (fourth highest in Argentina), compared to 1percent for San Juan. ThePatagonia The Patagonia region concentrates the largest valleys with pears and apples. About 40 percent o f total cropland i s irrigated inPatagonia. N o Negro province has the largest irrigated area (55 percent), followed by NeuquCn and Santa Cruz, with 26 and 27 percent o f total farm land irrigated, respectively. Fruits are the main irrigated crop in NeuquCn and Rio Negro, with 67 percent o f total irrigated farm land in each province. Fodder i s the main irrigated crop inChubut andin Santa Cruz with 76 and 95 percent o ftotal irrigated farm land ineach o f these provinces, respectively. InNeuquCn and N o Negro, fodder is the second main irrigated crop with 21 and 25 percent o f total irrigated farm land, respectively. As inCuyo, small farmers inthe Patagonia have been unable to modernize their farms and meet the requirements o f international demand. Many sold their farms to the new investors and 140 became temporary wage workers or migrated to other areas (Rofinan, 1999). InRio Negro, rural dispersed population decreased by 24 percent between 1991 and 2001 and the number o f farms shrank by 19 percent between 1988 and 2001. The introduction o f technology in these agro- industriesis also reportedto have diminishedlabor demand inurbanareas. Livestock activities are very important in Patagonia, mostly sheep and lambs. This region concentrated about 66 percent o f total sheep and lambs in Argentina. Nevertheless, farmers growing lamb have been seriously affected by low prices o f wool during the nineties. In the province o f Santa Cruz it was estimated that from the 1,102 farms with sheep in 1988, only 51 percent (567) remained active in 1997 (Lattuada, 2000). Considerable international capital was invested inPatagonia provinces following the reforms o f the ~O'S,particularly in modern agro-exporter plants and large farms. Neuqukn and Rio Negro became important producers o f pears, and in Chubut and Santa Cruz investment went into fine fruits, flowers and garlic. All five provinces exhibited rapid growth o f agricultural-based manufactured exports, over the period, and all five achieved substantially above average growth o f agricultural GDP (over the 1991-2001 period). Chubut, which has the fourth highest ratio o f agricultural-based manufactured exports to ag GDP inArgentina (7 percent), i s the only province inPatagoniato havereestablished itspre-crisis ag-basedmanufacturedexport levelby2003. Although pear production has expanded reasonably among the more commercially-oriented farms o f the Upper Rio Valley, apple production has stagnated, due to a mixture o f outdated rootstock o f low commercial value, and poor phytosanitary control. This is both a problem o f lack o f credit availability to upgrade rootstock, and lack o f an effective program o f Carpocapsa, control - especially failure to eradicate "abandoned" orchard. Outmigration from the agricultural sector has been rapid in Santa Cruz, from both the grouped and dispersed areas, due mostly to sustained low wool prices during the 1992-2001 period. Remarkably, Santa Cruz has among the best high school youth participation rates, in both grouped and rural dispersed areas, as well as the best record o f meeting basic needs o f the rural population (only 12 percent without basic needs met). These facts suggest that farmers leaving the country side inSanta Cruz are likely to be relatively well-prepared for life elsewhere. ThePampas Finally, the Pampeana region i s the richest rural area o f the country. The main crops are soy, wheat, and corn. Soy is the major crop and accounts for 12 million hectares, or 89 percent of the total surface. Sunflower i s also important in Santa FeyBuenos Aires, L a Pampa and Cbrdoba, while linenand rice, are also predominant inEntreNos. Most farms (approximately 75 percent) combine agricultural with livestock activities. The large majority o f the farms inthe Pampas produce for export. They are mainly medium and large farms: the 37 percent o f all farms with fewer than 100 hectares accounts for only 3 percent of all farm land. 141 Important decreases in the number o f farms have been observed in the Pampas. In 2001 there were 29 percent fewer farms than in 1988. Buenos Aires and C6rdoba showed the highest reductions with 32 and 36 percent losses between 2001 and 1988. However, it i s not clear if all these farmers have abandoned the activity or formed new enterprises (Bertolassi, 2004). In certain industries, such as the dairy industry, the extinction o f farms is evident. For example, between 1985 and 1995 the number o f dairy farms dropped by 40%, while output increased by 90 percent (Gutman, 2002). Those more negatively affected have been small dairy farms (Lattuada, 2000). Irrigationinthe Pampeana region i s supplemental, largely pivot irrigation based on groundwater. Grains constitute the main irrigated crop in every province, except L a Pampa, where fodder accounts for a large share. InEntreRios, the second main irrigated crop i s rice with 42 percent o f total irrigated farm land.There, fruits are also important. The Pampas is undergoing a rapid process o f farm consolidation, driven by cost reducing and labor saving technological change, and by new contractual instruments (generically labeled "pools") for combining land, machinery, and high quality management. These pooling arrangements have proved effective at improving the overall level o f management and risk diversification. Productivity indicators suggest that commercial farms inthe Pampas are world class in grain and oilseed production. Soybeans and grain production has boomed inthe past 15 years. Ironically, beef production has stagnated, in production, in exports, and by all available productivity measures. Since most operations in the Pampas are mixed crop-livestock operations it i s unlikely that this stagnation is a reflection o f management deficiency. It i s more likely to be a reflection o f (1) the very high and competing profitability o f grains and soybeans (especially since the wide-scale adoption o f no-till plantingand pooling arrangements), (2) the perception o f greater risk associated with cattle--especially since the hoof-and-mouth outbreak o f 2001,and(3) governance problems in the beef chain. Cap and Gonzalez (2005) reported that most producers operate at low technological levels and that the productivity gap between low and high technology packages i s high(65 percent on average). Relative to the other regions, indicators o f rural quality o f life are high: unmet basic needs and rural poverty are the lowest o f all the regions, and education levels the highest. Nonetheless, farm size concentration is taking place more rapidly inthe Pampas than elsewhere, and the rate o f out migration from dispersed areas is the most rapid. Inthe Pampas, higheducation levels (facilitating off-farm employment) indispersed rural areas, even among small farmers, combine with an active land market (for both sales and rentals) to lure small and mid-size farmers o f f the land. Inthese cases out migration should not be taken as prima facie evidence o f decreased well-being. On the contrary, migration may be evidence o f a farm family having achieved a critical level o f assets (human capital plus increasingly high valued land) to successfully make the transition to non-farm employment. Unfortunately, no data are available to evaluate whether or not these farm families are, in fact successful in non-farm life. 142 ANNEX 11: INDICATIVEANALYSIS OF EXPORT MARKET STRATEGIESOF THE REGIONALECONOMIES Inthis sectionwe look at two generally accepted `generic' approaches to market penetrationthat are widely believed to lead to significant success for a firm or an industry: 0 Offering the market a distinctive product (or service) for which consumers are willing to pay a premium price (be the "price leader"); 0 Producing a good quality, standardproduct at the lowest unit cost (be the "cost leader"). As a first approach, we examine the performance o f Argentinean industries based on irrigated landto see whether they mightbe attempting either to: 0 Getthe highestprice for its freshproduce or processedproducts; or 0 Maximize the sales o f `commodity' products at the lowest cost. We might, prima facie, expect to find evidence o f the latter strategy since Argentinean agribusiness has historically tended to favor a "commodity" stance for products such as wheat, beef andtea. Although a wide range o f permanent and annual crops can be grown on irrigated land, we look at Argentina's internationalperformance injust two products: wine andtable grapes. 0 First, we look at the relative position o f Argentina vis-&-vis other countries in recent years. 0 Secondly, we look at Argentina's performance over time to check for consistency in its strategy. 0 Finallywe look for a `fit' betweenprice and cost. The data used Since we only intend to illustrate a general approach, we use readily available data, mostly from the FAOSTAT (2005). O f necessity this means using average prices, a move that conceals the complexity of, say, Frenchwines. Since FA0 data does not distinguish between table grapes and grapes grown for wine-making, we use what amounts to a weighted average o f the two for consideration o f the cost o f agricultural production in the case o f both table grapes and wine. Despite these shortcomings, the analysis seems to provide a good enough approximation o f Argentina's situation inboth industries examined. Wine The World Market. Figure A2.1 displays the main wine exporting countries, ranging them according to the average free on board price (fob) value o f a ton o f wine exported (the Y axis) 68This note was kindlyprovidedby JohnYoung. 143 and the rate o f sales growth over the preceding five years (the X axis). The size o f the `bubble' indicates the dollar volume o f sales. The market may be dividedgenerally into three groups. The first "group" consists solely o f France, the historical market leader. France's average export price and export volume have both risen steadily since the 1960s. The price leveled out somewhat in the early eighties, and again in the nineties, but reached record levels in 2003. Sales growth has recently been slow - as i s often the case with a market leader. As is well known, French wine-makers export a range of products with a wide variation inprice. The second group comprises the other large European wine exporters (Italy, Spain, Portugal and Germany) plus the US, Chile and Uruguay. Their average prices ranges from half o f France's (Italy) to a quarter (Spain's). Their sales volume has also been slow. (The position of Chile i s shown based on its actual sales volume and sales growth rate for 2003 and its average export value for 2002). The third group is less compact and comprises countries such as New Zealand, Australia and South Africa. All have experienced recent dramatic sales growth but have different pricing strategies. New Zealand specializes in premium wines and earns a higher fob price than France's average (though not its top products). Australia prices its wines above the "European" average but below France. South Africa prices its wine above Spain but below Italy. All three have doubled or tripled sales over the last five-year period. Figure A2.1: World exports of wine World exports ofwine 2003 I I I -20% -IO% 0% 10% 20% 30% 40% 30% 60% 70% 80% 90% I W C 110% 120% 130% 140% 150% IM% 170% 180% 190% 1W% 210% I l l e l growth I-year n t c Source: FAOSTATS data (2005). Argentina falls below the "European" group, earning an average fob price below $1,000 per tonne with a more or less stagnant sales volume. 144 Insummary, we can identify France as the "price leader", and also tentatively identify Spain as the "cost leader". (The former `cost leader', Algeria, declined dramatically in the 1980s and i s now scarcely visible.) Argentina's Performanceover Time. Whereas most newcomersto the wine trade - like Chile - or revamped old-timers - like Australia and South Africa - initially positioned themselves to compete with the `cost leader', then gradually moved up-market, Argentina appears at times to have followed this approach, and, at others, to have reverted to its traditional `commodity' stance that adopted the `newcomer' strategy - with that o f Argentina. - where it appears to be in2003. Figures A2.2 and A2.3 contrast the trajectory o f New Zealand - New Zealand shows a steady increase in average unit price and sales volume with only one deviation in2000. This i s a typical New Zealand approach to agribusiness exports, well suited to a small economy, geographically distant from markets. Argentina seems to have moved up-market in the early 1970s, again briefly in the late 1970s - but then steadily dropped its prices until the mid-1980s. The average export price see-sawed throughout the 1990s, then climbed to a point above Chile in2001 - after which exporters again reverted to lower prices (that i s to say, attempted to penetrate markets, like Russia) where price i s the basis o f competition. Unit value changes depend munch on the variation in the composition o f the export mix o f common and quality wines. FigureA2.2: New Zealand exports ofwine New Zealand: export wine performance $7,000 I $6,000 P $5,000 e ;$4,000 -" n -3 3 $3,000 sh : 5 $2,000 \ Yource:FAOSTATSdata (2005). 145 Figure A2.3: Argentina Exports of wine Argentina: export wine performance $1,750 $1,500 P 8 - $1,250 P Y H i$1,000 - m C e $750 r $500 $250 e $0 1963 1965 1967 1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 Source: FAOSTATSdata (2005). The Relationship between Price and Cost. In the absence of cost data, we have used the agricultural yield of grapes as a crude proxy. FigureA2.4 plots the average export price in 2003 against the yield of grapes. Figure A2.4: Export wine prices and grape yields Export wine prices &grape yields $6,000 New Zealand $5,000 -. P e - $4,000 - Y P r0 7 - Y 3 $3,000 ~ s Aurunlis a 8 Uruguay eP$2,000 - ' M y , -Portugal -Gaecce L"l,cfm) Germany - *SPnln. USA $1,000 - *Algeria Moldova I *Argentina \ *Brazil 1 $0 4 I 20,000 40,000 60,000 80,000 100,000 120,000 140,000 160,000 agricultural yield of grapes hg/ha Iurce: FAOSTATSdata (2005). 146 The assumptions about the two generic strategies would lead to the expectation o f finding the `price leader' to have relatively higher raw material costs for two reasons: first, the emphasis would be on the quality of the raw material, not the quantity; secondly, there would not be any great imperative to drive down raw material costs, hence `traditional' methods would be likely to persist. On the other hand, one would expect to find the `cost leader' striving for lower costs in every component o f the value chain, especially the raw material that usually accounts for a high percentage o f the total cost. On examining the graph, one can imagine a line running from `New Zealand' (near the top left) ina south-easterly direction, passing close to France andAustralia, thenbisectingthe "USA" and "Brazil", Countries that lie close to the line may be thought of as successfully trading off cost andprice. The position o f Spain and Portugal requires comment. First, since Spain is, tentatively, the `cost leader', then its position in the south-west quadrant might suggest that other factors than agricultural yield account for low costs. The lower yields may also be the consequence o f using non-irrigated land with few alternative uses. A third hypothesis is that recent economic growth, accompanied by a rise in factor costs, i s inclining Spanish wine-makers to change their strategy and move up-market, a move that has its parallels in other sectors o f the Spanish food and beverage industry. The position of Argentina is less ambiguous. Its grape yield i s higher than most o f its wine- exporting competitors, but well short o f Brazil, an acknowledged exporter o f `commodity' wines. To put things another way, if Argentina wishes to compete at its current price level, then it would be logical to aim to increase its grape yieldby a third. If,on the other hand, it wishes to price its wine according to its current cost structure, then it would have to position itself close to Australia and Uruguay - but this would mean endeavoring to triple its price. Meanwhile, it i s stuck in the middle. Grapes The World Market. The international grape market includes most o f the players in the international wine market, though the relative role o f each i s different (with the possible exception o f Italy). The major exporters are shown inFigure A2.5. The `price leader' is the U S - and it may be the `cost leader' too for the northern hemisphere exporters. The `cost leader'- at least for the southern hemisphere - i s Chile. Since grapes, unlike wine, are perishable, small exporters are able to earn a premium price by supplying the U S or EU markets during the `windows' that occur between the end of the northern hemisphere crops (California and the regions bordering the Mediterranean Sea) and the southern hemisphere crop (Chile). Brazilian exporters have been able to exploit one such window, achieve a highprice and a fast rate o f sales increase, apparently at the expense o f Israel. Egypt, on the other hand, has yet to devise a strategy to obtainbetter than derisory prices - which nonetheless enable it to earn a modest profit by exploiting new, irrigated lands. 147 Argentina occupies the same position on the graph as Mexico with a smaller sales volume. Its price i s barely above the `cost leader', Chile, but with a slower rate o f sales growth. FigureA2.5: World Grape Exports World grape exports 2003 Argentina's Performance over Time. Figure A2.6 shows a similar series o f rises and falls in price to the graph depicting its perfonnance in exporting wine. Once again, it appears that, from time to time, Argentine exporters aim for a premium market, then change direction and decide instead to compete on the basis o fprice. Figure A2.6: Argentina: exports grapeperformance Argentina: export grape performance $1,600 $ 1,400 -dB 2 H $1,200 $1,000 a800 b f $600 a $400 $200 I I I I , I I j , , , , , , , 1960 1963 1966 1969 1972 1975 1978 1981 1981 1987 1990 1993 1996 1999 2002 2001 wce: FAOSTATSdata ( 2005). 148 The Relationship between Price and Cost. Figure A2.7 is harder to interpret than its `wine' counterpart. First, the `price leader', the US, is also a very efficient grower o f grapes. Secondly, certain small exporters- Brazil and Israel are able to exploit `windows' in order to earn a high - price. Thirdly, the `cost leader', Chile does not achieve outstanding grape yields comparedto the U S - with whom it avoids direct competition by beingina different hemisphere. Argentina's position, though, is rather clear. As with export wine, it is, once again, stuck in the middle. As a first cut, one may conservatively imagine a line starting at `France', bisecting `Greece' and `Australia' and proceeding south-east to `Egypt'. In order to `sit' on this line, Argentina would either have to increase its yield o f grapes by a quarter or its price by a half. A second, more aggressive cut would draw the northwest to south-east line from Israelto Egypt. This reading o f the data would suggest that Argentina (indeed all) exporters should endeavor to raise their agricultural yields to match those o f Israel, Brazil andthe US. Figure A2.7: Export grapeprices and yields Export grape prices and yields $2,000 Israel 11,800 \ 61,600 France . Australia Bmzil 0 d $1,400 8L k . SA \ $1,200A H Spain Y i161,000 Porlvgai *P Italy Uruguay : -*: $800 Mexico 7 , gb Chile Argentlns \ 1 I $600 S \\ I Turkey $400 QYTt Moldova $200- so I 20.000 40.000 60.000 80.000 100.000 120.000 140,000 l60.000 180.000 agricultural yield hgha I Source: FAOSTATSdata (2005). The foregoing is just a sketch o f why Argentina does not achieve as good a return on irrigated crop-land as some o f its competitors. Inboth the wine market and the table grapes market, the general strategy o f the various national industries and the relative success o f the strategy are readily apparent. There i s also a plausible fit between each leading player's price and cost profile. Not so for Argentina. The behavior o f Argentinean exporters o f wine and table grapes is suggestive of a trading mentality: the antithesis o f a marketing approach. The latter targets a specific market niche with a product designed or selected to appeal most to a particular segment o f the market, then adopts a long-term market penetration plan that `fits' or `aligns' all aspects o f the business to serving the 149 target customer. The countries perceived as following a successful strategy inthe wine and table grapes industries stayed with their marketing planover many years. This behavior o f Argentina's wine industry may be due to the transformation started inthe late 1980s when the preference o f domestic consumers for quality wines and the reduction o f per capita consumption together with the expanding opportunities for quality wines in world markets started being felt. Argentina's transition from a common quality to a high quality wine producer and exporter has been slowed down by capital constraints to renew grape plantations, ineffective collective action inthe wine chain and insufficient, although increasing, FDI. 150 ANNEX 111:THE NEWAGENDAFORFOODQUALITYAND SAFETY There has been a noticeable upsurge o f concern throughout the world over the last decades regarding food quality and safety issues. This i s essentially a demand driven phenomenon as it is mostly consumers' pressure what has given such relevance to quality and safety issues in the food and agriculture agenda. Also important i s the pressure o f trade-related legislations and process/product standards that are in flux because o f innovations in quality control. It i s also a relatively recent event; three or four decades ago governments andtheir phytosanitary andpublic health authorities were mostly driving the agenda. Governments today continue to be involved with regulatory and other functions, but the pressure to ensure quality and safety infood chains comes from consumer associations and consumers at large, and many o f the standards are currently set by the private sector. The change i s due to a combination o f factors among which the increase o f health consciousness among consumers, and the growing industrialization and differentiation o f food with the expansion o f processing, labeling and niche markets, as well as higher value added with incremental quality leading to higher prices. Developed, middle income and developing countries have all been affected by this trend although in different ways and to different extents. In middle income countries like Argentina there i s concern o f domestic consumers about food and feed quality and safety issues, but the strongest pressure for improvement comes from two sources: international trade and supermarkets. Argentinean, like other international producers, are subjected to the same or even more demanding standards in developed country markets when they try to export to these markets. Regulators require imported products to be subjected to the same control methods as local ones, andinternational agencies and agreements tendto adopt these methods as the international norm. Thus, for instance, HACCP has been adopted by the Codex Alimentarius as the international standard for food safety, and the US government has made it compulsory for imported seafood and juices.69 Meeting international food quality and safety standards becomes hence a requisite for export development. Supermarkets have also played their part. Many changes are taking place in food retailing in Latin America in recent times, with dramatic increases in the proportion o f food that i s sold in supermarkets. Argentina is ahead in this trend with a proportion o f 57 percent o f food consumption retailed through supermarkets (Reardon et al, 2002). The implications for the quality and safety o f foods are important. There are several reasons for this. One i s that many supermarket chains are multinational firms which bring with them international standards that they apply more or less commonly inthe countries where they operate. 69The Hazard Analysis and Critical Control Point system (HACCP) is an integrated food safety control method based on prevention. I t involves seven principles: analyze hazards; identify critical control points; establish preventive measures with critical limits for each control point; establish procedures to monitor de critical control points; establish corrective actions to be taken when monitoring shows that a critical limit has not been met; establish procedures to monitor that the system i s working properly; and establish effective record keeping to document the HACCP system. 151 The internalization o f international standards through supermarkets and foreign trade, and the new demands for food quality and safety are putting increasing pressure on all the elements o f the productiodmarketing chain: producers, processors and regulators. To the old phytosanitary agenda, mostly based on avoiding production losses through pest protection, a new one is thus added. It includes a broader concept o f quality incorporating inter alia concerns o f social and environmental production and processing conditions, the stability and homogeneity in the characteristics o f specific products, and the presence o f residuals from veterinary drugs and pest control chemicals as well as their safe use inthe field. The opportunities as well as the safety and commercial risks brought about by biotechnology and genetically modified events are also part o f the new agenda. This agenda brings new requirements o f traceability, certification and the use o f preventive systems like HACCP, as well as for the commercial release o f new varieties when they have been genetically modified. The areas requiring enhancement to meet the new challenges are field practices in crop and animal production and processing, the regulatory system, and the accreditation and certification system. Improving crop and animal production practices and processing i s essentially a private sector matter which, like environmental management, requires an adequate combination of economic incentives and command and control measures. Market signals are themselves a powerful incentive system, for instance the possibility o f accessing or not export markets according to product quality and safety. But there i s also plenty o f room for public action. One area is facilitating that market signals reach producers clearly, like for instance through information systems that alert producers to possible market opportunities which require meeting certain standards. Another is enabling producers to respond to these signals, for instance by training them in sanitary and phytosanitary regulations or assisting establishing HACCP systems inproductionchains. Butprobably the most important public role is infacilitating the operation of collective action. This i s essential because the benefits from good quality and safety practices rest on many producers and/or processors simultaneously following them either in a certain territory or along the production chain. Collective action can be ensured through facilitating the creation and working o f adequate economic coordination institutions and through command and control measures, such as the imposition of effective sanitarystandards. 152 h k V * 0 2 0 e, b Y i0 r2 u 0 v, v, 3 P a, 8a, c * L u c L n w e Q, 80L P c1 U Y .-a> Q) 0 0 S M =x L: P .- L L k 160 MAP SECTION