CIMMYT - AgEcon Search

Tropical and Subtropical Maize in AsiaProduction Systems, Constraints,and Research PrioritiesRoberta V. Gerpacio andPrabhu L. Pingali

CIMMYT® (www.cimmyt.org) is an internationally funded, not-for-profit organization that conductsresearch and training related to maize and wheat throughout the developing world. Drawing on strongscience and effective partnerships, CIMMYT works to create, share, and use knowledge and technology toincrease food security, improve the productivity and profitability of farming systems, and sustain naturalresources. Financial support for CIMMYT’s work comes from many sources, including the membersof the Consultative Group on International Agricultural Research (CGIAR) (www.cgiar.org), nationalgovernments, foundations, development banks, and other public and private agencies.© International Maize and Wheat Improvement Center (CIMMYT) 2007. All rights reserved. Thedesignations employed in the presentation of materials in this publication do not imply the expressionof any opinion whatsoever on the part of CIMMYT or its contributory organizations concerning the legalstatus of any country, territory, city, or area, or of its authorities, or concerning the delimitation of itsfrontiers or boundaries. CIMMYT encourages fair use of this material. Proper citation is requested.Correct citation: Gerpacio, R.V., and P.L. Pingali. 2007. Tropical and Subtropical Maize in Asia: ProductionSystems, Constraints, and Research Priorities. Mexico, D.F.: CIMMYT.Abstract: This book examines future technological and policy prospects for the sustainable intensificationof rainfed upland maize production in Asia, and derives R&D priorities for specific maize productionenvironments and markets. Village-level and farmer-group surveys were conducted to characterizeupland maize production environments and systems in China, India, Indonesia, Nepal, the Philippines,Thailand, and Vietnam. Survey findings, particularly farmer-identified constraints to maize production,complemented with other relevant data, were used in country-level, R&D priority-setting workshops.High on the list of farmer constraints was drought, estimated to affect three production environments thatare home to about 48 million rural poor and produce an estimated 16 million tons of maize, and otherssuch as downy mildew, stem borers, soil erosion/landslides, waterlogging, poor agricultural extension/technology transfer services, and poor access to low-interest credit and markets. Farmers felt thatsocioeconomic and policy-related constraints impact maize productivity more than technical constraintsdo. It is important to recognize that technology is not the only key to increasing productivity and betteringthe conditions of marginal maize farmers in Asia. There is a growing trend towards commercializingand intensifying maize production that is different from the staple food self-sufficiency paradigm thathas been the cornerstone of agricultural policy in most developing countries. Appropriate governmentpolicies could help alleviate the adverse consequences of commercialization and promote sustainableintensification of maize production, especially in marginal environments inhabited by resource-poorsubsistence farmers.ISBN: 978-970-648-155-9AGROVOC Descriptors: Maize; Agricultural development; Farming systems; Production policies;Environmental factors; Cropping systems; Research projects; Projectmanagement; AsiaAGRIS Category Codes: E10 Agricultural Economics and PoliciesDewey Decimal Classif.: 633.155Editing: Alma McNabLayout/design: Eliot Sánchez PinedaPrinted in Mexico.

ContentsPage No.Tables ................................................................................................................................................................vFigures ............................................................................................................................................................. viiAnnexes ............................................................................................................................................................ viiExecutive Summary .......................................................................................................................................viii1. Introduction ........................................................................................................................................................ 11.1 Background .......................................................................................................................................................... 11.2 Objectives ............................................................................................................................................................. 21.3 Key Activities, Methodologies and Partners .................................................................................................. 21.3.1 Detailed characterization of upland maize-based production systemsand identification of constraints ............................................................................................................ 21.3.2 Country-specific constraint prioritization and maize R&D plans .................................................... 31.3.3 Maize sector policy analysis and identification of key policy andinvestment options .................................................................................................................................. 31.4 Organization of this Book .................................................................................................................................. 42. Maize Sector Policies in Asia .......................................................................................................................... 52.1 Brief Overview of Maize Economy Trends ..................................................................................................... 52.1.1 Shifts in demand ...................................................................................................................................... 52.1.2 Production response ................................................................................................................................ 62.1.3 Trends in trade .......................................................................................................................................... 62.2 Maize Policy Development ............................................................................................................................... 72.2.1 Production policies .................................................................................................................................. 72.2.2 International trade policies ..................................................................................................................... 82.2.3 Macroeconomic policies on exchange rate ........................................................................................... 92.3 Impact of Policies on Maize Production ........................................................................................................ 102.3.1 NPC analysis using official exchange rates ........................................................................................ 102.3.2 NPC analysis using shadow exchange rates ...................................................................................... 112.3.3 What do they mean? .............................................................................................................................. 123. Maize Agro-ecosystems in Asia .................................................................................................................... 133.1 Maize Production Environments .................................................................................................................... 133.2 Biophysical Environment ............................................................................................................................... 143.2.1 Rainfall ..................................................................................................................................................... 153.2.2 Temperature ............................................................................................................................................ 163.2.3 Types of soil ............................................................................................................................................ 163.3 Institutional Environment ............................................................................................................................... 163.3.1 Line agencies ........................................................................................................................................... 163.3.2 Farmer cooperatives and user groups ................................................................................................ 163.3.3 Sources of material inputs .................................................................................................................... 173.3.4 Credit institutions .................................................................................................................................. 173.3.5 Input and output prices ........................................................................................................................ 183.3.6 Input-output price ratios ....................................................................................................................... 19iii

3.4 Infrastructure ..................................................................................................................................................... 203.4.1 Accessibility ............................................................................................................................................ 213.4.2 Irrigation .................................................................................................................................................. 213.4.3 Post-harvest facilities ............................................................................................................................. 213.4.4 Proximity to markets and marketing practices ................................................................................. 213.5 Socioeconomic Characteristics ........................................................................................................................ 223.5.1 Households and ethnicity ..................................................................................................................... 223.5.2 Literacy and level of education ............................................................................................................ 233.5.3 Landholdings and land tenure ............................................................................................................. 243.5.4 Maize utilization .................................................................................................................................... 243.5.5 Farmer classification .............................................................................................................................. 254. Maize Production Systems ............................................................................................................................ 274.1 Crops, Cropping Calendars, and Cropping Patterns .................................................................................. 274.2 Maize Varieties Grown and Farmer Preferences .......................................................................................... 294.3 Land Preparation and Crop Management Practices ................................................................................... 294.4 Labor and Material Input Use ........................................................................................................................ 304.5 Post-harvest Practices ...................................................................................................................................... 314.6 Yields and Yield Gap ........................................................................................................................................ 315. Maize Production Constraints ...................................................................................................................... 335.1 Biotic Constraints .............................................................................................................................................. 335.2 Abiotic Constraints ........................................................................................................................................... 345.3 Socioeconomic and Policy-Related Constraints ........................................................................................... 356. Priority Constraints for Maize Research and Development ................................................................... 376.1 Methodology for Identifying Priority Constraints ...................................................................................... 376.2 Production-based Priority Constraints .......................................................................................................... 396.2.1 Major findings ........................................................................................................................................ 396.2.2 Priorities by maize production environment and market orientation ........................................... 486.3 Area-based Priority Constraints ..................................................................................................................... 506.3.1 Major findings ........................................................................................................................................ 506.3.2 Priorities by maize production environment and market orientation ........................................... 586.4 Economics of Priority Constraints ................................................................................................................. 606.4.1 Production-based priority constraints ................................................................................................ 606.4.2 Area-based priority constraints ........................................................................................................... 616.5 Technology Options to Address Priority Constraints ................................................................................. 626.5.1 Technology interventions to address abiotic constraints ................................................................. 626.5.2 Technology interventions to address biotic constraints ................................................................... 666.6 Sources of Research and Technology ............................................................................................................. 696.6.1 Public and private sectors: delineation of research responsibilities ............................................... 696.6.2 Role of the public sector ........................................................................................................................ 696.6.3 Role of the private sector ...................................................................................................................... 706.6.4 Public and private sectors working together ..................................................................................... 706.6.5 Analyzing the challenges ...................................................................................................................... 726.7 Research and Policy Agenda ........................................................................................................................... 727. Summary and Conclusions ............................................................................................................................ 747.1 Maize Production Environments .................................................................................................................... 747.2 Maize Production Systems .............................................................................................................................. 757.3 Important Production Constraints ................................................................................................................. 767.4 The Role of Policy ............................................................................................................................................. 78References .............................................................................................................................................................. 79Annexes .............................................................................................................................................................. 83iv

Table 6.6. Top 30 priority constraints to maize production according to the production-based povertyindex, Asia (with and without STMA-commercial maize production in India and excludingSouthwest China’s low market surplus maize production systems). ............................................................ 45Table 6.7. Top 30 priority constraints to maize production according to the production-based marginalityindex, Asia (with and without STMA-commercial maize production in India and excludingSouthwest China’s low market surplus maize production systems). ............................................................ 47Table 6.8a. Top 30 priority constraints to maize production according to the production-based combinedindex, Asia (with STMA-commercial production areas of India and excluding SouthwestChina’s low market surplus maize production s ystems). ............................................................................... 48Table 6.8b. Top 30 priority constraints to maize production according to the production-based combinedindex, Asia (without STMA-commercial production areas of India and excluding SouthwestChina’s low market surplus maize production systems).................................................................................. 49Table 6.9. Production-based priority constraints to maize productivity by production environment andmarket orientation, Asia (with and without STMA-commercial maize production areas ofIndia and excluding Southwest China’s low market surplus maize production systems). ........................ 50Table 6.10. Top 30 priority constraints to maize production according to the area-based efficiency index,Asia (with and without STMA-commercial maize production areas of India and excludingSouthwest China’s low market surplus maize production systems). ............................................................ 51Table 6.11a. Selected indicators of impact of top 30 area-based priority maize productivity constraints inAsia (with STMA-commercial maize production areas of India and excluding SouthwestChina’s low market surplus maize production systems). ................................................................................ 53Table 6.11b. Selected indicators of impact of top 30 area-based priority maize productivity constraints inAsia (without STMA-commercial maize production areas of India and excluding SouthwestChina’s low market surplus maize production systems). ................................................................................ 53Table 6.12. Top 30 priority constraints to maize production according to the area-based poverty index,Asia (with and without STMA-commercial maize production areas of India and excludingChina’s Southwest low market surplus maize production systems). ............................................................ 54Table 6.13. Top 30 priority constraints to maize production according to the area-based marginality index,Asia (with and without STMA-commercial maize production areas of India and excludingSouthwest China’s low market surplus maize production systems). ............................................................ 55Table 6.14. Top 30 priority constraints to maize production according to the area-based combined index,Asia (with and without STMA-commercial maize production areas of India and excludingSouthwest China’s low market surplus maize production systems). ............................................................ 57Table 6.15. Area-based priority constraints to maize productivity by production environment andmarket orientation, Asia (with and without the STMA-commercial maize production areas ofIndia and excluding Southwest China’s low market surplus production areas). ........................................ 59Table 6.16. Estimated gross income loss in maize production across environments due to the top 30priority constraints, production-based and area-based combined indices, Asia (PPP $/ha). .................... 60Table 6.17. Estimated gross income loss in maize production by environment and market orientationacross the top 30 priority constraints, using production-based and area-based combinedindices, Asia (PPP $/ha). ....................................................................................................................................... 61Table 6.18. Technology and socioeconomic/policy-related options for addressing priority constraintsto maize production in Asia. ................................................................................................................................ 63vi

FiguresPage No.Figure 2.1. Mean utilization of maize as food, feed, and seed, Asia, 1961-2002. ............................................................ 5Figure 2.2. Area and yield contribution to maize production growth in Asia, 1961-2004. ........................................... 6Figure 3.1. Long-term rainfall distribution and maize cropping season in Cebu, Eastern Visayas(low market surplus production, top), and Bukidnon, Northern Mindanao (commercialproduction, bottom), Philippines. ................................................................................................................... 15Figure 3.2. Monthly rainfall and maize-based cropping patterns in the drylands of Jeneponto,............................... 15South Sulawesi, Indonesia.Figure 3.3. Seasonal rainfall distribution in China by mega-environment. .................................................................. 15Figure 4.1. Maize cropping calendars, Asia. ...................................................................................................................... 28AnnexesAnnex 1. Survey locations, their number by maize mega-environment and salient characteristics, Asia. ........... 83Annex 2. Matrix of RRA/PRA survey sites by maize agroecological environment and marketorientation of maize production. ....................................................................................................................... 85Annex 3. Summary of agro-climatic features of surveyed locations classified by agroecological regionand orientation of maize production, Asia. ..................................................................................................... 86Annex 4. Types of soil present and their farmer-reported advantages and disadvantages,across surveyed villages, Asia. ........................................................................................................................... 87Annex 5. Maize production and poverty parameters by country and agroecological zone, Asia. ......................... 88Annex 6. List of 286 prioritized maize productivity constraints identified by Asian maize farmers. .................... 89vii

Executive SummaryGlobally, 69% of all cereal areas is rainfed, and accounts for 58% of world cereal production.Rainfed areas include 40% of rice, 66% of wheat, 82% of maize, and 86% of other coarse grains(Rosegrant et al. 2002). This book examines future technological and policy prospects for thesustainable intensification of rainfed upland maize production in Asia, in light of its increasingimportance, particularly within the regional food system. Rapid rural appraisal/participatoryrural appraisal (RRA/PRA) techniques were used in conducting village-level and farmer-groupsurveys to obtain a detailed characterization of upland maize production environments andsystems in China, India, Indonesia, Nepal, the Philippines, Thailand, and Vietnam. Surveyfindings, particularly farmer-identified constraints to maize production, complemented withexhaustive literature searches and relevant published and unpublished secondary data, were usedin country-level, national maize technology R&D priority-setting workshops. The maize R&Dpriority-setting methodology developed by Pingali and Pandey (2001) at CIMMYT was applied toderive R&D priorities for specific maize production environments and markets. In each countrymaize sector policy studies were conducted alongside the characterization surveys and prioritysettingworkshops.Across Asia, farmers growing maize in different agroecological environments, for eitherhousehold use or commercial markets, experience very similar biotic, abiotic, socioeconomic,and policy-related constraints to maize production. Among these constraints, first addressingthe problem of drought in the rainfed lowland-commercial, rainfed upland-semi-commercial,and commercial production environments will provide the highest technical returns to maizeR&D investments in Asia when contribution to regional maize production and share in regionalmaize area are considered. Drought is estimated to affect about 6.8 million ha just in these threeproduction environments (53.5% of total regional maize area), where the affected areas producean estimated 16 million tons of maize, and about 48 million rural poor are located. Alleviatingdrought in these environments is estimated to improve yield by about 35%, and could havean enormous impact on maize production in Asia. Drought also appears to have the highesteconomic impact in terms of estimated gross income loss. Semi-commercial and commercialmaize farmers in rainfed upland environments are estimated to lose at least PPP$ 450/ha(purchasing power parity, PPP) when drought conditions persist.Apart from drought, downy mildew, stem borers, leaf blight, stalk rot, soil erosion/landslides,soil micronutrient deficiency, waterlogging, lack of capital, poor agricultural extension/technology transfer services, and poor access to low-interest credit and to input and outputmarkets commonly figured in the priority constraints lists. Socioeconomic and policy-relatedconstraints were estimated to affect up to 180 million of the rural poor across Asia. Alleviatingthese constraints could improve maize productivity by at least 18%.While constraints relating to low market surplus and subsistence maize production systemsdo not appear in the priority lists, continuing to invest (albeit modestly) in subsistence farmingresearch remains important, particularly in Indonesia, Nepal, and the Philippines. In general, thepriority constraints list based on combined efficiency, poverty, and marginality indices closelyfollows the priority list based on a poverty index alone, and reflects the farmers’ sentiments thatsocioeconomic and policy-related constraints impact maize productivity more than technicalconstraints do.viii

The public and private sectors each have unique capabilities, resources, and comparativeadvantages that can contribute to alleviating constraints to maize productivity in Asia, andlinks between the two sectors appear to be expanding. These include international germplasmexchanges, public-private germplasm transfers, collaborative varietal testing networks, geneticimprovement, and crop/resource management; they also exploit mutual advantages that worktowards the advancement of maize cultivars. Hence, public/private sector alliances couldpromote spillovers of research results from high- to low-potential environments and fromeconomically advanced to economically deprived areas.Finally, it remains important to recognize that technology—from the simple to the advanced—isnot the only key to increasing productivity, improving the sustainability of intensified maizeproduction, and bettering the conditions of marginal maize farmers in Asia. No amountof advanced public- or private-sector maize technology research and development willhelp the most disadvantaged farmers unless substantial parallel investments are made inrural infrastructure, agricultural training and extension, input and output distribution andmarketing systems, and harvest and post-harvest facilities. Hence, the growing trend towardscommercializing and intensifying maize production requires a paradigm in agricultural policyformulation and research priority setting that is different from the staple food self-sufficiencyparadigm that has been the cornerstone of agricultural policy in most developing countries.Appropriate government policies could help alleviate many of the possible adverse consequencesof commercialization and promote sustainable intensification of maize production, especiallyin marginal environments inhabited by resource-poor subsistence farmers, the eternal targetbeneficiaries of agricultural R&D in Asia.ix

1. Introduction1.1 BackgroundOver the next 20 years, Asian policy-makers willconfront unprecedented growth in the demand formaize. Recent projections by the International FoodPolicy Research Institute (IFPRI) indicate that by 2020the demand for maize in all developing countrieswill overtake the demand for wheat and rice, withAsia accounting for nearly 60% of the global demandfor maize (IFPRI 2003). Throughout much of theregion, rapid economic growth and acceleratingurbanization are causing notable changes in foodconsumption patterns. The trend is most evidentin Southeast and East Asia, where traditional ricediets are becoming increasingly diversified, leadingto greater consumption of fruits and vegetables, aswell as of bread, meat, poultry, and dairy products.Growing demand for the latter is expected, in turn,to generate an explosion in the demand for maize,which will rise from 295 million tons (Mt) in 1997to 514 million tons in 2020. China alone is expectedto experience a 94% increase in demand for maizeover this period. According to the report on currenttrends and future projections of livestock supply anddemand, developing countries are in the midst of ademand-driven ”Livestock Revolution.” Livestockproduction and consumption of both meat and milkproducts are expected to grow about four timesfaster in developing countries than in developedcountries up to 2020. By 2020, developing countrieswill produce 60% of the world’s meat products, andAsia, led by China, will account for the major share ofthe increase in meat demand in developing countries.In fact, China alone will account for 43% (51 milliontons) of additional meat demand worldwide between1997 and 2020 (Delgado et al. 1999).In some Asian countries, rising export demandmay also complement rising domestic demand formaize. In the last decade, maize exports have becomemore competitive following currency devaluations,particularly in Indonesia and Thailand, wherethis has placed an additional strain on domesticproduction capacity.Projected increases in demand for maize in Asia willhave significant implications for the sustainability ofagriculture in Asia’s rainfed upland environmentswhere maize is largely grown, and for householdfood security, especially among the poorest of thepoor. Future rapid growth in demand for maize,whether to meet domestic food and feed requirementsor to satisfy export demand, is expected to lead tocrop substitution, intensification of lands alreadyplanted to maize, commercialization of existingmaize-based production systems, and expansion ofmaize cultivation into lands not currently farmed.These processes are most likely to be observed inagriculturally disadvantaged areas, including easternIndia, the outer islands of Indonesia, the mid-hillsof Nepal, the island of Mindanao in the Philippines,northern Thailand, and the Central and NorthernHighlands of Vietnam. The response of these areasto future growth in demand for maize needs to beunderstood in terms of changes in farming systems,evolving land use patterns, and increasinglydiversified income growth. The need for a holistic,system-wide understanding of the intensificationprocess is especially important given that marginalupland regions in Asia are frequently home to poorrural communities, many of them ethnic minoritiesfor whom maize is a primary food staple. The foodsecurity implications for poor households of the rise inmaize demand need to be understood and addressed.This project was designed to strengthen thecapacity of research managers and policy makersto understand and respond to the intensificationof maize-based farming systems in the rainfeduplands of seven Asian countries, by providing themwith comprehensive, accurate data on the currentstate of upland maize farming systems, as well asinformation on the options available for promotingsustainable improvements in maize productivitygrowth. To date, these systems have receivedrelatively little attention; research and development(R&D) efforts have understandably focused on theirrigated lowland zones where most surplus food istraditionally produced. As the capacity of these morefavorable high-potential zones is exhausted, and asdiets diversify out of cereals, the rainfed uplands1

will play an increasingly important role in feedingthe region’s rapidly growing populations. Given thefragility of many of these rainfed upland systems, itwill be important that the intensification process beguided in a way that will not only be sustainable, butthat will also improve the incomes and welfare of localpopulations while protecting the resource base uponwhich agriculture depends.1.2 ObjectivesThe overall goal of this project is to promote thesustainable intensification of maize production systemsin the Asian uplands 1 and enhance maize supplieswhile ensuring income growth and improved foodsecurity for poor households that subsist on maize. Thestudy has four specific objectives:• To develop in-depth knowledge of upland maizebasedfarming systems, identify constraints tofuture productivity growth, and anticipate potentialenvironmental consequences likely to result fromintensification;• To conduct country-specific maize technologyresearch, set up development plans for the uplands,and promote their implementation;• To help IFAD and Asian governments identifypriority development projects for the uplands,particularly projects that enhance food security whileprotecting the environment and reduce povertyby promoting sustainable improvements in maizeproduction practices; and• To examine and analyze within the context of overallfood policies, country-specific macroeconomic andtrade policies influencing the maize sector, with thegoal of conveying information about key investmentopportunities to achieve sustainable and equitablemaize productivity growth in the uplands.1.3 Key Activities, Methodologies,and PartnersFunded by the International Fund for AgriculturalDevelopment (IFAD) and managed by theInternational Maize and Wheat Improvement Center(CIMMYT), the project was implemented through amulti-country, multi-institution, multi-disciplinarycollaborative effort involving China, India, Indonesia,Nepal, Philippines, Thailand, and Vietnam. Selectedagriculturists and economists from these sevencountries were actively involved in the project.Researchers from CIMMYT, IFPRI, and StanfordUniversity provided conceptual and methodologicalsupport. Senior officials working in the NationalAgricultural Research and Extension Systems (NARES)and/or Ministries of Agriculture in these countriesstrongly supported the project.Key activities fall into three main components: (1)detailed characterization of upland maize-basedproduction systems and identification of constraints tomaize productivity growth in Asia; (2) developmentof country-specific maize technology research plansand identification of related policy and investmentopportunities; and (3) macro- and microeconomicanalyses of each country’s maize sector. Results of thethird component have been included in a separatevolume and will not be discussed at length in thispublication.Two other important activities are implicit to theproject. First is strengthening of NARES capacityto conduct rapid rural assessment (RRA) andparticipatory rural appraisal (PRA) surveys andundertake prioritization and policy research,and instructing NARES researchers on thesocioeconomic aspects of maize production. Secondis the development and maintenance of a powerfuldatabase that incorporates farm-level quantitativeand qualitative information plus secondary data fromnearly 300 locations across the seven participatingcountries.1.3.1 Detailed characterization of uplandmaize-based production systems andidentification of constraintsA detailed characterization of upland maize-basedproduction systems across Asia was conductedthrough a two-stage survey: first, a RRA surveywith key village informants and farmer-grouprespondents; and second, a PRA component, again,with farmer-group respondents. The RRA asked moreof the production input-output questions typical ofagricultural economic surveys, while the PRA wasa semi-structured open discussion, gathering morequalitative than quantitative information. RRA wasused as a quick, effective, and low-cost method ofcollecting data at the village or community level1While generally taken as “high lands, ground elevated above the meadows,” “uplands” in this project refers to land that is generally dry.Also, marginal lands usually suffer from insufficient rainfall; some lands are considered marginal for other reasons, e.g. excessive slope orhigh soil acidity, but lack of water is by far the main cause. A rainfed environment is marginal when the water-limited potential yield of acrop falls to less than 40% of its potential yield (CIMMYT 1989).2

about local farming systems, production technologies,constraints to technology adoption and use, productutilization, etc. It was conducted systematically, withmultidisciplinary teams interviewing respondentgroups in an active, open-ended yet guided questionand-answerforum. Survey respondents includeddifferent sub-groups within the community, forexample, large farmers, small farmers, landlesslaborers, and female-headed households.The PRA discussions elicited information neededfor evaluating the potential for and constraintsto intensification of maize production systems,anticipating likely consequences of intensification (e.g.,declining soil fertility, soil erosion), and identifyingtechnological, institutional, and infrastructuralconstraints to the sustainable enhancement of maizeproductivity growth. These discussions also gavea glimpse of farmer-experienced environmentalpressures brought about by intensified maizeproduction in major upland environments across Asia.Many of the upland maize-growing environmentsare home to large numbers of low-income farmers.Thus, Asian policy makers will have to confront howintensification of maize-based production systemsis likely to affect the poor in these environments,particularly with regard to food security. Informationregarding maize utilization patterns (food and feed)was also collected, with an eye to identifying foodsecurity risks faced by vulnerable groups withinthe community. To elicit the perspective of farmersfrom different economic backgrounds and gender,the interviews/surveys were conducted usingfour groups of farmer-respondents: a mixed socialgroup—one male group and one female group—anda poor-farmer group—one male group and onefemale group. Information collected using RRA/PRAmethods was supplemented by information collectedthrough a detailed review of available published andunpublished literature.Between November 2000 and February 2002, a totalof 265 locations (villages) in primarily marginalupland regions across seven countries were visited,and detailed maize production and resourceuse information was collected using RRA/PRAtechniques. Annex 1 lists the specific study sites ineach country. The dataset generated from these fieldsurveys is the first extensive and comprehensivedataset documenting the wide range of maize-basedproduction systems across Asia.1.3.2 Country-specific constraintprioritization and maize R&D plansThe detailed country-specific maize R&D plansdeveloped emphasize the sustainable intensificationof maize production in the uplands, with the objectiveof alleviating the constraints identified through theRRA/PRA surveys. Country-specific R&D planningmainly involved organizing a national maize researchpriority-setting workshop, with the participation ofsenior maize researchers, national maize programdirectors, and other stakeholders from both thepublic and private sectors. The workshop allowed thecountry project teams to present findings from theRRA/PRA surveys; to inventory current and potentialtechnologies for alleviating the identified constraints;as well as to help identify technologies currently notavailable in a country but that may be brought in fromabroad. The workshop ranked the proposed solutionsbased on their potential for alleviating the identifiedconstraints, and identified policies needed for the rapidpromotion and adoption of the proposed solutions.1.3.3 Maize sector policy analysis andidentification of key policy and investmentoptionsConducted in collaboration with IFPRI and StanfordUniversity, the policy study component used an openeconomy framework, examining country- and maizesector-specific macroeconomic and trade policieswithin the context of overall food policies, and thechanging global trade environment. For example,given increasing globalization and trade liberalization,it examined the equity implications of opening thecountry’s maize sector and the potential effect ofmacroeconomic policies (e.g., trade, interest rate, laborpolicies) on maize supply and prices.This component identified and prioritized keyinfrastructure investment opportunities for selectedupland locations in Asia. In the context of decliningpublic sector investment, the need and potential forinvestments in micro-irrigation, transport, storage,and community level seed production systems and forinvestments in adaptive research and extension efforts,including the potential role of NGOs, were studied.This information will be useful not only for nationalpolicy makers, but also for IFAD regional staff andportfolio managers.3

1.4 Organization of this BookThis book provides comprehensive and updateddocumentation on upland maize-based productionsystems across Asia, and synthesizes characterizationand maize R&D prioritization reports from sevencountries. (Six country reports containing moredetailed country-specific information have beenpublished as separate volumes.) The book’s generalframework for analysis and discussion is based on amatrix of maize production environments and marketorientation, delineating the predominantly commercialand subsistence maize production systems, constraints,and research and development priorities.Chapter 2 summarizes findings from the countrylevelmaize sector policy studies, beginning witha brief description of the macro-level trends ofmaize utilization, production, and trade, to providea perspective of the crop’s economic importancein the region. Chapter 3 gives a description of therange of maize production environments in Asia,discussing their respective biophysical, institutional,infrastructure, and socioeconomic characteristics.Chapter 4 discusses the farm-level environmentof maize production, from cropping calendar andvarieties grown to crop management practicesto harvest and post-harvest practices. Chapter 5synthesizes farmer-identified maize productionconstraints elicited during the RRA/PRA surveys.These include biotic, abiotic, institutional, and inputsupply constraints, as well as other socioeconomicand policy-related factors. Chapter 6 discusses thepriority constraints systematically elicited at theregional level, and relates these to specific maizeproduction environments and to relevant technologydevelopment and dissemination options. All thesesections are brought together in a summary andconclusions chapter (Chapter 7), which highlightsmaize R&D recommendations for widely varyingmaize production areas, with special emphasis onAsia’s marginal uplands.4

2. Maize Sector Policies in AsiaThis chapter summarizes the findings of countryspecificmaize sector policy studies (publishedunder one cover; see Gulati and Dixon 2007 fordetails), and attempts to relate farmer situations,particularly maize production constraints, to variousagricultural and trade policies. Each country’s maizesector policy report provides an overview of recenttrends in maize demand, supply, and trade; analyzesgovernment maize policies; measures incentives formaize producers using nominal protection coefficients(NPCs); and concludes with impacts of tradeliberalization on the prevalence of poverty.2.1 Brief Overview of MaizeEconomy Trends2.1.1 Shifts in demandThe impact of Asia’s population growth in the past20 years on food demand has been compounded byincome growth and urbanization, which together haveshifted consumer preferences towards higher-pricedcalories and animal products. Statistics show thatAsia’s per capita consumption of staple grains (suchas maize) has declined, while per capita consumptionof livestock, fish, and other animal products hasincreased. Total meat consumption alone in the sevenMaize utilization (mill tons)9075604530150FoodFeedSeed1961-70 1971-80 1981-90 1991-2002Figure 2.1. Mean utilization of maize as food, feed, andseed, Asia, 1961-2002.Source: FAOSTAT Database, Utilization Domain, accessed March 2005.project countries rose by 280% between 1980 and2000. This in turn rapidly increased domestic meatproduction, particularly of pork and poultry, whoseproduction grew much faster than the global average.Between 1961 and 2001, Asia’s pork production grewnearly 8% annually (compared to 3.3% for the world),and its share in total world production grew from 10%in 1961 to 52% in 2001. In the same period, poultryproduction increased 7% annually (compared to 5% forthe world), and its share in world production doubledfrom 12% to 24%.As a consequence, demand for maize as feed grainquadrupled from 39 Mt in 1980 to 93.5 Mt in 2000, atan average growth rate of 6.9% per annum (see Figure2.1). In Asia as a whole, 62% (up from 47% in 1980) oftotal maize supply is currently used to feed livestock,about 22% is used as food, and the rest goes to otheruses (industrial input, seed). Not surprisingly given itsenormous size, China uses more maize than any otherAsian country, mainly to feed livestock. During 2000-02 in Asia, feed use of maize was highest in Thailand,where livestock producers used nearly 89% of allmaize (Table 2.1).The use of maize as an industrial input to produce foodand non-food products has also increased, althoughless dramatically. Food use of maize (includingindustrially processed food maize) rose from 23.0 Mtper year during the 1960s to 39.6 Mt per year duringTable 2.1. Maize utilization, Asia, 2000-2002.Human Proportion of totalper capita supply used forUtilization consumption(Mt) (kg) Food Feed OtherChina 116.6 14.9 16.5 67.7 15.8India 12.1 4.2 34.2 43.3 22.5Indonesia 10.7 35.4 68.7 5.2 26.1Nepal 1.5 42.1 68.4 17.7 13.9Philippines 4.8 5.2 8.5 79.7 11.8Thailand 4.3 6.2 9.1 88.8 2.1Vietnam 2.4 8.2 25.7 66.0 8.3Asia 152.2 12.1 22.0 61.6 16.4Source: FAOSTAT Database, Utilization Domain, accessed March 2005.5

the 1980s and to 39.8 Mt per year during the 1990sand 2000s (see Figure 2.1); it was highest in Indonesia,where people consumed 68.7% of the maize supply(Table 2.1). More specifically, however, per capita fooduse of maize in the region declined from 18.5 kg/headin 1980 to 13 kg/head in 2000. The decline in per capitafood maize consumption in China, Nepal, and thePhilippines outweighed the slight increases registeredin India, Indonesia, Thailand, and Vietnam. Althoughit remains important as a food grain for many ruralpoor, maize has shifted away from its traditional roleas a staple food towards being a commercially tradedingredient in livestock and poultry feed.From the 1960s to the 1990s and 2000s, the food use ofmaize grew at 1.6% per year, well below the 7% averageannual growth in feed use. Seed use meanwhiledoubled from 1.3 Mt during the 1960s to 2.6 Mt duringthe 1990s and 2000s (Figure 2.1), but its proportion ofthe total annual maize supply declined from 3.5% to1.8% between the two periods.2.1.2 Production responseWorldwide, maize ranks first in terms of productionamong cereals, just ahead of wheat and significantlyahead of paddy rice. In developing economies, maizeranks first in Latin America and Africa, but third inAsia, after rice and wheat (FAO 2001). Globally, 612 Mtof maize were harvested annually from 2000 to 2003from about 140 million ha (M ha). 2 During the sameperiod, nearly 160 Mt of maize grain (26% of worldproduction) were harvested in Asian countries fromabout 43 M ha (about 31% of global maize area). Thislevel of global production represented a 23% increasefrom 1990-93, when about 497 Mt of maize wereproduced annually from 133.5 M ha. However, Asia’scontribution to worldwide maize production markedlydecreased between the two periods (from about 28%to 26%) even though there was a slight increase in itscontribution to worldwide harvested maize area (from29.5% to 31%).Across Asia, domestic maize production respondedto the rapid growth in domestic feed maize demand,mostly through yield improvement rather than areaexpansion (see Figure 2.2 and Tables 2.2 and 2.3). Thearea planted to maize increased from 27 M ha duringthe 1960s to 39 M ha in 1991-2004 (1% average annualgrowth). During the same period, average maize yieldsgrew at 3.1% per annum from 1.4 t/ha to 3.8 t/ha. Dueto the combined effects of area expansion and yieldgains, maize production almost quadrupled from 37Mt in the 1960s to around 145 Mt in 1991-2004 (4.2%average annual growth).2Source: FAOSTAT Database, Production Domain, accessed March 2005.In 2000-2003, China led Asia in maize production.Chinese farmers produced nearly 118 Mt of maizeon 24.3 M ha, achieving average yields of 4.8 t/ha(Table 2.4). India was second after China in maize areaharvested and production, followed by Indonesia.Nepal was the smallest maize producer, with Nepalesefarmers producing 1.5 Mt of maize on about 825,700ha and achieving average yields of 1.8 t/ha. Exceptin China, maize yields in Asia are still significantlylower than the 4.4 t/ha world average. Maize yields areespecially low in India, Nepal, and the Philippines at1.9, 1.8, and 1.8 t/ha, respectively (Table 2.4).2.1.3 Trends in tradeDuring 1961-2003, maize exports across the sevenAsian countries grew 5.3% per annum, while importsgrew faster, 6.6% per annum. Trade moved fromnet imports of 13.2 Mt in 1995 to net exports of 10.5Mt in 2003. Across all of Asia, however, maize tradeconsisted of net imports, growing from an average of3.7 Mt in the 1960s to 31.8 Mt in the 1990s and 2000s.Furthermore, Asia is projected to be a net importer of atleast 60 Mt of maize grain by 2025—an enormous 330%increase from the 1997 net import level (IFPRI 2003).Table 2.2. Average maize area, yield, and production,Asia.76543210Area (M ha) Yield (t/ha) Production (Mt)1961 – 1970 27.0 1.4 37.11971 – 1980 31.9 2.0 62.81981 – 1990 34.3 2.8 95.41991 – 2004 38.7 3.8 145.51961 – 2004 33.5 2.7 90.7Source: FAOSTAT Database, Production Domain, accessed March 2005.Note: Provisional 2004 data.Average annual growth (%)4.41.942.1YieldArea1961-70 1971-80 1981-90 1991-2002 1961-2004Figure 2.2. Area and yield contribution to maizeproduction growth in Asia, 1961-2004.Source: FAOSTAT Database, Production Domain, accessed March 2005.31.31.21.03.11.16

China went from being a major net importer in 1980(importing 4.6 Mt) to a major net exporter in 2000-03(exporting 4.8 Mt) and has overtaken Argentina tobecome the world’s second largest maize exporter.Similarly, India moved from being a net importer in the1990s to being a net exporter in the 2000s. Supportedby improved production technologies and favorableagricultural policies (market reforms), the growth inmaize production in these two countries exceededthat of domestic utilization and allowed the changein trade status. Thailand was the world’s fourthlargest maize exporter, exporting 2 Mt annually inthe 1990s, but these exports have gradually declinedas domestic maize demand has grown. Problems ofaflatoxin contamination also hurt Thai maize exportsin the mid-1990s (Ekasingh et al. 2004). Indonesia, thePhilippines, and Vietnam moved from being more orless self-sufficient to being net importers, as domesticfeed maize demand grew. In Nepal, maize trade isnot particularly important because self-sufficiencyobjectives dominate.Table 2.3. Growth (%) in maize area, yield, andproduction, Asia, 1961-2004.Area harvested Yield ProductionChina 1.3 3.4 4.6India 0.7 1.8 2.4Indonesia 0.6 3.2 3.8Nepal 2.0 (0.2) 1.8Philippines 0.6 2.6 3.2Thailand 2.5 1.5 4.0Vietnam 3.1 2.5 5.6Asia 1.1 3.1 4.2Source: FAOSTAT Database, Production Domain, accessed March 2005.Note: Provisional 2004 data.Table 2.4. Maize production and net imports, Asia,2000-2003.Area harvested Yield Production Net trade(M ha) (t/ha) (Mt) (000 t)China 24.3 4.8 117.9 4,844.0India 6.7 1.9 13.0 147.7Indonesia 3.3 3.1 10.2 (926.3)Nepal 0.8 1.8 1.5 (4.6)Philippines 2.5 1.8 4.6 (249.3)Thailand 1.2 3.7 4.3 118.7Vietnam 0.8 2.9 2.4 (165.0)Asia 39.7 3.9 154.0 4,781.6Source: FAOSTAT Database, Production Domain, accessed March 2005.Note: Some columns do not add up due to rounding.2.2 Maize Policy DevelopmentFollowing the situation in the United States, EuropeanUnion, and Mexico, 3 many Asian developing countrieshave attempted to protect their domestic maizeproduction. This section briefly describes the policyenvironment under which maize is produced andmarketed in Asia, particularly policies on production(output and input), international trade, food, andgeneral macroeconomy.2.2.1 Production policiesOutput policies. Self-sufficiency has in general beenthe guiding principle for agricultural policy in Asia.However, except in Thailand, maize has been relativelyneglected by policy, compared to rice and wheat. Thisis partly explained by maize’s reputation as a poorman’s crop (production- and consumption-wise),reflecting an urban bias in policymaking. Only the risein maize’s importance as feed pushed the livestocksector as a major force in maize policy considerations.Government procurement of crops at a floor price is themost common method of supporting farmers. Exceptin Thailand (and Indonesia during the 1980s), maizesupport prices have not played a significant role inAsian maize policy. Floor prices in China and India,the largest producers in the group, were commonlyfar below actual market prices, and so did not affectproducer decisions. China seems to be reversing itstrend of non-support for maize farmers in recent yearsby raising quota procurement prices, although at theexpense of halting market reforms (Huang and Rozelle2005). Support prices in Nepal, the Philippines, andVietnam have been negligible or nonexistent. Indonesiahad a floor and ceiling price program for maizefrom 1978 until the early 1990s. Thailand employs alimited price support program for maize farmers, butprocurement programs for maize have been generallydysfunctional in these countries.Government support for R&D is an alternativepathway to supporting domestic maize production(see Chapter 2, Gerpacio 2001). Various research,development, and extension programs for maize existin all the seven countries, with the primary goal ofgenerally promoting the adoption of high-yielding(often hybrid) maize varieties, along with increasedfertilizer use. Maize production inputs increasinglyinvolve the private sector, and remain importantvehicles for providing support to maize farmers.3In these countries, maize production has historically been heavilysubsidized, resulting in domestic prices being twice as high as worldprices. If all such market distortions were removed, the world maize pricewould likely rise, with subsidized countries cutting back output more thanothers would be able to expand theirs.7

Input policies. Fertilizer and seed represent the largestinput costs in maize production, and were the focusof most input subsidy programs across Asia in the1980s to late 1990s (Table 2.5). However, they led to: (1)domestic fertilizer and seed prices being much higherthan world market prices; (2) inefficiencies in domesticmarkets; (3) illegal trading (particularly in Indonesia);and (4) a substantial financial burden in nationalbudgets, especially during the 1997 Asian economiccrisis.Since then, however, input markets in the sevencountries have generally been liberalized. A smoothtransition to a market-based system saw fertilizerprices gradually decline. Numerous private companieshave entered the seed market, and pricing policieshave been altered, resulting in the gradual increase ofthe seed-to-grain price ratio. Particularly in Thailand,input subsidies have continued, but with increasedparticipation of farmer organizations (Ekasinghand Thong-ngam 2005). In Nepal, the governmentgradually eliminated fertilizer and price subsidiesin the late 1990s, but continues to provide transportsubsidies for fertilizer and seed sales in areasinaccessible by roads. In 1998, Indonesia abolishedfertilizer subsidies immediately following the Asianfinancial crisis, doubling fertilizer prices. Rice farmerswere compensated for the large production costincrease, but maize farmers received no such benefit.there has been a general trend towards reducingmaize trade barriers under the Uruguay RoundAgreements Act (URAA). 4 India, China, Thailand,and the Philippines implemented tariff-rate quotas(TRQs) 5 for maize under World Trade Organization(WTO) commitments. Nepal and Indonesia havetariff-only policies (Table 2.6). Recently, however,depressed international grain prices prompted severalgovernments to reverse the trend towards grain tradeliberalization.Table 2.6. Tariff policies for maize, Asia.TRQ In-quota Over-quota Implementation(000 tons) rate (%) rate (%) periodChina 7, 200 1 65 2004India 500 15 60 1999-2003Indonesia -- 5 -- 1998Nepal -- 10 1 -- --Philippines 217 35 50 1995-2000Thailand 55 20 76-80 1995-2004Vietnam n.a. n.a. n.a. n.a.Source: IFAD-CIMMYT Asia maize sector country studies, WTO as cited inWada et al. (2005).1Exemption of tariff granted to products from India and Tibet.n.a. – not available.2.2.2 International trade policiesUnlike developed countries, developing countriesdo not have the fiscal resources to support producersthrough domestic measures, and most protectionoccurs at the border. Across the seven countries understudy, most maize trade is government arranged, andbarriers have historically remained high. However,4The URAA implements the Uruguay Round of the General Agreement onTariffs and Trade (GATT), which includes an agreement on Trade RelatedAspects of Intellectual Property (TRIPS).5A tariff-rate quota is a quota for a volume of imports at a lower tariff.After the quota is reached, a higher tariff (over-quota tariff) is applied onadditional imports. In principle, a TRQ provides more market access toimports than a quota. In practice, however, many over-quota tariffs areprohibitively high and effectively exclude imports in excess of the quota(Skully 2001).Table 2.5. Summary of agricultural input policies, Asia.Country Agricultural input policies ReferenceChina Fertilizer production and distribution government controlled. Dual marketing system Huang and Rozelle 2005with in-quota and out-of-quota prices. Fertilizer subsidy at 52-77%. Seed subsidies.India Fertilizer, power, and irrigation subsidies. Restricted seed imports. Gulati and Narayanan 2005Indonesia Subsidized fertilizers for food crops. Suhariyanto 2005Nepal Fertilizer, seed, and fertilizer/seed transport subsidies. Paudyal 2005Philippines Subsidized fertilizers and seed (mostly yellow hybrid maize seed). Costales 2005Import restrictions on both these inputs.Thailand Subsidized production of both hybrid maize seed and inorganic fertilizers. Ekasingh and Thong-ngam 2005Subsidized 90% of hybrid maize seed cost to farmers during 1994-98.Vietnam Subsidized hybrid maize seeds produced by government agencies. Khiem et al 2005Subsidized credit for fertilizer imports.8

A highly restrictive policy regime insulated India’seconomy from global markets. The Indian governmentheavily regulated food grains and related products,banned maize exports in the 1960s and 1970s, and inthe 1980s allowed maize imports only through theFood Corporation of India. Later, when the demandfor poultry feed began to grow, the feed industry wasallowed to import maize without license, and maizeexports were allowed under a ceiling. In 1999, a TRQfor maize of 15% in-quota and 60% out-of-quota wasset (Table 2.6). Trade liberalization remains slow,however, and any move toward liberalizing externaltrade policy is usually accompanied by measuresdesigned to counter the effects of such a move.In China, food self-sufficiency goals led policymakersto restrain imports of land-intensive grains. Nearly20 years of policy reforms gradually changed China’sforeign trade regime from a highly centralized,planned, and import substitution-oriented regime toa more decentralized, market-oriented, and exportpromotingone. Trade and other policy reformssignificantly altered China’s trade composition infavor of products for which China has a comparativeadvantage. China’s maize trade liberalization,however, was fairly minimal prior to WTO accession(Huang and Rozelle 2005). Maize imports were limitedto those who had licenses and import quotas, and thecurrent maize trading system results in considerabledistortions and inefficiency.China’s WTO accession commitments includedecreasing overall agricultural import tariffs fromabout 21% in 2001 to 17% by 2004; a maize in-quotatariff of only 1%, and decreasing the maize out-ofquotatariff to 65%; increasing maize TRQ volumesfrom 5.7 Mt in 2002 to 7.2 Mt in 2004; and phasingout all maize export subsidies (on average, 34% ofthe export price) (Huang and Rozelle 2005). Afterthe removal of maize export subsidies, maize pricesinitially fell by 20%, but exports continued duringChina’s first year as a WTO member.Thailand’s self-sufficiency objective restricted maizeexports during 1961-81, but increasing domestic feedmaize demand led to re-instituting export quotasfrom 1992 to date. Following the 1995 WTO minimumaccess requirements, Thailand implemented TRQs formaize, which actually represented a step backwardfrom liberalization in the early 1990s. The tariff rateis now 20% and 76-80% for in-quota and over-quotamaize, respectively (Table 2.6); in-quota rates can beannounced on a year-by-year basis.After opening up its economy considerably sincethe late 1980s, when it was nearly closed, Vietnam’sstate agencies still control imports and exports, andtariffs on maize are high. Prior to 2000, quantitativerestrictions and licensing requirements controlledmaize imports. Under the ASEAN Free TradeAgreement (AFTA), however, Vietnam has committedto reducing tariffs to 20% by 2003 and 5% by 2006.Indonesia too is reducing its tariffs on a number ofagricultural products, including maize, to a maximumof 5% under an agreement with the IMF. The WTOagreements on maize in the Philippines meanwhilerequire that out-of-quota tariffs be reduced to 50%by 2004. The in-quota tariffs are, however, stillrelatively high at 35%. In Nepal, maize trade isrelatively unimportant; hence there are no qualitativerestrictions on agricultural trade, and no customsduty is applied to imports from India, Nepal’sdominant trading partner (Paudyal 2005).2.2.3 Macroeconomic policieson exchange rateExchange rate policies. Macroeconomic policiescan significantly influence overall incentives toagricultural producers. In fact, agriculture may bemore sensitive than other sectors to macroeconomicpolicies (Bautista and Valdes 1993). One of the mostimportant factors influencing maize incentives is thenation’s exchange rate policy (Beghin and Fang 2002).Overvalued exchange rates depress incentives todomestic producers of tradable goods, and increaseincentives for imports. The real depreciation ofdomestic currency raises the local currency pricesof tradable relative to non-tradable goods, and thuscontributes to the price-competitiveness of domesticexports.The seven Asian governments fixed their exchangerates in the early 1980s, as did most developingcountries. There was also a general bias againstagriculture, with overvalued exchange rates andhigher import controls on industrial products. Sincethen, there has been a general movement towardfloating exchange rates, although applied rates inthese countries are, to some degree, still overvalued.China highly regulated its foreign exchange marketthrough the mid-1990s, and maintained a two-tieredforeign exchange rate system before 1994. Whilepolicy determined the official exchange rate, tradingat a swap center determined the other rate. In 1994,liberalization united the two exchange rates to createa single managed exchange rate system, resultingin an official devaluation of the yuan of over 50%.This rate has stayed remarkably constant since then,although it has depreciated slightly in recent years.It is widely believed, however, that the domesticcurrency has gradually become overvalued since1994, and as such has provided a disincentive to thetradable agricultural sector.9

In India, the real exchange rate declined by 53%between 1985 and 1991. A further devaluation in July1991 accompanied substantial liberalization of themanufacturing sector, and may have significantlyreduced the anti-agriculture bias in the country(Narayanan and Gulati 2005). Similar patterns ofgradual devaluation were observed in the other Asiancountries. The Philippine peso was overvalued priorto having a flexible exchange rate policy after the fallof the Marcos regime in 1986. The official value ofthe peso depreciated by 274% in 1985-2000, duringwhich a market exchange rate system prevailed, withperiodic Central Bank interventions in times of highfluctuation. Similarly, Vietnam’s exchange rate policysince 1990 involved making frequent adjustments tothe market rate. Nepal switched from a fixed exchangerate system to a floating system in 1983, resultingin substantial devaluation of the Nepali rupeeduring 1985-2001. Because India is Nepal’s largesttrading partner, Nepal maintained a special fixedexchange rate with the Indian rupee, and incentives toagricultural producers have remained fairly unaffectedby the depreciation of the Nepali rupee relative toother foreign currencies (Paudyal 2005).The Asian financial crisis. The 1997 Asian financialcrisis was a major factor that led to substantialdevaluation of most currencies in the region, hittingThailand and Indonesia especially hard. Prior to thecrisis, Thailand had a fixed exchange rate system, andthe prevailing rate had been 25 baht/US$ from the late1980s to 1997. In the 1990s, the overvalued exchangerate led to serious balance of payment deficit andcapital outflows, which contributed to the economiccrisis (Ekasingh and Thong-ngam 2005). In 2001-2002,Thailand adopted a managed floating exchange ratepolicy, which devaluated the baht to 40-45 baht/US$.In Indonesia, the exchange rate was tightly managedand quite stable before the economic crisis. The Asianfinancial crisis threw the economy into completedisarray, and the value of Indonesian rupiahplummeted upon the adoption of a floating exchangerate in August 1997. Within a year, the rupiah’s valuefell by over 80%, the largest devaluation in Asia inthat period. Prices of tradable commodities soared,resulting in economic hardship and social unrest. Theexchange rate has since then fluctuated considerably.The exchange rates in India and the Philippines werealso significantly affected by the Asian financial crisis.During 1997-2001, official exchange rates in thosecountries depreciated by 30% and 116%, respectively.In contrast, official exchange rates in China, Vietnam,and Nepal were largely unaffected.2.3 Impact of Policies on MaizeProductionThe net impact of government policies on maizeproduction (as incentives to maize producers) canbe measured by examining the difference betweendomestic prices and their corresponding internationalprices (the prices that farmers would have receivedunder a counterfactual free-trade situation), afteradjusting for transport and other costs. The nominalprotection coefficient (NPC) is a straightforward measureof this divergence between domestic and international(or reference) prices, and defined for maize as follows:NPC maize= (P d maize - Pw maize E) / Pw maize EwhereP d is the domestic price in local currency;maizeP w is the world reference price in US dollars; andmaizeE is the exchange rate (local currency/US dollar).If the NPC is larger than 1, the domestic price is greaterthan the world reference price, implying that maize isprotected by domestic policies, and that liberalizationwould bring greater imports, pushing down thedomestic price of maize. If the opposite is true, thenmaize is implicitly taxed or disprotected, and exportswould likely increase with liberalization. An NPCof near 1 implies a neutral protection structure, andproducers face domestic prices comparable to borderprices.Nominal protection coefficients are best examinedunder importable and exportable hypotheses, usingofficial and shadow exchange rates. When domesticallyproduced maize competes with imports (under theimportable hypothesis), international transport costsprovide a degree of protection for domestic maizeproducers. In contrast, if maize is an exportable item,the domestic producers’ price must be low enough tomake their maize competitive in foreign markets afterincluding the cost of transportation to that market.2.3.1 NPC analysis using official exchangeratesThe importable hypothesis is probably most relevantfor India, Indonesia, the Philippines, and Vietnam.The Philippines registered the highest NPCs formaize in Asia, ranging from 1.22 in 1988/89 to 1.69in 1996/97, which implies that government policieshave systematically and heavily protected domesticmaize producers. In Indonesia NPCs have declined10

since the 1980s, punctuated by a major fluctuationduring the Asian financial crisis (1.22 in 1985/86 to0.56 in 1997/98). NPCs for maize are presently verynear 1 (0.88 in 2000/01), indicating an overall lackof protection, or disprotection, to maize farmers inIndonesia. Government decisions to adopt a floatingexchange rate policy, reduce import tariff rates,and abolish the fertilizer subsidy, combined withthe decline of maize prices relative to other foodcrops, proved detrimental to maize farmers. Maizecultivation may hence shift to more marginal lowpotentiallands, as farmers reserve their productivelands for more profitable crops.Around 1990, India’s NPC for maize under theimportable hypothesis dropped to 0.86. In a relativelyshort time span, maize went from being highlyprotected (1.47 in 1987/88) to being implicitly taxed.Maize has thus become an efficient import substitutesince the end of the 1980s. India’s domestic maizeproducers, however, are non-competitive with regardto exports; Indian maize cannot be exported at currentlevels of prices and technology. To emerge as anefficient maize exporter, India would need to augmentyields, and reduce costs and prices. Higher priority toR&D, and improved seed production and distributionmay be the key to achieving this. Nepal is in a similarsituation: maize NPCs were below 1 during most ofthe 1990s under the importable hypothesis, but maizeproducers have been consistently non-competitivein the export market. Unless substantial investmentin maize technology generation and disseminationis made to improve resource productivity, Nepalesemaize cannot compete on the export market.In Thailand, maize producers are competitive underthe importable hypothesis, but have consistentlybeen protected under the exportable hypothesis. Thegovernment has always set maize price supports andactual import tariff rates to bring domestic prices inline with market (and world) prices. In China, maizehas been heavily taxed by procurement quotas, andNPCs for maize under both the importable andexportable hypotheses were much lower than 1 until1997/98. With the provision of subsidies and thereduction of procurement quotas, China has begunto protect its maize farmers in recent years. Underthe exportable hypothesis, only China approachescompetitiveness.Table 2.7 summarizes the NPCs for maize in Asia,calculated using official and shadow exchange ratesunder both the importable and exportable hypotheses.2.3.2 NPC analysis using shadowexchange ratesScenario analyses using social (shadow) exchange ratesshowed that NPCs for maize are significantly lowerthan those calculated using official exchange rates–45%lower, on average, in the importable hypothesis. Thesituation has gradually changed since the early 1990s,as exchange rate regimes have generally moved closerto market determination. Analyses also showed that theovervaluation of domestic currency creates the largestdisincentive to domestic maize production in Asia.In China, the net effect of all policies—includingmacroeconomic policies—has been heavy implicittaxation to the maize sector under both the importableand exportable hypotheses. Overvaluation of domesticcurrency in China caused much lower NPCs for maize(calculated under the shadow rate) until 1997, when theimpact of overvaluation became fairly small. Exchangerate policy represented a heavy tax on maize producersuntil this time. With further liberalization of exchangerate policies, maize prices may be expected to increasein China.Similarly, in India, overvaluation of the rupee resultedin moderately lower NPCs under the shadow exchangerate until the early 1990s. Since then, maize NPCs inIndia have been between 0.7 and 0.9 for both officialand social exchange rates under the importablehypothesis.Table 2.7. Summary of nominal protection coefficients (NPCs) for maize, Asia, 1997-2001.Using official exchange ratesUsing shadow exchange ratesImportable hypothesis Exportable hypothesis Importable hypothesis Exportable hypothesisChina Undistorted Subsidized Heavily taxed TaxedIndia Heavily taxed Subsidized Heavily taxed SubsidizedIndonesia Taxed Subsidized n.a. n.a.Nepal Undistorted Heavily subsidized Undistorted Heavily subsidizedPhilippines Heavily subsidized Heavily subsidized Heavily subsidized Heavily subsidizedThailand Undistorted Heavily subsidized Undistorted Heavily subsidizedVietnam Heavily subsidized Heavily subsidized n.a. n.a.Undistorted = NPCs between 0.95 and 1.05; Taxed = between 0.80 and 0.94; Heavily taxed = less than 0.80; Subsidized = between 1.06and 1.20; Heavily subsidized = above 1.20.n.a. – Data not available.11

The Philippine peso, meanwhile, was overvaluedbefore 1996/97, but has been undervalued since then,with significant impacts on production incentiveindicators. When the social exchange rate is used,NPCs were on average 18% lower during earlierperiods and 51% higher since 1996/97. It remains,however, that the Philippine maize sector has beenheavily protected, and that maize prices may decreaseif the official exchange rate approaches marketequilibrium.The impact of exchange rates on incentive indicatorsin Nepal and Thailand was relatively minor comparedto the other project countries. The effect in Nepal wasconsistently small, averaging 4% higher with the socialexchange rate during the study period owing to aslightly undervalued domestic currency. In Thailand,the major impact occurred mostly between 1991/92and 1995/96.2.3.3 What do they mean?The policy study indicates that maize producersin the Philippines and Vietnam, of all the Asiancountries examined, would be especially hurt bycomplete market liberalization. These countries,already moderate importers of maize, have protectedtheir domestic producers substantially through acombination of strict import controls and inputsubsidies. With market liberalization, Filipino andVietnamese maize producers would be largely unableto compete with cheap imports. Apart from supportivegovernment policies, substantial investment in maizetechnology generation and dissemination to improveproductivity and profitability will be needed to bringdown domestic prices of Filipino and Vietnamesemaize. In the Vietnamese uplands, where the maizearea recently expanded, the challenge will be toimprove the marketing system and infrastructure,and to disseminate and promote farmer adoption ofsustainable maize-based production systems (Khiem etal. 2005). In the Philippines, substantial improvementof rural roads, particularly in predominantly whitemaize-producing regions, would simultaneouslybetter farmers’ access to markets and institutions,which would, in turn, enhance productivity andcompetitiveness (Costales 2005).Indonesia is a maize importer with relatively fewmarket distortions; India and Thailand import andexport comparable quantities of maize, resulting innear zero net trade. Domestic maize production inthese countries appear to be fairly competitive as animport substitute, but not as an export commodity.As such, it pays for maize production in India toconcentrate on domestic markets, especially sincedemand from the food processing and feed industriesis likely to increase dramatically. Given the currentframework of self-sufficiency, if all maize demand isto be met through domestic production, then it has tocome from yield increases rather than from maize areaexpansion. Yield increases would have to come froma wide range of improved production technologiesas well as a whole gamut of government policies onagricultural production and marketing (Narayananand Gulati 2005).Similarly, in Thailand government policies andprograms related to agricultural credit, research, andextension services need to be strengthened to be ableto address the emerging problems of maize farmers.Government policies have tended to apply mostly tofarmers with good access to resources, markets, andinformation. It is time for resource-poor maize farmersto benefit more from government policies. Meanwhile,Nepal’s maize producers have had poor incentivesand are competitive as producers of import substitutes(though not as exporters). Since Nepal does not engagein much cross-border trade, and India is nearly its onlytrading partner, domestic maize producers are unlikelyto be affected much by market liberalization, unless thespecial exchange rate between the Nepalese rupee andthe Indian rupee changes significantly.Domestic producers in China, the world’s secondlargestmaize exporter, would be quite competitivewith imports, and nearly competitive in exports,assuming no change in the official exchange rate. Theovervalued exchange rate, however, has served as animplicit tax on producers (along with governmentprocurement policies), and devaluation of the yuanwould significantly benefit the maize sector.Given the uncertainties involved, it appears that whilesome maize producers in Asia may benefit from tradeliberalization, numerous others will face strongercompetition that may threaten their livelihoods.Food grain self-sufficiency policies have traditionallyfavored rice and wheat producers. To help alleviatethe economic hardship of maize producers, thispolicy priority must change. Economic liberalizationcombined with exchange rate reforms, ruralinfrastructure improvements, better market access formaize producers, and increased investment in research,development, and extension programs would go along way towards easing the transition of Asian maizeproducers to a globalized economy.12

3. Maize Agro-ecosystems in Asia3.1 Maize Production EnvironmentsMaize (Zea mays L.) is a versatile crop that adaptseasily to a wide range of production environments.Maize grows at latitudes ranging from the equator toslightly above 50 o North and South, from sea level toover 3,000 meters above sea level (masl), under heavyrainfall and in semi-arid conditions, in temperate andtropical climates. The maize growing cycle can rangefrom three months to more than a year (Dowswell etal. 1996). Maximum grain yields have been recordedin locations where temperatures reach 30-32 o C duringthe day and drop to 11-18 o C at night, but the crop cantolerate wide deviations from this ideal temperaturerange (Dayanand 1998).There is no universally recognized system forclassifying maize production environments. CIMMYT,which holds a global mandate for maize improvementin developing countries, has developed a classificationsystem based on the concept of mega-environments(CIMMYT Maize Program 1988a, Hartkamp et al.2000). 6 CIMMYT maize breeders commonly distinguishfour major mega-environments: the lowland tropics(or tropical lowlands), tropical highlands, subtropics/mid-altitude zones, and temperate zones (Table 3.1).These classifications do not correspond precisely togeographic definitions of the tropics, subtropics, andtemperate regions, but rather are based on agroclimaticcriteria that include minimum and maximum meantemperatures during the growing season, elevation,and, to a lesser extent, latitude (Dowswell et al. 1996).Steps are being taken to refine this classification toalso consider, among other factors, adaptability ofCIMMYT germplasm and incidence of maize insectpests and diseases.Across all developing countries and in Asia, maizeis grown mainly in tropical lowland and temperateenvironments. However, while tropical lowlandproduction environments are found in all theworld’s regions and in all Asian countries, temperateproduction environments are found mostly in EastAsia (87%), particularly in northern China (Table 3.1;CIMMYT 1988b as cited in Dowswell et al. 1996; Vasal1998). Within a given country, maize production maybe concentrated within a single mega-environmentTable 3.1. Major maize mega-environments.Tropical lowlands Tropical highlands Subtropics/mid-altitude zones Temperate zonesGeneral characteristicsLatitude 0-25º North and South 0-25º North and South 26º-36º North and South > 36º North and SouthElevation (masl) < 1,000 > 1,800 1,000 – 1,500 All elevationsDay length (hours) 11 – 12.5 11 – 12.5 12.5 – 13.4 > 13.4Mean temperature (ºC) > 24 < 18 18 – 24Description Largely high humidity, Equatorial highlands, Typically less than Highest latitude regionsrainfed systems. Includes typically over 1,800 masl. Usually where maize productionsome winter season 2,000 masl. rainfed but with large is possible.regions at higher latitudes.variation in rainfall.% of total maize area, all developing countries in Asia, late 1990sIncluding China 35.4 2.5 11.2 50.8Excluding China 53.6 3.8 16.9 25.6Source: Vasal (1998); CIMMYT (1989); Hartkamp et al. (2000).6A mega-environment (ME) is a broad, not necessarily contiguous production area, occurring in more than one country and frequently trans-continental,delineated by certain ecological conditions (e.g., temperature, rainfall, soils), crop characteristics (e.g., maturity cycle, grain color, grain texture), biotic andabiotic constraints, and socio-economic factors (e.g., production systems, cropping patterns, consumer preferences). MEs are also defined as the largestsubunits of a crop’s growing or target environment within which a particular variety or related practice is useful. Hartkamp et al. (2000) present a GISbasedapproach by CIMMYT to defining mega-environments for maize research.13

(as in Indonesia, Philippines, and Thailand, wheremaize is grown mainly in tropical lowland zones), orit may be distributed across several different megaenvironments(as in China, where maize is grown inlowland tropical, subtropical/mid-altitude, tropicalhighland, and temperate zones).The definition of environments, however, is subjective.For example, “mid-altitude” and “subtropical”are sometimes used interchangeably wheregermplasm requirements are similar, so results arenot easily reproducible, and the two types of megaenvironmentsare not well differentiated (Hartkampet al. 2000). A similar situation holds for the use ofelevation: perceptions of “lowland,” “mid-altitude,”and “highland” can vary among maize researchers,especially across regions and where countries haveestablished their own classifications. For example,India’s “plains” mistakenly give the impression ofbelonging to a lowland tropical environment, butupon closer scrutiny are more accurately part of asubtropical/mid-altitude environment, whereasthe “scarce rainfall zone” and “dry zone” of Indiado in fact fit into the lowland tropics classification.Unfortunately, updating the definitions of megaenvironmentshas been sketchy for lack of resources todo a more complete, methodical revision.This book concentrates on maize grown in tropicallowland and subtropical environments in Asia’sdeveloping countries, because these are the areaswhere most maize in the region is grown. It alsoincludes subtropical/mid-altitude maize productionareas in the region, as well as large temperatemaize production areas in countries such as China.Annex 1 shows the country and mega-environmentdistribution of villages surveyed in Asia.This book adapts the Asian paddy productionenvironment classification, wherein tropical lowlandenvironments can be subdivided into three morespecific agroecological zones, namely irrigatedlowlands, rainfed lowlands, and rainfed uplands (IRRI1984). In irrigated lowlands crops (usually rice) aregrown under irrigation in fields with well-maintaineddikes that can hold 30 cm of water. Availability ofirrigation water may be year-round or seasonal, andother non-rice crops, particularly high-value crops,may be grown under irrigation in rice-based patterns.In rainfed lowlands rainfed crops are grown onpuddled soil in fields bounded by dikes that pondwater to about 30 cm. Water depth seldom exceeds30 cm. Depending on the amount and distributionof rainfall over the year, one or two puddled cropsare grown. These areas receive little or no irrigation,and non-rice crops may be grown before or after therainfed rice crop. Rainfed uplands depend on rainfallfor crop production, but are not flooded. The runoffand infiltration of rainwater is so high that waterdoes not accumulate on the land. Most of these aresloping lands or lands with higher elevation than thesurrounding areas. Upland rice varieties are grown infields prepared and seeded under dry conditions, andnon-rice upland crops, such as maize, are importantand grown in the wet and dry seasons.At the country level, the districts and villagessurveyed had both large maize areas and highvolumes of maize produced (i.e., they had a significantshare of the district’s total maize area and production);also surveyed were districts and villages representingdifferent maize production environments and maizebasedproduction systems. Special attention was givento proper representation of marginal, less favorablemaize production environments. Surveyed villagesalso provided variation in household utilization ofmaize. For the purpose of the analysis, surveyedvillages were classified as low market surplus (orsubsistence), semi-commercial, and commercial,based on percentages of household utilization ofmaize as food, feed, and source of income throughsales. 7 Annex 1 shows the salient characteristics of thesurveyed villages, and Annex 2 classifies the specificsurvey sites within the matrix of maize productionenvironments and market orientation on whichsubsequent discussions in this book are based. Thefollowing sections describe and discuss the findingsand observations from the RRA/PRA surveys.3.2 Biophysical EnvironmentAnnex 3 describes the agro-climatic features of themaize growing areas surveyed and demonstrates theversatility of maize, i.e., its ability to flourish in justabout any environment. Maize can grow on the flatlands and plains of Indonesia and Vietnam, as wellas in the hills and steep slopes (up to 2,500 masl) inNepal, the Philippines, and Thailand. In India, maizeis grown in a wide range of environments, fromextreme semi-arid to sub-humid and humid regions,as well as in the low and mid-hills of the westernand northeastern regions. Even excluding temperategrowing environments, maize growing environmentsin China vary considerably in terms of rainfall,temperature, and topography over the tropicallowland and subtropical/mid-altitude regions.7Low market surplus (or subsistence) villages were defined as thosewhere the majority of households retained 70% or more of maizeoutput for household food and/or feed purposes; semi-commercialvillages were defined as those where most households retainedbetween 26% and 69% for household food/feed use; and commercialvillages were defined as those where most households retained 0 to25% for household food/feed use.14

3.2.1 RainfallIn most tropical environments, maize requires600-700 mm of moisture well distributed over thegrowing season. Annual rainfall in all study sites(with the exception of some areas of Karnataka andAndhra Pradesh, India) appears to be sufficientfor maize production and can support two or threecrops per year, but seasonal distribution regulatesplanting. Maize farmers who are partially or totallydependent on rainfall adjust the cropping calendarto coincide with months when precipitation willbe sufficient to supply crop water requirements.Supplementary water sources, such as wells, naturalstreams, reservoirs, and deep wells, can also befound. Supplementary water, however, is not usuallyused for irrigating field crops but rather for limitedvegetable and cash crop production. As an example,Figures 3.1 and 3.2, respectively, show how Filipinoand Indonesian farmers in low market surplus andcommercial maize-growing areas time their cropaccording to long-term rainfall distribution. Rainfalldistribution across growing seasons by megaenvironmentin China is presented in Figure 3.3.In the tropical rainfed lowland/rainfed uplandareas of Nusa Tenggara Barat, Indonesia, MindoroOccidental, Philippines, and Phichit and ChiangRai, Thailand, rice is grown during the wet season,followed by a maize planting supported by residualsoil moisture and supplementary irrigation, if sowingis timed properly. In these areas, maize is sownin October/November and harvested in March.In tropical lowland environments of China, two(spring and fall) or three (spring, fall, and winter)maize crops are possible, although in many areasdrought and competing crops have limited maizeproduction to spring. Spring and summer maizemm300mm250200250200150100500400350Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb150100500Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul AugMaizeMaizeMaizeCottonFigure 3.2. Monthly rainfall and maize-based croppingpatterns in the drylands of Jeneponto, South Sulawesi,Indonesia.Source: Swastika et al. (2004).250200150100500Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov DecFigure 3.1. Long-term rainfall distribution and maizecropping season in Cebu, Eastern Visayas (low marketsurplus production, top), and Bukidnon, NorthernMindanao (commercial production, bottom), Philippines.Rainfall data: Cebu, Eastern Visayas – 1971-2000; Bukidnon, NorthernMindanao – 1961-2000.Source of rainfall data: PAGASA.mm8007006005004003002001000Tropical LowlandSpring plantingSpring floweringSpring harvestingFall plantingFall floweringFall harvestingSummer plantingSummer floweringSummer harvestingSubtropical-midaltitudeFigure 3.3. Seasonal rainfall distribution in China bymega-environment.Source: IFAD-CIMMYT-CCAP RRA/PRA surveys, 2001-2002.15

crops are possible in some subtropical/mid-altitudegrowing environments. Meanwhile, annual rainfallin the subtropical/mid-altitude mid- and high-hillenvironments of Nepal allows only one crop of maize,often sown in late March/early April and harvested inSeptember/October.Maize can be affected by dry periods at critical growthstages in years when the dry season comes earlier thanexpected. Karnataka, India, has the lowest annualaverage rainfall (about 600 mm) among the study sites,and maize is grown only once a year, from May toSeptember. South Cotabato in Mindanao, Philippines,has one of the lowest levels of annual average rainfallreported (916 mm), yet three maize crops are planted.Rainfall in this province/region is lowest during theplanting of the first maize crop and highest duringthe third maize cropping. It helps that tropical stormsare a low risk factor in this southern island of thePhilippines.3.2.2 TemperatureA high mean temperature (around 28 o C) during thegrowing season generally characterizes tropical maizegrowing environments, with a high mean minimumtemperature (around 22 o C) and a high mean maximumtemperature (around 32 o C). Across the study sites,the range of air temperatures was widest in Indiaand relatively limited in Indonesia, Philippines, andThailand. In India, air temperatures can range from7-36 o C (in Bihar and Madhya Pradesh) to 2-46 o C (inRajasthan). Air temperature averaged 24-28 o C in themaize growing areas of the Philippines, and 25-27 o Cin Thailand, with March/April as the hottest monthsand December/January as the coolest months.Temperatures in China averaged 24.6 o C during thespring maize growing season and 26 o C during thefall growing season. In general, temperature variationduring the maize cropping season does not criticallyaffect the maize crop as much as rainfall variation.3.2.3 Types of soilAcross the survey areas, secondary data describingsoil characteristics were in general limited. Parent ororigin materials of soil were shown in soil maps, butdata on the current status of soil fertility, degradation,or management requirements in research sites werelimited. Primary soil data were derived almost entirelyfrom farmer interviews during the RRA/PRA surveys.In each country, farmer-respondents were asked todescribe the soil types in their farms and to assess theadvantages and disadvantages of each type based onfertility level, drainage characteristics, susceptibilityto erosion, and suitability for crop production or otheragricultural uses. The most common soil types in thesurvey sites are clay, clay loam, and sandy loam soils,consistently described by farmers as the most fertilesoils, generally suited to most crops (Annex 4).Farmer assessments of soils included the followingobservations:• Clay soils have good water-holding capacity and soilfertility, but can be wet/muddy and difficult to plowduring the wet season;• Clay loam soils retain water better and are suitableespecially for growing cereals, but drain poorly andare prone to soil erosion;• Sandy loams do not become waterlogged but aremore susceptible to drought and have poor soilfertility; and• Sandy clay loams are good for crop production andpasture, but susceptible to soil erosion.3.3 Institutional Environment3.3.1 Line agenciesAcross the region, government extension offices,mainly Departments (Ministries) of Agriculture (DA)and Departments of Agricultural Extension (DAE),and private seed companies play a vital role indisseminating technology information to farmers inmaize producing villages. Department of Agriculturetechnicians are responsible for conducting as regularlyas possible farmers’ field schools and trainingsessions to provide farmers with updated maizecropping technology. Seed and fertilizer companyfield technicians also provide information to maizefarmers, although generally limited to the productsbeing marketed by the company. Non-governmentorganizations (NGOs), nearby agricultural stateuniversities and colleges (SUCs), neighbors, and otherfarmers within the community share knowledgeand are seen as valuable sources of information.Television and radio programs also provide technologyinformation to farmers in China. In some countries(e.g., the Philippines and Vietnam), however, little orno extension services are provided by SUCs, and NGOsand international agricultural research centers (IARCs)have not yet reached many of the surveyed areas.3.3.2 Farmer cooperatives and user groupsFarmers in many parts of Asia realize the benefit ofworking together and have established formal andinformal groups for numerous objectives. Informal16

groups are based on mutual understanding only,community based, belong to the same faith, havespecific traditions, and share the same natural resources.Formal user groups, producer/marketing groups, andcooperatives reportedly have also been organized in thesurveyed villages with help from government or nongovernmentorganizations.In Nepal’s mid- and high–hills, for example, theexistence of forest, drinking water, and irrigationuser groups was reported. Vegetable and milkproducer/marketing groups are also present inrelatively accessible villages. In the Philippines, RuralImprovement Councils (RICs) provide technical,financial, and livelihood assistance to their membersin the form of livestock dissemination programs, smallhandiwork business assistance, and help in establishingsmall consumer (sari-sari) stores. In Lampung province,Indonesia, farmers organized themselves into groupsof 20-30 members; each group made a common maizecultivation plan covering the varieties to be grown,planting time, and level of fertilizer usage. In Thailand,farmer cooperatives often do not deal with outputmarketing, because individual farmers sell theirproduction through extensive networks of privatemerchants. There are several associations that facilitatethe management of water storage tanks and irrigationfor rice production in tropical lowland environmentsin China. Farmer associations for transplanting riceseedlings and for coordinated sales of agriculturalproducts such as fruit, ginger, and fish were alsodescribed. However, no associations related to maizeproduction or sale were reported.Across the region, women’s, youth, and churchorganizations are also active in running livelihood,leadership, and religious programs and seminars intheir respective communities. During the RRA/PRAsurveys, farmer-respondents rated most communityorganizations as helpful and performing satisfactorily,although several were reported to be ineffective andinactive.3.3.3 Sources of material inputsIn general maize farmers in Asia obtain material inputs(seeds, fertilizers, and pesticides) from the manyagricultural input outlets in the countryside and nearbycities. However, maize farmers are frequently unable topay cash for production inputs and instead obtain theirsupplies from private trader-financiers on a charge-tocropcredit arrangement.Concerned government agencies also play a role inensuring that maize seed and fertilizers are available tofarmers, sometimes free but more often at discountedor subsidized prices through agricultural productionintensification programs. When asked about problemsrelated to material inputs, a large number of farmersexpressed dissatisfaction with the high prices of seedand fertilizers purchased from private traders, whileothers complained that maize seeds supplied bygovernment programs often had poor germination,poor field performance, and low yields. Qualityproblems, including adulteration, were often reportedwith respect to fertilizers and other chemicals.3.3.4 Credit institutionsFarmers need not only appropriate agriculturaltechnologies, but also sufficient capital to purchasethem. A large percentage of farmer-respondents statedthat they lack the resources to support their farmoperations and thus depend heavily on borrowedcapital. Very few (e.g., 2% in the Philippines and 13%in Thailand) have the money to completely cover theircrop production costs. Many farmers are forced toborrow from private moneylenders or trader-financierson a charge-to-crop scheme with extremely highinterest rates (reported to be 3-5% a month in Thailandand 10-20% in the Philippines).In a charge-to-crop scheme, lenders (often privatetraders) sell agricultural inputs to farmers on credit butat higher than market prices, and later buy back theharvests at lower than market prices. In this scheme,loans for agricultural inputs, as well as interest due forone cropping season, are deducted from the total valueof the harvest sold to the lender. Farmers continue tosupport this seemingly costly and unfair arrangementbecause they find using the private trader-financiersmore accessible and/or convenient than going toformal credit institutions. These private loans do notrequire collateral and do not have the same credit limitsimposed by many formal credit institutions. Moreover,the process is less time-consuming, and private traderfinanciersare generally available at all times. Farmerrespondentsalso stated that they had to continue usingthe services of private trader-financiers because theincome from a harvest is usually insufficient to pay forloans taken during the previous cropping season.Formal credit programs are also available from somegovernment and private banks, yet very few farmerstake advantage of them because they find the paperwork too tedious and the requirements (especiallycollateral) prohibitive. Utilization of rural creditcooperatives varied across survey sites in China,but funds obtained were not necessarily targeted foragricultural production. An important source of creditacross all survey sites was informal credit obtainedfrom relatives, friends, other farmers, women’sassociations, or farmers’ credit groups. Some farmersreported their decision not to take on loans becausethey anticipated difficulties in repaying them.17

3.3.5 Input and output pricesTable 3.2 summarizes the estimated average pricesin purchasing power parity (PPP) dollars 8 of themost commonly used maize inputs and of maizegrain output in the surveyed maize productionenvironments. Price variations among environmentsare largely due to differences in transportation andhandling facilities provided by suppliers. Seed andfertilizer are the most common material inputs boughtby maize farmers. Seeds of local/traditional varietiesand improved OPVs are often saved from precedingharvests or purchased from/exchanged with otherfarmers in the community. If purchased, seed oflocal/traditional varieties costs anywhere from PPP$0.58/kg in the rainfed upland, semi-commercialproduction areas to PPP$ 0.95/kg in the subtropical/mid-altitude, low market surplus production areas.They were, surprisingly, most expensive in the rainfedupland, low market surplus production areas, at PPP$2.19/kg. Asian maize farmers, however, pay a muchhigher price for hybrid seed, about 10 to 16 times thatof local/traditional and improved OPVs. In contrast,hybrid seed costs from PPP$ 5.02/kg in rainfed uplandand subtropical/mid-altitude, low market surplusproduction areas to PPP$ 9.04/kg in the rainfedupland, semi-commercial areas.Farmer-respondents reported that the most commoninorganic fertilizers used in maize production are urea(46-0-0), complete fertilizer (14-14-14 or 15-15-15) andammonium phosphate (16-20-0). Prices of inorganicfertilizer did not vary much among agroecologies,but were expectedly higher in remote upland areaswith poor market access. In the irrigated and rainfedlowlands, ammonium phosphate was the cheapestinorganic fertilizer, and complete fertilizer the mostexpensive. In the rainfed uplands, urea was thecheapest inorganic source of plant nutrients, and againcomplete fertilizer the most expensive. Other inorganicfertilizers used in maize production included muriateof potash (0-0-60), diammonium phosphate (18-46-0),and ammonium sulfate (21-0-0). Some farmers appliedorganic fertilizers such as farmyard manure (FYM)and chicken dung either collected from their ownbackyards or purchased from input traders. In Nepal,however, manure and stover are normally not traded.Table 3.2. Average prices of the most commonly used maize production inputs and maize output(in PPP $ per unit), Asia.IrrigatedSubtropical/lowland Rainfed lowland Rainfed upland midaltitudeLow Semi- Low Semi- Low market Low market Semi-Commer- market commer- Commer- market commer- Commer- surplus surplus commercialsurplus cial cial surplus cial cial (excluding China) (China) cialMaize seed (per kg)Local 0.76 0.62 0.80 0.71 2.19 0.58 0.59 0.95 0.68 -Improved OPV 1.05 0.65 1.48 1.26 1.60 0.86 1.23 1.46 0.68 -Hybrid 7.45 5.06 8.96 8.02 5.02 9.04 8.22 5.02 3.82 4.45Inorganic fertilizers (per 50 kg)Urea 31.55 42.84 36.80 39.81 40.33 35.26 31.15 39.09 40.06 42.61Complete 42.51 30.76 42.06 40.93 38.30 39.45 39.38 --- 30.78 30.70Ammoniumphosphate 21.64 --- 27.02 31.23 --- 32.26 33.20 --- --- 34.09Labor wage rate (per person-day)Male 7.01 9.38 6.97 8.96 5.85 8.46 7.39 5.71 10.82 12.60Female 5.73 8.93 6.16 8.24 5.13 8.18 6.81 4.65 9.63 9.66Power rentalAnimal (per day) 15.00 290.70 a 12.78 13.62 12.76 12.19 13.72 12.95 301.05 a 340.95 aTractor (per ha) 41.35 291.15 74.39 50.24 --- 72.44 52.07 --- --- ---Maize grain price (per kg)Farm gate 0.66 0.49 0.57 0.47 0.64 0.49 0.44 0.82 0.48 0.49Nearest market 0.81 0.51 0.64 0.72 0.73 0.65 0.71 0.92 0.51 0.50Source: IFAD-CIMMYT RRA/PRA Surveys 2000-2001.aPPP$/ha.8Purchasing power parity (PPP) dollars are computed by converting national currency estimates to a common currency using a PPP conversion factordefined as the number of units of a country’s currency needed to buy the same amount of goods and services in the domestic market as a US dollarwould buy in the US. The use of PPP conversion allows a balanced comparison (and consolidation) of national currency prices across different countries orlocations. The source of conversion factors used in the calculations is 2002 World Development Indicators, World Bank.18

Some, but not many, farmers used pesticides in maizeproduction. The most commonly applied chemicalpesticides are field herbicides and insecticides. Somefarmers also treated their maize seed with chemicalsto protect it against storage pests such as weevils.Across Asia, average daily wages for agricultural farmlabor appear to discriminate against women: malelabor enjoys a higher average daily wage rate of PPP$5.71-8.96/person-day, while female labor receives anaverage daily wage of PPP$ 4.65-8.24/person-day(Table 3.2). Snacks and/or lunch for farm workers aresometimes provided, especially during planting andharvest operations. In Nepal, most farm operationsare done with family and exchange labor, and useof hired labor is not common except during peaktimes. In the Philippines, groups of farm laborerscan be hired for either planting or harvesting maize.The local contract arrangement is called the ‘pakyaw’system, where total labor is normally paid on a perhectarerate during planting. Harvesters are paideither in cash at a per-sack rate or in kind, with a shareof the harvested cobs. The most common harvestsharingscheme is 10:1, where for every 10 sacks ofharvested maize, the owner gets nine sacks and theharvesters get one sack.For land preparation, some farmers contract twoorfour-wheel tractors at a rate ranging from PPP$41.35/ha in irrigated lowland-commercial maizeareas to PPP$ 74.39/ha in rainfed lowland, semicommercialareas. Animal power with operators canalso be contracted for land preparation, especially inhilly areas, at PPP$ 12.19-15.00/person-animal day. Insome areas, farmers can rent draft animals at a dailyrental rate.Maize grain prices varied widely within and acrossthe maize production environments surveyed. Maizegrain prices in the nearest markets ranged from PPP$0.64 to 0.92/kg and were higher by PPP$ 0.07-0.27/kg than those at the farm gate (PPP$ 0.44-0.82/kg).Subsistence maize farmers in the rainfed upland andsubtropical/mid-altitude production areas (e.g., inNepal’s far and midwestern development regionand Cebu and Leyte provinces in the Philippines)kept most of their grain for home consumption.Where both are available, yellow maize (usuallyhybrid) commanded a higher price than white(local/traditional) maize. Farmers expressed that inrecent years they have observed that farm input pricestended to increase, while output prices have stayedmore or less at the same, if not declining, level.3.3.6 Input-output price ratiosUsing the average of all reported maize seed prices andthe average of all reported maize grain prices, the seedto-grainprice ratio across the surveyed areas in Asiawas 6.52 using farm-gate and 5.20 using nearest-marketgrain prices during 2000-2001. This means that Asianmaize farmers have to sell, on average, 5.20 kg of maizegrain to pay for one kilogram of maize seed. This ismuch lower than the seed-to-grain price ratio of 11.21derived in 1997/98 during CIMMYT’s Asia maize seedindustry study (Gerpacio 2001). As expected acrossproduction environments, the seed-to-grain priceratios were lower when farmers grew local/traditionalmaize or improved OPVs than when they used hybrids(Table 3.3). At farm-gate local/traditional maize grainprices, the seed-to-grain price ratio was lowest at 1.15in irrigated lowland, commercial areas, and highestat 3.40 in rainfed upland, low market surplus areas.For hybrids, the seed-to-grain price ratio was low at6.14 and 7.79 in subtropical/mid-altitude, low marketsurplus and rainfed upland, low market surplus areas,respectively. Ratios were consistently lower whenmaize grain prices at the nearest market were used inthe computation.To exploit the full genetic potential of modern maizevarieties, particularly hybrids, farmers must usecomplementary inputs, especially fertilizers, and theymust perform improved management operationsthat often require additional labor. The profitabilityof using complementary inputs and improvedmanagement practices depends primarily on their cost.Input-to-grain price ratios can be used to make interenvironmentcomparisons of the relative costlinessof key inputs, which, in turn, can serve as roughindicators of the profitability of adopting modernmaize varieties. Table 3.3 also shows the fertilizer-,labor- and power-to-grain price ratios prevailing inthe different maize production environments of Asiabetween late 2000 and early 2002.Across fertilizer grades, the fertilizer-to-grain priceratio at the farm gate was lowest at 1.08 in irrigatedlowland, commercial maize areas and highest at 1.60in rainfed upland, commercial areas. Ratios were muchlower using grain prices at the nearest market, rangingfrom 0.88 in irrigated lowland, commercial maizeareas to between 1.41 and 1.52 in rainfed upland lowmarket surplus and commercial areas in China andthroughout Asia and in subtropical/mid-altitude lowmarket surplus and semi-commercial maize productionareas in China. Inorganic fertilizers are more expensivein these areas, and farmers must sell more kilogramsof maize to compensate for the cost of one kilogram ofinorganic fertilizer.19

The most expensive input to maize productionappears to be mechanization (tractor power), whichrequires that farmers sell at least 63 kg of maize inirrigated lowland, commercial areas and up to 147kg in rainfed upland, semi-commercial environmentsto recover the cost of having one hectare of landmechanically prepared. Animal power is cheaper thanmechanization, especially in low market surplus andsemi-commercial maize production areas where it iswidely used by marginal, resource-poor farmers.As previously mentioned, input-output price ratiosare higher when maize grain farm-gate prices,rather than nearest-market grain prices, are used inthe calculations. This indicates that price ratios arerelatively more favorable to farmers selling in thenearest market than farmers selling at the farm gate.Farmers accept lower prices due to the convenienceand savings in time and labor of having the buyer cometo them. Meanwhile, across production environments,farm inputs tend to be cheaper in the irrigatedlowland, commercial maize production areas and moreexpensive in the rainfed upland, semi-commercial andcommercial environments. In particular, seed-to-grainprice ratios (for local/traditional and improved OPvarieties) tend to be highest in rainfed upland, lowmarket surplus maize production environments. Thissuggests that high (and rising) input prices constrainmaize production, especially for marginal, resourcepoorfarmers who reside mostly in less favorableproduction environments. Appropriate governmentprograms will be important to address this concernand contribute to alleviating rural poverty in the moremarginal areas of rainfed upland environments.3.4 InfrastructureA rural community’s economic development dependson a number of critical infrastructure components,particularly road and communication systems andpost-harvest and marketing facilities. The availabilityof this type of infrastructure plays an important role inpromoting the modernization and commercializationof agriculture—often the backbone sector—in thecommunity. However, most maize producing areasin the tropical lowlands have only poor, inadequateinfrastructure to support the expansion of agriculturalproduction.Table 3.3. Input-to-maize grain price ratios for seed, fertilizer, labor, and power, Asia.IrrigatedSubtropical/lowland Rainfed lowland Rainfed upland midaltitudeLow Semi- Low Semi- Low market Low market Semi-Commer- market commer- Commer- market commer- Commer- surplus surplus commercialsurplus cial cial surplus cial cial (excluding China) (China) cialInput-to-maize grain farm-gate price ratioSeed (all) 6.77 6.70 7.45 8.94 3.38 8.77 11.69 1.87 5.60 8.90Local 1.15 1.27 1.41 1.53 3.40 1.18 1.34 1.17 1.42 -Improved OPV 1.59 1.33 2.61 2.70 2.48 1.75 2.77 1.79 1.42 -Hybrid 11.31 10.33 15.81 17.22 7.79 18.38 18.56 6.14 7.96 8.90Fertilizer (all) 1.08 1.51 1.41 1.49 1.23 1.45 1.60 1.35 1.50 1.48Urea 0.96 1.75 1.30 1.38 1.25 1.43 1.41 0.96 1.67 1.70Complete 1.29 1.26 1.48 1.71 1.19 1.60 1.78 --- 1.28 1.23Labor 9.66 18.68 11.59 18.47 8.52 16.96 16.03 6.34 21.30 22.26PowerAnimal 22.76 28.63 22.54 29.26 19.81 24.78 30.97 15.85 53.00 48.98Tractor 62.77 44.76 131.22 107.89 --- 147.29 117.55 --- --- ---Input-to-maize grain nearest-market price ratioSeed (all) 5.53 6.30 6.59 5.81 2.98 6.60 7.30 1.66 5.51 9.08Local 0.94 1.22 1.24 0.99 2.99 0.89 0.84 1.03 1.33 -Improved OPV 1.30 1.27 2.30 1.76 2.18 1.32 1.73 1.59 1.33 -Hybrid 9.24 9.92 13.98 11.20 6.86 13.84 11.59 5.45 7.49 9.08Fertilizer (all) 0.88 1.45 1.24 1.49 1.08 1.09 1.00 1.20 1.41 1.52Urea 0.78 1.68 1.15 0.90 1.10 1.21 0.88 0.85 1.57 1.74Complete 1.05 1.21 1.31 1.11 1.05 1.20 1.11 --- 1.21 1.25Labor 7.89 17.95 10.25 12.01 7.49 12.77 10.01 5.62 20.05 22.71PowerAnimal 18.58 27.84 19.92 19.03 17.44 18.66 19.34 14.06 50.52 46.69Tractor 51.24 43.52 115.98 70.16 --- 110.91 73.39 --- --- ---Source: IFAD-CIMMYT RRA/PRA Surveys 2000-2001.--- No data available.20

3.4.1 AccessibilityMost of the surveyed villages have seasonal/fairweatherroads and fair-to-good asphalted roads thatallow motorized vehicles, from motorcycles to smallpick-ups and trucks, to pass (Table 3.4). Agriculturalextension agents can visit these areas to disseminateupdated agricultural technology information tofarmers. Sales agents of private input companies cancome and promote their new maize seed, fertilizers,or pesticides. Agricultural traders-financiers canalso come to collect, buy, and transport farmers’agricultural output in bulk.Some interior villages in the mountainous areasof northwest Vietnam and in Bukidnon province,Northern Mindanao region, Philippines, reportedpoor road conditions that were inaccessible to motorvehicles, especially during the rainy season. In mostvillages, tractors and small trucks are used to transportlarge loads, while motorcycles are used to transportsmall amounts of farm inputs and produce. In villageswhere vehicle access is most difficult, farmers usehorses, cattle, and carts for transport. In the mid- andhigh-hills of Nepal, farmers also carried inputs andproduce (and even construction materials) on theirshoulders and backs.In general, semi-commercial and commercial maizeproducingareas that are nearer to market or to feedprocessingcenters were reported to have relativelygood road and transportation systems. The villagessurveyed in East Lombok, Indonesia, however, areonly three kilometers away from the nearest market,but the poor transportation system makes it difficultfor farmers to transport their produce. In Vietnam’sMekong River Delta provinces, most transport ofagricultural produce is done via the waterways; henceboats are an important means of transportation.3.4.2 IrrigationIrrigated maize is grown mainly in lowland ecosystemswhere tubewells and communal irrigation systemssupply water mainly for rice production. In uplandenvironments, farmers largely depend on rainfall, andthe main source of irrigation water, if any, is springsand rivers, although irrigation for maize tends tobe minimal. One exception was in certain areas ofthe subtropical/mid-altitude maize environment inChina (Sichuan Province), where irrigated springmaize is cultivated. However, irrigation water, whenavailable, is reserved for rice production and cashcrops in southwestern China, high-value crops suchas vegetables in the Philippines, baby corn and sweetcorn in Thailand, and coffee and black pepper inVietnam.3.4.3 Post-harvest facilitiesThere are few drying and no large storage facilities inmost of the surveyed maize growing upland villages.Only a few farmer cooperatives have some kind ofdrying or storage facility for their members’ use. Inthe large commercial maize areas of Vietnam, a limitednumber of power-operated drying facilities is availableto local traders during the rainy season. Farmersusually dry their maize under the sun, on house roofs,flat cement floors or roads, drying baskets, or plasticsheets. In southern Philippines, rice and maize farmerswere observed to dry grain on one lane of a two-lanehighway over an 11-km stretch. Across Asia, sun-driedmaize grain is stored in sheds, storage barns, bamboobaskets, and wooden boxes or in cupboards, urns, orplastic sacks at home. Commercial maize farmers donot store maize for long periods to avoid high storagelosses caused by weevils.Most small Asian farmers shell their maize manually.In the commercial maize growing areas of Vietnam,large maize producers hire shelling machines toreduce labor and time, especially when the crop isharvested under unfavorable weather conditions. Inthe Philippines, a roving mechanical maize shellercomes to the villages during harvest season, or farmerscan have their maize shelled for a fee at a few privatelyoperated stationary shellers. Some farmers acrossthe region also commonly use small electric shellingdevices or manually operated maize shellers to processgrains intended for home use.In the hills of Nepal, the presence of power-operatedmaize-grinding mills in the production areas dependslargely on road access, since transportation costs canbe prohibitive. Only a few farmers in semi-commercialvillages grind their maize, mainly for farm animal feed.Traditional manual grinding stones are also used inremote villages. Maize farmers in the rainfed lowlandand semitropical/mid-altitude regions of Chinausually have access to power-operated milling facilitiesin their own villages.3.4.4 Proximity to markets and marketingpracticesMost of the lowland villages surveyed have smallregular markets within the community. Othervillages have only periodical (weekly) marketswhere agricultural products and livestock are sold.In contrast, farmers in Vietnam’s remote villagesin the northern uplands and central coast uplandareas walk 10-25 km to get to the nearest (primary)market. Similarly, in one upland village in NorthernMindanao, Philippines, people travel up to 57 km toreach the nearest market. Villages in commercial maizeproduction areas of the irrigated lowlands, rainfed21

lowlands, and rainfed upland environments were, onaverage, only 20 km away from the nearest marketand have better access to markets compared to thosein the rainfed uplands (Table 3.4). In commercial maizeareas, the marketing system involves small villagebasedassemblers, commission agents, middlemen, andindependent traders. Farmers have observed intensecompetition among these marketing agents to get themost farm produce to maximize their capital output.Maize farmers sell their grain and other farm productseither directly in public markets or to traders whocome to the villages. A barter system also prevails on alimited scale in Nepal. In general, self-financed maizefarmers sell their grain at more distant secondarymarkets, where prices are often higher than at marketscloser to the village. Maize farmers with loansfrom trader-financiers have to sell their grain to thefinanciers despite the low prices they often get. Traderfinancierscome to the villages during harvest and haulthe grains in volume. Trucking services (transportationcosts) may be charged to the farmers, shouldered bythe traders, or shared by both parties. In Thailand andthe Philippines, a few farmers sell their output to feedmills in nearby areas. Farmers consider feed mills goodmarket outlets because they pay higher than marketprices, and some do not have very strict grain qualitystandards. Chinese farmers sell maize to governmentgrain outlets to fulfill government grain procurementquotas, but recent policy changes have begun toeliminate this requirement.In Indonesia, how maize is marketed depends onthe type of maize. Yellow maize (mostly hybrid andrecycled hybrid) is sold directly to other farmers soonafter harvesting. White maize is usually stored as earswith husks after sun-drying. Farmers sell their whitemaize gradually, and the money it brings in is used tocover daily household expenditures. Part of the whitemaize harvest is also consumed as a staple food in thehousehold.In Vietnam’s commercial maize areas, farmers withsubstantial yield prefer to sell freshly shelled maizeto local traders right on the field at harvesting. Thismarketing practice is more popular in the rainy seasonbecause farmers do not have sufficient drying andstorage facilities. In villages with good road access,local/traditional and improved OPV maize is sold onthe field to local traders who harvest and transport it.3.5 Socioeconomic Characteristics3.5.1 Households and ethnicityTotal population and number of households variedwidely among participating countries, surveyedvillages, and maize production environments(Table 3.5). The most densely populated surveyedvillages were located in India, in the subtropical/mid-altitude,low market surplus, irrigated lowlandcommercial,and rainfed lowland, semi-commercialTable 3.4. Summary of the infrastructure in maize production environments across 265 surveyed locations in Asia.Maize production Distanceenvironment and to nearestmarket orientation market (km) Road conditions Means of transportation availableIrrigated lowlandCommercial 0 - 5.0 Relatively good to good; good and moderate Minicab; pick-up trucks; cartsasphalt; black top gravel; seasonal/gravelRainfed lowlandLow market surplus 0 – 20 Dirt road Bus; by foot; bicycle; motorcycle; tractor;3 wheeled vehicle (cyclecar); ox pulled cartSemi-commercial 3.0 – 57.0 Poor to fair to good; moderate asphalt andgravel; black top; seasonal/fair-weather gravel road;Minicab; jeepneys; motorcycles; tricycles;small trucksCommercial 1.5 – 10.0 Poor; relatively good to good asphalt; moderate gravel Minicab; jeepneys; tricycles; trucks;pick-up trucks; cartsRainfed uplandLow market surplus 4.0 – 25.0 Black top; seasonal/fair-weather road; gravel/asphalted road Jeepneys; tricycles; motorcyclesSemi-commercial 2.0 – 57.0 Good; gravel/asphalted road; seasonal/fair-weather road; Motorcycles; jeepneys; tricycles;small trucks; pick-up trucksCommercial 2.7 – 20.0 Fair to good; good asphalt; moderate gravel Minicab; jeepneys; tricycles;small trucks; pick-up trucksSubtropical/mid-altitudeLow market surplus 0.5 – 25.0 Dirt road; Seasonal/fair-weather road; gravel; black top Bus; by foot; bicycle; motorcycle; tractor;3 wheeled vehicle (cyclecar); truckSemi-commercial 10 Dirt road Bus; by foot; boatCommercialSource: IFAD-CIMMYT RRA/PRA Surveys 2000-2001.22

maize environments, with a maximum of 21,000,34,300, and 38,900 people, respectively. No consistentpatterns emerged based on production orientation, butvillages in rainfed upland maize production areas weregenerally observed to have the smallest populationsize.On the other hand, average household size observedacross the surveyed villages and productionenvironments did not vary much, ranging from 3-6members per household across all sites (Table 3.5).The rainfed upland-semi-commercial and -commercialenvironments tended to have higher averagehousehold sizes of up to 6 members per household ascompared to other production environments.Ethnic grouping is localized in the surveyed regions/provinces. In Vietnam, Kinh is the major ethnic group,accounting for nearly 90% of total population; the restare minor ethnic groups, living mostly in the country’supland areas. In Thailand, most maize farmers are Thaiand only a small number are members of hill tribes.In the Philippines, ethnic groupings are different ineach of the surveyed provinces. The Ibanag and Ilocanogroups are prominent in Isabela Province, whilethe Mangyans predominate in Mindoro Occidentaland Camarines Sur. The study areas in Mindanao(Bukidnon, South Cotabato, and Cotabato) had thehighest number of ethnic groups, with the Cebuanos,Lumads, and T’bolis being the top three. Althoughthe Han Chinese make up the majority ethnic groupacross China as a whole, the Zhuang and Yao are veryprevalent in the tropical lowland sites of GuangxiProvince. The largest group, the Zhuang, was themajority ethnic group in many of the survey sites.Males headed most households across all surveyedvillages. However, male and female usually play anequal role in making farm decisions, since agricultureis considered to be a family enterprise. An exceptionwas the surveyed villages in Nepal’s eastern mid-hillsand terai/inner terai, where the heads of 22% and 15%of households, respectively, were female. These figuresindicate that women are highly regarded in these areas,and also that men tend to seek employment outside thevillages, e.g., in urban areas. In Mindanao, Philippines,women headed 14% of all households.3.5.2. Literacy and level of educationEducation is a good indicator of farmers’ capability toadopt new technologies that will help improve farmproduction. In general, the literacy rate was highest inthe rainfed lowland-commercial and rainfed uplandsemi-commercialproduction environments. Themajority of the population across the surveyed villageshas attended or completed primary (elementary)education (Table 3.5). A good percentage also attendedand/or finished secondary school (high school),particularly in the irrigated lowland-commercial andrainfed upland-low market surplus maize areas.In the Philippines, most people consider educationto be a top priority, and even resource-poor familieswill strive hard to send their children to school. InNepal, passing the ‘School Leaving Certificate (SLC)’examination is an important indicator of the level ofeducation. Unfortunately, only a minimum number ofNepalese passed the examination in 2000/2001; for thisreason, an astounding 30-50% was reported as illiteratein the subtropical/mid-altitude hill maize areas.Table 3.5. Selected demographic information from across 265 surveyed locations, Asia.Educational level (range in %)Maize production Total Average Elementary High schoolenvironment and population Number of household level or level or College levelmarket orientation (000) households size Illiterate graduate graduate or graduateIrrigated lowlandsCommercial 2.4-34.3 467-3,800 4.0-5.7 0-30 44-57 35-54 5-16Rainfed lowlandsLow market surplus 1.0-3.1 250-811 3.8-5.1 0-5 63-94 5-30 1-12Semi-commercial 0.4-38.9 108-8,000 3.3-5.9 0-30 33–79 17-54 1-25Commercial 0.9-18.6 159-3,505 4.3-5.7 0–9 44–66 19-46 1-11Rainfed uplandsLow market surplus 0.5-9.3 96-1,759 4.3-5.0 8-38 32-48 32-55 6-9Semi-commercial 1.4-7.1 229-1,419 4.8-5.9 2–3 42-79 17-30 3-25Commercial 1.2-1.7 211-347 4.9-5.9 2–9 66-79 17-19 3-6Subtropical/mid-altitudeLow market surplus 0.9-21.1 209-4,063 3.2-4.2 0-50 57-95 4-20 0-15Semi-commercial 1.4 359 3.9 20 71 6 1CommercialSource: IFAD-CIMMYT RRA/PRA Surveys 2000-2001.23

3.5.3 Landholdings and land tenureMaize farmers across all production environmentsin Asia had an average farm size ranging from 0.3 to20.8 ha (Table 3.6). The range of farm sizes was widest(1.5-20.8 ha) in the rainfed upland-semi-commercialmaize areas and narrowest (0.3-1.3 ha) in the irrigatedlowland-commercial maize areas. In Thailand, 31% ofmaize farmers had large farms, averaging about 16 ha.The largest farm size reported was a landholding of 80ha in Nakorn Ratchaseema Province, Thailand.With the exception of farmers in China, mostmaize farmers in surveyed villages across Asia arelandowners, with the highest proportion (65-100%)living in the irrigated lowland-commercial productionenvironments (Table 3.6). Tenants/sharecroppers arelocated mostly (16-64%) in the rainfed upland-lowmarket surplus production environments. Fixedrentpayers are high in the rainfed lowland-semicommercial(up to 22%) and subtropical/mid-altitudelowmarket surplus (up to 19%) maize areas. In therainfed lowland- and rainfed upland-semi-commercialmaize production environments, however, up to 92%of maize farmers are landless.In China, under the household responsibility system(HRS) reform that began in 1979, collective land wasallocated to individual households based on householdsize or a combination of household size and labor.Since private ownership of agricultural land does notofficially exist in China, this so-called responsibilityland remains collectively owned and subject toperiodic reallocation by village leaders. Farmers aresupposed to have complete use and income rightsover the land during the contract period. Mostagricultural lands in surveyed villages in China fallunder the category of responsibility land. Villages canalso maintain an area of unallocated “contract land,”which is leased to households on a short-term basis.Households in all surveyed villages in both rainfedlowland and subtropical/mid-altitude regions werealso allocated small amounts of land for householdplots largely used to grow vegetables for householdconsumption.While most farmers in the lowlands of Vietnam havebeen provided the red book (land use certification)for the land they own, many farmers in the uplands,particularly those located near forest areas, still do nothave legal land use privileges. These farmers do nothave access to formal credit sources and have littleincentive to invest in land that is not theirs.3.5.4 Maize utilizationIn Asia, maize has become a major component ofhuman diets and the preferred substitute duringperiods of rice shortage. As expected, however, maizeutilization patterns differ depending upon the foodhabits of the local population. Where livestock, pig,and poultry production is developing rapidly, mostmaize (particularly hybrid) is grown not for humanconsumption, but for animal feed (Table 3.7). In semicommercialand commercial production systems,40-100% of all maize produced was sold, and up to60% was fed to barnyard animals. With the exceptionof surveyed villages in China that predominantly grewhybrids for feed (40-90% of total output), subsistencefarmers in the rainfed uplands and subtropical/mid-Table 3.6. Selected socioeconomic information from 265 surveyed locations, Asia.Maize productionTenure status of farmers (%) aenvironment and Average farm Tenant/ Fixed-rent Mortgage CLT/partmarketorientation size (ha) Land-owner share-cropper payer Landless payer owners bIrrigated lowlandsCommercial 0.3-1.3 65-100 15-35 0-6 0-7 0-1Rainfed lowlandsLow market surplus 0.2-0.8Semi-commercial 0.7-3.3 0-100 2-28 0-22 0-92 0-16 5-13Commercial 0.3-15.4 23-99 1-35 0-2 0-10 0-14 3-36Rainfed uplandsLow market surplus 0.7-1.8 36-81 16-64 0-3Semi-commercial 1.5-20.8 0-88 2-20 2-3 7-92 5-16 3-50Commercial 0.6-7.2 23-100 1-34 0-2 0-10 0-14 3-38Subtropical/mid-altitudeLow market surplus 0.1-2.1 44-92 1-33 0-19 1-4Semi-commercial 0.3-2.5Commercial 0.5-1.7Source: IFAD-CIMMYT RRA/PRA Surveys 2000-2001.aTenure status does not include surveyed villages in China, where all crop land is allocated by or contracted from the village.bCLT – Certificate of Land Transfer, issued to agrarian reform beneficiaries in the Philippines.24

altitude environments planted mostly local/traditionalmaize varieties and kept 59-90% and up to 73% ofoutput, respectively, for household consumption.Again, with the exception of subtropical/mid-altitudeenvironments in China, where approximately 20% wasretained for feed use, only 1-5% of maize produced insubtropical/mid-altitude-semi-commercial areas wasfed to barnyard animals. Although subsistence farmersin rainfed lowland environments reported marketingsome of their maize, most of it is used for householdconsumption and feed.In predominantly low-market-surplus villages, maizeears are dried with or without husk and stored, andlater milled as needed for immediate use. In Bukidnon,Northern Mindanao, Philippines, maize farmersreported that when the harvest is very poor, evenyellow maize is processed and consumed as food.Maize farmers in the rainfed uplands-low marketsurplus production environments retained a significantproportion of their grain for seed. In subtropical/midaltitude-semi-commercialenvironments, a portion ofthe maize grain was allocated for other purposes, e.g.,as the main ingredient in brewing the local beverage.3.5.5 Farmer classificationOne exercise conducted during this study was toclassify farmers within the community, in accordancewith local perceptions of wealth and poverty. Farmhouseholds were classified as either poor/small,medium/intermediate, or rich/large farmers, andfarmer-respondents were asked to provide descriptionsand general characteristics of each group. Respondentselectedparameters most frequently included farmsize, income by source, household size, and number oflivestock owned.Farmer groupings were characterized very similarlyacross production environments, except for slightdifferences in farm size and proportion of incomecontributed by maize production (Table 3.8). Thepoor/small farmer group was characterized as mostlytenants or sharecroppers with large households andlittle or no education. With the exception of China,where landholding size was believed to be relativelyuniform across the different farmer groups, if landwas owned, farmers tended to have small farms:0.5 ha or less in the irrigated lowland-commercialenvironments and up to 2.0 ha in the rainfed lowlandsemi-commercialand all the rainfed uplands maizeareas. Poor/small farmers in the rainfed lowlandcommercialmaize areas tend to have relatively largefarms of up to 3.2 ha. Most of their income comes frommaize production and other agricultural activities(they often hire themselves out as labor to other farms).Other types of off-farm income opportunities appearto be quite limited. Poor/small farmers in the rainfedlowland-commercial maize areas of the Philippinesearn an estimated 59-76% of their income from maizeproduction and none from non-agricultural activities.In South Lampung, Indonesia, poor farmers earnmost of their income from non-farming activities. Insubtropical/mid-altitude environments, poor farmersoften lag behind in adopting new technologies due tolack of capital.Medium/intermediate farmers, many of whom arelandowners, have more sources of income than poor/small farmers. Farm size ranges from 0.5 to 5.0 haacross all environments, except in the rainfed lowlandcommercialmaize areas, where farms covered 0.8-8.0ha. Income in this group comes mainly from maizeproduction and other agricultural crops/activities.In Thailand, medium/intermediate farmers earn anaverage of 50,000-80,000 baht (US$1,250-2,000). In theTable 3.7. Maize utilization (% of grain produced) across 265 surveyed locations, Asia.Maize production environmentMaize utilization (% of grain produced)and market orientation Sold Kept for food Animal feed Retained as seed OtherIrrigated lowlandsCommercial 40-100 0-41 0-60 0-3 0-9Rainfed lowlandsLow market surplus 0-28 1-60 20-94 0-10 0Semi-commercial 46-100 0-41 0-29 0-30 0-9Commercial 70-100 0-22 0-29 0.1-2 0-4Rainfed uplandsLow market surplus 1-3 59-90 2-34 4-27Semi-commercial 62-100 0-20 0-29 0-1 1-3Commercial 62-100 0-14 0-29 0-2 0-3Subtropical/mid-altitudeLow market surplus 0-35 1-73 1-95 0-4 0-24Semi-commercial 40-52 30-45 1-20 0 0-8Commercial 60-80 10-30 5-10 4-10Source: IFAD-CIMMYT RRA/PRA Surveys 2000-2001.25

Philippines, maize production provides an estimated48-65% of this farmer group’s total household income,and non-farm activities account for about 6-14%.The typical non-farm income-earning activities formedium/intermediate farmers include buy-and-sellenterprises, driving public vehicles for hire, workingin factories or stores in nearby cities, and working asconstruction laborers within or outside the community.Farmers in the relatively rich/large farmer group tendto have smaller households and larger farms (at least5.0 ha) where crop diversification, particularly to cashcrops, is more prevalent. These farmers (in China,Philippines, and Thailand) tend to be less economicallydependent on maize production and characteristicallyobtain the highest share of their income from cash cropsand non-agricultural activities. In several survey sites inthe Philippines and Thailand, the relatively richer/largerfarmers have education and skills that allow them ortheir family members to hold white-collar jobs in biggercities; some family members may work overseas and sendremittances home. In contrast, richer/larger farmers inSouth Sulawesi, Indonesia, still earn most of their incomefrom producing agricultural commodities (cotton, cocoa,and tobacco) and barnyard livestock raising. In Nepal,rich/large farmers often lead the community towardsadoption of new and improved technologies especiallythose pertaining to maize production.Table 3.8. Classification of maize farmers within the community, 265 surveyed locations, Asia.Maize productionFarmer group/classificationenvironment andmarket orientation Poor/Small Medium Rich/LargeGeneral characteristics, similar across environmentsSmall farm size; large family/ Medium-sized farms; more income Large farms growing more than one crop;household size; little or no sources; income mainly from maize small family/household size; income mainlyeducation; large portion of income and other agricultural products/ from maize and other agricultural products/from maize and other agricultural activities; easily adopt new activities; skills or education allows familyproducts/activities; lag in adopting agricultural technologies members to work in city or abroad; easilynew technologyadopt new technologiesSpecific characteristics (differences) reportedIrrigated lowlandsCommercial Average farm size up to 0.5 ha Farms of 0.5-2.0 ha; no income Farm size 2.0 ha or morefrom cash crops/business (Nepal)Rainfed lowlandsLow market surplus Large income share from maize Income share from maize Largest share of income from off-farm work(40-100%), small share from non- approximately 30-60%; young to (60-80%), smallest share from maizeagricultural sources; less educated middle-aged household heads (7-27%); less maize consumption as food;household head, older householdhousehold head better educated than thoseheadsof poor/small or medium households;middle-aged household headsSemi-commercial Farms up to 2.0 ha; about 59-76% of Farm size range 0.5-5.0 ha Average farm size >5.0 ha; in East Java,total income from maize productionIndonesia, large portion of income comesand none from non-agriculturalfrom other agricultural products/activitiesactivities; in East Java, Indonesia, largeand little from maizeportion of income from other agriculturalproducts/activities and little from maizeCommercial Farms at up to 3.2 ha; in Thailand, Farm size range 0.8-8.0 ha Farm size 1.0-8.0 haincome was mostly dependent onwage employmentRainfed uplandsLow market surplus Average farm size 0-2.0 ha Average farm size 2.0-5.0 ha Average farm size >5.0 ha; less proportionof income (29-56%) from maize production,and higher from non-agricultural activitiesSemi-commercial Average farm size 0–2.0 ha Average farm size 2.0-8.0 ha Average farm size >5.0 haCommercial Average farm size 0–2.0 ha Average farm size 2.0-5.0 ha Average farm size >5.0 haSubtropical/mid-altitudeLow market surplus Lag in adoption of new technologies Less dependent on agricultural Larger share of income from off-farmSemi-commercial due to lack of capital to buy seed, income than poor/small farmers employment; household head betterCommercial fertilizer, and other inputs; most educatedincome from agricultural production;little or no off-farm incomeSource: IFAD-CIMMYT RRA/PRA Surveys 2000-2001.26

4. Maize Production SystemsCrop production was the main agricultural enterpriseacross the surveyed villages in Asia, most of whichgrew maize as the primary crop. Cotton, wheat, chili,vegetables, legumes, root crops, and cash crops are alsorelay-cropped or intercropped with maize, usually ona small percentage of the total cultivated area. Mostfarm households also engage in small-scale (barnyard)poultry and livestock production to augment theirincome and supply home needs. In the Philippines,for example, non-farm activities and enterprises suchas small-scale trading, small sari-sari stores, or drivingtricycles and jeepneys also provide additional incomein many of the surveyed sites.Animals commonly raised by maize farmers includecattle, carabaos (water buffaloes), goats, swine, andpoultry, with relatively wealthier farmers owningmore animal types and units. Besides providing meat,water buffaloes and cattle are kept as work animals,while goats, swine, and poultry are kept to sell asnecessary, especially during lean months. In somevillages in Indonesia and Nepal, poultry is also kept forhome consumption, especially in lower-income farmhouseholds. Eggs are a source of both household foodand income in China.4.1 Crops, Cropping Calendars,and Cropping PatternsAcross all surveyed areas, the diversity in numberand types of crops is very large, with maize growneither as a main crop or as a second crop after rice orwheat (Table 4.1). In Nepal, where maize is the singlemost important crop in terms of production andTable 4.1. Major cropping patterns and calendars in maize environments, Asia.Country Major maize cropping patterns Maize cropping calendarIndia Maize-wheat/sugarcane; paddy/maize-wheat; maize; potato-wheat; Rainy season: June/July-Sep/Octmaize-potato+coriander-vegetables; maize-mustard-onion;Winter season: Oct/Nov-Marmaize-potato-cucumber; maize-chiliIndonesia Maize-maize; maize-cassava Wet season: Oct/Nov-Feb/MarRice-maize; maize-chiliDry season: March-JulyNepal Bari land: Maize+millet; maize+millet-wheat/barley; maize-wheat/barley; Summer: Mar/Apr-Sepmaize-pulses/oilseedsWinter: Sep-FebPhilippines Maize-maize-maize; maize+legumes-maize-vegetables; 1 st : May-Julyupland rice/maize-vegetables2 nd : Aug-Nov3 rd : Dec-MarThailand Maize-mungbean; 1 st : Apr/May-Aug/SeptMaize-sorghum/sunflower2 nd : Sept-Dec/Jan3 rd : Jan-AprVietnam Rice-rice-maize; Maize; Spring-summer: Jan-MayMaize-rice-riceSummer-autumn: May-SepAutumn-winter: Aug-DecWinter-spring: Sept-JanChina Spring maize: maize-soybean/cassava; maize-sweet potato; maize-rice-sweet potato; Feb/March-June/July(subtropical/ maize-maize/sweet potato; maize/vegetable-sweet potato; wheat-maize/soybean; maizemid-altitudeand rice-sweet potato; rice-wheat; wheat-maize; wheat-maize/green bean; rape/maize-rice.rainfed tropical Summer maize: maize/soybean-wheat; wheat-maize; wheat-maize/soybean; May/June-Sep/Octlowland) maize-soybean; maize/wheat; wheat/vegetable-sweet potato/maize/soybeanFall maize: maize-maize; maize/soybean-wheat; maize/sweet potato-soybean;July-Oct/Novmaize/cassava-soybean; maize-sweet potato-soybean; maize-maize/sweet potatoSource: IFAD-CIMMYT RRA/PRA Surveys 2000-2001.27

consumption, two main crops of maize are grown peryear, mostly on Bari land (rainfed uplands). Othercrops include millet (the second most important crop),wheat, soybean, and peas, the latter two usuallyintercropped with maize. Large cardamom, tea, broomgrasses, sugarcane, and jute are some of the majorcash crops planted in the mid-hills and terai of Nepal.Cropping calendars and cropping patterns differacross maize agroecologies, reflecting variation inenvironmental conditions such as soil, topography,irrigation, drainage, rainfall, and other climatic factors(Figure 4.1). In India, maize is largely grown underrainfed conditions during the rainy season. Themaize crop is sown after the onset of the monsoon inJune/July and invariably harvested before the firstfortnight of September. In winter, maize is grown infavorable and irrigated environments from Novemberto March. In Thailand, the first maize crop is grown inthe wet season from May to August/September. Thesecond maize crop is grown during the dry seasonfrom September to December, and gets only a fewweeks of rain; hence it is subjected to a high risk ofdrought. A third season of maize (from January toApril) after rice is only possible in irrigated paddyfields. A limited area can be planted with third seasonmaize, with the limited output often commandinga relatively high price. In Indonesia, the first crop(wet season) is planted in October/November andharvested in February/March; the second croppingcycle (dry season) (in Lampung and East Java) is fromMarch to June/July. Lampung maize farmers followa maize-maize cropping pattern, while others followa maize-cassava pattern. In East Java, rice is the mostimportant crop, and maize is only planted after rice.The Philippines and Thailand have two maizecropping seasons, with maize grown either as amain crop or as a dry season crop after transplantedrice. With sufficient rainfall and favorable weatherconditions, some areas grow a third crop of maize,usually in paddy fields. Legumes, vegetables, andcash crops such as tobacco, sunflower, and sorghummay also be planted as a relay or intercrop with maize.Maize, however, is second only to rice in Indonesiaand Vietnam. It is generally grown once a year, withfew farmers planting twice a year. Cassava, chili, sweetpotato, beans, tea, fruit trees, groundnut, tobacco,sugarcane, cotton, coffee, rubber, cashew, and blackpepper are the other major crops grown in these twocountries. In India, maize is most commonly grownin rotation with wheat in Bihar, Madhya Pradesh,Rajasthan, and Uttar Pradesh (traditional maize areas)and with chickpea in Karnataka and Andhra Pradesh(non-traditional maize areas). Black gram, green gram,or vegetables are commonly intercropped with maizeduring the rainy season. In the most marginal rainfedenvironments, farmers customarily fallow their landsafter the maize crop.In China’s subtropical/mid-altitude environment, theprimary maize production system is rainfed summermaize, although a much smaller irrigated spring maizesystem can be found in areas of Sichuan Province.A cropping pattern common in the rainfed summermaize system is winter wheat–summer maize–fallvegetable. The duration of summer maize is similar tothat of middle season rice, approximately 110 days. Inthe rainfed tropical lowlands, three cropping seasonsper year is common. The predominant maize systemis the rainfed spring maize system, but in manyJan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar AprIndiaIndonesiaNepalPhilippinesThailandVietnamChinaFigure 4.1. Maize cropping calendars, Asia.Source: IFAD-CIMMYT RRA/PRA Surveys 200-2001.28

areas the cultivation of fall and/or winter maize isalso possible. The winter maize crop is largely greenmaize, cultivated and consumed more as a vegetablethan as a grain crop. However, the cultivation of falland winter maize following spring maize has beendecreasing due to competition from other crops andto shifts in consumption preferences from maize torice. In subtropical/mid-altitude and tropical lowlandmaize regions of China, maize may be planted inrotation with a wide range of crops, including potato,rape, vegetables, melons, and wheat (in subtropical/mid-altitude maize environments).4.2 Maize Varieties Grown andFarmer PreferencesMaize farmers in Asia grew local/traditional,improved open-pollinated, and hybrid varieties.In much of Asia, local/traditional maize varietiesare grown in areas predominantly subsisting onthe crop, and modern yellow hybrids are grown inpredominantly commercial maize-growing areas.Farmers generally select maize varieties based onintended use. For home food and feed, farmersprefer to grow local/traditional white maizevarieties for their good eating quality, low materialinputs requirement (especially fertilizers), and lowproduction cost, as the seed can be recycled. If maizeis grown purely for cash income, farmers are morelikely to grow hybrid varieties, provided they haveaccess to enough capital for material and labor inputs.Farmers are aware these improved varieties willproduce higher yields when proper quantities ofinputs are supplied.When choosing hybrid varieties to plant, Asian maizefarmers prefer those with high yields, heavy grains,general resistance to pests and diseases, resistance todrought and other climatic stresses, and high shellingrecovery. In the mid-hills of Nepal and the centralisland of Visayas, Philippines, where white maize(mostly local/traditional and improved OPVs) isgrown largely for human consumption, farmers prefermaize varieties with early maturity, high millingrecovery, good eating quality, and general suitabilityto marginal soils. Other characteristics that influencefarmers’ choice of variety are grain weight, levelof productivity, maturity period, and quantity andquality of foliage.In India, the more commercial, non-traditionalmaize-growing areas of Karnataka and AndhraPradesh are generally planted to hybrids, for whichseed replacement is high (75-90%). Traditional maizegrowing areas (Bihar, Madhya Pradesh, Rajasthan,and Uttar Pradesh) are sown to more local/traditionalvarieties, especially during the rainy season, andseed replacement is very low. Winter maize is gainingimportance because it is, subject to less risk underassured irrigation and the best management practices,and gives higher yields than rainy season maize.Indian farmers often sow hybrids or composites duringthe winter season, and farm-saved seed of local/traditional varieties during the rainy season.Both subsistence and semi-commercial farmers inthe subtropical/mid-altitude maize environments ofChina cultivate hybrid varieties almost exclusively.Recycling of hybrid seed was reported in some lowmarket-surplusvillages, but they represented only asmall share of the total maize area. In rainfed tropicallowland environments, farmers grew hybrids, local/traditional varieties, and improved OPVs. Local/traditional varieties were preferred for cultivation onhilly terrain, for their suitability to marginal soils, andfor their consumption qualities. The performance ofimproved OPVs varied by location, but in sites wherethey continued to be cultivated, farmers preferredthem for their consumption qualities, good eardevelopment, drought tolerance, and earlier maturity.The fact that farmers can save seeds for planting wasanother advantage of improved OPVs.Part of the RRA/PRA survey was a researcher-ledfarmer exercise on ranking desirable characteristicsof maize varieties. As expected, farmer-respondentsranked high yield as the most desirable and importantcharacteristic, especially if the crop is grown for cashincome. Where maize is grown for food, farmersconsidered good eating quality as the most importantcharacteristic. Other maize characteristics preferredby Asian maize farmers included pest and diseaseresistance, heavy grain weight, and full maize ear(Philippines); early maturity, lodging resistance(Nepal); reasonable seed price (Thailand); insectresistance, firm stalks, and large ears (Vietnam); anddrought tolerance, insect and disease resistance,lodging resistance, and upright leaf structure forintercropping (China).4.3 Land Preparation and CropManagement PracticesIn general, land preparation for maize productionin Asia consists of one or two plowing operationsusing either machine or animals, or a combinationof both, mechanized plowing, and harrowing andfurrowing using draft animals. In some remote andsloping villages, land preparation is done eitherwith animals or manually. In Lam Dong and KaDo provinces, Vietnam, ethnic groups still practiceshifting cultivation. Their land preparation begins with29

slashing and burning forest, bush, or grassland andcontinues with hand tools. Land preparation is moreintensive in commercial maize-producing areas. In thePhilippines, land preparation consists of one or twoplowing operations, harrowing to level the field andreduce the size of soil clods, and furrowing. These landpreparation operations are often done with animaltraction, but may be mechanized on level terrain,especially if capital is available to pay for tractorrental. In Thailand, land preparation consisted of landclearing, burning crop residues, and tillage usinglarge tractors. In China’s subtropical/mid-altitudeenvironments where maize is often relay-cropped withwheat, no tillage operations are carried out beforeplanting maize. Farmers leave space for maize whensowing wheat, and maize is sown by hand in thespring. In rainfed lowland environments, land is tilledonce or twice (once is most common) before sowing.Maize seeds are commonly sown in plow marks orholes and then covered with soil. Hybrids are sownone to two seeds per hill, with 25-30 cm between hillsand 75-80 cm between furrows. Local/traditional andimproved OPVs are sown two to three seeds per hill,leaving 60 cm between furrows and 30-40 cm betweenhills. In Thailand, maize farmers used a higher seedrate when planting manually than when a mechanicalseeder is used (normal seed rate is two seeds per hill).Replanting and thinning are done as necessary.Weeds are commonly controlled twice in a cropseason, through combined manual or hand weedingand off-barring and hilling-up (in the Philippines).First weeding is done 25-30 days after sowing (DAS),while the second weeding is done at 50-60 DAS.Across Asia, herbicides are very seldom used in maizeproduction, although Thailand applies pre-emergenceherbicides after planting, followed by mechanicalweeding. Post-emergence herbicides are also usedwhen weeds are abundant. Fertilizers and otherpesticides are also applied but often at rates lowerthan recommended or than required by the crop, andonly when pest infestation is heavy.Hill farmers in Nepal apply farmyard manure (FYM)at 9-22 t/ha to increase soil fertility. Although farmershave been introduced to integrated pest management(IPM) in maize, its use has been limited. SomeFilipino maize farmers have been trained in the use ofTrichogramma ostriniae, a biological control for Asiancorn borer, yet only a few actually apply it, mainlybecause of its limited availability. Harvesting is donemanually at all sites.4.4 Labor and Material Input UseSeeds, fertilizers, and farmyard manure are the majormaterial inputs used by Asian maize farmers. Therange of seed rate used appears to be widest at 6-60kg/ha in the subtropical/mid-altitude-low marketsurplus maize environments, where (except in China)mostly local/traditional and improved OP varietiesare grown (Table 4.2). Farmers use high seeding ratesto ensure against low germination and pest problems.Meanwhile, in most irrigated lowland, rainfedlowland, and rainfed upland environments, maize isseeded at 13-34 kg/ha. Low market surplus areas inrainfed lowland environments had a somewhat largerrange for seeding rate.Table 4.2. Average use of labor and material inputs, by maize production environments, Asia.Manual laborMaize production Total Equivalent nutrients Farmyard Totalenvironment and Seed rate fertilizer manure (person- Male Femalemarket orientation (kg/ha) use (kg/ha) N P K (kg/ha) day/ha) (%) (%)Irrigated lowlandsCommercial 18-27 125-810 - - - 1125-9000 30-243 47 53Rainfed lowlandsLow market surplus 22.5-70 52-641 10.4-382 29-190 13.5-96 0-14611 176-397 47-50 49-53Semi-commercial 16-24 100-604 49-375 14-128 14-102 0-9000 28-195 47-86 14-53Commercial 13-24 100-725 42-375 14-128 2-102 0-4000 28-195 55-86 14-45Rainfed uplandsLow market surplus 16-34 50-325 30-235 7-135 7-121 0-22000 40-295 38-100 0-62Semi-commercial 16-24 50-604 30-235 7-135 7-121 0-10000 40-111 55-100 0-45Commercial 13-24 50-604 20-325 7-135 3-121 0-10000 40-181 55-100 0-45Subtropical/mid-altitudeLow market surplus 6-60 43-495 9-199 23-177 12-56 0-20000 65-295 40-53 47-60Semi-commercial 7-9 25-52 47-373 16-176 10-92 0-10000 172-321 36-49 51-64Source: IFAD-CIMMYT RRA/PRA Surveys 2000-2001.30

Across the region, urea (45-0-0), complete fertilizer(15-15-15), ammonium phosphate (16-20-0), andammonium sulfate (21-0-0) are the most popularfertilizers used in maize production. The averageamount of inorganic fertilizer used varied widelyacross the surveyed sites, mainly due to differencesin specific soil conditions and in farmers’ knowledgeof the maize plant’s basic nutrient requirements.Across all maize production environments surveyed,total inorganic fertilizer applied ranged from a low22 kg/ha in the subtropical/mid-altitude-low marketsurplus areas to a high 810 kg/ha in the irrigatedlowland-commercial environments (Table 4.2). Ingeneral, inorganic fertilizers are applied twice, firstduring sowing and then 25-30 DAS (in the Philippines)or 45-60 DAS (in Thailand). Subsistence maize farmersin rainfed upland and subtropical/mid-altitudeenvironments tend to apply the most organic fertilizer(farmyard manure) at up to 22 t/ha.Labor use in maize production was, on average, 28-295 person-days per hectare (PD/ha), with mostlymale agricultural labor (Table 4.2). Subsistence maizeproduction in the rainfed uplands and subtropical/mid-altitude areas tended to be labor-intensive usingmostly family labor. On average across maize farmsin these two environments, 62% of total labor is doneby women, who are hired mostly for planting, manualweeding, and harvesting. In Vietnam, maize shellingand drying are also regarded to be a woman’s job.4.5 Post-harvest PracticesIn general, maize is harvested manually using family,exchange, and hired labor. Farmers transport theharvested ears home to dry. Sun drying takes placein the yard, on drying pavements at home, or incommunity areas because mechanical dryers areseldom available in the villages. Shelling is doneeither manually or by a hired shelling machine. Insome cases, maize is first shelled and then sun-dried,before being sold to traders and feed mills. Sincemany households lack storage facilities, farmers havelittle choice but to sell their maize grain soon afterharvest. Farmers also find it difficult to store maizegrain long enough for prices to go up, due to variousreasons: weevils attack stored grain and lower itsquality; grain that is not dried to optimum moisturecontent, especially during the wet season, can developmold; and, in the Philippines, farmers have to selltheir harvest immediately to pay back loans fromtrader-financiers. Better-off farmers bring their grainto commercial maize mills for milling, while poorerfarmers use their own wooden or stone mills. Farmersreported that mechanical milling produces maize gritsof better quality.In Thailand, most farmers sold their grain tomerchants who provide mechanical mills and haulingservices. Milling of maize grain is done in the villageimmediately after harvesting. The merchants dryand store the milled output in their silos. Large feedmills to whom merchants sell the grain store it intheir facilities. In contrast, Indonesian farmers selltheir maize unshelled (still on the cob) right afterharvesting. Traders dry and shell the maize beforeselling it to large feed mills.Local/traditional and improved OP maize harvestedfor home use is hand-sorted at home; the smaller and“not-so-good” ears are milled, and the good ones arekept for seed. If planting the same variety the nextseason, farmers select the seed soon after harvest. Forseed, farmers choose big, bright-colored grains fromclean, pest-free ears with good husk cover. The earsare chosen from maize plants of average height andgood stand so that the second-generation crop will beresistant to lodging. The ears are sun-dried with orwithout husks and stored in dry areas in the house orin a separate storage shed. Farmers report that earswith husks store better because weevils, a commonstorage pest, do not easily infest them. The ears areshelled manually just before use.4.6 Yields and Yield GapAs expected, hybrid varieties yield more thanlocal/traditional and improved OP varieties, withrainfed lowland and rainfed upland environmentsgiving comparable levels of good yields (Table 4.3).Across production environments, hybrid maize yieldsranged from 1.6 t/ha in subtropical/mid-altitudelowmarket surplus areas to 6.5 t/ha in rainfedupland-commercial, and 7.0 t/ha in subtropical/midaltitude-commercialenvironments. In contrast, yieldsof local/traditional maize varieties ranged from a lowof 0.9 t/ha in rainfed lowland and subtropical/midaltitudelow market surplus areas to a high of 3.0 t/hain rainfed lowland low market surplus environments.Recycled hybrids, commonly used in Indonesia,yielded from 2.3 to 4.3 t/ha, which is more than theyield of hybrid maize grown in the subtropical/midaltitude-semi-commercialareas.In all surveyed sites, farmers are aware that theproductivity of their maize crop can be improvedand reported several reasons for the yield gap. Firstand foremost are the erratic, unpredictable weatherconditions that affect crop growth and yields.Tropical storms, such as often occur in the Philippines(Camarines Sur and Leyte), can easily destroy crops.In some areas, weather extremes (heavy rains andflooding early in the maize season and drought31

in the later stages of crop growth) adversely affectmaize production. Second, Asian maize farmerstend to use less than the recommended amounts offertilizers because they lack the capital to purchaseinputs. Farmers cited soil acidity, declining soilfertility, and the continued loss of fertile topsoil dueto erosion (particularly in the hills of Nepal) as otherprobable causes of maize yield gaps in Asia. In India,farmers reported that maize yield gaps are explainedby low seed replacement, poor seed quality, andineffectiveness of recommended agronomic practices.Other causes include pest incidence, poor agriculturalextension services that contribute to farmers’insufficient access to improved technology and/ortechnical information, and poor cultural managementpractices.Table 4.3. Maize yield by production environment, Asia.Maize production environment andRange of maize yield (t/ha)market orientation Local/ traditional Improved OPVs Hybrids Recycled hybridsIrrigated lowlandsCommercial 1.3 – 2.5 2.1 – 3.5 3.8 – 5.5 2.3Rainfed lowlandsLow market surplus 0.9 – 3.0 1.9 – 3.0 1.9 – 6.0 -Semi-commercial 1.3 – 2.3 1.7 – 3.0 3.3 – 5.5 2.3Commercial 1.2 – 1.7 1.5 – 3.0 3.5 – 5.6 2.5Rainfed uplandsLow market surplus 1.3 – 2.2 1.7 – 2.7 3.3 – 4.4 -Semi-commercial 1.3 – 2.3 1.7 – 2.5 3.3 – 4.8 -Commercial 1.3 – 2.3 1.7 – 3.5 2.8 – 6.5 3.5 – 4.3Subtropical/mid-altitudeLow market surplus 0.9 – 2.4 1.2 – 3.0 1.6 – 4.7 -Semi-commercial 1.2 – 1.6 1.4 – 1.9 2.0 – 4.5 -Commercial 1.5 – 2.5 1.9 – 2.9 3.0 – 7.0 -Source: IFAD-CIMMYT RRA/PRA Survey 2001.- Not reported.32

5. Maize Production ConstraintsThe increasing demand for feed grain from thelivestock and poultry industries coupled with the highyielding maize varieties now available on the marketgive Asian farmers good reason to intensify maizeproduction. However, as with any other agriculturalenterprise, maize production is constrained byproblems that depress yields and productiondespite the advance and availability of agriculturaltechnologies. The most commonly reported bioticconstraints are downy mildew, Asian corn borers,stem borers, stalk rot, weevils, and weeds. Abioticconstraints include declining soil fertility, soil erosion,and drought. Improper or unbalanced use of fertilizers,lack of capital, and poor technology transfer wereother maize production constraints identified.5.1 Biotic ConstraintsBiotic and abiotic constraints are similar across maizeproduction environments and differ only in intensityat each site. Maize downy mildew (Peronosclerosporaspp.), a major constraint to Asian maize production,can cause crop losses of more than 80%. Plants infectedearly in the growing season usually die within a monthor do not produce any ears, while plants infectedlater produce very small ears with scattered grains.Downy mildew was reportedly present in all maizegrowingprovinces of Indonesia, in South Cotabato andCotabato, Mindanao (Philippines), in the upper north,northeast, and lower north regions of Thailand, in theCentral and Western Development Regions of Nepal,and in Uttar Pradesh, Madhya Pradesh, and Rajasthan(India). In 1987, an epidemic of downy mildew inNepal caused yield losses as high as 50%. In areas ofIndonesia where maize follows maize or sugarcane,downy mildew incidence has been increasing. Duringthe 1996 cropping season in Central Lampung, Javadowny mildew (Peronosclerospora maydis) affecteda total of 7,665 ha, of which 2,880 (about 38%) weretotally damaged or heavily infected. Epidemics withan incidence of 20-90% are not uncommon, and annuallosses average 40% (Dalmacio 2000). In Indonesia,downy mildew is the most important biotic stressaffecting maize production; thus, to be commerciallyreleased, varieties must possess downy mildewresistance as well as high yield potential.Asian corn borer (Ostrinia furcanalis Guenee) isobserved in Isabela, Southern Tagalog, SouthCotabato, and Cotabato (Philippines); in thesoutheastern region, Mekong River Delta, and thecentral coasts and highlands of Vietnam; in therainfed summer maize systems of the subtropical/mid-altitude regions of Sichuan and southern ShaanxiProvince (China); in the rainfed spring and fall maizesystems of the rainfed lowland regions of China;and in the lower north and upper northeast regionsof Thailand. In the Philippines, Asian corn boreris considered the most destructive pest of maize(Morallo-Rejesus and Punzalan 2002); its incidencehas reportedly increased over the last 10 years,causing yield losses of 30-100%. Significant effortshave been invested in the development of biologicalcontrols, especially in the identification and massrearing of controls for field release. Trichogrammaostriniae, an egg parasitoid used to control Asiancorn borer, was introduced to maize farmers at a fewsurvey sites in the Philippines. Farmers, however,reported that the limited and unreliable supply ofTrichogramma ostriniae constrained its adoption andfield application.In Vietnam, farmers reported an upward trend ininsect and disease infestation, particularly stemborer and leaf blight. Depending on temperature andclimate, leaf blight is incited by Helminthosporiummaydis in the warmer lowland tropics and Exerohilumturcicum in the cooler climates at intermediateelevations and the highlands, and in the winterseason in tropical lowlands (Dowswell et al. 1996).Despite using more pesticides on hybrid varieties,farmers reported that these pest and diseases canmore than 30% of expected yields to be lost. Stemborers were reportedly present in Indonesia’s outerislands; Nepal’s terai, as well as its Eastern, Central,and Western Development Regions; India’s Karnatakaand Andra Pradesh Provinces, and Vietnam’s centralhighlands, central coasts, and southeastern andMekong River Delta regions. In China, Turcicumand maydis leaf blight, as well as banded leaf andsheath blight (BLSB), were reported by farmers asconstraints to maize production in surveyed locationsin both subtropical/mid-altitude (Sichuan andsouthern Shaanxi Provinces) and rainfed lowland33

production (Guangxi Province) regions. The besteconomical solution to leaf blight is resistant cultivars.Many tropical germplasm complexes have goodresistance to leaf blight, and many tropical inbredlines developed at CIMMYT and IITA are highlyresistant. Some resistance genes have been cloned andtagged, which has helped the private sector to quicklyintroduce resistance into susceptible but high-yieldinggenotypes.Stalk rots are another serious and widespread groupof maize diseases in Asia, and early maturing,highland floury germplasm is highly susceptibleto them. Stalk rot is associated with good growingconditions early in the season, followed by stressfulconditions later in the season. Foliar diseases,imbalanced soil fertility, and boring injury by variousinsect pests increase the likelihood of stalk rot. Stalkrot is the decay of the stalk’s internal pith tissues, butthe term is often used to indicate stalk breakage, stalklodging, premature plant death and, occasionally,root lodging. Stalk rots are broadly classified intopre-flowering (Erwinia and Pythium) and postflowering(Fusarium, Macrophomina, Cephalosporium,etc.) stalk rots (Lal et al. 2000). The nature of stalk rotis often complex, as a number of fungi, nematodesand, sometimes, bacteria are involved in causing thedisease. Yield losses due to stalk rots may result frompremature plant death or lodging, or ear rot causedwhen lodged plants come in contact with the soil.Stalk rot was reported in Isabela and Southern andCentral Mindanao (Philippines), Eastern DevelopmentRegion (Nepal), Central Plain (Thailand), Sichuan andGuangxi Provinces (China), and Southeastern andMekong River Delta regions (Vietnam).An overriding concern of maize farmers throughoutthe tropics is grain damage by storage pests,particularly the maize grain weevil (Sitophiluszeamais), in areas where maize is used for humanconsumption. Dent grain genotypes are morevulnerable than flints; the softer endosperm makesit easier for weevils to lay eggs in the grain and forthe larvae to damage the kernel (Dowswell et al.1996). The extent of grain damage, however, mayalso depend on the duration of storage. Weevilsare reported prevalent in Nepal, Cebu, and Leyte(Philippines), in India’s Central and Western UttarPradesh, Madhya Pradesh, and Rajasthan, and insouthwestern China’s maize producing regions.Although the ears are thoroughly sun-dried for twoto three days prior to storage, weevil infestation isstill very common. Tight, undamaged husk coversreduce weevil infestation in the field. Unfortunately,many of the highest-yielding improved tropical maizematerials have relatively poor husk cover, makingthem highly susceptible to weevil damage. CIMMYThas identified source materials that possess resistanceto these pests, and is now conducting cellular andDNA studies to single out the mechanisms underlyingresistance, with an eye to transferring this trait to elitemaize genotypes.Across Asia, weeds are a substantial problem not onlyin maize production but in other crops as well. Severalspecies (A. spinosus, C. odorata and Ipomoea triloba) arepersistent problems every cropping season, causingyield losses as high as 100% if no hand weeding orherbicide control is used. It is particularly prevalentin the irrigated areas of Indonesia, all surveyed sitesin the Philippines, Central and Western DevelopmentRegions of Nepal, and in Karnataka, Andhra Pradesh,Eastern Uttar Pradesh, and Bihar, India.5.2 Abiotic ConstraintsAmong abiotic constraints, low and declining soilfertility appears to seriously affect maize productionin Asia, particularly in Thailand’s unfavorableuplands, the Philippines’ sloping and hilly areas,Nepal’s Central and Western Development Regions,karst-dominated environments in China’s GuangxiProvince, and Karnataka and Andhra Pradesh, India.Decline in soil fertility is often a result of soil erosiondue to intensified land use and rapid decline in fallowperiods, coupled with the extension of agriculture intomarginal lands (Pingali and Pandey 2001). In the hillyareas of Bukidnon and South Cotabato, Philippines,farmers are keenly aware of the loss of fertile topsoildue to erosion, yet techniques and technologiesto reduce erosion are not widely practiced. Somefarmers in Leyte, also in the Philippines, establishrock walls and follow contour plowing, but theadoption of contour hedgerow technology is oftennot sustained because of the intensive labor requiredand the farmers’ perception that shading may affecttheir maize crop. In Thailand, most maize farmers usesingle-cross hybrids, yet yields were still low becauseof poor soil fertility. Some farmers cannot afford toapply appropriate fertilizer levels, which also results inpoor maize yields. Because labor is scarce, the crop isnot cared for adequately.Using high-quality data spanning 38 years (1961-98)and collected at 91 stations in 15 countries, Manton etal. (2001) found that the number of rainy days (at least2 mm of rain) has decreased significantly throughoutSoutheast Asia (and the western and central SouthPacific), and that the annual number of hot days andwarm nights has significantly increased. This climatictrend was felt in the surveyed villages, and droughtwas widely considered the most important abioticconstraint to maize production. Drought at any crop34

development stage affects production, but maximumdamage is inflicted when it occurs around flowering.Farmers may respond to drought at the seedlingstage by replanting their crop, and some yield maybe salvaged when drought occurs at later crop stages,but drought at flowering can be mitigated only byirrigation (Pingali and Pandey 2001). Drought wasexperienced in Indonesia, Philippines, Nepal, China,some parts of Thailand and Vietnam, as well as inKarnataka and Andhra Pradesh, India. In Vietnam,the second maize crop (planted at the end of the rainyseason) is usually affected by drought if the rains stopearly. Similarly, the area planted to fall maize in Chinahas declined partly due to problems with drought.In in rainfed regions of India, the problem is droughtduring the summer and waterlogging in times ofexcess rainfall. In the Philippines, a severe drought wasbrought on by the El Niño phenomenon in 1997, andsome crops, including maize, were severely damaged.5.3 Socioeconomic and Policy-Related ConstraintsAs is the case with biotic and abiotic stresses,socioeconomic and policy-related constraints to maizeproduction are similar across environments andinclude poor technology transfer; high input prices,and low output prices, as well as farmers’ inadequateinvestment in inputs such as fertilizers, lack of capitalor of low-interest credit sources, lack of information oncrop management, lack of appropriate varieties, andlack of post-harvest facilities.Maize farmers across production environmentscommonly reported poor technology transfer servicesor insufficient agricultural extension assistancefrom local government agencies. In the Philippines,Thailand, Nepal, and Vietnam, only a few localextension workers are available to serve big farmingcommunities and large numbers of farmers, and,as a result, some are not reached at all. In remoteareas, poor transportation systems and difficultworking conditions combined with low salarieslead to ineffective service and lack of motivationamong extension workers. There is also concern thatgovernment agricultural technicians need to be bettertrained to address specific production problemsand provide better, up-to-date information. Farmerscommented that the inability of extension servicesto provide adequate information on agriculturaltechnologies contributes to poor farm productivity.In Indonesia, public extension agencies have beenreorganized and now consist of researchers andextension personnel working together to better servethe farmers. Reorganization and policy changeshave also occurred in China, but policies allowingresearch/extension organizations to market seed andinputs, combined with policies requiring those sameorganizations to finance their own activities haveoften resulted in conflicting priorities and objectives.In Nepal, some non-government organizations(NGOs) provide updated and improved agriculturaltechnologies, but tend to be more concerned aboutsocial awareness; thus farmers still rely on extensionworkers for new maize production technologies.Vietnamese farmers, realizing the lack of extensionworkers to inform them of new technologies, rely ontheir interpersonal network of co-farmers, friends,and relatives.Farmers’ nutrient management practices (e.g.,inefficient fertilizer applications, imbalance innutrients provided) are serious concerns in the maizegrowingareas of Asia. Often due to lack of capital,farmers apply fertilizers at lower-than-recommendedrates and do not practice regular soil testing andmonitoring. The use of organic fertilizers such asfarmyard manure is decreasing because of time andlabor requirements and diminishing availability. Somefarmers also perceive them to be less effective thanchemical fertilizers. Meanwhile, incorporation of cropresidues is practiced only where land preparationis mechanized. In Indonesia, farmers apply largeamounts of fertilizers to realize the full potential ofhigh yielding maize varieties; however, the practiceleads to soil acidity in the long run. In recent years inVietnam, many useful agricultural extension activitieshave focused on crops, animals, and integrated pestmanagement (IPM), but hardly any have addressedfertilizers and soil improvement technologies. Peopleare worried that if soil quality improvement is notaddressed soon, it may be difficult to ensure futurefood self-sufficiency, especially in remote uplandcommunities.Asian farmers reported that prices of maizeproduction inputs such as seed, fertilizer, tractorrental, and hired labor have been increasing throughthe years, while output prices have either remainedthe same or decreased, resulting in lower profitsespecially for farmers living in remote areas. In thePhilippines, although material inputs are alwaysavailable, most farmer-respondents reported theyhave insufficient capital to purchase inputs directly,and instead obtain them from private traderfinancierswho provide inputs on loan, at high interestrates. This arrangement does not always allowfarmers much choice of materials within the availablesupply of fertilizers, pesticides, or even maize seed.With private traders, material inputs are often pricedhigher than the prevailing market retail price. Farmerswho use recycled seed may have less financial stress,35

and the lack of capital at planting time may notseriously hamper their planting schedule. However,they are aware that much lower yields are obtainedfrom recycled seed, especially if no fertilizer is applied.Farmer cooperatives are the best source of productioninputs, but few are successful enough to adequatelysupply the needs of their members. In Nepal, farmerscomplained that the improved maize varietiesavailable are not well-suited to their environments ortheir tastes.Farmers, especially those in remote areas, have hada perennial problem with low output prices. Becausemarkets are so distant, these farmers—served more byprivate trader-financiers—tend to pay more for theirinputs and receive less for their products. The poorerand more marginal among them are likely to be forcedto stop maize cultivation as a result of accumulatedlosses at the end of the crop season. Others whodepend on maize for household food security andanimal feed continue to cultivate maize, but at lowlevels of productivity and profitability.36

6. Priority Constraints for Maize Researchand Development6.1 Methodology for IdentifyingPriority ConstraintsGiven the many constraints reported in each countryand maize production environment, a way had tobe found of combining and comparing constraintsacross the region to put together a prioritized agendafor maize R&D in Asia. This study used a modifiedversion of the methodology developed by CIMMYT(Pingali and Pandey 2001) to prioritize maizeproductivity constraints that farmer-respondentsidentified during the country RRA/PRA surveys.Three criteria were used for prioritizing the list offarmer-identified constraints: efficiency, extent of ruralpoverty, and extent of marginality of the maize agroenvironment.Details of how each index was createdand the weights used for deriving a composite indexthat included all three criteria are found in Table 6.1.The efficiency index prioritizes constraints in terms ofgetting the biggest ”bang for the research buck” (i.e.,the highest return on the investment) and estimatesthe expected production gain associated withalleviating the constraint. In contrast to the countrylevel,prioritization-exercise efficiency index, whichwas based on actual volume of maize production,the regional maize R&D prioritization efficiencyindex uses a particular maize environment’s marketorientation and contribution (share) to total regionalmaize production (or maize area), 9 to determine theimportance of the typology in the region. The riskinherent to research investments is quantified in termsof the probability of success in finding a technologicalor policy solution that will alleviate the constraint,based on the maize scientists’ knowledge of thesolution.Even where appropriate technologies are available,their adoption by farmers is not guaranteed. Toquantify the probability that farmers in a particularlocation will adopt a technology, CIMMYT drew onfarmer history of technology adoption and patternsof adoption for that location. At the country-levelnational maize R&D priority-setting workshops, thisinformation was drawn from both historical data andthe most informed knowledge of maize scientistsparticipating in the exercise.The poverty index redirects the focus of the efficiencycriteria by targeting investments to areas whererural poverty is highest. The commonly acceptedmeasure of absolute poverty is the proportion of thepopulation living below the poverty line, measuredas the lowest annual income level required for acitizen to have the basic necessities: food, housing,and clothing. Secondary data on by-country, regional,Table 6.1. Prioritizing constraints across maize environments and geographic regions.Efficiency index Poverty index Marginality index Combined indexis a product of: is a product of: is a product of: is equal to:- Importance of constraint in the particular - The efficiency index and share - The efficiency index and 0.5*Efficiency indexecology and geographic region of the rural population living inverse of the average maize +- Yield gain associated with below the poverty line in the yield in the particular ecology 0.3*Poverty indexconstraint alleviation particular ecology and geographic and geographic region, as +- Share of the particular ecology and region to the region’s total obtained from the RRA/PRA 0.2*Marginality indexgeographic region in the region’s total number of rural poor surveysmaize production (or area)- Probability of success in finding asolution to the constraint- Adoption history (% farmers that haveadopted new technologies in the past)Source: Modified from Pingali and Pandey (2001).9This project used production-based and area-based indices in determining priority constraints to maize production in Asia.37

ural-urban poverty was used to estimate the numberof rural poor in Asia and calculate the poverty index.The marginality index modifies the efficiency indexby targeting investments toward more marginalproduction areas, on the assumption that morecommercial areas are being served by the privatesector. The inverse of the estimated average maizeyield obtained in a particular maize-producinggeographic region or ecology during the by-countryRRA/PRA surveys was used as a measure ofmarginality index.At the country level, the constraints identified fromthe RRA/PRA surveys were ranked across all maizeecologies and geographic regions using the abovethree indices and a composite index. The weightsused in computing the composite index may varydepending on the relative importance of each indexand on the mission and perspective of the user. As theobjective of this international study was to delineateresearch and development guidelines, efficiency wasused as the primary determining factor in allocatingscarce public (and perhaps even private sector)resources, with important consideration given to theextent of poverty and marginality of environmentwithin the ecology and geographic region (seedetailed discussion in the next section).The country-level planning process took place in anational maize R&D priority-setting workshop, withthe participation of senior maize researchers from thepublic and private sectors, regional maize programdirectors, and other stakeholders. The three-to-fourdaynational maize workshops presented findingsfrom the RRA/PRA work, inventoried current andpotential technologies for alleviating the identifiedconstraints, and identified technologies currently notavailable in the country, but that could be brought infrom abroad. The in-country workshop ranked theproposed solutions based on potential for alleviatingthe constraints, and identified country-level policiesneeded for the rapid promotion, deployment, andadoption of the proposed solutions.The regional priority-setting exercise: (a) classified theby-country geographic regions into the productionenvironment-market orientation matrix; (b) combinedthe by-country farmer-identified maize productivityconstraints; (c) combined similar constraints andcalculated appropriate parameters for each constraintin each cell of the production environment-marketorientation matrix, and (d) calculated the production(or area)-based indices and ranked the constraintsaccording to the computed indices.The approach taken in two countries, India and China,were somewhat different from those taken in the othercountries included in the regional study. Due to themagnitude of production levels in India, each set ofproritized constraints was analyzed with and withoutthose from the subtropical/mid-altitude-commercialmaize production systems of India to isolate theimpact of these high-production, high rural povertyenvironments on regional maize R&D priorities.Issues of great magnitude in both maize productionand area relative to other countries present evenmore of a consideration for the inclusion of Chinain regional priorities. Furthermore, although theemphasis of the broader seven-country projectwas primarily on tropical lowlands and, to a lesserextent, semitropical/mid-altitude maize productionenvironments, such a focus in China would havelimited the analysis to a relatively small percentageof total Chinese maize area in southwestern China.While maize production and utilization as food,feed, and income source are indeed important tofarmer livelihoods in this region, the limited focuswould have precluded the discussion of maizeresearch priorities on a national level. To betterrepresent the overall range of maize production inthe country and expand maize characterization inChina to include major maize production systems,the China team broadened the scope of its research toinclude temperate maize production environments.Therefore, much of the Chinese maize is grownin environments very different from most maizeenvironments investigated in the other six countries,to which only the tropical and subtropical/midaltitudeenvironments in southwestern China aresimilar. Moreover, although it is possible to reclassifythe geographic regions used in China based on theproduction environment-market orientation matrix,it was not possible to reclassify and re-rank priorityconstraints discussed by farmers and maize scientistsin the same way. Results of prioritization of maizeproduction constraints in the Southwest region ofChina, where the tropical lowland and semitropical/mid-altitude growing environments are located, willtherefore be presented parallel to those of the largersynthesis.Table 6.2 presents the geographic regions andenvironments by country as classified in the maizeproduction environment-market orientation matrixused as the framework of analysis in this book. Tables6.3a and 6.3b present the key parameters used inprioritizing regional constraints, respectively, with andwithout subtropical/mid-altitude-commercial maizeproduction systems in India and Southwestern China.(Annex 5 presents the key maize production and ruralpoverty parameters by country and agroecologicalzone that were used as the basis of parameters shownin Tables 6.3a and 6.3b.) All analyses are presented,compared, and contrasted in the discussion below.38

Annex 6 shows the full list of 286 prioritized maizeproduction constraints identified by farmers in thedifferent maize production environments surveyed. Itis important to note that the constraints given prioritymay vary depending on the index used.6.2 Production-based PriorityConstraints6.2.1 Major findingsBased on the efficiency index. Tables 6.4a and 6.4bshow the top 30 constraints, associated by maizeenvironment and market orientation, according to theproduction-based efficiency index, with and withoutsubtropical/mid-altitude (STMA)-commercial maizeproduction in India and the Southwest region of China,respectively. The numbers of priority constraints inthe different maize production and market orientationareas are similar with and without India’s STMAcommerciallists. Sixteen of the top thirty productionbasedpriority constraints were reported by farmersas prevalent in rainfed upland (semi-commercialand commercial) maize production environments(Table 6.4a). Ten to eleven constraints from the rainfedlowland (semi-commercial and commercial) systems,and two to three constraints from the irrigatedcommercial areas were included in the list. All of thehighest ranked production-based constraints in China’sSouthwest region pertained to the low-market–surplus,rainfed spring maize production system in rainfedupland environments (Table 6.4b).Interestingly, nine of the top ten priority constraintsin each list, with and without STMA-commercialmaize production in India, are exactly the same. Theseresults indicate that, based on the efficiency indexTable 6.2. Classification of by-country geographic regions in the maize environment-production orientation matrix.Maize productionMarket orientation of maize productionenvironment Low market surplus Semi-commercial CommercialIrrigated China: Sichuan Province Indonesia: All irrigated areas(corresponds to irrigated springThailand: Irrigated lowland areas in themaize system in China’supper north regionSouthwest maize agroecologicalVietnam: Lowland areas of central andregion)coastal highland, northern, andsoutheastern and Mekong Delta regionsRainfed lowland China: Guangxi Province Indonesia: Java and Bali and outer Indonesia: Rainfed lowland areas of(corresponds to rainfed spring islands outer islandsmaize system in China’s Nepal: Terai area of all regions Philippines: Broad plain and hilly areasSouthwest maize agroecological Philippines: Upland plains of Southern of Cagayan Valleyregion) Tagalog, Bicol, and Central Mindanao Thailand: Favorable upland in theVietnam: Upland areas of northernregioncentral, lower north, lower northeast,upper north, upper northeast regionsVietnam: Upland areas of central andcoastal highland, and southeastern andMekong Delta regionsRainfed upland Philippines: Rolling to hilly and Indonesia: Dryland areas of Java and India: All areas of Karnataka andupland areas of Central and Bali, and some outer islands Andhra PradeshEastern Visayas Philippines: Rolling to hilly areas of Indonesia: Dryland areas of outerBicol, all areas in Southern Mindanao, islandsupland, rolling to hilly and upland Philippines: Hilly areas of Cagayanareas of Northern Mindanao ValleyThailand: Unfavorable uplands in the Thailand: Hilly or unfavorable uplandcentral region and hilly areas in the areas in the lower north, lowerupper northeast regionnortheast, and upper north regionsSubtropical/ Nepal: Mid-hills of Central and Western, India: Low rainfall areas of Central India: Medium to high rainfall areas ofmid-altitude Eastern, and Far-Midwestern Development and Western Uttar Pradesh, Madhya Central and Western Uttar Pradesh,Regions; high hills of all development Pradesh, and Rajasthan Madhya Pradesh, and Rajasthan;regionsall regions of Eastern Uttar PradeshChina: Sichuan and southern Shaanxiand BiharProvinces (corresponds to rainfed summermaize system in China’s Southwestmaize agroecological region)39

Table 6.3a. Production, area, number of rural poor, and average yield by maize production environment and marketorientation (with and without subtropical/mid-altitude-commercial maize production in India and excludingSouthwestern China), Asia.Market orientationMaize production environmentof maize production Irrigated Rainfed lowland Rainfed upland Subtropical/mid-altitudeMaize production (000 tons)Low market surplus --- --- 201.5 1,031.9(0.8) (4.0)(0.9) (4.8)Semi-commercial --- 2,378.0 5,836.2 ---(9.2) (22.6)(11.0) (26.9)Commercial 2,318.3 5,317.2 4,610.1 4,184.2(9.0) (20.5) (17.8) (16.2)(10.7) (24.5) (21.2) (0.0)Maize area (000 ha)Low market surplus --- --- 316.1 598.2(2.5) (4.7)(3.4) (6.5)Semi-commercial --- 1,594.3 3,065.4 ---(12.6) (24.2)(17.3) (33.3)Commercial 764.9 1,420.0 1,436.0 3,465.2(6.0) (11.2) (11.3) (27.4)(8.3) (15.4) (15.6) (0.0)Number of rural poor (millions)Low market surplus --- --- 5.6 7.5(2.7) (3.5)(6.0) (8.0)Semi-commercial --- 19.9 6.2 ---(9.4) (3.0)(21.2) (6.7)Commercial 21.4 6.6 26.6 118.3(10.1) (3.1) (12.5) (55.8)(22.8) (7.0) (28.3) (0.0)Average maize yield (t/ha)Low market surplus --- --- 0.70 1.70Semi-commercial --- 1.80 2.82 ---Commercial 2.78 3.78 3.42 1.03Source: IFAD-CIMMYT country-level maize R&D prioritization workshops 2001-2002.Note: Figures in parentheses are shares of regional totals. Italicized proportions are for without STMA-commercial maize production in India.Table 6.3b. Production, area, number of rural poor, and average yield by maize productionenvironment, Southwestern China.Maize production environment Rainfed spring maize Rainfed summer maize Rainfed fall maizeMaize production (000 tons) 13,904.6 624.9 1,093.6Maize area (000 ha) 3,550.4 159.6 279.2No. of rural poor (millions) 7.3 3.9aAverage maize yield (t/ha) 3.96 3.60 2.64Source: IFAD-CIMMYT country-level maize R&D prioritization workshops 2001-2002.aRainfed fall maize production system overlaps with rainfed spring maize production system.40

Table 6.4a. Top 30 priority constraints to maize production according to the production-based efficiency index,Asia (with and without STMA-commercial maize production in India and excluding Southwest China’s maizeproduction systems).With STMA-commercialproduction, IndiaWithout STMA-commercial production, IndiaMaize production Farmer-identified Maize production Farmer-identifiedConstraint environment and maize production environment and maize productionranking market orientation constraint market orientation constraint1 RFLL-commercial Drought RFLL-commercial Drought2 RFUP-semi-commercial Drought RFUP-semi-commercial Drought3 RFUP-commercial Drought RFUP-commercial Drought4 RFUP-semi-commercial Inappropriate fertilizer use RFUP-semi-commercial Inappropriate fertilizer use5 RFUP-commercial Poor availability of quality seed RFUP-commercial Poor availability of quality seed6 RFLL-commercial Corn borers (ear/stem borers) RFLL-commercial Corn borers (ear/stem borers)7 RFUP-semi-commercial Lack of appropriate variety RFUP-semi-commercial Lack of appropriate variety8 RFUP-commercial Weeds and weed management RFUP-commercial Weeds and weed management9 RFUP-semi-commercial Soil erosion and landslides RFLL-commercial Rust10 RFLL-commercial Rust RFUP-commercial Downy mildew11 RFUP-commercial Downy mildew RFLL-commercial Waterlogging12 RFLL-commercial Waterlogging RFUP-commercial Stem borers13 RFUP-commercial Stem borers RFUP-commercial Soil erosion14 RFUP-commercial Soil erosion RFUP-semi-commercial Soil erosion/landslides15 RFUP-commercial Soil infertility RFUP-commercial Soil infertility16 RFUP-semi-commercial Soil infertility and acidity RFUP-semi-commercial Low output prices17 RFUP-semi-commercial Low output prices RFUP-semi-commercial Downy mildew18 RFUP-semi-commercial Downy mildew RFUP-semi-commercial Lack of post-harvest facilities19 RFUP-semi-commercial Lack of post-harvest facilities RFUP-semi-commercial Soil infertility and acidity20 RFLL-commercial Soil infertility RFLL-commercial Soil infertility21 STMA-commercial Post-flowering stalk rot RFLL-commercial Lack of capital22 RFLL-semi-commercial Lack of capital/low interest RFLL-commercial Banded leaf and sheath blightcredit sources (inadequatecredit support)23 RFLL-semi-commercial Drought Irrigated commercial Lack of capital24 RFLL-commercial Lack of capital Irrigated commercial Undeveloped irrigation system /water shortage25 RFUP-semi-commercial Limited capital / access to credit RFLL-commercial Lack of post-harvest facilities26 RFLL-commercial Banded leaf and sheath blight Irrigated commercial Inefficient fertilizer use/inappropriatefertilizer application27 STMA-commercial Lack of quality seed RFLL-semi-commercial Drought28 Irrigated commercial Lack of capital RFLL-commercial Rodents29 Irrigated commercial Undeveloped irrigation system / RFUP-commercial Post-flowering stalk rotwater shortage30 RFLL-commercial Lack of postharvest facilities RFLL-commercial FloodingBy production Frequency count By production Frequency countenvironment and (no. of priority environment and (no. of prioritymarket orientation constraints) market orientation constraints)Irrigated-commercial 2 Irrigated-commercial 3RFLL-semi-commercial 2 RFLL-semi-commercial 1RFLL-commercial 8 RFLL-commercial 10RFUP-semi-commercial 9 RFUP-semi-commercial 8RFUP-commercial 7 RFUP-commercial 8STMA-commercial 2By type of constraintBy type of constraintDrought 4 Drought 4Biotic 8 Biotic 9Abiotic 6 Abiotic 7Socioeconomic 12 Socioeconomic 10Source: IFAD-CIMMYT RRA/PRA Surveys 2000-2001 and country-level maize R&D prioritization workshops 2001-2002.Note: RFLL – rainfed lowland; RFUP – rainfed upland; STMA – subtropical/mid-altitude.41

alone, first addressing the problem of drought in therainfed lowland-commercial, rainfed upland-semicommercial,and commercial production environmentswould provide the highest returns to maize R&Dinvestments in Asia. Drought is estimated to affectabout 6.8 M ha in these three production environmentsalone (53.5% of the total regional maize area), and thealleviation of this particular constraint has enormouspotential impact on maize production. Drought-proneareas include the rainfed lowland areas of CagayanValley and Central Mindanao in the Philippines,the southeastern region and Mekong River Deltaof Vietnam, and the rainfed upland areas of Java(Indonesia), Southern Mindanao (Philippines) andthe upper and lower northeast and central regions ofThailand. In the rainfed upland-low market surplusproduction environments of Southwest China,alleviation of production constraints caused bydrought is also the top priority (Table 6.4b).Next on the list of priority constraints areinappropriate fertilizer use in the rainfed uplandsemi-commercialareas, poor availability ofappropriate varieties and quality seed in therainfed upland-commercial maize areas, and earand stem borers in the rainfed lowland-commercialenvironments (Table 6.4a). Other maize productionconstraints included biotic constraints such as rust,Table 6.4b. Top 30 priority constraints to maize production according to the production-based efficiency index,maize production systems of Southwest China.Constraint Maize production environment andranking market orientation Farmer-identified maize production constraint1 Rainfed spring maize (subsistence) Drought2 Rainfed spring maize (subsistence) Low soil fertility3 Rainfed spring maize (subsistence) Cultivated varieties susceptible to insects and diseases4 Rainfed spring maize (subsistence) Low level of investment in inputs5 Rainfed spring maize (subsistence) Difficulties in purchasing seed of desired varieties (few available outlets)6 Rainfed spring maize (subsistence) Poor grain quality7 Rainfed spring maize (subsistence) Few opportunities for farmers to sell maize due to undeveloped market8 Rainfed spring maize (subsistence) Banded leaf and sheath blight9 Rainfed spring maize (subsistence) Poor knowledge of cultivation techniques and crop management10 Rainfed spring maize (subsistence) Corn borer11 Rainfed spring maize (subsistence) Lack of functioning dissemination system for information on new varieties12 Rainfed spring maize (subsistence) Lodging caused by high winds13 Rainfed spring maize (subsistence) Maydis leaf disease14 Rainfed spring maize (subsistence) Turcicum leaf blight15 Rainfed spring maize (subsistence) Small production scale (low per capita land)16 Rainfed spring maize (subsistence) Low seed quality17 Rainfed spring maize (subsistence) Soil erosion18 Rainfed spring maize (subsistence) Low-quality or fake fertilizers19 Rainfed spring maize (subsistence) Storage rodents20 Rainfed spring maize (subsistence) Ear rot21 Rainfed spring maize (subsistence) Flooding22 Rainfed spring maize (subsistence) Low maize price23 Rainfed spring maize (subsistence) High seed price24 Rainfed spring maize (subsistence) Grain weevils25 Rainfed spring maize (subsistence) Mineral deficiencies in soil26 Rainfed spring maize (subsistence) Poor transportation infrastructure27 Rainfed spring maize (subsistence) Fake or low-quality seed28 Rainfed spring maize (subsistence) Cutworm29 Rainfed spring maize (subsistence) Labor shortage30 Rainfed spring maize (subsistence) Field rodentsBy production environment andmarket orientationFrequency count (no. of priority constraints)Rainfed spring maize (subsistence) 30By type of constraintDrought 1Biotic 10Abiotic 5Socioeconomic 1442

downy mildew, post-flowering stalk rot (PFSR),and banded leaf and sheath blight (BLSB), andabiotic constraints such as soil erosion/landslides,waterlogging, soil infertility, and soil acidity(each group listed in order of priority). The list ofpriorities also included socioeconomic and policyrelatedconstraints such as low output prices, lack ofpostharvest facilities, lack of capital, poor access tolow-interest credit sources, and difficulties in marketingmaize.Among the top 30 priority constraints in all Asiaaccording to the production-based efficiency indexalone, abiotic constraints are estimated to affect about5.1 M ha, biotic constraints (pests and diseases) 8.5 Mha, and socioeconomic and policy-related constraints10.4 M ha (Table 6.5a). Excluding the STMA-commercialareas of India and Southwest China’s maize productionsystems, these constraints affect 5.6, 5.3, and 5.1 M ha,respectively, in Asia (Table 6.5b). In both cases, droughtalone affects about 47.7 M rural poor people in maizegrowingareas across Asia; thus alleviating this toppriorityconstraint could potentially add another 35%to maize yields in the region. Including the droughtproneareas of rainfed upland maize environments inChina, the area affected by drought increases by anadditional 3.6 M ha. Biotic constraints and, particularly,socioeconomic, infrastructural, and policy-relatedconstraints in China’s Southwest region, however, alsocontribute significantly to reduced maize yields andproductivity (Table 6.5c).Based on the poverty index. According to theproduction-based poverty index alone, post-floweringstalk rot received the highest ranking in the STMAcommercialmaize areas (Table 6.6), including themedium-to-high rainfall areas of India’s UttarPradesh, Madhya Pradesh, Rajasthan, and Bihar.Maize production needs in the STMA-commercialenvironments, reflected in 12 constraints, dominatedthe top 30 priority constraints. The rural poorpopulation in this environment totals about 118.3million, or 55.8% of the total rural poor populationacross all environments surveyed in Asia (see Table 6.3).Analysis without STMA-commercial India shows thatdrought received top priority in the rainfed uplandcommercialareas (Table 6.6). These areas includedrylands in Indonesia’s outer islands, the unfavorableuplands of Thailand’s Lop Buri in the Central Plainsand Phetchabun in the Lower North region, and India’sKarnataka and Andhra Pradesh. Constraints from therainfed upland-semi- and commercial maize productionareas dominated the top 30 priority list.Table 6.5a. Selected indicators of impact for the top 30 priority maize productivity constraints according to theproduction-based efficiency index in Asia (with STMA-commercial areas of India and excluding Southwest China’smaize production systems).Farmer-identified maize productivity constraintsFarmer-identified maize productivity constraintsOther Biotic Socioeco- Other Biotic Socioecoabiotic(pests and nomic/policy- abiotic (pests and nomic/policy-Impact indicators Drought problems diseases) related Drought problems diseases) relatedBased on efficiency indexBased on poverty indexEffective total areaaffected (000 ha) 6,767.4 5,098.8 8,542.1 10,438.6 7,482.0 --- 3,449.3 6,596.8Estimated maize production acrossareas affected (000 t) 16,344.7 12,851.1 17,637.6 20,187.7 15,474.3 --- 11,827.2 12,639.4Average yield gain when constraintis alleviated (%) 35.0 28.8 14.1 31.2 31.9 --- 11.9 18.8Estimated no. of rural poor affected(millions) 47.7 15.3 155.0 185.0 162.1 --- 76.1 179.4Based on marginality indexBased on combined indexEffective total area affected (000 ha) 6,767.4 3,934.7 9,702.6 9,339.9 5,335.0 6,335.2 6,255.3 4,776.4Estimated maize production acrossareas affected (000 t) 16,344.7 8,426.1 18,521.8 16,515.6 14,130.6 9,260.3 14,154.2 8,527.7Average yield gain when constraintis alleviated (%) 35.0 37.6 15.6 37.5 34.2 24.4 11.9 18.1Estimated no. of rural pooraffected (millions) 47.7 10.5 168.1 166.3 37.0 124.5 171.9 162.3Source: IFAD-CIMMYT RRA/PRA Surveys 2000-2001 and country-level maize R&D prioritization workshops 2001-2002.Note: The geographic regions and environments where specific constraints (except drought) under each of the above groups are reported are not necessarilymutually exclusive. However, careful calculations were made to avoid double counting in the above parameters.43

In contrast to priorities based on the efficiency index,those according to the production-based povertyindex generally consisted of biotic constraints. Theseconstraints include downy mildew, stem borers, andbanded leaf and sheath blight in the rainfed uplandcommercialareas, and Turcicum leaf blight, stemborers, and Maydis leaf blight in the STMA-commercialproduction areas. These biotic constraints wereestimated to affect 3.4-4.9 M ha across Asia; thus theiralleviation could improve maize production in theregion by an estimated 12% (Tables 6.5a and 6.5b).Looking more closely at the list including STMAcommercialIndia in Table 6.6, only 12 of the 30priority constraints according to the productionbasedefficiency index alone are included in thepriority constraint list based on the poverty indexalone, while 18 constraints ranked well below 30based on the efficiency index were recently addedto the top 30 poverty-based priority constraints.These new entrants are constraints reported mostlyfrom the STMA- and the rainfed upland-commercialenvironments, which together are estimated toTable 6.5b. Selected indicators of impact for the top 30 priority maize productivity constraints according to theproduction-based efficiency index in Asia (without STMA-commercial areas of India and excluding Southwest China’smaize production systems).Farmer-identified maize productivity constraintsFarmer-identified maize productivity constraintsOther Biotic Socioeco- Other Biotic Socioecoabiotic(pests and nomic/policy- abiotic (pests and nomic/policy-Impact indicators Drought problems diseases) related Drought problems diseases) relatedBased on efficiency indexBased on poverty indexEffective total area affected(000 ha) 6,767.4 5,608.1 5,264.4 5,100.9 6,767.4 4,679.6 4,874.1 3,711.6Estimated maize production acrossareas affected (000 t) 16,344.7 14,540.3 13,947.2 13,453.4 16,344.7 10,838.6 13,701.2 8,981.6Average yield gain when constraintis alleviated (%) 35.0 26.8 15.0 26.7 35.0 27.4 12.5 24.4Estimated no. of rural pooraffected (millions) 47.7 19.2 36.7 54.8 47.7 29.0 83.4 67.8Based on marginality indexBased on combined indexEffective total area affected(000 ha) 6,767.4 7,101.6 7,050.7 6,155.0 6,767.4 4,791.1 4,697.7 3,711.6Estimated maize production acrossareas affected (000 t) 16,344.7 12,851.1 16,491.0 12,449.5 16,344.7 11,185.1 12,805.1 8,981.6Average yield gain when constraintis alleviated (%) 35.0 28.8 19.7 36.1 35.0 31.1 12.5 24.1Estimated no. of rural pooraffected (millions) 47.7 15.3 56.9 52.8 47.7 29.1 67.6 67.8Source: IFAD-CIMMYT RRA/PRA Surveys 2000-2001 and country-level maize R&D prioritization workshops 2001-2002.Note: The geographic regions and environments where specific constraints (except drought) under each of the above groups are reported are not necessarilymutually exclusive. However, careful calculations were made to avoid double counting in the above parameters.Table 6.5c. Selected indicators of impact for the top 30 priority maize productivity constraints according to theproduction-based efficiency index, Southwest China.Farmer-identified maize productivity constraintsOther abiotic Biotic (pests Socioeconomic/Impact indicators Drought problems and diseases) policy-relatedBased on efficiency indexEffective total area affected (000 ha) 3,550.4 3,550.4 3,550.4 3,550.4Estimated maize production across areas affected (000 t) 13,904.6 13,904.6 13,904.6 13,904.6No. of rural poor (millions) 7.3 7.3 7.3 7.3Average yield gain when constraint is alleviated (%) 28.0 20.0 39.5 48.544

Table 6.6. Top 30 priority constraints to maize production according to the production-based poverty index, Asia (withand without STMA-commercial maize production in India and excluding Southwest China’s low market surplus maizeproduction systems).With STMA-commercial production, India45Without STMA-commercial production, IndiaConstraint Maize productionMaize productionpoverty environment and Farmer-identified maize Efficiency environment and Farmer-identified maize Efficiencyranking market orientation production constraint ranking market orientation production constraint ranking1 STMA-commercial Post-flowering stalk rot 21 RFUP-commercial Drought 32 STMA-commercial Lack of quality seed 27 RFUP-commercial Poor availability of quality seed 53 STMA-commercial Improper use of fertilizers 53 RFUP-commercial Weeds and weed management 84 STMA-commercial Post-harvest losses 58 RFUP-commercial Downy mildew 10(due to weevils in storage)5 STMA-commercial Improper cropping systems 64 RFUP-commercial Stem borers 126 RFUP-commercial Drought 3 RFUP-commercial Post-flowering stalk rot 297 RFUP-commercial Poor availability of quality seed 5 Irrigated-commercial Lack of capital 238 RFUP-commercial Weeds and weed management 8 Irrigated-commercial Inefficient/inappropriate 26fertilizer application9 RFUP-commercial Downy mildew 11 RFUP-commercial Storage pests 3310 STMA-commercial Transplanting maize under 45 RFUP-commercial Turcicum leaf blight 36late sown conditions11 STMA-commercial Turcicum leaf blight 46 RFLL-commercial Drought 112 RFUP-commercial Stem borers 13 RFUP-commercial Banded leaf and sheath blight 3813 STMA-commercial Moisture stress (drought) 129 Irrigated-commercial Undeveloped irrigation system 24(water shortage)14 STMA-commercial Lack of appropriate maturity 130 Irrigated-commercial Rodents 39varieties15 STMA-commercial Stem borer 134 Irrigated-commercial High production costs/input prices 3516 STMA-commercial Maydis leaf blight 139 RFLL-commercial Corn borers (ear/stem borers) 617 RFUP-commercial Post-flowering stalk rot 34 RFLL-semi-commercial Lack of capital/low interest credit, 31inadequate credit support18 Irrigated-commercial Lack of capital 28 RFLL-semi-commercial Lack of suitable (hybrid) varieties 3219 Irrigated-commercial Inefficient/inappropriate 31 Irrigated-commercial Waterlogging during crop 46fertilizer applicationestablishment20 RFUP-commercial Storage pests 40 RFUP-semi-commercial Soil erosion/landslides 1421 RFUP-commercial Turcicum leaf blight 42 RFUP- commercial Soil erosion 1322 RFLL-commercial Drought 1 RFUP-semi-commercial Drought 223 RFUP-commercial Banded leaf and sheath blight 44 RFLL-semi-commercial Drought 2724 Irrigated-commercial Undeveloped irrigation system 29 Irrigated-commercial Lack of labor 53(water shortage)25 STMA-commercial Inappropriate crop 119 Irrigated-commercial Downy mildew 56establishment method26 Irrigated-commercial Rodents 47 Irrigated-commercial Leaf blight/rust (foliar diseases) 5927 Irrigated-commercial High production costs/ input prices 41 RFLL-semi-commercial Rodents 4528 RFLL-commercial Corn borers (ear/stem borers) 6 RFUP-semi-commercial Soil infertility and acidity 1929 RFLL-semi-commercial Drought 23 RFLL-semi-commercial Inadequate post-harvest 65technologies/facilities30 RFLL-semi-commercial Lack of suitable (hybrid) varieties 39 RFLL-semi-commercial Poor marketing system, input/ 49output market access andundeveloped transport systemBy maize productionBy maize productionenvironment and Frequency count environment and Frequency countmarket orientation (no. of priority constraints) market orientation (no. of priority constraints)Irrigated-commercial 5 Irrigated-commercial 9RFLL-semi-commercial 2 RFLL-semi-commercial 6RFLL-commercial 2 RFLL-commercial 2RFUP-semi-commercial 0 RFUP-semi-commercial 3RFUP-commercial 9 RFUP-commercial 10STMA-commercial 12By type of constraintBy type of constraintDrought 4 Drought 4Biotic 14 Biotic 12Abiotic 0 Abiotic 4Socioeconomic/policy 12 Socioeconomic/policy 10Source: IFAD-CIMMYT RRA/PRA Surveys 2000-2001 and country-level maize R&D prioritization workshops 2001-2002.Note: RFLL – rainfed lowland; RFUP – rainfed upland; STMA – subtropical/mid-altitude.

support about 144.9 million (68.3%) of Asia’s ruralpoor people. Meanwhile, 14 constraints ranked below30 based on the efficiency index were included inthe list of poverty-based priorities without STMAcommercialIndia. The new entrants mostly came fromthe irrigated-commercial and rainfed lowland-semicommercialproduction areas where about 41.3 millionrural poor are located.Based on the marginality index. Including India’sSTMA-commercial production areas, 17 of the 30 toppriority constraints according to the production-basedmarginality index alone are those reportedly prevalentin the rainfed upland semi-commercial and commercialareas, and nine are from rainfed lowland areas (Table6.7). Similar to results using the efficiency index,drought in the rainfed lowland and rainfed uplandsemi-commercial and commercial maize environmentsranked within the top five priority constraints basedon the marginality index alone. These areas includedIndonesia’s drylands, Philippines’ Cagayan Valleyand Southern Mindanao regions, and Vietnam’scentral coastal highlands and southeastern MekongRiver Delta areas. Other top priorities includedabiotic constraints (soil erosion, infertility, and acidityproblems) and socioeconomic and policy-relatedconstraints (lack of quality seed and post-harvestfacilities, limited access to technical information, lowadoption of improved technologies, and low outputprices). All together, the marginality-based priorityabiotic constraints (excluding drought) impact aneffective total maize area of 3.9 M ha, which producesabout 8.4 M tons of maize. Alleviation of this groupof constraints could add an average of almost 38% tomaize grain yield. However, it appears that, among thepriority constraints based on the marginality of maizeenvironments, biotic constraints (pests and diseases,plus weeds) impact a high number of rural poor (about168 million), yet once alleviated would contribute theleast average gain to yield (15.6%) (see Table 6.5a).Similarly, in the list without India’s STMA-commercialproduction constraints, 19 of the 30 top priorityconstraints came from the rainfed upland semicommercialand commercial areas; drought rankedas the first priority constraint that maize R&D in Asiashould address. More constraints reported from therainfed lowland-commercial production environments,as well as more abiotic constraints, made it to thispriority list. These abiotic constraints are estimated toaffect about 7.1 M ha, on which nearly 13 M tons ofmaize are produced. Alleviating those contraints couldadd an average of about 29% to maize grain yields(Table 6.5b).Table 6.7 also shows that more farmer-identifiedmaize production constraints that ranked lowerbased on the efficiency and poverty indices rankedhigher based on the marginality index. For example,soil erosion/landslides in rainfed upland-semicommercialenvironments ranked second based on themarginality index, ninth based on the efficiency index,and thirty-first based on the poverty index. Similarly,low adoption of improved technology was number27 in rainfed lowland-semi-commercial areas basedon the marginality index, number 57 based on theefficiency index, and number 74 based on the povertyindex. Moreover, only 13 and 20, respectively, of thetop 30 priority constraints based on efficiency andpoverty alone were included in the priority list basedon marginality alone. All these observations show thatpriorities will depend on the parameter (efficiency,poverty, marginality of environment) that is perceivedto need more emphasis, according to the overall goalsand objectives of the prioritization exercise.Based on the combined index. Aggregating the abovethree indices (criteria) generated a production-basedcomposite index and ranking using a set of arbitraryweights: 50% for efficiency, 30% for poverty, and 20%for marginality of the environment (see Table 6.1). Thetop 30 priority constraints, according to the productionbasedcombined index, that should be addressedby R&D are shown in Tables 6.8a and 6.8b. Of the30 priority constraints, including those from India’sSTMA-commercial production areas, 11 are specificto the rainfed upland environments, 2 to the rainfedlowlands, 5 to the irrigated (commercial) lowlands,and 12 to STMA (commercial) environments (Table6.8a). On the other hand, the top 30 priority constraintslist, excluding India’s STMA-commercial productionareas, consists of 14 from the rainfed uplands, 7from the rainfed lowlands, and 9 from the irrigatedcommercialareas (Table 6.8b). Moreover, while bioticand abiotic constraints dominate the list in Table 6.8a,biotic constraints and socioeconomic and policy-relatedconstraints dominate the list in Table 6.8b.Maize area and production in low market surplusfarming environments are relatively low across Asia;consequently, this maize production system does notappear in priority listings based on the combined index(nor in any of the priority listings based on efficiency,poverty, and marginality). The two top 30 constraintspriority listings according to the production-basedcombined index closely follow the priority listing basedon the poverty index, which in turn is strongly linkedto the number of rural poor by geographic region andproduction environment. Low market surplus farmingnonetheless remains important at the country level,particularly in Indonesia, Nepal, and the Philippines.It is therefore important for these countries to continueinvesting (even if modestly) in low market surplusfarming research. As such, mechanisms to promotespillovers from research on more commercial areas tolow market surplus farming environments ought to beestablished and encouraged.46

Table 6.7. Top 30 priority constraints to maize production according to the production-based marginality index, Asia(with and without STMA-commercial maize production in India and excluding Southwest China’s low market surplusmaize production systems).With STMA-commercial production in India47Without STMA-commercial production in IndiaConstraint Maize productionMaize productionmarginality environment and Farmer-identified maize Efficiency environment and Farmer-identified maize Efficiencyranking market orientation production constraint ranking market orientation production constraint ranking1 RFUP-semi-commercial Drought 2 RFUP-semi-commercial Drought 22 RFUP-semi-commercial Soil erosion/landslides 9 RFUP-semi-commercial Soil erosion/landslides 143 STMA-commercial Post-flowering stalk rot 21 RFUP-commercial Drought 34 RFUP-commercial Drought 3 RFLL-semi-commercial Drought 15 RFLL-commercial Drought 1 RFUP-semi-commercial Low output prices 166 RFUP-semi-commercial Low output prices 17 RFUP-semi-commercial Inappropriate fertilizer use 47 RFUP-semi-commercial Inappropriate fertilizer use 4 RFUP-semi-commercial Lack of post-harvest facilities 188 STMA-commercial Lack of quality seed 27 RFUP-commercial Poor availability of quality seed 59 RFLL-semi-commercial Lack of capital/low interest 22 RFUP-semi-commercial Downy mildew 17credit, inadequate credit support10 RFUP-semi-commercial Downy mildew 18 RFUP-semi-commercial Weeds 3711 RFUP-semi-commercial Lack of post-harvest facilities 19 RFLL-commercial Corn borers (ear/stem borers) 612 RFUP-commercial Poor availability of quality seed 5 RFUP-commercial Downy mildew 1013 RFUP-semi-commercial Weeds 37 RFUP-commercial Stem borers 1214 RFUP-semi-commercial Soil infertility and acidity 16 RFUP-commercial Weeds and weed management 815 RFLL-semi-commercial Drought 23 RFLL-semi-commercial Lack of capital/low interest credit, 31inadequate credit support16 RFLL-commercial Corn borers (ear/stem borers) 6 RFUP-commercial Soil infertility 1517 RFUP-commercial Downy mildew 11 RFLL-semi-commercial Lack of suitable (hybrid) varieties 3218 RFUP-commercial Stem borers 13 RFUP-semi-commercial Soil infertility and acidity 1919 RFUP-commercial Weeds and weed management 8 RFLL-semi-commercial Drought 2720 RFUP-semi-commercial Limited capital/access to credit 25 RFUP-commercial Soil erosion 1321 RFUP-commercial Soil infertility 15 RFLL-commercial Rust 922 RFLL-semi-commercial Lack of suitable (hybrid) varieties 39 RFUP-semi-commercial Limited capital/access to credit 3423 RFLL-semi-commercial Rodents 43 RFLL-commercial Waterlogging 1124 RFLL-semi-commercial Post-harvest pests and diseases 49 RFLL-semi-commercial Poor marketing system, input/ 49output market access andundeveloped transport system25 STMA-commercial Improper/inadequate use of fertilizers 53 RFLL-semi-commercial Rodents 4526 RFUP-semi-commercial High price of inputs 36 RFUP-semi-commercial High price of inputs 40(including seed, transport)(including seed, transport)27 RFLL-semi-commercial Low technology adoption 57 RFUP-semi-commercial Limited farmer access to 41information and technology dueto poor extension28 STMA-commercial Post-harvest losses 58 RFLL-semi-commercial Post-harvest pests and diseases 61(weevils during storage)29 RFLL-semi-commercial Low fertilizer use 61 RFUP-semi-commercial Stalk rot 4330 RFUP-semi-commercial Limited farmer access to 38 RFLL-commercial Soil infertility 20information and technologydue to poor extensionBy maize productionBy maize productionenvironment and Frequency count environment and Frequency countmarket orientation (no. of priority constraints) market orientation (no. of priority constraints)Irrigated-commercial 0 Irrigated-commercial 0RFLL-semi-commercial 7 RFLL-semi-commercial 6RFLL-commercial 2 RFLL-commercial 5RFUP-semi-commercial 11 RFUP-semi-commercial 12RFUP-commercial 6 RFUP-commercial 7STMA-commercial 4By type of constraintBy type of constraintDrought 4 Drought 4Biotic 10 Biotic 10Abiotic 3 Abiotic 6Socioeconomic/policy 13 Socioeconomic/policy 10Source: IFAD-CIMMYT RRA/PRA Surveys 2000-2001 and country-level maize R&D prioritization workshops 2001-2002.Note: RFLL – rainfed lowland; RFUP – rainfed upland; STMA – subtropical/mid-altitude.

6.2.2 Priorities by maize productionenvironment and market orientationTo gain a better picture of priority constraints by maizeproduction environment and market orientation acrossAsia, the list of top 30 priority constraints according toa production-based combined index was augmentedwith a few selected lower-priority constraints reportedprevalent in the other maize production and marketorientation areas. These lower-priority constraintswere intentionally selected to show the type andextent of specific constraints experienced by Asianmaize farmers. Table 6.9 shows the supplemented listof priority maize constraints; constraints not includedamong the top 30 regional priorities have their regionalranking according to a production-based combinedindex indicated in parenthesis. Constraints that enteredTable 6.8a. Top 30 priority constraints to maize production according to the production-based combined index,Asia (with STMA-commercial production areas of India and excluding Southwest China’s low market surplus maizeproduction systems).Maize production environmentRanking by index usedand market orientation Farmer-identified maize production constraint Efficiency Poverty Marginality CombinedSTMA-commercial Post-flowering stalk rot (PFSR) 21 1 3 1STMA-commercial Lack of quality seed 27 2 8 2STMA-commercial Imbalanced/improper/inadequate fertilizer use 53 3 25 3RFUP-commercial Drought 3 6 4 4STMA-commercial Post-harvest losses (due to weevils during storage) 58 4 28 5STMA-commercial Improper cropping systems (mixed cropping/intercropping) 64 5 36 6RFUP-commercial Poor availability of appropriate varieties and quality seed 5 7 12 7RFUP-commercial Weeds and weed management 8 8 19 8RFUP-commercial Downy mildew 11 9 17 9RFUP-commercial Stem borers 13 12 18 10STMA-commercial Transplanting maize under late sown condition 45 10 34 11STMA-commercial Turcicum leaf blight 46 11 35 12STMA-commercial Moisture stress (drought) 129 13 83 13RFLL-commercial Drought 1 22 5 14STMA-commercial Lack of appropriate maturity varieties 130 14 86 15STMA-commercial Stem borers 134 15 95 16STMA-commercial Maydis leaf blight 139 16 107 17RFUP-commercial Post-flowering stalk rot (PFSR) 34 17 49 18Irrigated-commercial Lack of capital 28 18 43 19Irrigated-commercial Inefficient fertilizer use/inappropriate fertilizer application 31 19 45 20RFUP-commercial Storage pests 40 20 53 21RFUP-semi-commercial Drought 2 33 1 22RFUP-commercial Turcicum leaf blight 42 21 55 23Irrigated-commercial Undeveloped irrigation system (water shortage) 29 24 52 24RFLL-commercial Corn borers (ear/stem borers) 6 28 16 25RFUP-commercial Banded leaf and sheath blight 44 23 62 26RFUP-semi-commercial Soil erosion/landslides 9 31 2 27Irrigated-commercial Rodents 47 26 65 28Irrigated-commercial High production costs/input prices 41 27 59 29STMA-commercial Inappropriate crop establishment method 119 25 138 30By maize production environmentand market orientationFrequency count(no. of priority constraints)Irrigated-commercial 5RFLL-semi-commercial 0RFLL-commercial 2RFUP-semi-commercial 2RFUP-commercial 9STMA-commercial 12By type of constraintDrought 3Biotic 14Abiotic 11Socioeconomic/policy 2Source: IFAD-CIMMYT RRA/PRA Surveys 2000-2001 and country-level maize R&D prioritization workshops 2001-2002.Note: RFLL – rainfed lowland; RFUP – rainfed upland; STMA – subtropical/mid-altitude.48

the priority list that did not include India’s STMAcommercialmaize production areas are listed in italics.Some of the specific constraints added may alreadyhave been on the regional priority constraints list, butassociated with a different production environmentor market orientation. For example, poor quality ofhybrid seed was added to the rainfed upland-semicommercialenvironment, but its equivalent—pooravailability of quality seed—was already listed as apriority constraint in the rainfed upland-commercialareas. Noticeably, important constraints, such aspoor farm-to-market roads in rainfed upland-lowmarket surplus areas and nematodes in STMA-semicommercialenvironments, were close to the bottom ofthe full priority list based on the combined index.Table 6.8b. Top 30 priority constraints to maize production according to the production-based combined index, Asia(without STMA-commercial production areas of India and excluding Southwest China’s low market surplus maizeproduction systems).Maize production environmentRanking by index usedand market orientation Farmer-identified maize production constraint Efficiency Poverty Marginality CombinedRFUP-commercial Drought 3 1 3 1RFUP-commercial Poor availability of appropriate varieties and quality seed 3 5 2 2RFUP-commercial Weeds and weed management 8 3 14 3RFUP-commercial Downy mildew 10 4 12 4RFUP-commercial Stem borers 12 5 13 5RFLL-commercial Drought 1 11 4 6RFUP-commercial Post-flowering stalk rot (PFSR) 29 6 36 7Irrigated-commercial Lack of capital 23 7 31 8Irrigated-commercial Inefficient fertilizer use/inappropriate fertilizer application 26 8 33 9RFUP-commercial Storage pests 33 9 40 10RFUP-commercial Turcicum leaf blight 36 10 42 11Irrigated-commercial Undeveloped irrigation system (water shortage) 24 13 39 12RFUP-commercial Banded leaf and sheath blight 38 12 46 13RFLL-commercial Corn borers (ear/stem borers) 6 16 11 14Irrigated-commercial Rodents 39 14 52 15RFUP-semi-commercial Drought 2 22 1 16Irrigated-commercial High production costs/input prices 35 15 44 17RFUP-semi-commercial Soil erosion/landslides 14 20 2 18RFUP-commercial Soil erosion 13 21 20 19RFLL-semi-commercial Lack of capital/low interest credit, inadequate credit support 31 17 15 20RFLL-semi-commercial Lack of suitable (hybrid) varieties 32 18 17 21RFLL-semi-commercial Drought 27 23 19 22RFUP-semi-commercial Soil infertility and acidity 19 28 18 23Irrigated-commercial Waterlogging 46 19 65 24Irrigated-commercial Lack of labor 53 24 70 25Irrigated-commercial Downy mildew 56 25 117 26RFUP-commercial Soil infertility 15 31 16 27RFLL-semi-commercial Rodents 45 27 25 28Irrigated-commercial Leaf blight/rust (foliar diseases) 59 26 79 29RFLL-semi-commercial Inadequate post-harvest technologies/facilities 65 29 34 30By maize production environmentand market orientationFrequency count(no. of priority constraints)Irrigated-commercial 9RFLL-semi-commercial 5RFLL-commercial 2RFUP-semi-commercial 3RFUP-commercial 11By type of constraintDrought 4Biotic 12Abiotic 5Socioeconomic/policy 9Source: IFAD-CIMMYT RRA/PRA Surveys 2000-2001 and country-level maize R&D prioritization workshops 2001-2002.Note: RFLL – rainfed lowland; RFUP – rainfed upland; STMA – subtropical/mid-altitude.49

6.3 Area-based PriorityConstraints6.3.1 Major findingsBased on the efficiency index. Table 6.10 shows thetop 30 regional constraints, associated by maizeproduction environment and market orientation,according to an area-based efficiency index. In bothpriority lists (with and without India’s STMAcommercialmaize production areas), most of thetop 30 area-based priority constraints were reportedby farmers to be prevalent in the rainfed upland(semi-commercial and commercial) maize productionenvironments. Both lists are dominated by bioticconstraints, as well as socioeconomic and policyrelatedconstraints to maize production.Table 6.9. Production-based priority constraints to maize productivity by production environment and marketorientation, Asia (with and without STMA-commercial maize production areas of India and excluding SouthwestChina’s low market surplus maize production systems).MaizeMarket orientation of maize productionproductionenvironment Low market surplus Semi-commercial CommercialIrrigatedLack of capitalInefficient fertilizer useUndeveloped irrigation system (water shortage)RodentsHigh production costs/input pricesWaterloggingLack of laborDowny mildewLeaf blight and rust (foliar diseases)Waterlogging/crop establishment (37)Stem borers (94)Lack of post-harvest facilities (158)Rainfed lowland Lack of capital/low interest credit, Droughtinadequate credit supportCorn borers (ear/stem borers)Drought Rust (42)Rodents Soil infertility (58)Inadequate post-harvest technologiesand facilitiesLack of suitable (hybrid) varieties (34)Poor marketing system (50)Low technology adoption (72)Poor technology transfer system (81)Rainfed upland Ineffective financial scheme (190) Drought DroughtLow soil fertility (194) Soil erosion/landslides Poor availability of appropriate varieties and quality seedPoor farm-to-market roads (195) Soil infertility and acidity Weeds and weed managementPoor quality of hybrid seeds (91)Downy mildewRodents (99)Stem borersPoor farm-to-market roads (102)Leaf blight (120)Post-flowering stalk rotStorage pestsTurcicum leaf blightBanded leaf and sheath blightSoil erosionSoil infertilitySubtropical/ Declining soil fertility (108) Inadequate availability of quality seeds (218) Post-flowering stalk rotmid-altitude Lack of suitable improved Stem borers (222) Lack of quality seedvarieties (116) Improper maize-based intercropping Imbalanced use of fertilizersTermite/white grubs (138) system (225) Post-harvest lossesSoil acidity (152) Nematodes (231) Improper cropping systemsTransplanting maize under late sown conditionsTurcicum leaf blightMoisture stress (drought)Lack of appropriate maturity varietiesStem borersMaydis leaf blightInappropriate crop establishment methodNote: In the above table, constraints when including India’s STMA-commercial maize production areas are listed in regular font; those in the scenarioexcluding these production areas are listed in italics.50

Table 6.10. Top 30 priority constraints to maize production according to the area-based efficiency index, Asia (with andwithout STMA-commercial maize production areas of India and excluding Southwest China’s low market surplus maizeproduction systems).Area- With STMA-commercial production in India Without STMA-commercial production in Indiabased Maize production Farmer-identified Production- Maize production Farmer-identified Productionefficiencyenvironment and maize production based efficiency environment and maize production based efficiencyranking market orientation constraint ranking market orientation constraint ranking1 RFUP-semi-commercial Drought 2 RFUP-semi-commercial Drought 22 RFUP-semi-commercial Inappropriate fertilizer use 4 RFUP-semi-commercial Inappropriate fertilizer use 43 RFUP-semi-commercial Lack of appropriate variety 7 RFUP-semi-commercial Lack of appropriate variety 74 STMA-commercial Post-flowering stalk rot (PFSR) 21 RFUP-semi-commercial Soil erosion and landslides 145 RFUP-semi-commercial Soil erosion and landslides 9 RFUP-semi-commercial Downy mildew 176 STMA-commercial Lack of quality seed 27 RFUP-commercial Drought 37 RFUP-semi-commercial Downy mildew 18 RFUP-semi-commercial Low output prices 168 RFUP-commercial Drought 3 RFLL-commercial Drought 19 RFUP-semi-commercial Low output prices 17 RFUP-semi-commercial Soil infertility and acidity 1910 RFLL-commercial Drought 1 RFUP-semi-commercial Lack of post-harvest facilities 1811 RFUP-semi-commercial Soil infertility and acidity 16 RFUP-commercial Poor availability of quality seed 512 RFUP-semi-commercial Lack of post-harvest facilities 19 RFLL-semi-commercial Lack of capital/low interest credit 31source (inadequate credit support)13 RFUP-commercial Poor availability of quality seed 5 RFLL-semi-commercial Drought 2714 RFLL-semi-commercial Lack of capital/low interest credit 22 RFUP-commercial Weeds and weed management 8source (inadequate credit support)15 RFLL-semi-commercial Drought 23 RFUP-commercial Downy mildew 1016 RFUP-commercial Weeds and weed management 8 RFLL-commercial Corn borers (ear/stem borers) 617 RFUP-commercial Downy mildew 11 RFUP-semi-commercial Limited capital / access to credit 3418 RFLL-commercial Corn borers (ear/stem borers) 6 RFUP-commercial Soil erosion 1319 RFUP-semi-commercial Limited capital / access to credit 25 RFUP-commercial Soil infertility 1520 RFUP-commercial Soil erosion 14 RFUP-commercial Stem borers 1221 RFUP-commercial Soil infertility 15 RFLL-semi-commercial Lack of suitable (hybrid) varieties 3222 RFUP-commercial Stem borers 13 RFUP-semi-commercial High input prices 40(especially seed, transport)23 STMA-commercial Imbalanced/improper/ 53 RFUP-semi-commercial Weeds 37inadequate use of fertilizers24 STMA-commercial Post-harvest losses (due to 58 RFUP-semi-commercial Farmers’ limited access to information 41weevils during storage)and technology due to poor extension25 RFUP-semi-commercial Pests and diseases 32 RFLL-semi-commercial Rodents 45(ear rot, stalk rot)26 STMA-commercial Inadequate cropping systems 64 RFLL-commercial Rust 9(mixed cropping/ intercropping)27 RFLL-semi-commercial Lack of suitable (hybrid) varieties 39 RFUP-semi-commercial Stalk rot 4328 RFUP-semi-commercial High input prices (especially 36 RFLL-commercial Waterlogging 11seed, transport)29 RFUP-semi-commercial Weeds 37 RFLL-semi-commercial Post-harvest pests and diseases 6130 RFUP-semi-commercial Farmers’ limited access to 38 RFLL-semi-commercial Sloping land and soil erosion 62information and technologydue to poor extensionBy maize productionBy maize productionenvironment and Frequency count environment and Frequency countmarket orientation (no. of constraints) market orientation (no. of constraints)Irrigated-commercial 0 Irrigated-commercial 0RFLL-semi-commercial 3 RFLL-semi-commercial 6RFLL-commercial 2 RFLL-commercial 4RFUP-semi-commercial 13 RFUP-semi-commercial 13RFUP-commercial 7 RFUP-commercial 7STMA-commercial 5By type of constraintBy type of constraintDrought 4 Drought 4Biotic 9 Biotic 10Abiotic 4 Abiotic 6Socioeconomic/policy 13 Socioeconomic/policy 10Source: IFAD-CIMMYT RRA/PRA Surveys 2000-2001 and country-level maize R&D prioritization workshops 2001-2002.Note: RFLL – rainfed lowland; RFUP – rainfed upland; STMA – subtropical/mid-altitude.51

A similar exercise was carried out for maize systemsin China’s Southwest agroecological region. Due tothe similarity of the information and magnitudesconveyed by area and production data across maizesystems, there was no difference in efficiency rankingsbetween the production-based priority ranking resultspresented in Table 6.4b and the results of area-basedprioritization. The top 30 constraints in China’sSouthwest all pertained to drought and to bioticconstraints and socioeconomic and policy-relatedproblems in the rainfed upland low market surplusspring maize production system.For the rest of Asia, area-based prioritizationindicated that addressing the problem of drought inthe rainfed upland-semi-commercial environmentsfirst will provide the highest returns to maize R&Din Asia. Alleviating the drought problem will impacta maize area of around 3 M ha (24% of the regionaltotal), with an average maize yield of 2.8 t/ha, thatsupports an estimated 6.25 million rural poor inAsia. These environments cover the drylands of Javaand Bali (Indonesia), the upland, rolling-to-hillyareas of Southern Mindanao (Philippines), and theunfavorable uplands in the lower northeast regionof Thailand. Other maize production constraintsincluded in the priority list were biotic constraintssuch as post-flowering stalk rot, downy mildew,weeds and stem borers, and abiotic constraints suchas soil erosion/landslides and soil infertility andacidity. Socioeconomic and policy-related constraints,however, dominated the priority lists and included,among others, lack of quality seed, low output prices,lack of post-harvest facilities, lack of capital or of lowinterestcredit, and inadequate credit support from thegovernment.A closer look reveals that eight to ten constraintsnot included in the priority list according to theproduction-based efficiency index (discussed insection 6.2.1) are included in Table 6.10. The newentrants are mostly socioeconomic and policy-relatedconstraints from the rainfed upland-semi-commercialand STMA maize environments. They range fromlack of suitable (hybrid) varieties to inappropriatecropping systems and farmers’ limited access toinformation and technology, all of which reflectinadequate or poor agricultural extension services tomaize farmers in these areas.Of the top 30 priority constraints in Asia (includingIndia’s STMA-commercial maize areas and excludingSouthwest China’s rainfed upland low marketsurplus areas) according to the area-based efficiencyindex alone, socioeconomic and policy-relatedconstraints are estimated to impact 9.3 M ha (73.8%of the regional total) and about 78.3% of Asia’s ruralpoor. The set of biotic constraints in the priority listcomes next, affecting 8.9 M ha and 161.7 million ruralpoor Table 6.11a). However, without India’s STMAcommercialmaize production environments andSouthwest China’s rainfed upland low market surplusareas, the set of biotic constraints appear to have thehighest estimated impacts, affecting about 7 M ha(76.7% of the total maize area) and 56.9 million ruralpoor (43.4% of the regional total). It is estimated thatalleviating that set of biotic constraints would improvemaize yields by an average of nearly 20% (Table 6.11b).Based on the poverty index. When included,constraints reported from Asia’s STMA-commercialmaize production environments composed the top 10priority constraints based on the poverty index alone(Table 6.12), led by post-flowering stalk rot, lack ofquality seed, and imbalanced or inadequate fertilizeruse. These environments include all maize productionzones in India’s Uttar Pradesh, Madhya Pradesh,Rajasthan, and Bihar, and account for about 16% ofAsia’s maize production, grown on 27% of the maizearea. However, nearly 56% of the region’s rural poorlive in these environments, where average maize yieldsare barely 1.03 t/ha (see Table 6.3).In contrast to priorities determined using the areabasedefficiency index, which were dominated bysocioeconomic and policy-related constraints, thosebased on the poverty index consisted mostly of bioticconstraints (15 in all) such as post-flowering stalkrot, Turcicum leaf blight, stem borers, Maydis leafblight, and downy mildew. These biotic constraintswere estimated to affect about 6 M ha of maize area(47.8% of the regional total) and 171.2 million ruralpoor (nearly 81% of the regional total) (Table 6.11a).Noticeably, 17 of the priority constraints in Table 6.12are new entrants that received lower priority accordingto the area-based efficiency index. Most (14) of thesenew entrants are constraints reported from the STMAcommercialmaize production environments of Asia.Meanwhile, 21 of the constraints on the priority listaccording to the production-based poverty indexare included in Table 6.12, and again most (seven)of the nine new entrants are prevalent in the STMAcommercialmaize production environments. The newentrants are mostly biotic constraints (banded leaf andsheath blight, brown stripe downy mildew, termites,and nematodes). These results consistently show theinfluence on the poverty-based priority constraints listsof the STMA-commercial production environments’share in the regional number of rural poor.Without India’s STMA-commercial maize productionenvironments, constraints from the rainfed uplandcommercialmaize production environmentsdominated the top 30 priority list, led by drought,poor availability of quality seed, and weeds. Theseconstraints were reportedly prevalent in the low tohigh rainfall zones of Karnataka and Andhra Pradesh52

Table 6.11a. Selected indicators of impact of top 30 area-based priority maize productivity constraints in Asia (withSTMA-commercial maize production areas of India and excluding Southwest China’s low market surplus maizeproduction systems).Farmer-identified maizeproductivity constraintsFarmer-identified maizeproductivity constraintsOther Biotic Socioeco- Other Biotic Socioecoabiotic(pests and nomic/policy- abiotic (pests and nomic/policy-Impact indicators Drought problems diseases) related Drought problems diseases) relatedBased on efficiency indexBased on poverty indexEffective total area affected (000 ha) 6,767.4 4,253.9 8,876.0 9,339.9 2,730.4 6,255.3 6,056.8 6,484.4Estimated maize production across areasaffected (000 t) 16,344.7 9,513.4 17,817.8 16,515.7 6,350.3 9,260.3 12,533.1 12,283.8Average yield gain when constraint isalleviated (%) 35.0 33.7 17.1 31.0 38.8 24.4 8.8 18.8Estimated no. of rural poor affected (millions) 47.7 12.6 161.7 166.3 37.3 124.5 171.2 178.1Based on marginality indexBased on combined indexEffective total area affected (000 ha) 5,432.7 3,904.3 7,780.5 7,987.5 6,767.4 6,482.2 6,188.0 6,943.9Estimated maize production across areasaffected (000 t) 11,387.9 6,960.8 11,415.9 12,230.2 16,344.7 10,003.6 10,495.2 12,692.2Average yield gain when constraint isalleviated (%) 37.6 29.5 16.8 33.5 35.0 29.9 10.2 21.1Estimated no. of rural poor affected (millions) 41.2 14.5 142.2 139.8 47.7 124.8 158.1 180.0Source: IFAD-CIMMYT RRA/PRA Surveys 2000-2001 and country-level maize R&D prioritization workshops 2001-2002.Note: The geographic regions and environments where specific constraints (except drought) under each of the above groups are reported are not necessarilymutually exclusive. However, careful calculations were made to avoid double counting in the above parameters.Table 6.11b. Selected indicators of impact of top 30 area-based priority maize productivity constraints in Asia(without STMA-commercial maize production areas of India and excluding Southwest China’s low market surplusmaize production systems).Farmer-identified maizeproductivity constraintsFarmer-identified maizeproductivity constraintsOther Biotic Socioeco- Other Biotic Socioecoabiotic(pests and nomic/policy- abiotic (pests and nomic/policy-Impact indicators Drought problems diseases) related Drought problems diseases) relatedBased on efficiency indexBased on poverty indexEffective total area affected (000 ha) 6,767.4 5,937.9 7,050.7 5,995.2 6,767.4 4,679.6 6,902.0 4,649.1Estimated maize production across areasaffected (000 t) 16,344.7 13,975.5 16,491.0 12,449.5 16,344.7 10,838.6 16,386.7 10,405.8Average yield gain when constraint isalleviated (%) 35.0 28.8 19.7 35.7 35.0 27.4 15.8 28.3Estimated no. of rural poor affected (millions) 47.7 23.3 56.9 52.8 47.7 29.0 74.9 73.8Based on marginality indexBased on combined indexEffective total area affected (000 ha) 6,767.4 4,773.7 7,617.0 5,995.2 6,767.4 4,142.4 6,918.8 6,550.1Estimated maize production across areasaffected (000 t) 16,344.7 9,550.5 15,202.9 12,449.5 16,344.7 9,166.9 16,437.0 14,074.6Average yield gain when constraint isalleviated (%) 35.0 33.5 22.3 38.9 35.0 29.7 15.5 30.4Estimated no. of rural poor affected (millions) 47.7 18.5 50.6 52.8 47.7 12.5 74.8 74.0Source: IFAD-CIMMYT RRA/PRA Surveys 2000-2001 and country-level maize R&D prioritization workshops 2001-2002.Note: The geographic regions and environments where specific constraints (except drought) under each of the above groups are reported are not necessarilymutually exclusive. However, careful calculations were made to avoid double counting in the above parameters.53

Table 6.12. Top 30 priority constraints to maize production according to the area-based poverty index, Asia (with andwithout STMA-commercial maize production areas of India and excluding China’s Southwest low market surplus maizeproduction systems).With STMA-commercial production, India54Without STMA-commercial production, IndiaArea- Prod’n- Area- Prod’n- AreabasedMaize production Farmer-identified based based Maize production Farmer-identified based basedpoverty environment and maize production poverty efficiency environment and maize production poverty efficiencyranking market orientation constraint rank rank market orientation constraint rank rank1 STMA-commercial Post-flowering stalk rot 1 4 RFUP-commercial Drought 1 62 STMA-commercial Lack of quality seed 2 6 RFUP-commercial Poor availability of quality seed 2 113 STMA-commercial Imbalanced /improper/ 3 23 RFUP-commercial Weeds and weed 3 14inadequate use of fertilizersmanagement4 STMA-commercial Post-harvest losses (due to 4 24 RFUP-commercial Downy mildew 4 15weevils during storage)5 STMA-commercial Improper cropping systems 5 26 RFUP-commercial Stem borers 5 20(mixed cropping/intercropping)6 STMA-commercial Transplanting maize under 10 33 RFLL-semi-commercial Lack of capital/low interest 17 12late sown conditionscredit, inadequate credit support7 STMA-commercial Turcicum leaf blight 11 34 Irrigated-commercial Lack of capital 7 368 STMA-commercial Moisture stress (drought) 13 72 RFLL-semi-commercial Lack of suitable (hybrid) varieties 18 219 STMA-commercial Lack of appropriate maturity 14 75 RFLL-semi-commercial Drought 23 13varieties10 STMA-commercial Stem borers 15 84 Irrigated-commercial Inefficient fertilizer use 8 3811 RFUP-commercial Drought 6 8 RFLL-semi-commercial Rodents 27 2512 STMA-commercial Maydis leaf blight 16 90 RFUP-commercial Post-flowering stalk rot 6 4913 RFUP-commercial Poor availability of quality seed 7 13 RFUP-semi-commercial Soil erosion/landslides 20 414 RFUP-commercial Weeds and weed management 8 16 Irrigated-commercial Undeveloped irrigation system 13 39(water shortage)15 RFUP-commercial Downy mildew 9 17 RFUP-commercial Storage pests 9 5416 RFUP-commercial Stemborers 12 22 RFUP-semi-commercial Drought 22 117 STMA-commercial Inappropriate crop 25 105 RFUP-commercial Turcicum leaf blight 10 55establishment method18 STMA-commercial Lack of location-specific transfer 42 117 RFUP-commercial Banded leaf and sheath blight 12 57of technology for rainfedconditions, esp for farm women19 STMA-commercial Banded leaf and sheath blight 43 119 RFLL-commercial Drought 11 820 STMA-commercial Broadleaf and grassy weeds 44 120 Irrigated-commercial High production costs/input prices 15 4621 STMA-commercial Brown stripe downy mildew 46 124 RFUP-semi-commercial Soil infertility and acidity 28 922 Irrigated-commercial Lack of capital 18 43 RFLL-semi-commercial Inadequate postharvest 29 51technologies/facilities23 STMA-commercial Rodents 48 125 RFLL-semi-commercial Poor marketing system, input/ 30 31output market access andundeveloped transport system24 STMA-commercial Termites 49 126 Irrigated-commercial Rodents 14 6825 RFLL-semi-commercial Lack of suitable (hybrid) varieties 30 27 RFUP-semi-commercial Weeds 32 2326 RFLL-semi-commercial Drought 29 15 RFLL-commercial Corn borers (ear/stem borers) 16 1627 Irrigated-commercial Inefficient fertilizer use 19 45 Irrigated-commercial Waterlogging 19 5928 STMA-commercial Nematodes 50 130 RFUP-semi-commercial Limited capital/access to credit 38 1729 RFLL-semi-commercial Rodents 32 31 RFUP-commercial Soil erosion 21 1830 RFUP-semi-commercial Soil erosion/landslides 31 5 Irrigated-commercial Lack of labor 24 70By maize productionBy maize productionenvironment and Frequency count environment and Frequency countmarket orientation (no. of priority constraints) market orientation (no. of priority constraints)Irrigated-commercial 2 Irrigated-commercial 7RFLL-semi-commercial 3 RFLL-semi-commercial 6RFLL-commercial 0 RFLL-commercial 2RFUP-semi-commercial 1 RFUP-semi-commercial 5RFUP-commercial 5 RFUP-commercial 10STMA-commercial 19By type of constraintBy type of constraintDrought 2 Drought 4Biotic 15 Biotic 11Abiotic 2 Abiotic 4Socioeconomic/policy 11 Socioeconomic/policy 11Source: IFAD-CIMMYT RRA/PRA Surveys 2000-2001 and country-level maize R&D prioritization workshops 2001-2002.Note: RFLL – rainfed lowland; RFUP – rainfed upland; STMA – subtropical/mid-altitude.

states of India and in the drylands of Indonesia’s outerislands. Biotic as well as socioeconomic and policyrelatedconstraints dominated this priority scenario.These groups of constraints impact an estimated 74.9and 73.8 million rural poor, respectively, and theiralleviation would improve maize yields by about 16%and 28%, respectively (Table 6.11b).Based on the marginality index. Including India’sSTMA-commercial areas, the 30 top priority constraintsbased on the marginality index alone come almostequally from rainfed lowland- and rainfed uplandsemi-commercialproduction environments, andare mostly (16) socioeconomic and policy-relatedconstraints (Table 6.13). All together, the marginalitybasedpriority socioeconomic and policy-relatedconstraints impact an estimated total maize area ofnearly 8 M ha, whose average maize yield is onlyabout 1.9 t/ha, and an estimated 140 million rural poorpeople, or nearly 66% of Asia’s rural poor population(Table 6.11a).Similar to results using the poverty index, postfloweringstalk rot and lack of quality seed in theSTMA-commercial maize areas were the top twopriority constraints that maize R&D in Asia oughtto address based on the marginality index alone.Also similar to area-based priorities accordingto the efficiency index, the top 30 marginalitybasedproductivity constraints included mostlysocioeconomic and policy-related constraints suchas lack of capital or of low-interest credit and lowtechnology adoption in rainfed lowland-semicommercialareas, and low output prices in rainfedupland-semi-commercial areas. Other top prioritiesincluded abiotic (drought, soil erosion, and landslides)and biotic constraints (downy mildew, weevils duringstorage, weeds, and rodents).Interestingly, the above general trend in constraintprioritization remains even when the STMAcommercialmaize production environments of Indiaare excluded from the analysis. That is, constraintsTable 6.13. Top 30 priority constraints to maize production according to the area-based marginality index, Asia (withand without STMA-commercial maize production areas of India and excluding Southwest China’s low market surplusmaize production systems).With STMA-commercial production, IndiaWithout STMA-commercial production, IndiaArea- Prod’n- Prod’nbasedbased Area- based Areamargina-Maize production Farmer-identified margina- based Maize production Farmer-identified margina- basedlity environment and maize production lity efficiency environment and maize production lity efficiencyranking market orientation constraint rank rank market orientation constraint rank rank1 STMA-commercial Post-flowering stalk rot 3 4 RFUP-semi-commercial Drought 1 12 STMA-commercial Lack of quality seed 8 6 RFUP-semi-commercial Soil erosion/landslides 2 43 RFUP-semi-commercial Drought 1 1 RFLL-semi-commercial Lack of capital/low interest creditsource (inadequate credit support) 15 124 RFUP-semi-commercial Soil erosion/landslides 2 5 RFUP-semi-commercial Downy mildew 9 55 RFLL-semi-commercial Lack of capital/low interest 9 14 RFUP-semi-commercial Low output prices 5 7credit, inadequate credit support6 STMA-commercial Inadequate fertilizer use 25 23 RFUP-semi-commercial Inappropriate fertilizer use 6 27 RFUP-semi-commercial Downy mildew 10 7 RFUP-semi-commercial Lack of post-harvest facilities 7 108 RFUP-semi-commercial Low output prices 6 9 RFUP-semi-commercial Weeds 10 239 STMA-commercial Post-harvest losses 28 24 RFLL-semi-commercial Drought 19 13(weevils during storage)10 STMA-commercial Inappropriate cropping systems 36 26 RFLL-semi-commercial Lack of suitable (hybrid) varieties 17 21(mixed cropping/inter-cropping)11 RFUP-semi-commercial Inappropriate fertilizer use 7 2 RFLL-semi-commercial Post-harvest pests and diseases 28 2912 RFUP-semi-commercial Lack of post-harvest facilities 11 12 RFUP-semi-commercial Limited capital / access to credit 22 1713 RFUP-semi-commercial Weeds 13 29 RFLL-semi-commercial Rodents 25 2514 RFLL-semi-commercial Drought 15 15 RFUP-semi-commercial Soil infertility and acidity 18 915 RFLL-semi-commercial Lack of suitable (hybrid) varieties 22 27 RFLL-semi-commercial Low technology adoption 47 3216 RFLL-semi-commercial Post-harvest pests and diseases 24 36 RFUP-semi-commercial High input prices (including seed, 26 22transport)17 RFUP-semi-commercial Limited capital/access to credit 20 19 RFLL-semi-commercial Low fertilizer use 51 3318 RFLL-semi-commercial Rodents 23 31 RFUP-semi-commercial Farmers’ limited access to 27 24information and technology dueto poor extension19 RFUP-semi-commercial Soil infertility and acidity 14 11 RFLL-semi-commercial Poor marketing system, input/ 24 31output market access andundeveloped transport systemSource: IFAD-CIMMYT RRA/PRA Surveys 2000-2001 and country-level maize R&D prioritization workshops 2001-2002.Note: RFLL – rainfed lowland; RFUP – rainfed upland; STMA – subtropical/mid-altitude.55

Table 6.13. Top 30 priority constraints to cont’d....With STMA-commercial production, IndiaWithout STMA-commercial production, IndiaArea- Prod’n- Prod’nbasedbased Area- based Areamargina-Maize production Farmer-identified margina- based Maize production Farmer-identified margina- basedlity environment and maize production lity efficiency environment and maize production lity efficiencyranking market orientation constraint rank rank market orientation constraint rank rank20 RFLL-semi-commercial Low technology adoption 27 RFUP-commercial Drought 3 621 STMA-commercial Transplanting maize under late 34 33 RFUP-semi-commercial Stalk rot 29 27sown conditions22 STMA-commercial Turcicum leaf blight 35 34 RFLL-semi-commercial Sloping land and soil erosion 32 3023 RFUP-semi-commercial High input prices (including 26 28 RFLL-semi-commercial High input prices (including 63 45seed, transport)transport, credit)24 RFLL-semi-commercial Low fertilizer use 29 40 RFUP-commercial Poor availability of quality seed 8 1125 RFUP-semi-commercial Limited access to information 30 30 RFLL-commercial Drought 4 8and technology at farmers’ leveldue to poor extension26 RFLL-semi-commercial Poor marketing system, input/ 39 38 RFLL-semi-commercial Weeds (sedges and broad-leaf) 67 48output market access andundeveloped transport system27 RFUP-commercial Drought 4 8 RFUP-commercial Soil infertility 16 1928 RFLL-semi-commercial Sloping land and soil erosion 32 37 RFUP-commercial Downy mildew 12 1529 RFUP-semi-commercial Pests and diseases (ear rot, 38 25 RFUP-commercial Stem borers 13 20stalk rot)30 RFUP-semi-commercial High input prices (including 40 51 RFUP-commercial Weeds and weed management 14 14seed, transport)By maize productionBy maize productionenvironment and Frequency count environment and Frequency countmarket orientation (no. of constraints) market orientation (no. of constraints)Irrigated-commercial 0 Irrigated-commercial 0RFLL-semi-commercial 10 RFLL-semi-commercial 11RFLL-commercial 0 RFLL-commercial 1RFUP-semi-commercial 12 RFUP-semi-commercial 12RFUP-commercial 1 RFUP-commercial 6STMA-commercial 7By type of constraintBy type of constraintDrought 3 Drought 4Biotic 8 Biotic 9Abiotic 3 Abiotic 4Socioeconomic/policy 16 Socioeconomic/policy 13Source: IFAD-CIMMYT RRA/PRA Surveys 2000-2001 and country-level maize R&D prioritization workshops 2001-2002.Note: RFLL – rainfed lowland; RFUP – rainfed upland; STMA – subtropical/mid-altitude.reported from the semi-commercial maize productionareas of Asia’s rainfed lowlands and rainfed uplands,and those related to socioeconomics and policydominate the top 30 priorities in this scenario (Table6.13). Included in these production areas are thedrylands in Java and Bali, Indonesia, the uplandto rolling to hilly areas in Northern and SouthernMindanao, the Philippines, and the rolling to hillyareas of Thailand’s lower north region. Droughtand soil erosion/landslides in the rainfed uplandsemi-commercialproduction areas are the top twoconstraints to be addressed when marginality of theproduction environment is emphasized. Droughtalone impacts about 3.9 million rural poor spreadacross 2.7 M ha that, on average, yield about 2.6 t/ha.Drought alleviation could add an estimated 35% tomaize yields in these areas.Except for five to seven new entrants, the set ofmarginality-based priority constraints using shareof regional maize area is the same as that based oncontribution to regional maize production (Table 6.13).This indicates results are consistent regardless of theparameter used in prioritizing maize R&D based onmarginality of the maize production environment.Based on the combined index. Table 6.14 shows the top30 priority constraints that, based on the combinedindex using share of regional maize area, ought to beaddressed by maize R&D in Asia. The table also showshow the constraints ranked when contribution toregional maize production was used in prioritization.Nine of the top ten priority constraints were reported56

Table 6.14. Top 30 priority constraints to maize production according to the area-based combined index, Asia (withand without STMA-commercial maize production areas of India and excluding Southwest China’s low market surplusmaize production systems).Area-based With STMA commercial production, India Without STMA commercial production, Indiacombined Maize production Production- Maize production Productionindexenvironment and Farmer-identified maize based environment and Farmer-identified maize basedranking market orientation production constraint ranking market orientation production constraint ranking1 STMA-commercial Post-flowering stalk rot 1 RFUP-commercial Drought 12 STMA-commercial Lack of quality seed 2 RFUP-commercial Poor availability of appropriate 2varieties and quality seed3 STMA-commercial Imbalanced/improper/inadequate 3 RFUP-commercial Weeds and weed management 3use of fertilizers4 STMA-commercial Post-harvest losses (due to weevils 5 RFUP-commercial Downy mildew 4during storage)5 STMA-commercial Improper cropping systems (mixed 6 RFUP-commercial Stem borers 5cropping/ intercropping)6 STMA-commercial Transplanting maize under late 11 RFLL-semi-commercial Lack of capital/low-interest credit, 20sown conditionsinadequate credit support7 STMA-commercial Turcicum leaf blight 12 RFUP-semi-commercial Drought 168 STMA-commercial Moisture stress (drought) 13 RFUP-semi-commercial Soil erosion/landslides 189 STMA-commercial Lack of appropriate maturity varieties 15 RFLL-semi-commercial Drought 2210 RFUP-commercial Drought 4 RFLL-semi-commercial Lack of suitable (hybrid) varieties 2111 STMA-commercial Stem borers 16 Irrigated-commercial Lack of capital 812 STMA-commercial Maydis leaf blight 17 RFLL-semi-commercial Rodents 2813 RFUP-commercial Poor availability of appropriate 7 Irrigated-commercial Inefficient fertilizer use/inappropriate 9varieties and quality seedfertilizer application14 RFUP-commercial Weeds and weed management 8 RFLL-commercial Drought 615 RFUP-commercial Downy mildew 9 RFUP-semi-commercial Soil infertility and acidity 2316 RFUP-commercial Stem borers 10 RFUP-commercial Post-flowering stalk rot 717 RFUP-semi-commercial Drought 22 Irrigated-commercial Undeveloped irrigation system 12(water shortage)18 STMA-commercial Inappropriate crop establishment method 30 RFUP-commercial Storage pests 1019 RFUP-semi-commercial Soil erosion/landslides 27 RFUP-commercial Turcicum leaf blight 1120 RFLL-semi-commercial Drought 31 RFUP-commercial Banded leaf and sheath blight 1321 RFLL-semi-commercial Lack of suitable (hybrid) varieties 34 Irrigated-commercial High production costs (input prices) 1722 RFLL-semi-commercial Lack of capital/low-interest credit, 35 RFLL-semi-commercial Inadequate post-harvestinadequate credit support technologies/ facilities 3023 RFLL-semi-commercial Rodents 36 RFUP-semi-commercial Weeds 3424 RFLL-commercial Drought 14 RFLL-semi-commercial Poor marketing system, input/output 35market access and undevelopedtransport system25 Irrigated-commercial Lack of capital 19 RFUP-semi-commercial Downy mildew 3226 RFUP-semi-commercial Soil infertility and acidity 33 RFLL-commercial Corn borers (ear/stem borers) 1427 STMA-commercial Lack of location-specific technology 60 RFUP-semi-commercial Limited capital/access to credit 39transfer for rainfed conditions,especially for farm women28 Irrigated-commercial Inefficient fertilizer use/inappropriate 20 RFUP-commercial Soil erosion 19fertilizer application29 STMA-commercial Banded leaf and sheath blight 64 Irrigated-commercial Rodents 1530 STMA-commercial Broadleaf and grassy weeds 65 RFUP-semi-commercial Low output prices 36By maize production environment Frequency count By maize production environment Frequency countand market orientation (no. of constraints) and market orientation (no. of constraints)Irrigated-commercial 2 Irrigated-commercial 5RFLL-semi-commercial 4 RFLL-semi-commercial 6RFLL-commercial 1 RFLL-commercial 2RFUP-semi-commercial 3 RFUP-semi-commercial 7RFUP-commercial 5 RFUP-commercial 10STMA-commercial 15By type of constraintBy type of constraintDrought 4 Drought 4Biotic 11 Biotic 12Abiotic 3 Abiotic 3Socioeconomic/policy 12 Socioeconomic/policy 11Source: IFAD-CIMMYT RRA/PRA Surveys 2000-2001 and country-level maize R&D prioritization workshops 2001-2002.Note: RFLL – rainfed lowland; RFUP – rainfed upland; STMA – subtropical/mid-altitude.57

from the STMA-commercial maize productionenvironments in the Indian states of Uttar Pradesh,Madhya Pradesh, Rajasthan, and Bihar, and are ledby post-flowering stalk rot, lack of quality seed, andimbalanced/inadequate fertilizer use. Of the top30 priority constraints, 15 are specific to the STMAcommercialenvironments, 8 to the rainfed uplands, 5to the rainfed lowlands, and 2 to irrigated-commercialenvironments. Most of these priority constraintsare socioeconomic and policy-related problems andbiotic constraints. Together, these socioeconomic andpolicy-related constraints are reported to impact anestimated 6.9 M ha and 180 million rural poor, whichrepresent 55% and 85% of regional totals, respectively(Table 6.11a). Alleviating these constraints couldcontribute 21% to maize yields in the region. Bioticconstraints, meanwhile, impact about 54% of theregional maize area and 22.5% of Asia’s rural poor, andit is estimated their alleviation would improve maizeyields by 10%. These results confirm that addressingsocioeconomic and policy-related constraints wouldsignificantly impact maize productivity much morethan technology.When constraints from STMA-commercialenvironments are excluded from the analysis,constraints reported from the rainfed upland-semicommercialand commercial maize productionenvironments dominated the list of top 30 priorityconstraints. Drought is the first priority that shouldbe addressed, followed by poor availability ofappropriate varieties and quality seed, then by threebiotic constraints—weeds and weed management,downy mildew, and stem borers (Table 6.14). All topfive priority constraints were reportedly prevalentin rainfed upland-commercial maize productionareas. In this scenario, biotic and socioeconomic andpolicy-related constraints were estimated to impact6.9 and 6.5 M ha of maize area, respectively, and about74-75 million rural poor people. Similarly, it appearsthat addressing socioeconomic and policy-relatedconstraints could have a greater impact (30%) thanaddressing biotic constraints only (16%) (Table 6.11b).Table 6.14 also shows that most of the area-based top30 priority constraints also appear in the productionbasedtop 30 list. Including India’s STMA-commercialareas, eight constraints that were not on theproduction-based top 30 list made it to the area-basedlist. Most of these new entrants are socioeconomicand policy-related constraints reported from therainfed lowland-semi-commercial maize productionenvironments, including the rainfed lowlands of Javaand Bali in Indonesia, the terai region of Nepal, andthe upland plains of Southern Tagalog, Bicol, andMindanao in the Philippines. Meanwhile, withoutIndia’s STMA-commercial areas in the analysis, fiveconstraints that ranked lower based on contribution toregional maize production made it to the area-basedtop 30 priority constraints list. Most of these newentrants were, again, socioeconomic and policy-relatedconstraints reported this time from Asia’s rainfedupland-semi-commercial maize production areas.Classified in these environments are the drylandsof Java and Bali in Indonesia, the upland, rolling tohilly areas of Bicol and Northern Mindanao in thePhilippines, and the northern region of Vietnam.6.3.2 Priorities by maize productionenvironment and market orientationTable 6.15 shows the area-based priority constraintsacross Asia by maize production environment andmarket orientation, both with and without India’sSTMA-commercial maize areas, excluding SouthwestChina’s low market surplus production areas, andwith the addition of a few selected lower-priorityconstraints reported in the other maize productionenvironments. Constraints that entered the prioritylist when India’s STMA-commercial maize productionareas were included are in regular font; the priorityconstraints when these production areas have beenexcluded are in italics. As in the production-basedset shown in Table 6.9, some of the specific lowerpriorityconstraints added may already appear on theregional priority constraints list, but associated withdifferent maize production environment and marketorientation. In this case, stem borers in the rainfedupland-commercial maize production environmentswas among the top 30 priority constraints with andwithout India’s STMA-commercial maize productionareas, but the same constraint reported in the irrigatedcommercialareas ranked only 68 in the latter scenario.Discussions in Sections 6.2 and 6.3 show that:(a) research prioritization (in this case, of maizeproductivity constraints) can be done systematicallyon a scientific basis; (b) priorities for Asian maize R&Dcan vary depending on the production parameteremphasized; and (c) the top priority constraints to beaddressed by Asian maize R&D are similar, whetherprioritization was done based on contribution toregional maize production or on share of regionalmaize area. Results presented here, however, are notintended to be recommendations written in stone;the overall picture has been presented to help guideAsian maize researchers and policymakers. It is alsoacknowledged that regional and national maize R&Dcoffers are not unlimited, so all priority constraintscannot be addressed at the same time, and thatlocal politics could revise the priority constraints“recommended” to receive early attention andalleviation.58

Table 6.15. Area-based priority constraints to maize productivity by production environment and marketorientation, Asia (with and without the STMA-commercial maize production areas of India and excluding SouthwestChina’s low market surplus production areas).Maize productionMarket orientation of maize productionenvironment Low market surplus Semi-commercial CommercialIrrigatedLack of capitalInefficient fertilizer useUndeveloped irrigation system (water shortage)High production costs/input pricesRodentsDowny mildew (42)Stem borers (68)Lack of short-cycle (winter-crop) varieties (114)Rainfed lowland Drought DroughtLack of suitable (hybrid) varietiesCorn borers (ear/stem borers)Lack of capital/low interest credit, inadequate credit support Waterlogging (51)Rodents Flooding (61)Inadequate post-harvest technologies and facilities Cutworms/armyworms (85)Poor marketing system, input/output market access andundeveloped transport systemLow technology adoption 44)Rainfed upland Limited knowledge Drought Droughtof proper crop Soil erosion/landslides Poor availability of appropriate varieties and quality seedmanagement (132) Soil infertility and acidity Weeds and weed managementHigh production and Weeds Downy mildewpost-harvest pest Downy mildew Stem borersincidence (149) Limited capital/access to credit Post-flowering stalk rotLow output pricesStorage pestsPoor farm-to-market roads (80)Turcicum leaf blightLack of labor (125)Waterlogging (160)Banded leaf and sheath blightSoil erosionZinc/micronutrient deficiency (66)Low quality land preparation (166)Subtropical/ Lack of suitable Post-flowering stalk rotmid-altitude improved varieties (86) Lack of quality seedSoil erosion (95)Imbalanced use of fertilizersSoil acidity (117)Post-harvest losses (weevils in storage)Inadequate cropping systemsTransplanting maize under late sown conditionsTurcicum leaf blightMoisture stress (drought)Lack of appropriate maturity varietiesStem borerMaydis leaf blightInappropriate crop establishment methodLack of location-specific technology transfer,especially for farm womenBanded leaf and sheath blightBroadleaf and grassy weedsNote: In the above table, constraints when including India’s STMA-commercial maize production areas are listed in regular font; those in the scenarioexcluding these production areas are listed in italics.59

6.4 Economics of PriorityConstraints6.4.1 Production-based priority constraintsTable 6.16 shows the estimated economic value(in PPP dollars) of addressing the top 30 regionalpriority constraints and constraint groups identifiedusing the combined indices and the top 30 priorityconstraints for Southwest China using the efficiencybasedmeasure. Both with and without India’sSTMA-commercial areas, drought appeared to havethe highest economic impact on maize production(in terms of estimated gross income loss), i.e., at leastPPP$ 503/ha using farm-gate maize grain prices.When India’s STMA-commercial areas were included,socioeconomic and policy-related constraints causedthe next highest gross income loss; other abioticconstraints came in second when India’s STMAcommercialareas were excluded. Biotic constraints(pests and diseases) appear to have the leasteconomic impact among the farmer-identified maizeproductivity constraints in both scenarios.Including India’s STMA-commercial areas, maizefarmers in rainfed upland-semi-commercial and-commercial areas were estimated to lose aboutPPP$ 450/ha and PPP$ 588/ha, respectively, whendrought conditions persist. Among the maize areasaffected by drought, the highest economic impact ofthe constraint is felt in the rainfed upland-commercialareas of Indonesia’s outer islands, India’s Karnatakaand Andhra Pradesh states, and in the upper andlower north regions of Thailand. Looking at morespecific production constraints, irrigated-commercialmaize farmers, particularly in Vietnam’s southeasternregion and Mekong River Delta, tend to lose the mostincome (about PPP$ 708/ha) due to limited capitaland/or poor access to low-interest credit. Thesefarmers perceive that they lose as much as 35% oftheir gross income from maize production due to thisconstraint, which keeps them from applying neededfarm inputs such as fertilizers. Turcicum leaf blight inlow-to-high rainfall zones of Karnataka and AndhraPradesh (India) has the least economic impact onmaize production, PPP$ 135/ha based on farm-gategrain prices.In China’s Southwest region, the economic impact onmaize production of drought alone may range betweenPPP$ 499/ha and PPP$ 510/ha. Aggregate effects ofbiotic constraints together with socioeconomic andpolicy-related constraints have a significant economicimpact in terms of estimated gross income losses, PPP$704-720/ha and PPP$ 864-883/ha, respectively.At the environment and market orientation level, thetop 30 priority constraints appear to have the highestcombined economic impact on maize production inAsia’s irrigated-commercial areas including India’sSTMA-commercial maize environments, and in therainfed upland-semi-commercial areas when thoseenvironments are excluded. Using farm-gate maizeprices, commercial maize farmers in irrigated areascan lose about PPP$ 447/ha (Table 6.17) due to all30 top priority constraints in their area—downymildew, foliar diseases, rodents, waterlogging,,and inefficient/inappropriate fertilizer application,among them. Rainfed upland-semi-commercial maizefarmers, meanwhile, can lose about PPP$ 411/ha,Table 6.16. Estimated gross income loss in maize production across environments due to the top 30 priorityconstraints, production-based and area-based combined indices, Asia (PPP $/ha).Gross income loss estimates (PPP $/ha)Production-basedArea-basedPriority constraint/ Using farm- Using nearest- Using farm- Using nearestconstraintgroups gate prices market prices gate prices market pricesWith STMA-commercial areas of IndiaDrought 511.78 760.36 440.22 607.81Other abiotic constraints 199.05 264.04 339.03 449.74Biotic constraints (pests and diseases) 156.41 241.49 176.61 258.84Socioeconomic and policy-related constraints 284.30 364.60 221.39 275.59Without STMA-commercial areas of IndiaDrought 503.36 694.99 503.36 694.99Other abiotic constraints 487.27 682.57 415.91 588.72Biotic constraints (pests and diseases) 174.71 250.89 186.88 273.83Socioeconomic and policy-related constraints 372.02 455.98 431.21 531.19Southwest China’s maize production systemsDrought 498.96 510.05 a aOther abiotic constraints 356.40 364.32 a aBiotic constraints (pests and diseases) 703.89 719.53 a aSocioeconomic and policy-related constraints 864.27 883.48 a aaIdentical to corresponding production-based results.60

primarily due to abiotic constraints led by drought,soil erosion, and soil infertility and acidity. The latterareas include Java and Bali in Indonesia, Northernand Southern Mindanao Provinces in the Philippines,and the central, upper, and lower northeast regions inThailand.6.4.2 Area-based priority constraintsSimilar to production-based priority constraints,drought has the highest economic impact on maizeproduction, causing gross income losses of at leastPPP$ 440/ha using farm-gate maize grain prices (Table6.16). The next highest loss in gross income is causedby all other abiotic constraints if the STMA-commercialareas of India are included, and by socioeconomicand policy-related constraints if those environmentsare excluded. Abiotic constraints include mainly soilerosion, infertility, and acidity, while socioeconomicand policy-related constraints include lack of capitalor low-interest credit, lack of quality seed, and poortechnology transfer, among others. Biotic constraints(e.g., downy mildew, stem borers, and post-floweringstalk rot) again appear to cause the lowest economiclosses of gross income obtained from maize production.These biotic constraints were reported from thedrylands of Indonesia’s outer islands and from UttarPradesh, Madhya Pradesh, Rajasthan, and Bihar, thenon-traditional maize-growing states of India.Table 6.17. Estimated gross income loss in maize production by environment and market orientation across the top30 priority constraints, using production-based and area-based combined indices, Asia (PPP $/ha).Estimated income loss usingEstimated income loss usingMaize production environment maize farm-gate price maize price at nearest marketand market orientation Production-based Area-based Production-based Area-basedWith STMA-commercial areas of IndiaIrrigated lowlandCommercial 447.49 661.84 549.20 812.26Rainfed lowlandSemi-commercial --- 291.19 --- 326.95Commercial 361.78 485.12 554.21 743.16Rainfed uplandLow market surplus --- --- --- ---Semi-commercial 348.36 411.25 462.11 545.53Commercial 259.86 293.76 419.32 474.03Subtropical/mid-altitudeLow market surplus --- --- --- ---Semi-commercial --- --- --- ---Commercial * * * *Without STMA-commercial areas of IndiaIrrigated lowlandCommercial 385.27 447.49 472.83 549.20Rainfed lowlandSemi-commercial 260.61 256.13 292.61 287.58Commercial 361.78 361.78 554.21 554.21Rainfed uplandLow market surplus --- --- --- ---Semi-commercial 411.25 370.58 545.53 491.59Commercial 301.18 270.01 485.99 435.70Subtropical/mid-altitudeLow market surplus --- --- --- ---Semi-commercial --- --- --- ---Commercial • • • •China’s Southwest maize production systems (gain by addressing all top 30 priority constraints in each production system)Rainfed spring maize (low market surplus) 2,423.52 ** 2,477.38 **Rainfed summer maize (low market surplus) 2,086.56 ** 2,136.24 **Rainfed fall maize (low market surplus) 1,478.72 ** 1,512.32 **--- Not computed; no constraint from this environment and production system was included in the top 30.* Not computed; price data not available.• Not computed; excluded from the analysis.** Same as corresponding production-based results.61

At the production environment and market orientationlevel, the top 30 priority constraints appear to havethe highest combined economic impact on maizeproduction in Asia’s irrigated-commercial areas, bothwith and without India’s STMA-commercial maizeareas. At farm-gate prices, commercial maize farmersin the irrigated areas could lose at least PPP$ 447/ha(Table 6.17) due to the combined effect of the top30 priority constraints. Irrigated-commercial maizeproduction areas include the irrigated environmentsof Indonesia and the lowland areas of Vietnam’ssoutheastern region and Mekong River Delta.Estimated losses of maize production gross incomedue to the various constraints are even higher whennearest-market grain prices are used in the calculations.This confirms the earlier observation that Asianmaize farmers may lose more income when they selltheir products at the farm gate, and not at the nearestmarket, where they can get better prices.6.5 Technology Options to AddressPriority ConstraintsThe country-level national maize R&D prioritizationworkshops each identified technological as well associoeconomic and policy-related options to addresspriority constraints. The workshops also identifiedsuppliers with a comparative advantage in deliveringa particular research product. The country-leveltechnology and policy-related options identifiedhave been consolidated into one regional list for thisreport (Table 6.18). The sub-sections below discusstechnological options for alleviating abiotic and bioticconstraints. Country-level maize sector policies andpolicy-related options for alleviating the socioeconomicconstraints of Asian maize farmers are summarizedand discussed in a separate chapter (see Chapter 2).6.5.1 Technology interventions to addressabiotic constraintsDrought/moisture stress. Most yield losses due todrought are caused by moisture stress during floweringand grain-filling, and the rest by pre-floweringstress. Maize is unusually susceptible to droughtat flowering, when it depresses yield potential byirreversibly limiting the number of kernels and earsthat develop; if stress is severe, yields may be reducednearly to zero. At this stage in crop development, it istoo late for farmers to adjust management practicesbecause fertilizers and/or pesticides have alreadybeen applied, and the season is far too advanced toconsider replanting (Edmeades et al. 1994). In thisstudy, drought was reported as prevalent in the rainfedlowland and rainfed upland environments surveyed.Technologies that reduce the effects of drought includesmall-scale irrigation, rainwater harvesting, cultivarsthat either escape or tolerate the stress, and better cropand water management strategies such as conservation(zero/minimum) tillage.Small-scale irrigation/water harvesting. In India,water has been harvested since antiquity, when theart of water management was perfected. In the faceof adversity, communities have revived or creatednew water harvesting systems. Water harvestingstructures and water conveyance systems specific tothe ecological region and culture have been developed.Raindrops are harvested directly, and checkdams,called johads, 10 and other structures have been builtto harvest water that runs off the soil. Water fromrooftops is collected and stored in tanks built in thecourtyards. Rain from open community lands iscollected and stored in artificial wells. Monsoon runoffis harvested by catching water from swollen streamsand flooding rivers and stored it in various kinds ofreservoirs. Traditional and contemporary technologiesallow water harvesting in both rural and urban areas.In many locations in India these technologies havehelped people withstand recurring drought, and thuslook promising for adaptation and promotion in otherparts of Asia.Although many water harvesting case studies haveshown increases in yield and water use efficiency, it isnot clear that the widespread use of these technologiesis feasible (Rosegrant et al. 2002). The high initialcost of building the water harvesting structure oftenprovides a disincentive for adoption. These structuresare usually built during the dry season when laboris cheaper, but also scarce due to worker migration.Maintenance costs, on the other hand, often occurin the rainy season, when labor costs are higher dueto competition with conventional agriculture. Thuswhile many case studies show positive results, waterharvesting methods have yet to be widely adopted byfarmers.In the Philippines, the government’s Bureau of Soilsand Water Management (BSWM) carried out a 10-year small-scale irrigation project (official title: SmallWater Impounding Project or SWIP) that covered theconstruction of rainwater harvesting structures, smalldiversion dams, small farm reservoirs, and shallowtubewells. Under SWIP, structures were built across10Johads, traditionally used in Rajasthan, India, are small earthen check dams or crescent-shaped bunds, built across a sloping catchment to captureand conserve surface runoff. Water accumulating in the johad percolates into the soil and augments the groundwater. See www.rainwaterharvesting.orgfor the full list of water harvesting structures and conveyance systems in India.62

narrow depressions or valleys to hold back water,and reservoirs were created to store rainfall andrunoff during the rainy season. SWIP usually worksin partnership with local government units (LGUs)that have the technical capability to implement suchprojects. In most cases, the LGUs provide matchingfunds and, as a result, both national and localgovernments are held accountable for the project.Small diversion dams are concrete or rockfill structuresconstructed across a channel or river with continuousflow to raise the water level and divert the water bygravity from the source to the point of use. Primarilyintended for supplemental irrigation, diversion damsserve farmers who cultivate areas adjacent to smallrivers or creeks. A small farm reservoir (SFR) is asmaller version of SWIP structures, designed to collectand store rainfall and runoff for use on a single farm,for which water is delivered to the canals throughsiphons. A shallow tubewell (STW) is a tube or pipevertically set into the ground to a depth of 20-60 feetTable 6.18. Technology and socioeconomic/policy-related options for addressing priority constraints to maizeproduction in Asia.Technology supplierOptions identified to alleviate Concerned Private LocalPriority constraints priority constraint government agencies IARCs sector universities NGOsAbiotic constraintsDrought Small-scale irrigation * • •Rainwater harvesting • * •Tolerant maize variety • * *Early maturing maize variety • * *Conservation tillage (zero/minimum tillage) • * * • •Soil acidity Soil amelioration (organic matter/lime) * • • •Balanced fertilization * • • •Tolerant maize variety • * *Soil erosion Sloping agricultural land technology * * • •Conservation tillage (zero/minimum tillage) * * • •Biotic constraintsCorn/ear/stem borers Biological control * * •Tolerant/resistant maize variety • * *Integrated pest management * * •Biotechnology • * *Downy mildew Tolerant maize variety • * *Judicious pesticide application * • •Weeds Judicious herbicide application * • •Weed control management practices (proper land preparation) * • •Post-flowering stalk rot Tolerant/resistant maize variety • * *Integrated pest management (IPM) * • • •Socioeconomic/policy-related constraintsLimited access to Conduct of farmer field schools (FFS) / on-farm * • • •technical information research / other farmer trainingDevelop and promote school-on-the-air programs * • •Production and distribution of agricultural information materials * • • • •Lack of post-harvest Investment in appropriate, low-cost post-harvest * •facilitiesfacilities and technologiesPromotion/deployment of available post-harvest technologies * • • •Lack of capital/ Improved/increased access to formal credit * •inadequate credit Train farmers to become entrepreneurs * • •supportInadequate farm-to- Investment in rural transport facilities * •market transport systemPoor access to input Market matching * • •and output markets /inadequate marketingopportunitiesDevelop appropriate marketing network * • •Investment in / development of maize processing industry * •and livestock industrySource: IFAD-CIMMYT country-level National Maize R&D Priority-Setting Workshops, 2001-2002.Note: * - principal actor; • - secondary actor.63

(6-18 meters) to raise water by suction from shallowaquifers. Such small-scale irrigation systems consist ofone or more fully developed STWs equipped with acentrifugal pump powered by a 5-10 hp diesel engineor electric motor that can serve a contiguous area of 3-5ha. The system is designed for supplemental irrigationand as an irrigation water augmentation source at theend of existing systems. As of December 2003, 53,800of these four small scale irrigation projects had beenconstructed, with a total service area of about 176,500ha and 130,000 farmer-beneficiaries (http://bswm.da.gov.ph/swip.html, accessed 7 October 2004).Early maturing cultivars for drought avoidance.In Asia, 63% and 76% of maize varieties releasedby public and private maize breeding programs,respectively, by late 1998/early 1999 were extraearly(

lands is classified as acidic, and 38% of those landsare in Asia. The area of acid soils planted to maize inAsia was estimated to be around 2.5 M ha (de Leonet al. 2000). Acid uplands cover from 33% of totalland area in Indonesia and the Philippines to as highas 66% in Laos (Garrity 2000). The excess Al in acidsoils interferes with root cell division, increases cellwall rigidity, fixes P in less available forms in thesoil and on root surfaces, decreases root respiration,and interferes with the uptake, transport, and useof essential elements. Roots in acid soil becomeinefficient at absorbing and utilizing water andnutrients, even when the root zone is moist, quicklydeveloping wilting and nutrient deficiency symptoms(Pandey et al. 1994). Soil acidity, often due to longtermapplication of inorganic fertilizers, was widelyreported by farmer-respondents from the rainfedupland environments of Karnataka and AndhraPradesh (India) and Northern and Central Mindanao(Philippines). The options identified to help alleviatethis priority constraint included soil amelioration withorganic matter or lime, balanced fertilizer application,and the use of tolerant maize varieties. In general,applying adequate amounts of lime to increase soilpH to approximately 5.5 is recommended. Very fewof the surveyed villages practiced liming, whichneeds to be repeated every few years and is tooexpensive for resource-poor maize farmers. Moreover,liming subsoils deeper than 30 cm is difficult andalso incompatible with the current trend towardsconservation tillage on sloping lands (Pandey et al.1994).The development and use of germplasm tolerantto both high Al toxicity and low P levels wouldprovide a less expensive, permanent solution. Acidtolerant cultivars would, for example, help minimizethe amount of amendments (lime) applied to thesoil. Tolerant varieties also offer an ecologicallyclean, cost-effective way to increase maize yields inaffected areas. They would allow sustainable maizecropping systems to be established on less-utilizedacid hillsides, thus reducing the pressure on morefragile lands. CIMMYT researchers have found thatacid tolerance is present both in germplasm materialsthat originated in acid areas and in those that havenever been grown in acid soils. Also, yield in acidenvironments was found to be positively correlatedwith yield in non-stress environments, so that it ispossible to develop a variety that would yield well inboth acid and normal soils. Tolerance to soil aciditydoes not necessarily mean lower performance underhigh management conditions (Pandey et al. 1994).The national programs of Indonesia and Colombia,two countries with extensive acid soil problems,have released varieties derived from CIMMYT MaizeProgram materials: Antasena in Indonesia and Sikuaniin Colombia (de Leon et al. 2000).Soil erosion. Sloping uplands cover about 60-90% ofthe total land area of each country in Southeast Asia(Garrity and Sajise 1993). Soil erosion, as estimated byriver sediment load per hectare of watershed, is muchmore serious in Southeast Asia than in any other regionof the world (Milliman and Meade 1983). The densestpopulations in the world are transforming thesewatersheds at a tremendous rate, and exacerbatingtheir degradation.Soil erosion in the rainfed upland environmentssurveyed (particularly in Cagayan Valley, Philippines,and the outer islands of Indonesia) can be substantiallyreduced by the adoption of sloping agricultural landtechnologies (SALT) and/or of conservation (zero orminimum) tillage systems. SALT is a package of soilconservation and food production technologies thatintegrates several soil conservation measures in justone setting. With SALT field and permanent crops aregrown in 3-to-5-meter-wide bands between contouredrows of nitrogen fixing trees, to minimize soil erosionand maintain soil fertility. The trees are thickly plantedin double rows to make hedgerows. When a hedge is1.5 to 2.0 meters tall, it is cut down to about 75 cm, andthe cuttings (tops) are placed in alley-ways to serve asorganic fertilizer.A simple, easily applied, low-cost, and timely methodof farming uplands, SALT was developed for Asianfarmers that have few tools, limited capital, and littletraining in agriculture. With the SALT system, farmerscan grow familiar crop varieties and use traditionalfarming patterns. If farmers leave the SALT farm, assome tribal groups do, the nitrogen fixing trees andshrubs (NFTS) will continue to grow and overshadowthe cropping area. By the time the land is cultivatedagain, the soil has been enriched by the large amountof NFTS leaves, and there is no erosion to contendwith. The trees may also be harvested for firewoodor charcoal (ARLDF 1997). Extensive data from theIBSRAM Sloping Lands Network trials in six countrieshave confirmed that annual soil loss with hedgerowsystems is typically reduced by 70-99% (Sajjapongseand Syers 1995).Unfortunately, farmer adoption of these systems isstill very low. Constraints include the tendency forperennials to compete for growth resources, therebyreducing yields of associated annual crops and theinadequate amounts of phosphorus that are cycledto the crop in the prunings. The major problem isthe extra labor needed to prune and maintain thehedgerows. Farmers’ labor investment to pruneand maintain the hedgerows was about 31 days perhectare, or 124 days of annual labor for four prunings,which increased labor for a maize crop by 90% (ICRAF1996). Such an increase in production costs was seldomrewarded by a commensurate increase in returns.65

Conservation tillage (CT) involves minimal soildisturbance and maintains at least 30% of cropresidues on the soil surface at planting time (CTIC1994). The residue cover helps minimize wind and soilerosion, preserve soil structure, and conserve water byreducing runoff. Four specific tillage systems can beclassified as CT: no-till (zero tillage), ridge till, mulchtill, and minimum/reduced tillage systems (Monsanto2004). The no-till or zero tillage system involves noseedbed preparation other than opening the soil (a slitor punched hole) to place seed at the intended depth.In the ridge till system, row crops may be planted onthe ridge top, in the furrow, or along both sides of thefurrow as a way to manage soil moisture. The mulchtill system retains crop residues on the soil surface,providing a protective cover. The minimum/reducedtillage system involves minimal soil disturbance forcrop production, and minimal use of conventionaltillage equipment.A primary advantage of CT is that no additionalinvestment in land conservation measures suchas terraces, contour bunds, or soil conservationbarriers is required, making the technology equallyaccessible to small- and large-scale farmers. Thetechnology also offers (a) easier land preparationand weeding operations, (b) substantial savings byreducing mechanical power and labor requirements,(c) improved soil quality and moisture retention,(d) reduced soil erosion, and (e) increased yields.Conservation tillage systems, however, are notuniversally applicable. They are adapted for use inareas with sandy to clay loam soil types, in flat tohilly and slightly rolling terrain, and in areas with noconventional tillage implements. Adoption of reducedtillage systems is limited by two factors: competitionfor crop residues and input availability (Pingali andPandey 2001). Where crop residues are important forlivestock feed, it is difficult for farmers to leave evena part of those residues on the soil. Herbicides andmachinery are crucial to CT, and limited access to theseinputs can constrain adoption in remote low marketsurplus production systems.6.5.2 Technology interventions to addressbiotic constraintsStem and corn borers. Stem borer resistance hasbeen developed through conventional breeding andgenetic engineering. Entomology research at CIMMYThas aided in identifying inbred lines that possesstolerance to fall armyworm and maize stem borers. Inthe field, on-farm tests of CIMMYT’s insect resistantmaize populations have shown that, in addition tohigh yields, the populations have better grain qualitybecause they avoid Stenocarpella maydis ear rot. Onearea that warrants further research is germplasm thatresists “second-generation” attack, i.e., larvae thatattack during flowering. Selection for resistance atflowering has been slow because borers feed on diverseplant tissues at this time. Historically, selection focusedon increasing stalk strength to withstand tunneling,thereby facilitating mechanical harvesting. To reducesecond generation damage, researchers are nowscreening plants for reduced feeding damage in thetissues first fed upon by larvae, specifically the sheath,husk, and ear.In eastern, western, and central Africa, where farmerscan lose 15-45% of their annual maize harvest to stemborers (equivalent to at least 400,000 tons of maize,valued at US$ 90 million) CIMMYT’s Africa MaizeStress (AMS) Project helped the Kenya AgriculturalResearch Institute (KARI) to identify insect-resistantmaize and develop testing sites where thousands ofborer larvae are reared and applied each crop cycle.Maize that resists one or more borer species has beenidentified and is being made available to farmers. Leaftoughness, a trait identified at CIMMYT, has been usedto screen maize for insect resistance without havingto infest plants in the field. Leaf toughness has beenobserved to lower insect feeding and is not negativelycorrelated with yield. Molecular research at CIMMYThas found five markers that account for nearly twofifthsof the phenotypic variation for leaf toughness. Inwestern and central Africa, intercropping maize withcassava or cowpeas has been found to reduce yieldlosses from stem borers.In 1984, Taiwan’s Council of Agriculture launched aprogram to biologically control the Asian corn borer,Ostrinia furnacalis Guenee, on both field corn and sweetcorn using an egg parasitoid, Trichogramma ostriniae.For higher effectivity in controlling the Asian cornborer, T. ostriniae was combined with two or threeinsecticide treatments on field corn, and with threetreatments of Bacillus thuringiensis and one low-toxicityinsecticide for sweet corn. This system of integratedcontrol significantly reduced the proportion ofdamaged plants and increased grain yield of field cornby 11%. The long-term mass release of T. ostriniae alsoshowed an accumulative effect in suppressing cornborer population density (Tseng 2000).More recently, genetic engineering has been used toincorporate genes derived from Bacillus thuringiensis(Bt) into maize, which could provide effective inherentcontrol of stem and corn borers, allowing maize to begrown using fewer chemical pesticides. The use ofbiotechnology crop varieties has also been documentedto increase fertilizer efficiency, provide more flexibleweed control, promote conservation tillage, protectwater quality, and aid in soil conservation (Phillips2001). By 2004/2005, Bt maize had been adopted by12 countries (US, Canada, Spain, Germany, Argentina,66

Honduras, South Africa, Uruguay, Portugal, France,Czech Republic, and the Philippines) (ISAAA 2005),that cover nearly 20 M ha, which is equivalent to 13.6%of the global maize area of 146.8 M ha (FAOSTATProduction Domain, accessed 13 December 2006).There is also significant potential for Bt maize in India,Indonesia, and Thailand, where Asiatic corn borers,spotted stem borers, Asian pink stem borers, andarmyworms are important pests. In China, early fieldtrials indicate that Bt maize can increase yields by over15%, which is significant because China has around25 M ha of maize. Prior to commercialization in thePhilippines, significant increases in productivity of25% in the dry season and 40% in the wet season havebeen reported from Bt maize field trials (ISAAA 2004).As early as 2001, Monsanto Philippines reported anaverage 40% yield advantage for Yieldgard ® , its Btprotectedmaize variety, over traditional hybrid maizebased on results of their multi-location trials across thePhilippines (Monsanto Philippines 2001). Extensiveresearch on Yieldgard ® in the US also showed that itis nutritionally equivalent and performs comparablyto conventional varieties for livestock and poultryfeed; it is also effective for controlling targeted insectpests without harming humans, fish, wildlife, andbeneficial insects, and has improved grain quality,thus enhancing food and feed safety (Monsanto 2001).A year after commercialization in the Philippines(with approximately 20,000 ha planted in 2003),Yorobe (2004) reported that Bt maize yielded 37%more than conventional maize crops. Although itsproduction costs were higher (due to more expensiveseed), Bt maize also generated a higher net income.Filipino farmers who planted Bt maize saved PhP168/ha (about US$ 3/ha) on pesticide and earned anadditional PhP 10,132/ha (about US$ 170/ha).However, this technology does raise a major concern:the development of Bt resistant insects. Effective insectresistance management (IRM) strategies must beestablished to counter this natural adaptation. Refugiaare needed to maintain populations of susceptibleinsects that will mate with resistant insects anddelay the development of Bt resistance (Pingali andPandey 2001). Refugia are a central component of abroader IRM strategy that includes integrated pestmanagement and the combination of multiple sourcesof insect resistance in the maize plant. Stacking orpyramiding Bt genes to ensure that multiple toxins areexpressed in plants will also prolong resistance, whichmay be further enhanced through the incorporation ofconventional resistance.The application of DNA markers has the potentialto improve selection efficiency and accelerate thedevelopment of germplasm with good agronomicperformance and adequate levels of host plantresistance. Restriction fragment length polymorphism(RFLP) markers are being used to identifychromosomal regions that control quantitative traitloci (QTL) affecting insect resistance. Cost-savings andefficiency gains are expected if RFLP markers can beused in practical breeding to facilitate the incorporationof insect resistance into elite germplasm (Dowswell etal. 1996).Downy mildew. Several strategies are available forthe management of downy mildew diseases of maize.Cultural practices have been suggested, includingroguing or eliminating diseased plants and plantinga wide area within two weeks to eliminate thecontinuous presence of inoculum. These practices,while practical and economical, require perfect timing,early monitoring, and prompt destruction of infectedseedlings. Seed dressing with a metalaxyl slurry hasbeen the standard treatment to control downy mildewfor more than two decades. Metalaxyl, also knownas Apron, SD, and Ridomil, is a systemic fungicidecapable of protecting maize crops from downy mildewinfection. Complete reliance on metalaxyl is, however,not a sound practice; should a metalaxyl-resistant strainof the downy mildew fungus arise, all gains achievedthrough maize breeding could be wasted. The chemicalhas actually begun to show signs of declining efficacybased on observations of plants from treated seed lots(Raymundo 2000).A broad range of high-yielding germplasm withadequate downy mildew resistance is now availablefor most areas where the disease is prevalent. Maizeresearch institutions in Thailand and the Philippinessignificantly contributed to the development of thesematerials. International germplasm exchange in Asiaand cooperative testing for downy mildew resistancehas led to the rapid development and deploymentof resistant varieties, including Thailand’s Suwan 1(released in 1973), the DMR (downy mildew resistant)composite series released in the Philippines in the1970s, Indonesia’s Penjalinan, Ganjah Kretek, andArjuna, and India’s Ganga 5 (Raymundo 2000). ADB-CIMMYT’s Asian Maize Biotechnology Network(AMBIONET) Project used molecular markers toaccelerate breeding for traits of interest, includingdowny mildew resistance. CIMMYT’s now-defunctAsian Regional Maize Program (ARMP) used to workwith national systems in the region to develop anddisseminate high yielding varieties and hybrids, withspecial emphasis on resistance to downy mildew, aregionally important maize disease. 11 In addition toadapting germplasm from CIMMYT-Mexico for usein Asia, the ARMP developed several broad-based,11The Thailand-based Asian Regional Maize Program (ARMP) no longerexists. Today a few CIMMYT scientists are stationed in the region(India, Nepal, Bangladesh), but none of them works specifically ondowny mildew.67

downy mildew-resistant populations to meet a rangeof local needs. It also pyramided genes conferringresistance to other types of stress (such as corn borers,acid soils, and waterlogging) into germplasm thatalready possessed downy mildew resistance.The private sector has been selling, in mildew-proneareas, seed of commercial cultivars treated withRidomil, a systemic fungicide; it has also beganto develop resistant cultivars. Seed treated withRidomil is, however, too expensive for resource-poorfarmers. While the public sector has been relativelysuccessful in developing resistant cultivars throughtraditional breeding (de Leon and Lothrop 1994), manysuch cultivars are unfortunately not reaching farmersdue to insufficient seed production and distribution.Stalk rots. 12 Considerable research has been done onfactors that influence stalk rots, the most importantof which is the inherent susceptibility of the cultivar,which may be altered by weather conditions, moistureavailability, grain fill and kernel number, culturalpractices, plant densities, leaf disease damage, cloudcover, and insect damage. In general, stalk rot is mostsevere and incidence is greater with increased fertility.Several studies suggest that a balanced, continuoussupply of nitrogen throughout the maize growingseason helps reduce stalk rot incidence, and that whenadequate potassium is present, stalk rot severity isreduced. Tillage practices have also been related to theoccurrence of stalk rot diseases. In the absence of leafblights, reduced tillage has been shown to actuallydiminish most stalk rot diseases, possibly because itdecreases soil moisture evaporation and results in lessstress.A systematic and comprehensive breeding programon stalk rot resistance began in India (Delhi, Ludhiana,Hyderabad) in the early 1990s, and in Thailand (FarmSuwan) in 1993, in collaboration with CIMMYT’sARMP. During the last 30 years, a wide collection ofmaize has been screened at research stations coveringmost disease-prone agro-climatic regions in India. Earlymaturing composites developed in the 1970s and early1980s in Pantnagar have shown high levels of resistanceto both downy mildew and stalk rot. ‘Rudrapur local’contributed resistance to Erwinia and Pythium (preflowering)stalk rots to the highly resistant hybridGanga Safed-2. Pioneer Hi-Bred Inc. has apparentlyfound resistance to Pythium stalk rot and is using it inbreeding its materials in Egypt (Dowswell et al. 1996).However, when there is severe environmental stressand disease pressure, stalk rots can affect even the mostresistant germplasm (De Leon and Pandey 1989).Since there are few sources of resistance to stalk rotcomplexes in maize, breeders have been examiningrelated species and genera such as Zea diploperennis,Z. mexicana, Z. perennis, Z. luxurians, Tripsacum spp.,Coix, Chlonachne, and others for the presence ofgenes conferring resistance to stalk rots. Teosinte,for example, has excellent resistance to many of thestalk rot pathogens. Research is still needed, however,in the following areas (a) inoculation techniques todistinguish resistant plants from escapes; (b) studies onappropriate inoculum loads for individual pathogens;and (c) genetic studies on resistance to differentpathogens.Post-harvest (storage) pests. Damage from insectsand diseases that attack stored grain commonlycauses losses of more than 20% of harvested grainin least developed countries (LDCs), where tropicalconditions allow insects and disease pathogens toreproduce rapidly and colonize unprotected grain(Meikle et al. 1999, as cited in Bergvinson and Garcia-Lara 2004). Among the important storage insect pestsof maize in developing countries are the maize weevil(Sitophilus zeamais), larger grain borer (Prostephanustruncatus), and red flour beetle (Tribolium castaneum).Control of storage pests has largely focused on hygieneand includes thorough cleaning of the storage area,removing old grain from the area, repairing storagestructures, filling cracks where insects can hide, andapplying recommended insecticides to storage wallsand ceilings (especially if infested) (Bergvinson 2001a).One of the most successful means of preventing postharvestpests is good husk cover, the result of farmerselection over time (Dobie 1977, as cited in Bergvinson2001a) and seed hardness (Ransom 2001). Selectinghard grain types and types known to slow the buildupof insects may reduce losses due to insects. However,for kernel hardness to be effective, grain moisturecontent must be below 16% for the existing resistantsources (Bergvinson 2001b).To capitalize on genetic diversity for storage pestresistance, researchers have made significant progressin understanding the biochemical, biophysical, andgenetic bases of host plant resistance to ensure that thetraits being selected meet consumer demands. A newapproach called “targeted allele introgression” usingmarker-assisted selection (genetic engineering) is beingused to introduce storage pest and disease resistanceinto farmers’ varieties, while preserving their foodand processing qualities and enhancing their geneticdiversity. The impact of this technology will be feltmost in LDCs because it is packaged in the seed andbeing designed to ensure that farmers have the optionto recycle seed, a common practice among subsistencefarmers (Bergvinson and Garcia-Lara 2004).12This section draws heavily from Lal et al. (2000).68

6.6 Sources of Research andTechnology 13This section discusses the roles and responsibilitiesof the public and private sectors (including, amongothers, IARCs, such as CIMMYT, NARESs, andnational and multinational companies) that supplyresearch and technology products. Each of theseplayers has unique capabilities, resources, andcomparative advantages that can help alleviate maizeproduction constraints in Asia.6.6.1 Public and private sectors:delineation of research responsibilitiesWhen prioritizing future public sector maize research,it is important to accurately anticipate prospectiveprivate sector activities in order to avoid duplicatingefforts and identify potential areas of collaboration.The private sector has been active in maize research,development, and dissemination since the 1930s and1940s. The private sector has been active in tropicalmaize systems that support commercial production,developing and selling hybrids adapted to particulargeographic and ecological regions. It has been wellacknowledged that the private sector is far moreeffective than the public sector in providing seed tofarmers in most developing countries (see Morris 1998and Gerpacio 2001).During the past decade, private sector investment intropical maize has increased substantially, mainly dueto four factors:1) The rapid growth in feed maize demand andconsequent commercialization of maize productionsystems have provided an impetus for privatesector investment;2) Global amalgamation of agribusiness has broughtsignificant resources to bear on the problems oftropical maize systems;3) The emergence of biotechnology as a strategicforce in developing agricultural technology andenormous private sector investments in exploitingthis technology; and4) The increased use of intellectual property rights(IPR), which allows developers of a technology toappropriate the profits it generates.6.6.2 Role of the public sectorWhen national research systems are initially formed,state-sponsored organizations almost always play adominant role in organizing the development anddissemination of improved technologies. Over time,however, the role of the public sector typically declines,and its functions are gradually taken over by theprivate sector. In Asia today, public sector dominanceof agricultural R&D is largely a thing of the past. OnlyChina and India retain sizeable public agriculturalresearch and extension systems. In Southeast Asiatoday, public maize breeding research is carried outonly in two or three organizations per country.Nevertheless, a key role of national and internationalpublic sectors has been training and human resourcedevelopment, which, by lowering the costs of learningand capacity building, has encouraged privatecompanies to become involved in R&D. The publicsector will continue to enjoy a strong comparativeadvantage in human resource development, especiallyin the developing world.Many national and international public sectors alsoopt to move upstream in the germplasm developmentprocess, for example, by concentrating on geneticresource conservation and management and prebreedingactivities designed to produce basicgermplasm that can be used as source materials bycommercial breeding programs. Public sector effortsin the collection, characterization, and preservationof genetic resources have resulted in significant socialand private sector benefits. Social benefits are gainedin terms of conserving the rich genetic heritage oflandraces and wild relatives of maize (and other crops)that are in danger of disappearing from developingcountry farming systems. Private sector benefits accruein terms of free access to genetic resource collectionsthat private companies can use to enhance their cropbreeding activities. Pre-breeding research to produceelite breeding materials that can be used to developlocally adapted varieties will remain an importantpublic sector activity. Although some believe that it willbecome obsolete if anticipated advances in genomicsare achieved, pre-breeding will remain an importantcomponent of maize research in developing countriesfor the next 5 to 20 years.Within the realm of genomics and biotechnology,advanced national research institutes (such as thoseof China and India) and multinational companieswill probably maintain their dominance in basic andapplied research. Nevertheless, the international publicsector could act as a conduit that provides access tothese technologies by training developing countryscientists in their use.13This section draws heavily from Pingali and Pandey (2001)and Gerpacio (2001).69

Perhaps most important to the world’s poorest farmersand communities, the public sector will continue to bethe sole source of research and technology (particularlydeveloping and evaluating varieties) for geographicareas that the private sector considers unprofitable.These include predominantly low market surplusproduction areas with low market potential or that aremarginal in terms of crop production (e.g., droughtproneenvironments). For these environments, a fewpublic agencies produce maize seed and distributeit through agricultural extension services, especiallyin less favorable areas where maize is an importantcrop. One may expect private sector involvement tobe relatively low in the Central and Eastern Visayasregions of the Philippines and some parts of SouthAsia. Hence, rather than producing hybrid seed indirect competition with private companies, manypublic seed agencies concentrate on promotinginformal seed production and distribution systems(Dowswell et al. 1996).6.6.3 Role of the private sectorPrivate sector participation in maize R&D in Asiahas grown steadily since the early 1990s, when awave of policy reforms broke up what had effectivelybeen state monopolies in many countries. At thattime, governments began framing new policiesto encourage the growth of national private seedcompanies, and provide a better business environmentfor multinationals. Most multinational companieshave strong maize R&D and seed production anddistribution infrastructure, which allows them tooperate efficiently in major maize production zones.The private sector began to invest more in breedingmaize hybrids (and varieties in some instances) fordeveloping countries, particularly tropical maizeproduction systems and areas where secure profitscan be realized. Maize R&D activities pursued by theprivate sector in Asia vary depending on the size ofthe company and the volume of maize seed it sells.Generally speaking, the larger the company, the greaterits ability to establish its own breeding program. Manysmaller seed companies that lack in-house researchcapacity contract public research programs andsometimes even large private companies to multiplyand distribute seeds of improved OPVs and hybridsdeveloped by others. Most multinationals operatingin Asia are large enough to have ventured intobiotechnology research.The private sector will continue to be the dominantplayer in genomics and biotechnology, both in termsof investment and as a source of technology andbioinformation. Through consortia and alliances,these resources will be made available to national andmultinational companies in the developing world.Having developed transgenic maize, the private sectorpromises to provide maize cultivars that tolerate orresist a wide range of stresses and offer improvednutritional quality. This could broaden the range ofenvironmental conditions under which maize canbe grown and increase its productivity and stability.Developing world maize farmers and consumers,however, are just beginning to reap the full benefitsof these technologies (for example, through the recentcommercialization of Bt maize in some Asian countries)because the private sector, in light of inadequate IPRprotection, farmers’ inability to afford the product, andbiosafety concerns, has moved cautiously and slowly inextending these technologies to the developing world.The private sector also dominates the growing fieldsof genomics and proteomics, research areas that allowidentifying and studying a multitude of individualgenes, their interaction and expression under diverseenvironmental conditions. In addition, the discoveryof syntenies among species promises to revolutionizeplant breeding by allowing scientists to capitalizeon the basic similarity across all cereal genomesto quickly apply advances in one species to all theothers. Coupled with the ability to transfer genes ofinterest through genetic engineering, advances in thesefields will undoubtedly change the pace and scope ofagricultural research and development.6.6.4 Public and private sectors workingtogetherInternational maize R&D is carried out by a complexsystem made up of diverse organizations, large andsmall, public and private, national and multinational.Many of the organizations that participate in theglobal maize breeding system are linked through theexchange of products, services, or information. Howare these linkages playing out in the Asian context?Collaborative activities discussed below illustrate howpublic-private sector linkages are growing in Asia.International germplasm exchange. The establishmentof the Consultative Group on International AgriculturalResearch (CGIAR) in 1969 provided a formal systemor mechanism through which the global breedingcommunity could access research products frompublic institutions. In Asia, maize germplasm is madeavailable by CIMMYT (a CGIAR center), which alsofacilitates international germplasm exchange, whereinany bonafide maize breeder can request samples ofexperimental materials (provided free of charge) foruse in his or her breeding program. Today, most publicand private breeding programs have received, andare using, germplasm materials from CIMMYT. Newinfusions of CIMMYT germplasm remain important70

to them as a source of genetic diversity and new traits(e.g., stress tolerance), but only new or emergingprograms rely heavily on direct requests of CIMMYTgermplasm.Public-private germplasm transfer. To reducegovernment expenditures, many Asian countrieshave scaled back investments in seed productionand distribution, activities that are readily assumedby private companies because they offer clear profitopportunities. Hence, to move improved materialsinto farmers’ fields, public breeding programs havemade their germplasm products available to privateseed companies, often on a commercial basis. Toillustrate, Thailand’s two main public breedingprograms (at Kasetsart University and the Departmentof Agriculture) provide elite breeding lines tomultinational and domestic private companies for usein developing commercial hybrids (see Ekasingh etal. 2001). Traditionally, these lines were provided freeof charge, but beginning in the mid-1990s, KasetsartUniversity started collecting royalties from privatesectorrecipients, who, in turn, are assured exclusiveuse of the germplasm.Genetic improvement. Strategic alliances in severalareas of genetic improvement—those that do notrequire the proprietary protection associated withgenetic engineering—would be beneficial to boththe public and private sectors. One example is thedevelopment of early maturing maize varieties andhybrids to fit the intensive cropping systems in thetropical lowlands of Asia. Both the public and privatesectors are particularly keen to develop hybrids forthe feed industry in the lowlands of Southeast Asia;the public sector is also interested in OPVs that couldenhance food supplies and food security in Asia. Thepublic and private sectors could also play mutuallysupportive roles in the development of maize that isresistant to economically important pests and diseases(such as downy mildew and borers) or tolerant toabiotic constraints such as drought, acid soils, andwaterlogging.Crop/resource management. Public/private sectoralliances are also possible in the realm of crop andresource management technologies. For example,public sector interest in promoting sustainable landuse, together with private sector interest in promotingRoundUp, an effective, safe, and inexpensiveherbicide, gave rise to a very successful partnershipaimed at developing and promoting zero tillagesystems in Argentina and Brazil (Ekboir 2000; Ekboirand Parellada 2000, as cited in Pingali and Pandey2001). Clearly, it would be constructive to exploresimilar win-win alliances in the geographic areasand agroecologies of Asia. A multi-sector, multi-actoralliance similar to the Rice-Wheat Consortium in theIndo-Gangetic Plains (covering Bangladesh, India,Nepal, and Pakistan; see www.rwc.cgiar.org for details)can be explored for maize in Asia. Within its nutrientmanagement objectives, the Consortium aims to:(a) develop a better understanding of the extent of yielddeclines, stagnating yields, and/or factor productivitydeclines in the rice-wheat systems; (b) gain a betterunderstanding of changes in soil biological activity andphysical properties in intensive rice-wheat productionsystems; and (c) develop effective soil nutrient supplyassays/tests that would lead to more effective fertilizerrecommendations/practices for rice-wheat systems.Mutual advantages. Public/private sector allianceswould help narrow the science/technology gapbetween rich and poor nations and also help delivernew technologies to farmers. For the private sector,participation in such alliances would acceleratethe progress of subsistence farmers towardscommercialization and increase their client base. Thepublic sector would benefit by gaining easier accessto private sector technologies and more sophisticatednetworks and techniques for technology dissemination.At the research level, the relative strength of the privatesector in biotechnology and genomics, and of thepublic sector in germplasm (especially informationand expertise related to desirable traits and germplasmimprovement for developing countries) could providea strong basis and considerable impetus for the creationof alliances.In low market surplus maize production areas, thepublic sector will continue to be the leading sourceof technology, although the need for private sectorsupport will increase. Public/private sector alliancescould promote spillover of research results from highpotential to low potential environments and fromeconomically advanced to economically deprived areas.Private sector innovations for more favorable areascould be shared with (or licensed to) the public sectorfor use in less favorable areas. Such arrangementscould provide the private sector opportunities tocontribute to the social good and promote the longtermcommercialization of less favorable low marketsurplus environments.In the high-potential commercial maize productionareas, the public sector can actively complement theactivities of the private sector. Pre-breeding researchand the provision of source germplasm would reducethe cost of private sector development of hybridssuited to particular ecological and geographic niches.Public sector research aimed at developing maizewith improved tolerance and resistance to abioticand biotic stresses for low-potential agroecologicalzones could also provide considerable benefits tohigh-potential environments. Similarly, the public71

sector could play a crucial complementary role to theprivate sector in developing appropriate crop andresource management technologies for high-potentialenvironments. Indeed, it would be beneficial for theprivate sector to fund such efforts.6.6.5 Analyzing the challengesSpielman and von Grebner (2004) assessed theopportunities for, and challenges to, creating andsustaining public-private partnerships between theIARCs and leading multinational, research-basedagribusiness companies, and tentatively concludedthat public-private partnerships are significantlyconstrained by insufficient accounting of the actualand hidden costs of partnership; persistent negativeperceptions across sectors; undue competition forfinancial and intellectual resources; and a lack ofworking models from which to draw lessons andexperiences. Despite these constraints, however, thereis sufficient common space in which to generate greateropportunities for public-private partnerships in propooragricultural research. To create an environmentmore conducive to public-private partnerships, thefollowing steps can be taken:a) Compile and maintain an analytical inventory/database of public-private partnerships in theIARCs and, more generally, in developing countryNARSs from which lessons may be learned.b) Identify feasible research problems andopportunities that require research inputs from boththe public and private sectors and are immediatelyrelevant to small-scale, resource-poor farmers andother vulnerable agents in developing countries.c) Increase the frequency and technicality of dialoguebetween the sectors to reduce negative perceptionsand foster understanding of potential researchopportunities; make the dialogues attractive andconstructive for decision-makers from both sectors.d) Improve the quality of cost-benefit analyses ofpartnerships and make available information onterms and conditions included in agreements tomanage risk and liability.e) Explore the creative use of third-party brokers andother mechanisms to separate research prioritysettingand financing from research execution.f) Engage in multi-stakeholder discussion on publicprivatepartnerships and agricultural biotechnologyresearch with a wider audience, despite thecontroversy and conflict that such interactions maycause.Spielman and von Grebner (2004) concluded thatif public- and private-sector actors are willingand able to take these steps, both may realize thepotentially significant benefits of greater intersectoralcollaboration, including improved access to scientificand financial resources, new synergies in research,access to new and emerging markets, and greatercapacity to solve problems that cannot be addressed bya single actor. Most important, greater public-privatepartnership may contribute to the improvement oflivelihoods for small-scale, resource-poor farmers,food-insecure urban and rural households, and othervulnerable individuals and households in developingcountries.6.7 Research and Policy AgendaIn Chapters 2, 5, and 6 we identified and prioritizedtechnical and socioeconomic constraints to maizeproduction, discussed possible solutions to alleviatethose constraints, and presented the country-specificpolicy environments in which players in the maizeindustry are embedded. The primary objective ofmaize research, development, and extension systemsin Asia is to generate new and improved technologiesthat will increase maize productivity and maizefarmers’ incomes as well as promote longer-termproduction resource sustainability. Appropriateinvestments in research, technology generation,and policy reforms could help alleviate many maizeproductivity constraints and improve maize farmers’livelihoods. In responding to commercializationtrends in maize production, however, research andpolicy should not emphasize just one technology ortechnology aspect. The focus of research should be toprovide maize farmers the flexibility to make decisionsacross technology choices (and perhaps even acrosscrop choices), amidst supportive agricultural policies.Both substantial crop-specific and system-levelresearch efforts will be important in providingmaize farmers the flexibility to make these choices.Crop-specific research includes increases in yieldpotential, shorter-cycle cultivars, improved qualitycharacteristics, and greater tolerance to biotic andabiotic stresses. System-level research would includeland management and tillage systems, farm-levelinput management, and post-harvest technologies.Also important is research on the impacts ofintensification in terms of buildup and incidence ofpest populations and soil degradation (depletion ofsoil nutrients, soil acidity). Including households,communities, and markets in the system, researchshould study the impacts of intensification onhousehold and community dynamics, on labor marketsand availability, as well as on output marketing anddistribution.72

Given growing populations and income-induceddemand for increased consumption, there continuesto be a strong need to achieve higher productivitylevels for the staple cereals. The more high-potential,favorable areas are diverted to non-cereal production,the higher the need for increasing productivity ofcereals, including maize. The question to ask is: towhat extent should the research system be concernedwith technological developments in marginal, lessfavorable environments, where maize often is grownby resource-poor farmers? In large countries, suchas India and China, with high domestic demand forcereals, the answer is relatively clear: investments inmarginal environments are absolutely essential forensuring food security, even if the country is integratedinto the global economy (Pingali 2004). Cost-effectiveresearch investments would occur in areas wherespillover benefits from favorable environments arehigh. Identifying strategies for diversifying the incomeand livelihood base of farm households in theseenvironments should also be an important area forresearch and policy.Commercialization trends require a paradigm shift inagricultural policy formulation and research prioritysetting, as presented earlier. The paradigm of staplefood self-sufficiency that has been the cornerstone ofagricultural policy in most developing countries needsto be updated given changing economic scenarios.Asian governments are facing a challenging task: toassure continued food security for ever-increasingpopulations on the one hand, and to diversify researchand infrastructural investments out of the primarystaples, on the other. Appropriate government policiescan alleviate many potentially adverse consequencesof commercialization and promote sustainableintensification of maize production, especially inrelatively marginal, less favorable environments. Longtermstrategies include investment in rural markets;transportation and communication infrastructure tofacilitate integration of the rural economy; investmentin crop improvement research to increase productivity,and crop management and extension to increasefarmers’ flexibility and reduce possible environmentalproblems from high input use; and establishment ofsecure rights to land and water to reduce farmers’ riskand provide incentives for investment in sustaininglong-term productivity.73

7. Summary and Conclusions7.1 Maize Production Environments• Maize (Zea mays L.) is a versatile crop that is adaptedto a wide range of production environments.Maize grows in diverse agroecological conditions—inareas from 30 to 2,500 m above sea level, with annualrainfall of as little as 500 mm to as much as 2,600mm, air temperatures as cool as 2 o C to as hot as 36 o C,and a growing cycle ranging from three monthsto more than a year. High grain yields have beenrecorded in locations with extreme temperatures. Themost common soil types where maize is planted areclay, clay loam, and sandy loam, which have beenconsistently described by farmers as well suited togrow most crops.• Most maize farmers in Asia are landowners ortenants/sharecroppers who have attended orcompleted primary education and whose farmsaverage 0.3-20.8 ha in size.Across the surveyed locations, maize farm sizes werelarger in rainfed upland-semi-commercial areas andsmaller in irrigated lowlands-commercial areas. InThailand, 31% of maize farmers had large farms thataveraged about 16 ha. Most landowners are locatedin irrigated lowlands-commercial areas, while mosttenants/sharecroppers live in rainfed uplands-lowmarket surplus production environments. Up to 92% ofmaize farmers in the rainfed upland-semi-commercialareas were landless laborers. In Vietnam, many uplandfarmers do not have legal land use privileges, andhence do not have access to formal credit sources norincentives to invest in the land they till. This indicatesthat land use or land reform policies may help improvemaize productivity and sustainability in the Asianuplands.• Most maize-producing upland villages are wantingin terms of better infrastructure (transport systems,irrigation, markets), post-harvest facilities, andsources of capital/credit.Most maize-producing upland villages have onlyseasonal/fair-weather or fair-to-good weatherasphalted roads. Small vehicles such as motorcyclesand pickup trucks in accessible areas, and animaldrawncarts in more remote villages, are used for farmtransport. Relatively better roads and transportationsystems are found in semi-commercial and commercialmaize-producing areas that are closer to markets or tolivestock feed processing centers.Lowland villages have small regular markets withinthe community, while other villages only have once-aweekmarkets. In Vietnam and the Philippines, somefarmers travel 15-57 km to reach the nearest primarymarket. In Nepal, a barter system prevails on a limitedscale. Irrigation, drying, and storage facilities formaize are not common, and farmers consider this animportant constraint to improved maize production inAsia.Most Asian maize farmers reported they did not haveenough capital and depended heavily on high-interestcredit provided by informal sources in a charge-to-cropscheme. In this lending scheme, trader-financiers sellagricultural inputs to farmers at higher than marketprices, and later buy back their harvests at lowerthan market value. Loans and interest are deductedfrom the total value of the harvest. Farmers tended topatronize these informal credit sources because theyfind them more reliable, accessible, and convenientthan formal credit institutions.• Across the region, government extension offices,private seed companies, and formal and informalfarmers’ organizations play a vital role indisseminating updated technology and informationto maize farmers.Mainly government extension agencies are responsiblefor conducting farmers’ field schools and trainingsessions. Seed and fertilizer companies also providetechnology information to maize farmers, but it isgenerally limited to the product being sold. Farmercooperatives and informal organizations give farmerstechnical information, in addition to acting as sourcesof capital and material inputs, and outlets for farm74

output. Youth and women’s organizations present inthe communities are also channels for disseminatingupdated agricultural and other information. Allthese channels, however, need to be improved, andthe agricultural extension system could be betterorganized. SUCs, NGOs, and IARCs are minimallyinvolved in local agricultural extension.• Asian maize farmers pay much higher pricesfor hybrid seed, about 9-16 times that of local/traditional and improved OP varieties.Farmers often recycle or exchange seed of local/traditional and improved OP varieties with otherfarmers within the community. If puchased, seed oflocal/traditional maize varieties was priced at PPP$0.58-0.95/kg, and hybrids at PPP$ 5.02-9.04/kg.Surprisingly, maize seed was most expensive in rainfeduplands-low market surplus and semi-commercialproduction areas. Transportation and handling chargesaccounted for most variation in maize seed pricesbetween production environments. Improvement intransport infrastructure and marketing systems mayhelp close the seed price gaps (as well as those of otherinputs), in favor of smallholder maize farmers in therainfed uplands.• Average daily wages for farm labor in the Asianuplands favor men over women.Male laborers earn, on average, PPP$ 0.72-1.06/persondaymore than female laborers. Men commonlyperform land preparation, planting, fertilizerapplication, and harvesting. Female labor is usedfor planting, weeding, harvesting, and post-harvestoperations such as shelling and drying. In Nepal, mostfarm operations are done by family and exchangelabor, and use of hired labor is not common, exceptduring peak periods within the crop season. In suchinstances, the development and promotion of smallfarm implements, specifically designed for use in therainfed uplands, could help alleviate the drudgery offarm operations for all agricultural labor.• Seed-to-grain price ratios are lower with local/traditional or improved OPV maize than in the caseof hybrids, and are higher when calculated based onfarm-gate prices, rather than nearest-market prices.Seed-to-grain price ratios were highest in rainfeduplands-low market surplus maize productionenvironments.Based on averages of all reported maize seed and allreported maize grain prices, the seed-to-grain priceratio was 6.52 and 5.20 across Asia during 2000-2001,using farm-gate and nearest-market grain prices,respectively. This indicates that Asian maize farmershave to sell, on average, 5.2 kg of maize grain to payfor one kilogram of maize seed. More specifically,seed-to-grain price ratios ranged from 1.15 to 3.40 forlocal/traditional and improved OPV maize, and from6.14 to 18.56 for hybrids.The computed input-to-grain price ratios were lowerwhen using nearest-market grain prices than whenusing farm-gate grain prices, indicating that Asianmaize farmers receive better prices for their grain ifthey sell at the nearest market than at the farm gate. Onaverage, farmers received PPP$ 0.44-0.82/kg of maizegrain at the farm gate, compared to PPP$ 0.64-0.92/kgat the nearest market. Where both are available, yellowmaize (usually hybrids) commanded a higher marketprice than white (local/traditional) maize.Across maize production environments, average pricesof farm inputs were higher in the rainfed upland areasthan in the irrigated lowlands, to the disadvantageof marginal resource-poor farmers in less favorableareas. Appropriate government agricultural supportprograms will be important to address this concernand alleviate rural poverty in marginal rainfed uplandenvironments.• Maize has become a major component of people’sdiets and the preferred substitute during periods ofrice shortage.Subsistence farmers in the rainfed uplands andsubtropical/mid-altitude environments kept as muchas 90% of their maize for household consumption.Maize utilization, however, differs from one locationto another, depending on people’s food habits.Where livestock and poultry production is booming,maize (mostly hybrids) is grown to sell to feed millindustries, while in predominantly low market surplusareas, maize (mostly local/traditional varieties) isgrown for household consumption.7.2 Maize Production Systems• Asian maize farmers choose varieties depending onintended use.For home food and animal feed, farmers prefer togrow local/traditional white maize varieties for theirgood eating quality, low input requirements, and lowproduction costs. If maize is grown purely for cashincome, farmers are more likely to grow hybrids, aslong as they have access to capital to pay for materialand labor inputs. Farmers are aware that higher yieldscan be expected from improved varieties if adequateamounts of inputs are supplied.75

When selecting hybrids, maize farmers prefer thosehaving high yields, heavy grains, general resistanceto pests and diseases, tolerance to drought and otherclimatic stresses, and high shelling recovery. Whenselecting white maize (mostly local/traditionaland improved OPVs), grown largely for humanconsumption, farmers prefer varieties with earlymaturity, high milling recovery, good eating quality,and general suitability to marginal soils. Othercharacteristics that influence farmers’ choice of varietyare grain weight, level of productivity, maturity period,and quantity and quality of foliage.• Farming households in the Asian uplands followtraditional maize production practices.Asian farmers grow maize as the main crop andintercrop or relay crop vegetables, root crops, legumes,or cash crops. They augment their income withsmall-scale livestock and poultry production. Maizeis generally grown twice a year, with the first cropusually planted after the first rain. Land preparationconsists of one or two plowing operations, andmaize seeds are generally sown manually. Weeds arecommonly controlled twice in the crop season, througha combination of hand-weeding, off-barring, andhilling-up. Fertilizers are also applied twice in a cropseason but often at rates lower than recommended.Pesticides are used only when infestation is heavy.Harvesting is done manually, and most farmers tendto sell yellow maize right after harvest, while local/traditional and improved OPVs are sun-dried andstored mainly for home use. Grains consumed at homeare shelled and milled as need arises. Maize yieldgaps can be due to erratic and unpredictable weatherconditions, lack of fertilizers, pest incidence, poorextension services, and poor management practices.farm machinery for maize-producing areas couldhelp ease tasks often performed by female laborers.Hybrids yielded 3.3-5.6 t/ha in irrigated and rainfedenvironments, and up to 7 t/ha in the STMA areas.Local/traditional and improved OPV maize yielded1.3-3.5 t/ha across irrigated and rainfed environments.7.3 Important ProductionConstraints• Maize farmers in different production environmentsacross Asia experience very similar biotic, abiotic,and socioeconomic and policy-related constraints tomaize production.The important biotic constraints identified weredowny mildew, Asian corn borers, stem borers, leafblights, and stalk rots in the field, and weevils duringstorage. Weeds (A. spinosus, C. odorata, and Ipomoeatriloba) are persistent problems every cropping season,causing yield losses as high as 100% if no handweeding or herbicide control is used. Drought, soilacidity, and declining soil fertility were the importantabiotic constraints identified. Soil fertility declineis due to intensified land use and rapid decline infallow periods, while soil acidity is often caused byinappropriate nutrient management practices.Poor technology transfer, inappropriate fertilizeruse, lack of capital or low-interest credit, lack ofappropriate varieties, lack of post-harvest facilities,high input prices, and low output prices wereamong the socioeconomic and policy-related maizeproduction constraints identified by Asian farmers,who considered them perennial problems hinderingimproved agricultural productivity.• Levels of agricultural input use and maize outputdid not vary significantly across the four maizeproduction environments identified in this study.Maize seeding rate ranged from 13-24 kg/ha in theirrigated lowlands, rainfed lowlands, and rainfeduplands to 6-60 kg/ha in the subtropical/mid-altitude(STMA) environments. Total fertilizer use was lowestin the STMA areas (22-142 kg/ha), and highest inthe irrigated lowlands (125-810 kg/ha). Total laborused ranged from 28 to 295 person-days/ha, with thelowest level used in the irrigated lowlands and thehighest in the uplands, reflecting the labor intensityof maize production in more marginal environments.Female farmers provided anywhere from 53% of farmlabor in the irrigated and rainfed lowlands to 62% inthe STMA environments. The development of small• Addressing the problem of drought in therainfed lowland-commercial, rainfed uplandsemi-commercial,and commercial productionenvironments first would provide the highesttechnical returns to maize R&D investments in Asia,when contribution to regional maize production andshare in regional maize area are considered.Drought is estimated to affect about 6.8 M ha in thethree above-mentioned production environments(53.5% of the total regional maize area), where anestimated 16 Mt of maize are produced and about 48million rural poor are located. Alleviating droughtin these environments could improve yield by anestimated 35%, which would have an enormous impacton maize production in Asia.76

Apart from drought, other abiotic constraintsand socioeconomic and policy-related constraintsdominated the priority lists. These included soilerosion/landslides, soil micronutrient deficiency,waterlogging, lack of capital or poor access to lowinterest credit, poor agricultural extension/technologytransfer services, and poor access to input and outputmarkets. Socioeconomic and policy constraints wereestimated to affect up to 180 million rural poor peopleacross the region (including the STMA commercialareas of India), and alleviating them could improvemaize productivity by at least 18%.• Constraints reported from STMA (commercial)environments tend to dominate the top 30 priorityconstraints based on the combined index; whenSTMA environments are excluded from theanalysis, constraints reported from rainfed upland(semi-commercial and commercial) environmentsdominated the constraint priority lists for Asia.The top 30 priority constraints based on the combinedindex focus more on semi-commercial and commercialmaize areas. Due to low maize production, the lowmarket surplus (subsistence) system does not appearon any of the priority lists. Subsistence farming,however, remains important, particularly in Indonesia,Nepal, and the Philippines and, consequently, it isimportant to continue investing (relatively modestly)in subsistence farming research. Mechanisms thatpromote research spillovers from more commercialareas to subsistence farming environments ought to beestablished.• Drought appears to have the highest economicimpact in terms of estimated gross income loss,followed by socioeconomic and policy-relatedconstraints, and other abiotic constraints. Bioticconstraints (pests and diseases) have the leasteconomic impact of the identified maize productivityconstraints.Semi-commercial and commercial maize farmers inrainfed upland environments were estimated to loseat least PPP$ 450/ha when drought conditions persist.Irrigated lowland-commercial maize farmers perceivethat they lose the most income in maize productiondue to limited capital and/or poor access to lowinterestcredit.The top 30 priority constraints have the highestcombined economic impact on maize production inAsia’s irrigated lowland-commercial areas if India’sSTMA commercial maize areas are included, and inthe rainfed upland-semi-commercial areas if thoseenvironments are excluded.• A wide range of technology options that addresspriority constraints to maize productivity wasidentified across Asia.Technologies identified to address drought includedsmall-scale irrigation, rainwater harvesting, resistantor tolerant cultivars, and better crop and watermanagement strategies (e.g., conservation tillage). Soilamelioration with organic matter or lime, balancedfertilizer application, and the use of tolerant maizevarieties were suggested to help alleviate soil acidity,while SALT and/or conservation tillage systems werenamed to address soil erosion. In promoting thesetechnologies, however, compatibility with productionenvironment, location, farming system, and farmerresources and management capabilities need to beconsidered, given that technologies can be costly todevelop and disseminate to farmers.Strategies to help alleviate biotic constraints rangefrom the use of varieties developed throughconventional breeding and genetic engineering tochemical pesticide application and biological control.Cultural practices (e.g., elimination of diseased plants,synchronization of planting, reduced tillage) have alsobeen suggested. These practices, while practical andeconomical, require perfect timing, early monitoring,and prompt destruction of infected seedlings. Inaddition, many resistant maize cultivars developedthrough traditional breeding are unfortunately notreaching farmers due to lack of seed production anddistribution.• The public and private sectors have uniquecapabilities, resources, and comparative advantagesthat could contribute to alleviating constraints tomaize productivity in Asia, and links between thetwo sectors appear to be expanding.Hoping to avoid wasteful competition with theprivate sector, the public sector in general hasbegun to concentrate on activities and geographicareas that are unappealing to profit-oriented firms(e.g., genetic resource conservation, pre-breeding,population improvement, OPV development, aswell as subsistence production areas and marginalenvironments), while also continuing to advance theirhybrid maize research. National and internationalpublic sectors enjoy a strong comparative advantagein training and human resource development, whichlower the costs of learning and capacity building forprivate sector research and development.The private sector, meanwhile, has emphasizedinbreeding, hybrid development, hybrid seedproduction, marketing, and distribution. It has alsorecently become the dominant player in genomics77

and biotechnology. On the heels of transgenic maize,modern stress tolerant/resistant cultivars fromthe private sector promise to broaden the range ofenvironmental conditions under which maize canbe grown and increase its productivity and stability.Despite this increasing specialization, however, publicand private sectors working on maize continue to belinked through various types of collaborative activities.In Asia, these include international germplasmexchange, public-private germplasm transfer, geneticimprovement, crop/resource management, and mutualadvantages that work towards the advancement ofmaize cultivars. Hence, public/private sector alliancescould promote spillover of research results from highpotential to low potential environments and fromeconomically advanced to economically deprivedareas.7.4 The Role of PolicyThe growing trend towards commercializationand intensification of maize production requires aparadigm shift in agricultural policy formulation andresearch priority setting, different from the staplefood self-sufficiency paradigm that has been thecornerstone of agricultural policy in most developingcountries. Appropriate government policies canalleviate many of the possible adverse consequencesof commercialization and promote sustainableintensification of maize production, especially inmarginal environments. Long-term governmentstrategies address a wide range of activities frominvestments in rural infrastructure and agriculturalextension to providing farmers with secure rightsto land and water, as well as incentives to invest insustaining long-term productivity.As Rosegrant et al. (2002) aptly stated, emergingevidence shows that the right kind of investment canboost agricultural productivity far more effectivelythan previously thought in many less-favored lands (inthe case of maize). Increased public investment in theseareas has the potential to generate more competitive,if not greater, agricultural growth on the marginthan comparable investments in many high-potentialareas, and could have greater impact on povertyand environmental problems in less-favored areas.Although rainfed areas differ greatly from regionto region depending on biophysical characteristics,certain development strategies may work equallywell in many rainfed areas. Key strategies includeimproving technology and farming systems;ensuring equitable, secure access to natural resources;promoting effective risk management; investing inrural infrastructure; providing a policy environmentthat does not discriminate against rainfed areas (and/or against maize); and improving coordination amongstakeholders: farmers, NGOs, and public institutions.Finally, it remains important to recognize thattechnology—both simple and advanced—is not theonly key to increasing productivity, improving thesustainability of intensified maize production, andimproving the conditions of marginal maize farmersin Asia. No amount of advanced public- or privatesectormaize R&D will help the most disadvantagedfarmers, unless substantial parallel investments aremade in rural infrastructure, agricultural trainingand extension, input and output distribution andmarketing systems, and harvest and post-harvestfacilities. The returns to farmer investments in highyieldingvarieties will not be maximized if farminputs cannot be applied and outputs cannot bemarketed because of poor rural transport and marketsystems. Sustainability-promoting technologies willnot be attractive to farmers who have no incentiveto adopt them because they do not own the landsthey till. Improved technologies will remain on theR&D shelves if the agricultural extension systemhas no resources to effectively conduct its work anddisseminate research products to targeted end-users.In the end, appropriate government policies play abigger role and have greater impact than technologyalone does, a fact that has always been perceived byresource-poor farmers, the eternal target beneficiariesof agricultural R&D in Asia.78

ReferencesARLDF (Asian Rural Life Development Foundation). 1997.How to Series (1). Kinuskusan, Bansalan, Davao del Sur,Philippines. 20 p.Banzinger, M., S. Mugo, and G.O. Edmeades. 2000. Breedingfor drought tolerance in tropical maize: Conventionalapproaches and challenges to molecular approaches. In J.M.Ribaut and D. Poland (eds.), Molecular Approaches for theGenetic Improvement of Cereals for Stable Production in Water-Limited Environments. A Strategic Planning Workshop heldat CIMMYT, El Batan, Mexico, 21-25 June 1999. Mexico, D.F.:CIMMYT.Bautista, R.M., and A. Valdes. 1993. The Bias Against Agriculture:Trade and Macroeconomic Policies in Developing Countries.Institute for Contemporary Studies Press, San Francisco, CA.Beghin, J., and C. Fang. 2002. Protection and trade liberalizationunder incomplete market integration. American Journal ofAgricultural Economics 84:3.Bergvinson, D. 2001a. Global post-harvest losses in maize:An overview. In D.N. Manandhar, J.K. Ransom, and N.P.Rajbhandari (eds.). Developing and Disseminating Technologyto Reduce Post-harvest Losses in Maize. Proceedings of aWorking Group Meeting of the Hill Maize Research Projectheld in Khumaltar, Lalitpur, Nepal, on 25-27 September2000. Nepal Agricultural Research Council (NARC) andCIMMYT, Kathmandu, Nepal.Bergvinson, D. 2001b. Advances in post-harvest researchat CIMMYT. In D.N. Manandhar, J.K. Ransom, and N.P.Rajbhandari (eds.). Developing and Disseminating Technologyto Reduce Post-harvest Losses in Maize. Proceedings of aWorking Group Meeting of the Hill Maize Research Projectheld in Khumaltar, Lalitpur, Nepal, on 25-27 September2000. Nepal Agricultural Research Council (NARC) andCIMMYT, Kathmandu, Nepal.Bergvinson, D., and S. Garcia-Lara. 2004. Genetic approachesto reducing losses of stored grain to insects and diseases.Current Opinion in Plant Biology 2004, 7: 480-485.CIMMYT Maize Program. 1988a. Maize production regions indeveloping countries. (Internal document). InternationalMaize and Wheat Improvement Center, Mexico, D.F.CIMMYT (International Maize and Wheat ImprovementCenter). 1988b. Recent Advances in the Conservation andUtilization of Genetic Resources. Mexico, D.F.CIMMYT (International Maize and Wheat ImprovementCenter). 1989. Toward the 21 st Century: CIMMYT’s Strategy.Mexico, D.F.Costales, A.C. 2004. Changes in the incentives, infrastructuraland institutional environments impacting on maizecultivators in the Philippines. In N.C. Wada, C. Fang,and A. Gulati (eds). Maize in Asia: Changing Markets andIncentives. IFAD/CIMMYT/IFPRI, CIMMYT, Mexico D.F.Crawford, I.M. 1997. Rapid rural appraisal. In MarketingResearch and Information Systems (Marketing andAgribusiness Texts 4). Rome: Food and AgricultureOrganization of the United Nations. Retrieved 30 June2004 from www.fao.org/docrep/W3241E/w3241e00.htmCTIC (Conservation Technology Information Center). 1994.Definition of conservation tillage. Conservation Impact 12(11): 6.Dalmacio, S. 2000. Importance of and growing concernsfor maize diseases in the Asian region. In S.K. Vasal, F.Gonzalez Ceniceros, and Fan XingMing (eds.). Proceedingsof the Seventh Asian Regional Maize Workshop, PCARRD,Los Baños, Philippines (February 23-27, 1998), pp. 267-276.Dayanand. 1998. Principles governing maize cultivationduring rainy season. Indian Farming 48 (1): 84-87 (April).De Leon, C., and J. E. Lothrop. 1994. Lowland tropicalgermplasm development in Asia with emphasis onselection for downy mildew. In S. K. Vasal and S. McLean(eds.). The Lowland Tropical Maize Subprogram. MaizeProgram Special Report. Mexico, D.F.: CIMMYT. pp 30-45.De Leon, C., and S. Pandey. 1989. Improvement of resistanceto ear and stalk rot and agronomic traits in tropical maizegene pools. Crop Science 29: 12-7.De Leon, C., S. Pandey, and L. Narro. 2000. Developmentof acid soil tolerant maize germplasm. In S.K. Vasal, F.Gonzalez Ceniceros, and Fan XingMing (eds). Proceedingsof the Seventh Asian Regional Maize Workshop, PCARRD,Los Baños, Philippines (February 23-27, 1998), pp. 240-248.Delgado, C.L., M.W. Rosegrant, H. Steinfeld, S. Ehui, and C.Courbois. 1999. Livestock to 2020: The next food revolution.Food, Agriculture and the Environment. Discussion PaperNo. 28. IFPRI, Washington, D.C.Dobie, P. 1977. The contribution of the Tropical StoredProducts Centre to the study of insect resistance in storedmaize. Trop. Stored Prod. Inf. 34:7-22.Dowswell, C.R., R.L. Paliwal, and R.P. Cantrell. 1996. Maize inthe Third World. Westview Press, Boulder, Colorado.79

Edmeades, G., J. Bolaños, and S.C. Chapman. 1997a. Value ofsecondary traits in selecting for drought tolerance in tropicalmaize. In G. Edmeades, M. Banzinger, H.R. Mickelson, andC.B. Peña-Valdivia (eds.). Developing Drought- and Low N-Tolerant Maize. Proceedings of a Symposium, March 25-29, 1996,CIMMYT, El Batan, Mexico. CIMMYT/UNDP, Mexico, D.F.Edmeades, G., M. Banzinger, J. Bolaños, S.C. Chapman,A. Ortega C., H.R. Laffite, K.S. Fischer, and S. Pandey.1997b. Recurrent selection under managed drought stressimproves grain yields in tropical maize. In G. Edmeades, M.Banzinger, H.R. Mickelson, and C.B. Peña-Valdivia (eds.).Developing Drought- and Low N-Tolerant Maize. Proceedings ofa Symposium, March 25-29, 1996, CIMMYT, El Batan, Mexico.CIMMYT/UNDP, Mexico D.F. pp 263-273.Edmeades, G., H.R. Lafitte, J. Bolanos, S. Chapman, M.Banziger, and J. Deustch. 1994. Developing maize thattolerates drought or low nitrogen conditions. In G.Edmeades and J. Deutsch (eds.). Stress Tolerance Breeding:Maize that Resists Insects, Drought, Low Nitrogen, and AcidSoils. Mexico, D.F.: CIMMYT.Ekasingh, B., P. Gypmantasiri, K. Thong-ngam, and P.Grudloyma. 2004. Maize Production Potentials and Constraints,Research and Development Prioritization: The Case of Thailand.Mexico, D.F.: CIMMYT.Ekasingh, B., and K. Thong-ngam. 2004. Maize policies inThailand. In N.C. Wada, C. Fang, and A. Gulati (eds). Maizein Asia: Changing Markets and Incentives. Mexico, D.F.: IFAD/CIMMYT/IFPRI.Ekasingh, B.S., P. Gypmantasiri, and K. Thong-ngam. 2001.Impact of maize breeding research in Thailand. In R.V.Gerpacio (ed.). Impact of Public-and Private-Sector MaizeBreeding Research in Asia, 1966-1997/98. Mexico, D.F.:CIMMYT.Ekboir, J. 2000. Innovation systems and technology policy:Zero tillage in MERCOSUR. Paper presented at the XXIVConference of the International Association of AgriculturalEconomists. Tomorrow’s Agriculture: Incentives, Institutions,Infrastructure and Innovations. August 13-18, 2000, Berlin,Germany.Ekboir, J., and G. Parellada. 2000. Continuous complexinnovation processes: public-private interactions andtechnology policy. Paper presented at the XXIV Conferenceof the International Association of Agricultural Economists.Tomorrow’s Agriculture: Incentives, Institutions, Infrastructureand Innovations. August 13-18, 2000, Berlin, Germany.FAO (Food and Agriculture Organization of the UnitedNations). 2001. FAOSTAT Database. Rome, Italy.FAO (Food and Agriculture Organization of the UnitedNations). 2004. FAOSTAT Database. Rome, Italy.FAO (Food and Agriculture Organization of the UnitedNations). 2006. FAOSTAT Database. Rome, Italy.Garrity, D.P. 2000. The farmer-driven landcare movement: aninstitutional innovation with implications for extension andresearch. World Agroforestry Center, Nairobi, Kenya.Garrity, D.P., and P. Sajise. 1993. Sustainable land use systemsresearch in Southeast Asia: A regional assessment. InProceedings of a Workshop on Sustainable Land Use SystemsResearch. p. 59-72. Rodale Institute, Kutztown, Penn., USA.Gerpacio, R.V. (ed.). 2001. Impact of Public-and Private-SectorMaize Breeding Research in Asia, 1966-1997/98. Mexico, D.F.:CIMMYT.Gerpacio, R.V., J.D. Labios, R.V. Labios, and E.I. Diangkinay.2004. Maize Production Systems in the Philippines: Constraintsand Priorities for Research. Mexico, D.F.: CIMMYT.Gulati, A., and S. Narayanan. 2003. Subsidy Syndrome in IndianAgriculture. Oxford University Press, New Delhi.Gulati, A., and Dixon, J. 2007 Maize in Asia: Changing Marketsand Incentives. IFAD/CIMMYT/IFPRI, CIMMYT, MexicoD.F.Ha, D.T., T.D. Thao, N.T. Khiem, M.X. Trieu, R.V. Gerpacio,and P.L. Pingali. 2004. Maize Production Systems in Vietnam:Constraints and Priorities for Research. Mexico, D.F.: CIMMYT.Hartkamp, A.D., J.W. White, A. Rodriguez Aguilar, M.Banzinger, G. Srinivasan, G. Granados, and J. Crossa.2000. Maize Production Environments Revisited: A GIS-basedApproach. Mexico, D.F.: CIMMYT.Huang, J., and S. Rozelle. 2004. China’s maize economy andpolicy. In N.C. Wada, C. Fang, and A. Gulati (eds). Maizein Asia: Changing Markets and Incentives. IFAD/CIMMYT/IFPRI, CIMMYT, Mexico D.F.Huang, J., and S. Rozelle. 2002. The nature of distortions toagricultural incentives in China and implications of WTOaccession. Paper presented at the final project meeting onWTO Accession, Policy Reform, and Poverty in China,World Bank and the Development Research Center, Beijing,28-29 June 2002.ICRAF (International Center for Research on Agroforestry).1996. Annual Report for 1995. Nairobi, Kenya: ICRAF.IFPRI (International Food Policy Research Institute). 2003.2025 Projections, International Model for Policy Analysis ofAgricultural Commodities and Trade (IMPACT) Special Project:Global Trends in Food Supply and Demand. IFPRI, WashingtonD.C.IFPRI (International Food Policy Research Institute). 2002.2025 Projections, International Model for Policy Analysis ofAgricultural Commodities and Trade (IMPACT) Special Project:Global Trends in Food Supply and Demand. IFPRI, WashingtonD.C.IRRI (International Rice Research Institute). 1984. CroppingSystems in Asia: On-Farm Research and Management. IRRI, LosBanos, Laguna, Philippines.ISAAA (International Service for the Acquisition of Agri-biotechApplications). 2004. Global Adoption of Bt Maize, 1996-2003.Global Knowledge Center on Crop Biotechnology, ISAAAc/o IRRI, Los Banos, Laguna, Philippines.80

ISAAA (International Service for the Acquisition of AgribiotechApplications). 2005. Global Area of Biotech Crops,by Crop, 1996-2005. Available online at http://www.isaaa.org/kc/CBTNews/press_release/images/briefs34/figures/hectares/by%20crop_hectares.jpg.Joshi, P.K., N.P. Singh, N.N. Singh, R.V. Gerpacio, and P.L.Pingali. 2004. Maize in India: Production Systems, Constraints,and Priorities for Research. Mexico, D.F.: CIMMYT.Khiem, N.T., M.X. Trieu, T.D. Thao, and D.T. Ha. 2004. Policyanalysis of maize in Vietnam. In N.C. Wada, C. Fang, and A.Gulati (eds). Maize in Asia: Changing Markets and Incentives.IFAD/CIMMYT/IFPRI, CIMMYT, Mexico D.F.Lal, S., C. De Leon, V. K. Saxena, S. B. Singh, N. N. Singh, andS. K. Vasal. 2000. Maize stalk rot complexes: Innovativebreeding approaches. In S.K. Vasal, F. Gonzalez Ceniceros,and Fan XingMing (eds). Proceedings of the Seventh AsianRegional Maize Workshop, PCARRD, Los Baños, Philippines(February 23-27, 1998). pp. 290-306.Manton, M.J., P.M. Della-Marta, M.R. Haylock, K.J. Hennessy,N. Nicholls, L.E. Chambers, D.A. Collins, G. Daw, A. Finet,D. Gunawan, K. Inape, H. Isobe, T.S. Kestin, P. Lefale, C.H.Leyu, T. Lwin, L. Maitrepierre, N. Ouprasitwong, C.M. Page,J. Pahalad, N. Plummer, M.J. Salinger, R. Suppiah, V.L. Tran,B. Trewin, I. Tibig, and D. Yee. 2001. Trends in extreme dailyrainfall and temperature in Southeast Asia and the SouthPacific: 1961-1998. International Journal of Climatology 21:269-284.Meikle, W.G., N. Holst, and R.H. Markham. 1999. Populationsimulation model of Sitophilus zeamais (Coleoptera:Curculionidae) in grain stores in West Africa. EnvironmentalEntomology 1999, 28: 836-844.Milliman, J. D., and R. H. Meade. 1983. World-delivery of riversediment to the oceans. Journal of Geology 91:1-21.Monsanto. 2004. Conservation Tillage. (Internetaccessed information from Monsanto Company –http://212.100.230.153/reports/brochure/index.html)Monsanto. 2001. Monsanto reveals results of field trials. Agri AsiaYearender.Monsanto. 2001. YieldGard ® corn: food and feed safety.Biotechnology Insight.Morallo-Rejesus, B., and E. M. Punzalan. 2002. Mass Rearingand Field Augmentation of the Earwig, Euborollia annulata,Against Asian Corn Borer. Terminal report, Departmentof Entemology, University of the Philippines Los Baños,College, Laguna, Philippines. pp 1-19.Morris, M.L. 1998. Maize Seed Industries in Developing Countries.Boulder, Colorado: Lynne Rienner Publishers, Inc. andMexico, D.F.: CIMMYT.Narayanan, S. and A. Gulati. 2004. The corn economy of India.In N.C. Wada, C. Fang, and A. Gulati (eds.). Maize in Asia:Changing Markets and Incentives. IFAD/CIMMYT/IFPRI,CIMMYT, Mexico D.F.PAGASA – Philippine Atmospheric, Geophysical andAstronomical Services Administration. Monthly total andannual climatic data. Agham Road, Diliman, Quezon City.Pandey, S., H. Ceballos, and G. Granados. 1994. Developmentand improvement of lowland maize populations byCIMMYT’s South American regional maize program. InS.K. Vasal and S. McLean (eds.). The Lowland Tropical MaizeSubprogram. Maize Program Special Report. Mexico, D.F.:CIMMYT.Paudyal, K. 2004. Policy environment of maize productionand trade in Nepal. In N.C. Wada, C. Fang, and A. Gulati(eds.). Maize in Asia: Changing Markets and Incentives. IFAD/CIMMYT/IFPRI, CIMMYT, Mexico D.F.Paudyal, K.R., J.K. Ransom, N.P. Rajbhandari, K. Adhikari, R.V.Gerpacio, and P.L. Pingali. 2001. Maize in Nepal: ProductionSystems, Constraints, and Priorities for Research. NARC andCIMMYT, Kathmandu, Nepal.Phillips, M.J. 2001. The Future of Agricultural Biotechnology.Paper delivered at the 2001 D.W. Brooks Lecture, Collegeof Agriculture and Environmental Science, Universityof Georgia. Biotechnology Industry Organization (BIO).(October 1).Pingali, P.L. 2004. Westernization of Asian Diets and theTransformation of Food Systems: Implications for Research andPolicy. ESA Working Paper No. 04-17, Agricultural andDevelopment Economics Division, FAO of the UnitedNations, Rome, Italy. (September).Pingali, P.L. and S. Pandey. 2001. Meeting world maize needs:Technological opportunities and priorities for the publicsector. In P.L. Pingali (ed.). CIMMYT 1999-2000 World MaizeFacts and Trends. Mexico, D.F.: CIMMYT.Ransom, J.K. 2001. Factors affecting post-harvest lossesin maize. In D.N. Manandhar, J.K. Ransom, and N.P.Rajbhandari (eds.). Developing and Disseminating Technologyto Reduce Post-harvest Losses in Maize. Proceedings of aWorking Group Meeting of the Hill Maize Research Projectheld in Khumaltar, Lalitpur, Nepal on 25-27 September 2000.Nepal Agricultural Research Council (NARC) and CIMMYT,Kathmandu, Nepal.Raymundo, A.D. 2000. Downy mildew of maize in Asia: Newperspective in resistance breeding. In S.K. Vasal, F. GonzalezCeniceros, and Fan XingMing (eds.). Proceedings of theSeventh Asian Regional Maize Workshop. PCARRD, Los Baños,Philippines (February 23-27, 1998). pp. 277-283.Rosegrant, M.W., X. Cai, S. Cline, and N. Nakagawa. 2002.The role of rainfed agriculture in the future of global foodproduction. EPTD Discussion Paper No. 90 (February).Environment and Production Technology Division,International Food Policy Research Institute (IFPRI),Washington, D.C. USA.Sajjapongse, A., and K. Syers. 1995. Tangible outcomes andimpacts from ASIALAND management of sloping landnetwork. In Proceedings of the International Workshop onConservation Farming for Sloping Lands in Southeast Asia:Challenges, Opportunities and Prospects. Manila, 20-26 Nov.1994. Proceedings No. 14, Bangkok, Thailand: IBSRAM.81

Skully, D. 2001. AoA issues series: Tariff-rate quota administration.Economic Research Service, US Department of Agriculture(ERS-USDA). (http://www.ers.usda.gov/briefing/WTO/TRQ.htm)Spielman, D.J., and K. von Grebner. 2004. Public-privatepartnerships in agricultural research: An analysis of challengesfacing industry and the Consultative Group on InternationalAgricultural Research. EPTD Discussion Paper No. 113(January). Environment and Production TechnologyDivision, International Food Policy Research Institute(IFPRI), Washington, D.C. USA.Suhariyanto, K. 2004. Maize production and policy issues inIndonesia. In N.C. Wada, C. Fang, and A. Gulati (eds.). Maizein Asia: Changing Markets and Incentives. IFAD/CIMMYT/IFPRI, CIMMYT, Mexico D.F.Swastika, D.K.S., F. Kasim, K. Suhariyanto, W. Sudana, R.Hendayana, R.V. Gerpacio, and P.L. Pingali. 2004. Maize inIndonesia: Production Systems, Constraints, and Priorities forResearch. Mexico, D.F.: CIMMYT.Tseng, C.T. 2000. Use of Trichogramma ostriniae (HymenopteraTrichogrammidae), to control the Asian corn borer, Ostriniafurnacalis (Lepidoptera Pyralidae). In S.K. Vasal, F. GonzalezCeniceros, and Fan XingMing (eds.). Proceedings of theSeventh Asian Regional Maize Workshop. PCARRD, Los Baños,Philippines (February 23-27, 1998). pp. 339-356.Vasal, S.K. 1998. Hybrid Maize in Asia-Pacific. Paper presentedat the Fifth Annual Conference, Asian Seed ’98, Manila,Philippines (23-25 September).Yorobe, J.M. 2004. Economic Impact of Bt Corn in the Philippines.Paper presented at the 45 th National PAEDA Convention,12-13 October 2004, Bureau of Soils and Water Management(BSWM), Diliman, Quezon City, Philippines.Internet info: www.rainwaterharvesting.org, http://bswm.da.gov.ph/swip.html.82

AnnexesAnnex 1. Survey locations, their number by maize mega-environment and salient characteristics, Asia.China India Indonesia Nepal Philippines Thailand Vietnam Total % TotalLocations of sites Sichuan: Bazhong, Bihar: Begusarai, Lampung: Marga Tiga High hills: Cagayan Valley: Isabela Upper north: Chiang Northwest: Son La and 47 provinces/Guangan, Lezhi, Munger, and Siwan (Central Lampung) Sankhuwasabha, Southern Tagalog: Rai, Tak and Chiang Mai Vinh Phuc states/ regionsShehong, Xuanhan M. P.: Chhindwara, and Jati Agung Sindhupalchok, Mindoro Occidental Lower north: Nakorn Northeast: Ha Giang,Guangxi: Debao, Jhabua, and Mandsaur (South Lampung) and Bajhang Bicol: Camarines Sur Sawan, Phetchabun, Bac Giang and Phu ThoDuan, Longan, Rajasthan: Banswara, East Java: Kerek Mid-hills: Panchthar, Central Visayas: Cebu Kamphangphet, Pichit Red River Delta: Ha TayTiane, Wuming Bhilwara and Udaipur (Tuban) and Plemahan Nuwakot, Lamjung, Eastern Visayas: Leyte and Uthai Thani Northern central coast:Shanxi: Hongtong, U. P.: Behraich, and Kepung (Kediri) Baglung, Pyuthan, Northern Mindanao: Upper NE: Loei Thanh HoaLingshi, Lishi, Qingxu, Bulandshar, and Hardoi West Nusa Tenggara: Salyan, Dailekh, Bukidnon Lower NE: Nakorn Southern central coast:Shouyang A.P.: Mahboobnagar, Pringgabaya (East Achham, and Baitadi Southern Mindanao: Ratchaseema Quang NamShaanxi: Ankang- Karimnagar, Nizamabad Lombok) and Alas Terai/inner terai: South Cotabato Central plain: Lopburi Central highland: Dakhanbing, Luochuan, Karnataka: Chitradurga, Barat (Sumbawa) Jhapa, Udayapur, Central Mindanao: and Sra Kaew Lak, Lam Dong andShen-mu, Yanan, Ziyang Belgaum, Dharwad South Sulawesi: Klara Bara, Dang, and North Cotabato Gia LaiJilin: Changling, and Bangkala Bardiya Southeast: Dong NaiGongshuling (Jeneponto) and Mekong Delta:Shandong: Jiaxiang, Ulaweng and Soc TrangNingjing, Zhucheng Kajuara (Bone)No. of villages Two villages each from Two villages each, from Four villages each, Three village Three villages Two to three sub- One to three villages 265 villagesand provinces two to five counties two sub-districts from two districts in development (barangays) per region, districts each from each from one to two across 84per province blocks) from each each province committees (VDCs) with each village one to two districts districts from each counties/ districts(district from each district in representing a different from each province provinceTotal of 50 villages Total of 32 villages the high and mid hills maize productionacross six provinces Total of 72 villages across four provinces and two VDCs from eco-zone Total of 32 sub-districts Total of 19 villagesacross 18 districts each terai/ inner terai (tambons) across 18 across 13 provincesdistrict Total of 24 villages districts in 12 provincesacross eightTotal of 46 villages geographical regionsacross 17 districtscont’d.....83

Annex 1. Survey locations, Cont’d...No. of villagesby CIMMYTmaize megaenvironmentChina India Indonesia Nepal Philippines Thailand Vietnam Total % TotalTropical lowlands 10 36 32 10 24 32 19 163 61Tropical highlands 0 0 0 0 0 0 0 0 0Subtropical/mid-altitude 14 36 0 22 0 0 0 72 26Temperate 26 0 0 9 0 0 0 35 13Total 50 72 32 46 24 32 19 275 100% Total 18 26 12 13 9 12 7 100Salient characteristicsPredominantenvironment Rainfed upland a Upland plain Undulating upland Mid-hills Rainfed upland Rainfed upland Rainfed uplandPredominantmarketorientation Low market surplus a Semi-commercial Semi-commercial Low market surplus a Semi-commercial Commercial Semi-commercial% Maize areaplanted to MVs 80.0 a 51.0 70.0 64.0 36.0 99.7 55.5Mean farm size (ha) 0.29 a 1.91 1.03 0.78 1.00 7.30 1.09Mean maizeyield (t/ha) 3.1 a 1.4 2.9 1.7 1.6 3.7 2.5Source: IFAD-CIMMYT RRA/PRA Surveys, 2000-2001.Note: Proportions may not add up to 100% due to rounding.aAlthough sites located in temperate environments predominated in the China survey, for the purposes of the Asian synthesis, we focus only on survey sites located in tropical lowland and subtropics/midaltitudeenvironments.84

Annex 2. Matrix of RRA/PRA survey sites by maize agroecological environment and market orientation of maizeproduction.AgroecologicalMarket orientation of maize productionenvironment Low market surplus Semi-commercial CommercialIrrigated lowlandsIndonesia: Kediri and Tuban (East Java);Bone (South Sulawesi)Thailand: Pichit (Lower North)Vietnam: Ha Tay (Red River Delta);Quang Nam (Southern Central Coast);Soc Trang (Mekong Delta)Rainfed lowlands Indonesia: Tuban (East Java) India: Mahboobnagar andNepal: Dang and Bardiya (MWDR);Karimnagar (Andhra Pradesh)Bara (CDR); Jhapa and Udayapur (EDR)Indonesia: East Lombok (NTB);Philippines: Sablayan, Sta. Cruz andBone (South Sulawesi)Abra de Ilog (Mindoro Occidental);Philippines: Tumauini andTigaon (Camarines Sur); KadingilanSan Mariano (Isabela)and Talakag (Bukidnon); Tampakan(South Cotabato); Carmen, M’langThailand: Sra Kaew (Central Plain);and Tulunan (Cotabato)Chiang Rai and Tak (Upper North);Loei (Upper Northeast); Nakorn Sawan,Vietnam: Ha Giang and Phu Tho (Northeast); Phetchabun, Kamphangphet,Thanh Hoa (Northern Central Coast);Uthai Thani (Lower North);Quang Nam (Southern Central Coast);Nakorn Ratchaseema (Lower Northeast)Lam Dong (Central Highland)Vietnam: Dak Lak and Gia Lai (CentralHighland); Dong Nai (Southeast)Rainfed uplands Nepal: Lamjung (WDR) Indonesia: Tuban (East Java) Indonesia: South Lampung;Philippines: Iriga City and Buhi (Camarines Sur); East Lampung; Kediri (East Java);Philippines: San Remigio, Sibonga and Malaybalay City (Bukidnon); Sumbawa (NTB);Toledo City (Cebu); Mahaplag, Matalom T’Boli and Surallah (South Cotabato) Jeneponto (South Sulawesi)and Tabango (Leyte)Thailand: Lop Buri (Central Plain);Philippines: Benito Soliven (Isabela)China: Debao, Duan, Longan, Tiane, and Nakorn Ratchaseema (Lower Northeast);Wuming (Guangxi) Chiang Mai (Upper North) Thailand: Chiang Rai (Upper North);Loei (Upper Northeast);Vietnam: Vinh Phuc (Northwest)Phetchabun (Lower North)India: Nizamabad (Andhra Pradesh); Chitradurga,Dharwad and Belgaum (Karnataka)Vietnam: Son La (Northwest)Subtropical/mid-altitude India: Jhabua (Madhya Pradesh); India: Chindwara and Mandsaur (Madhya Pradesh); India: Munger, Siwan andBhilwara (Rajasthan) Udaipur (Rajasthan) Begusarai (Bihar); Banswara (Rajasthan);Behraich, Hardoi andNepal: Bajhang, Achham, Baitadi (FWDR); China: Ankanghanbing (Shaanxi) Bulandshar (Uttar Pradesh)Pyuthan, Salyan and Dailekh (MWDR);Baglung (WDR); Sindhupalchok andNuwakot (CDR); Sankhuwasabha andPanchthar (EDR)Source: IFAD-CIMMYT RRA/PRA Surveys 2000-2001.China: Ankanghanbing, Ziyang (Shaanxi);Bazhong, Guangan, Lezhi, Shehong,Xuanhan (Sichuan)85

Annex 3. Summary of agro-climatic features of surveyed locations classified by agroecological region andorientation of maize production, Asia.PredominantagroecologicalRange ofregion and Range annual averageproduction Elevation of annual temperatureorientation (masl) rainfall (mm) (ºC) Topography Soil types presentIrrigated lowlandCommercial 1,200 – 1,885 25 - 28 Relatively flat to flat with Sandy loam with good drainage;wet lands, plains, and meadows sandy alluvialRainfed lowlandSemi-commercial 17 - 1,500 916.5 – 2,556.3 17 – 29 Upland plain; plain undulating; Clay, loam, clay loam, sandy loam,rolling to hillysandy clay loam, volcanicCommercial 34 – 1,500 900 – 2,019 24 – 32 Upland plain; plain undulating; Clay, clay loam, sandy soil, sandyhilly; hill plains; highland plateau;medium to high slopesloam, alluvial soilRainfed uplandsLow market surplus 35 – 823 1,300 – 1,776.6 19.5 – 28 Rolling to hilly areas; mid-hills; Clay, clay loam, sandy loam,steep slopescalcareousSemi-commercial 30 – 1,350 966 – 2,028 22 – 32 Highland plateau, rolling to Clay, clay loam, sandy, sandy loamhilly areas, steep slopesCommercial 35 – 2,000 500 – 2,455 24 – 28 Plain and hill plains, rolling to Clay, clay loam, red loam, sandy soil,hilly, undulating land,alluvial soil, sandy alluvial, latosolsloping hillside, ridges,steep slopesSubtropical/mid-altitudeLow market surplus 360 – 2,500 600 – 1,546 14.7 -17.6 Mid-hills, hill Sandy, sandy loam,sandy clay, alluvialSemi-commercial 700 – 1,400 2 – 36 Plateaus, hills Clay, clay loam, sandy,sandy loam, loam, alluvialCommercial 700 – 1, 400 3 – 33 Diara belt, Doab (India) a Clay, clay loam, sandy loam,loam, alluvialSource: IFAD-CIMMYT RRA/PRA Surveys in Asia, 1999-2001.aDiara belt is a shallow riverbed that is submerged during the rainy season; crop cultivation is done only when rainwater recedes.Doab area is the land between the Ganga and the Yamuna rivers in Uttar Pradesh.86

Annex 4. Types of soil present and their farmer-reported advantages and disadvantages, across surveyedvillages, Asia.Soil type Advantages DisadvantagesClay Good/high water-holding capacity/water retention; good soil fertility; Difficult to plow; poor drainage; prone to soil erosion;less fertilizer requirement; more soil nutrients; good maize yield; wet/muddy in the rainy season causing poor yield; lessgood for contour farming and has potential to support forest trees; porous soil making it difficult to plow and lumpy in the rainygenerally suitable for growing many crops; suitable for pasture season; tend to lose moisture rapidly when there is no rain;needs proper soil fertility management practicesClay loam Suitable for growing any crop, especially cereal crops; Poor drainage; low soil fertility; difficult to plow; wet/muddygood water-holding capacity; fair soil fertility; easy to plowin the rainy season; prone to soil erosion; limited source ofmoisture for cropsClay + lateritic soil Good yield if there is no drought; can grow cassava withbig roots and good weight; easy harvestingPoor water holding capacity; requires much chemical fertilizer;drought-prone; low percentage of starch in cassava yield;difficult to plowSand No hard pan; easy to plow; minimal soil clodding; Poor water holding capacity; not suitable for maize;good drainagepoor soil fertilitySandy loam Easy to plow; good soil if there is enough rain; seeds germinate quickly; Poor water holding capacity, especially in the dry season;good drainage; suitable for paddy rice, maize, legumes, and vegetables; poor soil fertility; poor crop yield; requires much fertilizer;does not become waterloggedpoor drainage; requires organic fertilizers; susceptible to droughtSandy clay Suitable for growing maize; low soil erosion; good maize yield; Holds too much water and waterlogs during rainy season,easy for seed planting; allows rapid seed germinationmaking maize crop prone to stem rot disease; prone todrought in the dry seasonSandy clay loam Good for crop production and pasture Slightly susceptible to soil erosionLoam Easy plowing; suitable for growing maize, mungbean, and ground nut; Poor water-holding capacity in the dry season; muchallows easy harvesting of ground nutweevil; poor groundnut yield (poor weight)Source: IFAD-CIMMYT RRA/PRA Surveys 2000-2001.87

Annex 5. Maize production and poverty parameters by country and agroecological zone, Asia.Maize production (000 t) Maize area (000 ha) RRA/PRA Rural povertyaverage No. ofAgroecological % Country % Country maize rural poor % CountryGeographical region zone Production total Area total yield (t/ha ) (million) totalINDIAC&W Uttar Pradesh, Madhya Low rainfall 16.9 48.4 0.3 57.1 21.2Pradesh, RajasthanC&W Uttar Pradesh, Madhya Medium to high rainfall 2,652.1 2,535.4 1.0Pradesh, RajasthanEastern Uttar High (kharif)/medium (kharif)/ 1,515.2 881.4 1.7 61.2 22.7Pradesh and Biharspring/winter (irrigated)Karnataka High rainfall 1,363.3 391.6 3.5 22.5 8.4Karnataka Medium rainfall 473.9 138.0 3.4Karnataka Low rainfall 3.0 1.3 2.3INDONESIAJava and Bali Dryland 3,668.8 1,901.0 1.9 0.2 0.6Outer islands Dryland 2,243.2 757.8 3.0 3.9 13.3Java and Bali Irrigated 896.1 176.4 5.1 15.8 54.1Outer islands Irrigated 775.6 360.7 2.2 0.6 2.1Java and Bali Rainfed lowland 817.0 507.5 1.6 0.5 1.7Outer islands Rainfed lowland 523.3 210.2 2.5 2.1 7.2NEPALAll regions High hills 104.4 66.1 1.6 1.2 8.5Central and western Mid-hills 530.9 301.7 1.8 3.2 22.9Eastern DR Mid-hills 212.5 122.1 1.7 1.3 8.9Far and mid-western DR Mid-hills 184.1 108.3 1.7 1.8 13.1All regions Terai 335.4 175.6 1.9 6.6 46.6PHILIPPINESReg. 2-Cagayan Valley Upland plains to hilly 907.2 283.5 3.2 2.1 3.5Reg. 4-Southern Tagalog Rainfed lowland 101.1 72.2 1.4 4.9 9.8Reg. 5-Bicol Upland plains to hilly 62.8 89.7 0.7 4.3 13.7Reg. 7-Central Visayas Rolling to hilly 154.0 256.7 0.6 3.0 8.1Reg. 8-Eastern Visayas Rolling to hilly 47.5 59.4 0.8 2.6 7.1Reg. 10-Northern Mindanao Rolling to hilly 798.7 363.1 2.2 2.6 7.6Reg. 11-Southern Mindanao Upland plains to hilly 919.0 574.4 1.6 3.4 8.6Reg. 12-Central Mindanao Upland plains 919.0 459.5 2.0 1.9 5.4THAILANDUpper north Irrigated lowland 76.0 19.5 3.9 0.08 6.9Upper north Favorable upland 507.5 107.3 4.7 0.13 11.6Upper north Unfavorable upland 202.0 63.7 3.2 0.08 7.7Upper northeast Favorable upland 183.3 46.1 4.0 0.16 14.6Upper northeast Hilly 50.3 16.8 3.0 0.03 2.3Lower north Favorable upland 1,763.1 396.2 4.5 0.20 18.3Lower north Unfavorable upland 144.5 47.8 3.0 0.03 2.7Lower north Hilly 180.2 35.8 5.0 0.04 3.9Lower northeast Favorable upland 290.4 65.7 4.4 0.06 5.3Lower northeast Unfavorable upland 295.8 87.5 3.4 0.22 19.8Central Favorable upland 360.4 85.2 4.2 0.04 3.4Central Unfavorable upland 397.2 122.6 3.2 0.04 3.4VIETNAMCentral coast and highlands Lowland commercial 239.2 96.4 2.5 2.1 25.3Central coast and highlands Upland commercial 380.1 103.0 3.7 1.3 15.7North Lowland commercial 279.6 92.9 3.0 1.2 13.8North Upland semi-commercial 653.3 289.8 2.3 1.8 21.6Southeast and Mekong River Delta Lowland commercial 51.8 19.0 2.7 1.6 18.6Southeast and Mekong River Delta Upland commercial 401.9 122.8 3.3 0.4 5.0CHINASouthwest 11.2 38.2Southwest Rainfed spring maize 13,904.6 11.5 3,550.4 14.4 3.96Southwest Rainfed fall maize 1,093.6 0.9 279.2 1.1 3.60Southwest Rainfed summer maize 624.9 0.5 159.6 0.6 2.64Source: IFAD-CIMMYT country-level RRA/PRA surveys and National Maize R&D Priority-Setting Workshops, 2001-2002.88

Annex 6. List of 286 prioritized maize productivity constraints identified by Asian maize farmers.Region Ecozone Farmer-identified constraintUP,MP,Rajasthan,Bihar Medium-high rainfall/winter Post-flowering stalk rotUP,MP,Rajasthan,Bihar Medium-high/winter/spring Lack of quality seedUP,MP,Rajasthan,Bihar All ecozones Imbalanced/improper/inadequate use of fertilizersUP,MP,Rajasthan,Bihar All ecozones Postharvest losses (weevils during storage)UP,MP,Rajasthan,Bihar Medium-high/winter/spring Improper cropping systems (mixed cropping/ intercropping)EUP & Bihar Winter (irrigated) Transplanting maize under late sown conditionsEUP & Bihar Winter (irrigated) Turcicum leaf blightUP,MP,Rajasthan,Bihar Medium-high rainfall Moisture stress (drought)UP,MP,Rajasthan,Bihar Medium-high rainfall Lack of appropriate maturity varietiesOuter islands/KAP/EC367 Dryland/Unfavorable upland DroughtUP,MP,Rajasthan,Bihar All ecozones Stem borerUP,MP,Rajasthan,Bihar Medium-high rainfall/spring Maydis leaf blightKAP/Outer islands/EC3 Low-high rainfall/dryland/ unfavorable upland Poor availability of quality seedKAP/Outer islands Medium-high rainfall/dryland Weeds and weed managementOuter islands/EC3/KAP Dryland/Unfavorable upland Downy mildewKAP/Outer islands Low-high rainfall/dryland Stem borersJ+B/RP11/EC91113 Dryland/PURHM/Unfavorable upland DroughtEUP and Bihar All ecozones Inappropriate crop establishment methodJ+B/RP1011 Dryland/PURHM/rolling to hilly Soil erosion/landslidesAll Nepal/RP12/J+B/North TR/Upland plains/ RFLL/UPSC DroughtJ+B/All Nepal/RP5/North RFLL/TR/Upland plain/UPSC Lack of suitable (hybrid) varietiesRP4512/J+B/North RFLL/Upland plains/ UPSC Lack of capital/low interest credit, inadequate credit supportNorth/RP4512/J+B UPSC/RFLL/Upland plains RatsRP2/CCH/SEMK/ Outer islands/EC25810 Broadplains and hilly/ UPC/RFLL/Favorable upland DroughtAll Indonesia/SEMK/CCH Irrigated/LWC Lack of capitalRP1011/EC91113 PURHM/rolling to hilly/ Unfavorable upland Soil infertility and acidityC&W UP,MP, Rajasthan Medium-high rainfall Lack of location-specific TOT for rainfed conditions,esp for farm womenAll Indonesia/CCH/SEMK Irrigated/LWC Inefficient fertilizer use/inappropriate fertilizer applicationC&W UP, MP, Rajasthan Medium-high rainfall Banded leaf and sheath blightC&W UP, MP, Rajasthan Medium-high rainfall Broadleaf and grassy weedsC&W UP, MP, Rajasthan Medium-high rainfall Brown stripe downy mildewC&W UP, MP, Rajasthan Medium-high rainfall RatsC&W UP, MP, Rajasthan Medium-high rainfall TermitesC&W UP, MP, Rajasthan Medium-high rainfall NematodesKAP Low-high rainfall Post-flowering stalk rotAll Indonesia/EC1/North Irrigated/LWC Undeveloped irrigation system/water shortageKAP Low-high rainfall Storage pestsRP11/J+B/EC9 PURHM/Dryland/Hilly Downy mildewKAP Low-high rainfall Turcicum leaf blightRP511/J+B PURHM/Dryland WeedsKAP High rainfall Banded leaf and sheath blightAll Indonesia/North Irrigated/LWC High production costs/input pricesRP2/CCH/SEMK/ Outer islands/EC58 UP,BP&hilly/UPC/RFLL/ favorable upland Corn borers (ear/stem borers)J+B/RP45/North RFLL/Upland plain/UPSC Poor marketing system, input/output market access andundeveloped transport systemRP45/All Nepal/North RFLL/Upland plain/ TR/UPSC Inadequate post-harvest technologies/facilitiesRP1011 PURHM/rolling to hilly Limited capital/access to creditRP2/EC7/Outer islands Hilly/dryland Soil erosionJ+B/RP10 Dryland/rolling to hilly Low price of outputRP1011/EC9 Hilly/PURHM Pests and diseases (ear rot, stalk rot)J+B/RP10 Dryland/rolling to hilly Lack of post-harvest facilitiesRP1011 PURHM/rolling to hilly High price of inputs (including seed, transport)RP1011 PURHM/rolling to hilly Limited access to information and technology at farmerslevel due to poor extensionNorth/EC1/CCH/J+B/SEMK Irrigated/LWC RatsJ+B/RP512 RFLL/Upland plains Post-harvest pest and diseases (weevils)North/RP512 UPSC/Upland plains Sloping land and soil erosionJ+B/EC1113 Dryland/Unfavorable upland Inappropriate fertilizer useEC36/Outer islands Unfavorable upland/dryland Soil infertilityAll Indonesia Irrigated Waterlogging/crop establishmentAll Nepal/RP412 TR/RFLL/Upland plain Stem borers/corn borerJ+B/EC1113 Dryland/Unfavorable upland Lack of appropriate variety89

Annex 6. List of 286 ....cont’d.Region Ecozone Farmer-identified constraintAll Indonesia Irrigated Lack of laborAll Indonesia Irrigated Downy mildewRP4512 RFLL/Upland plains Soil acidity/poor soil fertilityRegion Ecozone Farmer-identified constraintRP512 Upland plains Low adoption of technologyRP4512 RFLL/Upland plains Weeds (sedges and broad leaves)RP512 Upland plains Misuse of fertilizer/low fertilizer useAll Indonesia Irrigated Leaf blight / rust (foliar diseases)EUP & Bihar Medium-high rainfall (kharif) Weed problemsEUP & Bihar High (kharif) Excess water (waterlogging)EUP & Bihar High-medium (kharif) Bacterial stalk rotRP512 Upland plains High costs of inputs (including transport, credit)CCH/SEMK/Outer islands/RP2 UPC/RFLL/BP&hilly RatsOuter islands/EC25810 RFLL/Favorable upland RustOuter islands/EC25812 RFLL/Favorable upland WaterloggingRP12/North Upland plains/UPSC Poor technology transfer systemRP11/EC9 PURHM/Hilly Poor quality of hybrid seedsCCH/SEMK/RP2 UPC/broad plains and hilly Lack of capitalEC212/CCH/SEMK/ Outer islands Favorable upland/UPC/RFLL Banded leaf and sheath blightCCH/SEMK/RP2 UPC/ broad plains and hilly Lack of post-harvest facilitiesOuter islands/EC25 RFLL/Favorable upland Soil infertilityRP2/CCH/SEMK BP, UPC FloodingRP11/EC9 PURHM/Hilly Corn borerRP11/EC9 PURHM/Hilly RatsKAP Medium rainfall Early maturing hybrids (irrigated environment)RP12/J+B RFLL/Upland plain Downy mildewOuter islands Dryland Post-harvestCCH/Outer islands/SEMK Irrigated/LWC Stem borersOuter islands Dryland Acid soilsOuter islands Dryland Purchasing powerS. Mindanao PURHM Flash floodsS. Mindanao PURHM Limited support from DA-LGUsOuter islands Dryland InfrastructureOuter islands Dryland Low priceKAP Low-medium rainfall Zn/micronutrient deficiencyRP2/Outer islands UP,BP&hilly/RFLL WeedsOuter islands Dryland Lack of laborS. Mindanao PURHM Strong windsRP5/J+B Rolling to hilly/Dryland Accessibility to input & output markets(poor farm-to-market roads)All Nepal Mid- and high-hills Declining soil fertilityS. Mindanao PURHM Poor grain qualityS. Mindanao PURHM Whorl maggotOuter islands/RP2 RFLL/UP,BP&hilly Cutworms/armywormsS. Mindanao PURHM Banded leaf and sheath blightS. Mindanao PURHM Leaf blightAll Nepal Mid- and high-hills Ear rotALL Nepal/CWDR/EDR/FMW Mid- and high-hills Lack of suitable improved varieties (HYVs, food maize, fodder,varieties for maize/millet, potato+maize system)S. Mindanao PURHM ArmywormS. Mindanao PURHM RustRP2/EC12 UP,BP&hilly/Favorable upland Stalk rotALL Nepal/CWDR/EDR/FMW Mid- and high-hills Turcicum leaf blightRP45 RFLL/Upland plain Unstable price of corn (lack of price support)Cagayan Valley BP & hilly Absence of pricing policy for input and outputRP45 RFLL/Upland plain Bad weather, flooding, siltationC. Mindanao Upland plains WeedsEC28/Outer is Favorable upland/RFLL Downy mildewALL Nepal/CWDR/FMW Mid- and high-hills Soil erosionNorth/RP4 UPSC/RFLL Insufficient technical know-how of cultural practices(pest/nutrient management, sowing distance)C. Mindanao Upland plains Limited support of DA-LGUC. Mindanao Upland plains Use of poor quality seeds90

Annex 6. List of 286 ....cont’d.Region Ecozone Farmer-identified constraintALL Nepal Terai Inadequate crop management technologiesRP45 RFLL/Upland plain Absence of strong farmers organization (coops)ALLCWDR/EDR/ FMW Mid- and high-hills Termites/white grubsCagayan Valley UP, BP & hilly AphidsALL Nepal Terai Lack of seed supplyCWDR Mid-hills Low plant populationCCH/EC1012 UPC/Favorable upland Lack of appropriate varietiesALL Nepal/CWDR/EDR Mid- and high-hills Stem borersRegion Ecozone Farmer-identified constraintC. Mindanao Upland plains Ear rotC. Mindanao Upland plains Poor grain qualityC. Mindanao Upland plains Stalk rotC. Mindanao Upland plains Whorl maggotsCWDR Mid-hills WeedsJava+Bali Dryland Low purchasing powerCCH/EC1012 UPC/Favorable upland Inefficient/inappropriate use of fertilizers & pesticidesC. Mindanao Upland plains Egg blightOuter islands/CCH RFLL/UPC Poor access to input & output markets(undeveloped infrastructure)CCH/North LWC Lack of short-duration (winter-crop) varietiesCCH/North LWC Wind, typhoons, floodsJava+Bali Dryland Lack of laborJava+Bali Dryland Post-harvest diseases and insectsALL Nepal/CWDR Mid- and high-hills Soil acidityOuter islands/SEMK RFLL/UPC Lack of laborCCH/SEMK UPC Declining soil fertilityCCH/SEMK UPC Lack of information on technology & marketsJava+Bali Dryland Foliar diseasesJava+Bali Dryland Seedling flyJava+Bali Dryland White grubBicol Region Upland plains Limited farm work/job opportunitiesCCH UPC Typhoons/westerly winds (dry, high temp winds)CCH/SEMK UPC Lack of draft powerBicol Region Upland plain Poor farming systemsJava+Bali RFLL WaterloggingEC1113 Unfavorable upland Inappropriate land preparationRP78 Rolling-hilly, uplands Ineffective financial schemeCCH/SEMK LWC FloodingJava+Bali RFLL Foliar diseasesEC81012 Favorable upland AflatoxinOuter islands Irrigated Ear rotsOuter islands Irrigated Late supply of inputsOuter islands Irrigated Shoot flyCCH/SEMK LWC Lack of information on technology and marketsOuter islands Irrigated WeedsOuter islands Irrigated ArmywormRP78 Rolling-Hilly, Uplands Limited knowledge on proper crop management(fertilization, variety use, planting density)RP78 Rolling-Hilly, Uplands Low priceRP78 Rolling-Hilly, Uplands Low soil fertilityRP78 Rolling-Hilly, Uplands Poor farm-to-market roads/limited access to market outletsCCH/SEMK LWC Lack of post-harvest facilitiesCCH/EC1 Irrigated/LWC Lack of draft powerCCH/SEMK LWC BlightCCH/SEMK LWC Ear borersRP78 Rolling-Hilly, Uplands High production and post-harvest pest incidenceRP78 Rolling-Hilly, Uplands Lack of improved post-harvest facilities/equipmentEC1/CCH Irrigated lowland/LWC InsectsEC28 Favorable upland Poor seed qualityEC1012 Favorable upland Inappropriate land preparationEC1012 Favorable upland Plant density too highCentral EC13 Unfavorable upland Insect at seedling timeCentral EC13 Unfavorable upland Plant density too high91

Annex 6. List of 286 ....cont’d.Region Ecozone Farmer-identified constraintCentral EC13 Unfavorable upland WaterloggingUpper north EC2 Favorable upland LodgingCCH UPC Lack of inputsCentral EC13 Unfavorable upland AflatoxinEC67 Unfavorable upland/hilly RustEC37 Unfavorable upland/Hilly Poor land preparationNorth LWC Lack of technology to plant maize on wet soilNorth LWC No land availableUpper north EC2 Favorable upland TermitesOuter islands RFLL Acid soilsAll Nepal/EDR Mid- and high-hills Storage grain loss (due to moths and weevils)Central EC12 Favorable upland Ear rotOuter islands RFLL WeevilsOuter islands RFLL Poor extensionRegion Ecozone Farmer-identified constraintOuter islands RFLL GrasshoppersOuter islands RFLL Wild pigsLower NE EC10 Favorable upland Lack of soil improvementS. Tagalog Rainfed lowland Insufficient water supplyUpper north EC3 Unfavorable upland Ear rotC&W UP,MP, Rajasthan Low Moisture stress (drought)ALL Nepal/EDR Mid- and high-hills Lack of tech. know-how (husbandry, seed maintenance)S. Tagalog Rainfed lowland Trader monopolyALL Nepal/EDR Mid- and high-hills Human drudgery (lack of improvement in local implements)C&W UP,MP, Rajasthan Low Inadequate availability of quality seedsS. Tagalog Rainfed lowland Sandy soilsEDR Mid-hills Labor shortage for first weedingC&W UP,MP, Rajasthan Low Imbalanced/improper use of fertilizersC&W UP,MP, Rajasthan Low Lack of early maturing varietiesALL Nepal/EDR Mid- and high-hills Loose husk coverAll Nepal/EDR Mid- and high-hills Stalk rotALL Nepal/EDR Mid- and high-hills Field cricketsC&W UP,MP, Rajasthan Low Chilo partellus (stem borer)C&W UP,MP, Rajasthan Low Lack of package for sloping & eroded landsC&W UP,MP, Rajasthan Low Post-flowering stalk rotC&W UP,MP, Rajasthan Low Improper maize-based intercropping systemC&W UP,MP, Rajasthan Low Broadleaf and grassy weedsC&W UP,MP, Rajasthan Low Brown stripe downy mildewC&W UP,MP, Rajasthan Low Lack of location-specific TOT for rainfed conditions,especially for farm womenC&W UP,MP, Rajasthan Low Maydis leaf blightC&W UP,MP, Rajasthan Low TermitesC&W UP,MP, Rajasthan Low NematodesC&W UP,MP, Rajasthan Low Weevils during storageEDR Mid-hills Silk beetlesEDR Mid-hills AphidsC&W UP,MP, Rajasthan Low Ear borer (Helicoverpa armigera)C&W UP,MP, Rajasthan Low RatsEDR Mid-hills Lack of market and good priceUpper north EC1 Irrigated lowland Poor seed qualitySEMK LWC DroughtSEMK LWC Poor soilSEMK LWC Stalk rotUpper north EC1 Irrigated lowland Lodging/strong windUpper north EC1 Irrigated lowland Post-harvest fungusUpper north EC1 Irrigated lowland ThripsALL Nepal High hills LodgingKAP Low Improper nutrient managementALL Nepal High hills Lack of post-harvest technologyALL Nepal High hills Seed not availableSouthwest Rainfed spring, summer, fall DroughtSouthwest Rainfed spring, summer, fall Low soil fertilitySouthwest Rainfed spring, summer Strong wind92

Annex 6. List of 286 ....cont’d.Region Ecozone Farmer-identified constraintSouthwest Rainfed spring, summer, fall Soil erosionSouthwest Rainfed spring, summer, fall FloodingSouthwest Rainfed spring Lack of soil micronutrients (such as Cu, Zn)Southwest Rainfed spring, summer HailSouthwest Rainfed spring, summer, fall Banded leaf and sheath blightSouthwest Rainfed spring, summer, fall Corn borerSouthwest Rainfed spring, summer, fall Turcicum and Maydis leaf blightSouthwest Rainfed spring, summer Ear rotSouthwest Rainfed spring, summer CutwormsSouthwest Rainfed spring, summer Southern rustSouthwest Rainfed spring, summer, fall ArmywormsSouthwest Rainfed spring, summer Stalk rotSouthwest Rainfed spring AphidsSouthwest Rainfed spring Head smutSouthwest Rainfed spring, summer, fall Corn silkwormSouthwest Rainfed spring, summer LocustsSouthwest Rainfed spring WeevilsSouthwest Rainfed spring, summer RodentsSouthwest Rainfed spring, summer Storage mothsSouthwest Rainfed spring, fall Cultivated varieties susceptible to insects and diseasesRegion Ecozone Farmer-identified constraintSouthwest Rainfed spring, summer, fall Poor grain qualitySouthwest Rainfed spring, fall Low seed germination rateSouthwest Rainfed spring Poor fertilizer qualitySouthwest Rainfed spring, summer, fall Lack of suitable production technologySouthwest Rainfed spring, fall OPV degradationSouthwest Rainfed spring, summer Lack of machinery suitable for hillside plotsSouthwest Rainfed spring, summer Lack of knowledge to distinguish qualityfertilizers and pesticidesSouthwest Rainfed spring LodgingSouthwest Rainfed spring, summer, fall Lack of desirable varieties (e.g., suitable for highdensity planting, early maturing, etc.)Southwest Rainfed spring, summer, fall Low level of investment in inputsSouthwest Rainfed spring Low purchasing powerSouthwest Rainfed spring, fall Problems with maize marketingSouthwest Rainfed spring, summer, fall Lack of well functioning dissemination systemSouthwest Rainfed spring, fall Low maize priceSouthwest Rainfed spring, summer High seed costSouthwest Rainfed spring, summer, fall Low per capita land – scale of production too smallSouthwest Rainfed spring, summer, fall Poor road networkSouthwest Rainfed spring, summer, fall Labor shortagesSouthwest Rainfed spring, summer Fake seed and pesticideSouthwest Rainfed spring, fall High cost of shelling machineSouthwest Rainfed spring, fall Few outlets to purchase seedSouthwest Rainfed spring Limited time period to purchase seedSouthwest Rained spring High deposit required for seed purchaseSouthwest Rainfed fall Low temperaturesSouthwest Rainfed summer Shortage of animal traction due to lack of grassSouthwest Rainfed summer High fertilizer price relative to maize priceRegions:India–Bihar, Madhya Pradesh (MP), Rajasthan, Uttar Pradesh (UP), Karnataka (K), Andhra Pradesh (AP), Central and western UP (C&W UP), Eastern UP (EUP).Indonesia–Outer islands, Java and Bali (J+B).Nepal development regions–Eastern (EDR), western (WDR), far and mid-western (FMW), central (CDR), terai.Philippines–Cagayan Valley (Region 2), Southern Tagalog (Region 4), Bicol (Region 5), Central Visayas (Region 7), Eastern Visayas (Region 8), Northern Mindanao(Region 10), Southern Mindanao (Region 11), Central Mindanao (Region 12).Thailand–Irrigated lowland Chiang Rai (EC1), rainfed upland favorable Chiang Rai and Tak (EC2), rainfed upland unfavorable Chiang Rai and Chiang Mai(EC3), irrigated lowland Phichit (EC4), rainfed upland favorable Nakorn Sawan, Phetchabun, Kamphangphet and Uthai Thani (EC5), rainfed upland unfavorablePhetchabun (EC6), Hilly land Phetchabun (EC7), rainfed upland favorable Loei (EC8), hilly land Loei (EC9), rainfed upland favorable Nakorn Ratchaseema (EC10),rainfed upland unfavorable Nakorn Ratchaseema (EC11), rainfed upland favorable Sra Kaew (EC12), rainfed upland unfavorable Lop Buri (EC13).Vietnam–Coastal and central highlands (CCH), Southeastern and Mekong River Delta (SEMK), northwest/northeast and Red River Delta (north).China–Southwest rainfed spring maize system, Southwest rainfed summer maize system, Southwest rainfed fall maize system.93

ISBN: 978-970-648-155-9Apdo. Postal 6-641, 06600 Mexico, D.F., Mexicowww.cimmyt.org