WNORLD BANK TECHNICAL PAPER NUMBER 87 WTP- 87 Aid and Agricultural Extension Evidence from the World Bank and Other Donors Michael Baxter, Roger Slade, and John Howell ~~~~~~~ .r~~~~~~~~~~~~~~~~~~~~~~. ~~~~~~~~~~~~~~~~~ -~~~~~~~~~~~~~~~~ RECENT WORLD BANK TECHNICAL PAPERS No. 50. Turtiainen and Von Pischke, Investment and Finance in Agricultural Service Cooperatives No. 51. Shuval and others, Wastewater Irrigation in Developing Countries: Health Effects and Technical Solutions No. 52. Armstrong-Wright, Urban Transit Systems: Guidelines for Examining Options (also in Spanish, 52S) No. 53. Bamberger and Hewitt, Monitoring and Evaluating Urban Development Programs: A Handbook for Program Managers and Researchers (also in French, 53F) No. 54. Bamberger and Hewitt, A Manager's Guide to "Monitoring and Evaluating Urban Development Programs: A Handbook for Program Managers and Researchers" (also in French, 53F; Spanish, 53S; and Chinese, 53C) No. 55. Technica, Ltd., Techniques for Assessing Industrial Hazards: A Manual No. 56. Silverman, Kettering, and Schmidt, Action-Planning Workshops for Development Management: Guidelines No. 5 7. Obeng and Wright, The Co-composting of Domestic Solid and Human Wastes No. 58. Levitsky and Prasad, Credit Guarantee Schemes for Small and Medium Enterprises No. 59. Sheldrick, World Nitrogen Survey No. 60. Okun and Ernst, Community Piped Water Supply Systems in Developing Countries: A Planning Manual No. 61. Gorse and Steeds, Desertification in the Sahelian and Sudanian Zones of West Africa No. 62. Goodiand and Webb, The Management of Cultural Property in World Bank-Assisted Projects: Archaeological, Historical, Religious, and Natural Unique Sites No. 63. Mould, Financial Information for Management of a Development Finance Institution: Some Guidelines No. 64. Hillel, The Efficient Use of Water in Irrigation: Principles and Practices for Improving Irrigation in Arid and Semiarid Regions No. 65. Hegstad and Newport, Management Contracts: Main Features and Design Issues No. 66F. Godin, Preparation of Land Development Projects in Urban Areas (in French) No. 67. Leach and Gowen, Household Energy Handbook: An Interim Guide and Reference Manual (also in French, 67F) No. 68. Armstrong-Wright and Thiriez, Bus Services: Reducing Costs, Raising Standards No. 69. Prevost, Corrosion Protection of Pipelines Conveying Water and Wastewater: Guidelines No. 70. Falloux and Mukendi, Desertification Control and Renewable Resource Management in the Sahelian and Sudanian Zones of West Africa (also in French, 70F) No. 71. Mahmood, Reservoir Sedimentation: Impact, Extent, and Mitigation No. 72. Jeffcoate and Saravanapavan, The Reduction and Control of Unaccounted-for Water: Working Guidelines No. 73. Palange and Zavala, Water Pollution Control: Guidelines for Project Planning and Financing No. 74. Hoban, Evaluating Traffic Capacity and Improvements to Road Geometry No. 75. Noetstaller, Small-Scale Mining: A Review of the Issues No. 76. Noetstaller, Industrial Minerals: A Technical Review (List continues on the inside back cover) WORLD B3ANK TECHNICAL PAPER NUMBER 87 Aid and Agricultural Extension Evidence from the World Bank and Other Donors Michael Baxter, Roger Slade, and John Howell The World Bank Washington, D.C. Copyright (© 1989 The International Bank for Reconstruction and Development/THE WORLD BANK 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. All rights reserved Manufactured in the United States of America First printing April 1989 Technical Papers are not formal publications of the World Bank, and are circulated to encourage discussion and comment and to communicate the results of the Bank's work quickly to the development community; citation and the use of these papers should take account of their provisional character. The findings, interpretations, and conclusions expressed in this paper are entirely those of the author(s) and should not be attributed in any manner to the World Bank, to its affiliated organizations, or to members of its Board of Executive Directors or the countries they represent. Any maps that accompany the text have been prepared solely for the convenience of readers; the designations and presentation of material in them do not imply the expression of any opinion whatsoever on the part of the World Bank, its affiliates, or its Board or member countries concerning the legal status of any country, territory, city, or area or of the authorities thereof or concerning the delimitation of its boundaries or its national affiliation. Because of the informality and ro present the results of research with the least possible delay, the typescript has not been prepared in accordance with the procedures appropriate to formal printed texts, and the World Bank accepts no responsibility for errors. The material in this publication is copyrighted. Requests for permission to reproduce portions of it should be sent to Director, Publications Department, at the address shown in the copyright notice above. The World Bank encourages dissemination of its work and will normally give permission promptly and, when the reproduction is for noncommercial purposes, without asking a fee. Permission to photocopy portions for classroom use is not required, though notification of such use having been made will be appreciated. The complete backlist of publications from the World Bank is shown in the annual Index of Publications, which contains an alphabetical title list and indexes of subjects, authors, and countries and regions; it is of value principally to libraries and institutional purchasers. The latest edition is available free of charge from the Publications Sales Unit, Department F, The World Bank, 1818 H Street, N.W., Washington, D.C. 20433, U.S.A., or from Publications, The World Bank, 66, avenue d'1ena, 75116 Paris, France. Michael Baxter is chief of the Agricultural Unit in the World Bank's Field Office in New Delhi, Roger Slade is a senior economist in the India Country Department of the World Bank, and John Howell is director of the Overseas Development Institute in London. Library of Congress Cataloging-in-Publication Data Baxter, Michael W. P. Aid and agricultural extension. (World Bank technical paper, ISSN 0253-7494 ; no 87) Bibliography: p. 1. Agricultural extension work. 2. Agricultural extension work--Economic aspects. 3. Agricultural assistance. 4. World Bank. I. Slade, Roger, 1941- . II. Howell, John, 1941- III. International Bank for Reconstruction and Development. IV. Title. V. Series. S544 1989 630'.7'15 89-5604 ISBN 0-8213-1193-X ABSTRACT The paper opens with brief examinations of the process of agricultural extension, extension as a public service and the different forms of publicly provided extension services. A more extensive discussion of the more common ways of organising extension emphasises the similarities and differences between, and strengths and weaknesses of, rural extension, commodity based extension and extension services founded in universities and those organised and managed b y Ministries of Agriculture. The World Bank is shown to have supported all types of extension. The scale of that support and the Bank's role and aims in providing it are examined at some length, drawing on the results of extensive, but simple, statistical analysis. The achievements and objectives of other donors, particularly USAID and FAO, are similarly examined although the statistical evidence is less exhaustive. Drawing on this background several important aspects of extension are identified which require careful treatment in designing or modifying extension systems. The importance of a consistent and stable agricultural policy environment is emphasised along with the need for a robust yet flexible research system, capable of responding to constantly changing farmer requiremnents, and the central need to apply proven, simple and replicable methods of organisation and management. Further drawing on work done in the World Bank as well as elsewhere some of the more important economic issues in developing, implementing, monitoring and evaluating extension projects and performance are reviewed. On the basis of earlier chapters the paper closes by drawing a series of conclusions relevant to the formulation of agricultural extension policy and operations and notes some areas where further study and analysis are needed. - iv - ACKNOWLEDGMENTS This paper has benefited greatly from help, guidance, and advice from many colleagues both within and outside the Bank. We particularly wish to acknowledge the careful attention to detail and long hours that characterized the work of Macdonald Benjamin and Ravi Venkataraman who, working as summer interns, laboriously extracted the data for the statistical annex from hundreds of Bank "black books" and to Ursula Alebrand, our research assistant, and Macdonald Benjamin who converted these data into usable tables at a later date. We are also grateful to the Agriculture and Rural Development Department in the World Bank for making John Howell's participation in writing this paper possible. Nevertheless, the authors alone remain responsible for all surviving errors of omission and commission. Table of Contents 1. THE NATURE OF AGRICULTURAL EXTENSION ................................... 1 The Process of Agricultural lExtension ...................... I" ............... 2 Agricultural Extension as a Public Service .............................................2 Approaches to Agricultural E,xtension .................................................. 4 Agricultural Extension in Historical Perspective ...................................... 6 Public Funding and Extension ................................................ 7 2. THE ORGANIZATION OF AGRICULTURAL EXTENSION .........................9 Rural Extension ................................................ 10 Commodity Extension ................................................... 11 University-based Extension .................... ............ 113 Ministry of Agriculture Extension Services ............................................ 13 Different Organisational Methods: Strengths and Weaknesses ...................... 14 3. THE WORLD BANK AND AGRICULTURAL EXTENSION ......................... 16 The Dimensions of Bank Support for Agricultural Extension ....................... 16 Trends in Bank Support for Agricultural Extension .................................. 19 4. OTHER DONORS AND AGRI'CULTURAL EXTENSION ............................ 22 Types of Support ................................................... 22 IFAD and the Regional Development Banks .......................................... 24 United States Agency for International Development (USAID) ..................... 25 The Food and Agriculture Organisation (FAO) ....................................... 28 5. THE DESIGN OF AGRICULTURAL EXTENSION SYSTEMS ...................... 32 The Policy Framework ................................................... 32 The Technological Base ................................................... 34 The Organization of Extension ................................................ 35 6. ECONOMIC ASPECTS OF AGRICULTURAL EXTENSION ....................... 39 Bank Practice ................................................... 40 Cause and Effect in Extension Project Analysis ....................................... 41 Studies in the Ex-Post Evaluation of Extension ....................................... 43 The Productivity Effects of Extension .................................................. 44 Monitoring and Evaluation in Extension ............................................... 47 Cost Recovery and the Role, of the Private Sector .................................... 49 7. CONCLUSIONS ................................................... 51 REFERENCES ......... ... 55 STATISTICAL ANNEX . ................ 59 - vi - LIST OF ABBREVIATIONS ADB Asian Development Bank CDC Commonwealth Development Corporation CFDT Compagnie Francaise pour le Developpment de Textiles FELDA Federal Land Development Authority HYV High Yielding Variety IDA International Development Association IDB Inter-American Development Bank IFAD International Fund for Agricultural Development KTDA Kenya Tea Development Authority M&E Monitoring and Evaluation ODI Overseas Development Institute Oxfam Oxford Committee for Famine Relief PVO Private Voluntary Organisation SAR Staff Appraisal Report T&V Training and Visit UN United Nations UNDP United Nations Development Programme USAHD United States Agency for International Development VEW Village Extension Worker VLW Village Level Worker CHAPTER 1 THE NATURE OF AGRICULTURAL EXTENSION The term 'extension' was first used in connection with agricultural improvement over a century ago,' and what we now call 'agricultural extension activities' pre-dates that usage by many years. In recent years, however, there has been much interest in the role of extension in agricultural development and in the organization and effectiveness of agricultural extension programs in developing countries.2 Many countries have made substantial investments in agricultural extension and many international development organizations have provided financial and technical assistance to agricultural extension in developing countries. These facts make it appropriate to take stock of recent experience with agricultural extension and to analyze its importance to developing countries. This paper draws extensively on the World Bank's experience with the design and implementation of agricultural extension projects, but it also surveys other donors experience of, and approaches to, agricultural extension. Taking stock of extension requires some care in determining precisely the nature of this recent experience in agricultural extension. Investment in extension has taken a number of different forms and the form and purpose of agricultural extension activity varies considerably even within countries. The first part of this opening chapter is concerned with clarifying the nature of agricultural extension, and the different agricultural strategies, extension approaches and methods which determine the range of activities that typically constitute the work of a public agricultural extension service. In subsequent chapters we ifrst examine the main approaches to the organization of agricultural extension. From this diversity we attempt to isolate some of the main components of effective agricultural extension and identify those components that appear to offer the greatest scope for improvement through external support. Chapter 3 examines, in some detail, the experience of the World Bank in providing assistance to extension services, and describes some of the operational issues which have become evident in the design, appraisal, monitoring and evaluation of the several different types of extension project with which the Bank has been associated. The experience of other major donors in 1 The first account of the origins of extension in Europe and the USA was by A.C. True, A History of Agricultural Extension Work in the Uniled States 1785-1923. US Dept. of Agriculture, 1929. 2 See, for example, William M. Rivera and Susan 0. Schram, Agriculture Extension Worldwide, Croom Helm, London, 1987, and Gwyn E. Jones, Investing in Rural Extension. Elsevier, London and New York, 1986. Both books are compendiums of recent work. - 2- assisting extension is then examined in Chapter 4. In Chapter 5 a summary account is given of the principal factors that should be taken into account in designing agricultural extension systems. Thereafter, in Chapter 6, we examine some of the more important economic aspects of extension. Finally, in Chapter 7, we draw on the evidence in this paper and consider some of the main outstanding difficulties and opportunities for future donor investment in agricultural extension. The Process of Agricultural Extension Farmers seek and acquire useful agricultural knowledge and develop their farm management skills in a large number of ways. The most common way is through farmer- to-farmer contact and this can be facilitated by government initiatives such as encouragement of farmer associations. Another important means of acquiring knowledge is the private agricultural supply industry. In the private sector, the provision of technical advice can be an established activity of major companies drawing, for example, upon contract growers. Management advice can be a component of agricultural financial services, such as insurance. More generally, advice may simply be the guidance provided by small local traders licensed to deal, for example, in pesticides. The emphasis upon extension as the acquisition of technical knowledge has led many professional extensionists to retreat from the view that extension can be usefully considered separately from research and from the knowledge of farmers themselves3. Few would now disagree with the proposition that extension is one part of an 'agricultural knowledge system', and it is rare to find extension projects which do not in some way involve both strengthening agricultural research and measures to involve farmers in technology development. Furthermore, if extension is seen primarily as part of a process of transferring knowledge, then the work not only of government field services but also the wide range of non-government agencies, universities and farmers associations and private suppliers must be considered.4 Agricultural Extension as a Public Service Yet, however wide the ambit of the extension process, agricultural extension as a specific publicly-provided service can be more narrowly defined. In this context, extension refers to the agencies of government which are explicidy concerned with the provision of technical advice to farmers. Of course, 'extension services', (the field services of 3 See N.G. Roling, Extension Science: Agricultural Information Systems in Agicultural Development. Cambridge University Press (forthcoming). 4 See F.A.O. Agicultural Extension: A Reference Manual. Rome, 1972 (first edition prepared by Derek Bradfield) and 1984 (much revised second edition prepared by Interpaks, University of Illinois). - 3 - ministries of agriculture or other agencies of government) are frequently involved in what would strictly be termed 'non-extension' duties such as the sale of production inputs. But 'extension' is a convenient and generally accepted term which describes the function of providing advice to farmers. It is investment in government-financed agricultural extension services with which this paper is largely concerned5 as it is the government itself which provides the most widespread and important institutional mechanism for providing advice to farmers in developing countries. Governments use varied ways of providing extension services and these different approaches reflect different objectives for extension and differences in production environments. These differences in strategy and approach are discussed below, but these also spill over into different extension methods. As so much of the literature on extension is concerned with methods,6 it is worth briefly reviewing the range of methods currently deployed. In some countries, a relatively formal style of extension is used whereby farmers are selected for instruction at Fanner Training Centers and may then be designated (for example, as model farmers or master farmers) to provide (or become the focal point of) local demonstrations and instruction. Another important extension method in some countries is the use of mass media such as radio or television programs. Mobile film and video units linked to the establishment of local groups visited by extension agents who have some knowledge of communications and adult education are another method. Another is the organization of farmer visits to other areas. Yet the staple method of agricultural extension in most developing countries remains the individual farm visit undertaken by the field staff of the Ministry of Agriculture. And it is this direct advisory work of Ministries of Agriculture, and the support services which are necessary to make such work effective, that has been the primary concern of recent World Bank involvement in agricultural extension. This, in general, is a reflection of the Bank's view that there are, in many countries, large numbers of field staff whose work can be made much more efficient, potentially high pay-offs to individualised advice in diverse farming systems and, the difficulties of using mass-media for extension unless linked to a strong field service. 5 Peter von Blankenburg, Ariculturid Extension Systems in Some African and Asian Countries, FAO Economic and Social Paper 46, Rome, 1984, identifies three different structures: government service, intervention service (mainly commodity-based) and unified development service. 6 For a wide ranging overview, see Gwyn E. Jones and Maurice Rolls, Progress in Rural Extension and Community Development. vol. 1, John Wiley, New York, 1982. -4 - Approaches to Agricultural Extension Governments adopt a number of different approaches to the organization of their field extension services. These often represent different policy objectives, although they may also represent different responses to production systems. We identify four main approaches.7 First, are those governments, particularly in East Asia and Latin America, who regard the dissemination of agricultural knowledge as only one part of a wider government involvement in changing rural attitudes and promoting community self-reliance. Normally, this tradition of extension is termed 'rural extension' and stresses human resource development alongside technology development within agriculture.8 Operationally, extension may be administered separately from the work of technical agricultural departments although the staffing and support structure of separate rural extension organizations draws upon national systems of agricultural training and research. The differences between this approach to agricultural extension and the approach based on Ministry of Agriculture field services are twofold. Firstly, rural extension involves a wider range of functions than the provision of technical advice on crop production and animal husbandry. Secondly, rural extension is often specifically targeted on poorer groups to improve their incomes, welfare or possibly access to services and assets such as land or water. Second, some Governments concentrate most of their extension effort on specific commodities providing a relatively intensive service in order to ensure, for example, quality control, adequate supplies to processing plants or regulation of planting. In such instances, the extension service is effectively an inspectorate with a range of input supply responsibilites and is also responsible for such matters as ensuring that disease control measures are taken. A third approach to extension is to give the responsibility to a university-based research system. This approach, modeled in part on the US Land Grant experience, sees extension as primarily a mechanism for carrying research results to farmers and linking farmer requirements to the research system. Scientists thus become extension workers themselves. Additional extension agents are used to provide research station scientists with a field service and are not employed direcdy by the Ministry or Department of Agriculture. 7 See Derek Byerlee, 'Extension in Post Green Revolution Agriculture', in John Howell (ed.), Agricultural Extension in Practice. ODI, London, (forthcoming). 8 See Maurice J. Rolls, Gwyn E. Jones and Christopher Garforth, 'The Dimensions of Rural Extension', in Jones (ed.), op. cit.. The fourth approach is one in which the Ministry of Agriculture establishes a professional (at least nominally) extension service whose primary purpose is to link farmers and research scientists in a two way transfer process. Such an approach often has, as a series of adjuncts, elements of the three other approaches noted above - rural extension, commodity specific extension, and university research-extension. But this system is found more in principle than in practice as the field services of Ministries of Agriculture - while often termed the 'extension services' - frequently cover a wider range of activities than simply 'extension'. Indeed, it can be argued that it is misleading to characterize this situation as a distinctive approach as there is a general absence of a distinctive philosophy. In some ministries, field staff may have some non-agricultural duties such as the organization of public works prograams or famine relief measures but such duties normally occur only in countries with either a relatively weak government infrastructure at local level, or a specific extension mandate to support measures aimed at poverty alleviation. Typically it is the 'agricultural' category of activities that provides most of the claims on an extension agent's time but few of these activities are really 'extension'. First, are the general ministry duties of data collection and reporting; second, are a range of direct production activities such as supervising school farms, seed multiplication plots and research trials; and third, are duties related to the provision to selected farmers of inputs provided by the Ministry itself, such as tractor hire, veterinary services or even seed and fertilizer. In short, the provision of technical advice to farmers may be only part (and sometimes a small one) of the duties of an agricultural field service agent. No matter what the basic approach, any extension service should be organized in response to local conditions such as administrative structure and tradition, agro-ecological and cultural factors, technology availability, and the educational levels of staff and farmers. These conditions change with time, thus requiring adjustments in the intensity and objectives of publicly-supported extension. Furthermore, within developing countries there are major differences in what farmers demand from extension. For example, there is likely to be a need for fairly rudimentary improvements in crop husbandry and soil conservation in the drier regions of Africa where the use of improved inputs is low (and which makes little sense until appropriate cultural techniques are adopted) and market opportunities are restricted. In contrast, farmers in some of the irrigated areas of Asia are now in what could be tenned a post-green revolution era where extension should focus on upgrading technical and management skills in the application of a diversity of available varieties, treatments and cropping patterns.9 9 See Robert Chambers, Setalement Schemes in Tropical Africa. R.K.P., London, 1969. - 6- Agricultural Extension in Historical Perspective The experience of some developed countries with institutionalized agricultural research and extension underscores the need to keep in mind the differences in the historical and technological context of agricultural extension in developed and developing countries. Today's agricultural extension activities in Western Europe and the United States have their origin in general agricultural promotion and education activities that date from the early nineteenth century. 10 These initiatives were largely privately sponsored, and included the creation of agricultural societies. It was largely demand from farmers and their organizations which led to increasing government involvement in agricultural technology generation and transfer towards the end of the nineteenth century, usually first in agricultural research and then in extension. Thus, government departments of agriculture in the United States and Western Europe date from the latter part of that century. Agricultural extension activities in developing countries, on the other hand, largely grew out of the rural administration of colonial governments. Agricultural officers and their staff were closely bound to the work of the administrative service and much of their work was concerned with revenue collection and enforcing regulations on such matters as soil conservation and animal health. Where they were involved in giving advice on how to improve agricultural production, extension activities normally focused on the introduction of new crops for export and this often entailed persuading producers to adopt recommended practices. Extension methods were generally imperious and regulative. Training was centered on 'progressive' farmers. The establishment of settlement schemes reinforced this style.11 Broadly speaking, the extension systems inherited by former colonies from the mid- 1960s onwards shared a number of common features. They were primarily government services, subject to a high degree of central direction (especially through 'production targets,') and hierarchical in structure. Generally they had poor linkages with agricultural research and there was little farmer participation in the design and content of extension work. As a consequence, the technologies promoted by extension services often had little relevance to traditional farming systems nor did they pay much attention to farmers' access to resources.12 10 See, for example, Michael Collinson, Farm Management in Peasant Agriculture. Westview, Boulder, 1983. 11 See Jon Moris, Managing Induced Rural Development. International Development Institute, Bloomington, 1981. 12 M. Ann Judd, et al, Investing in Agricultural Supply. Economic Growth Centre, Yale University, 1983. - 7 - The difficulty of establishing an institutional structure responsive to farmer demands was further exacerbated by post-independence developments in rural development administration.13 The agricultural field service in many countries was given a range of additional functions as governments increased their role in the rural economy. These functions included input supply anid credit disbursement as governments (often with external assistance) instigated a number of crop production programmes supported by subsidized credit and inputs. Public Funding and Extension The period after 1960 also saw an increase in national expenditures on agricultural extension in the wake of the new cereal production technologies now widely terrmed the 'green revolution'. The scale of this increase is difficult to establish precisely as there are differences in definition of extension expenditures and a variety of budget heads against which extension operations are oiften charged, but the one series of studies that has ventured in this area concludes that national investment in agricultural extension and research has roughly doubled in real terms since 1960.14 An important factor contributing to this upward trend in public investment in agricultural extension has been the impact attributed to extension services in a number of cases where significant technical advances have been made (as in the adoption of improved mnaize and wheat varieties in Mexico, the rapid expansion of fertilizer use on irrigated wheat and rice in the Punjab, and the spread of vegetative propagation methods among Kenyan tea growers). Although there were always factors other than the existence of a strong extension effort, it was generally accepted that extension had been a necessary condition of rapid agricultural development More generally, however, the case for public investment in extension has not always enjoyed the same degree of acceptance largely because of the intangible nature of the benefits of extension. Extension has an intermediary role in the generation, dissemination and adoption of technology. Thus, well-organized extension should draw farmers' attention to technological and management developments that will improve their farm profitability, and it should ensure that the services that develop technology (and those that supply other farmer services) are aware of farmer requirements. 13 For an account of then available evidence, see James K. Boyce and Robert E. Evenson, National and International Agricultural Research and Extension Programs. Agricultural Development Council, New York, 1975. 14 Evenson has been mainly responsible for this work. See for example Robert E. Evenson, 'IARC Investmnent, National Research and Extension Investment and Field Crop Productivity.' Yale University, New Haven, Connecticut, undated. (processed). - 8 - In some cases, the adoption of improved technology - such as new plant varieties - may appear to spread without the involvement of the agricultuml extension service and thus 'benefits' are difficult to attribute. However, this overlooks the general educational role of extension in facilitating adoption (and adaptation) by farmers of new technology. Even in cases where extension's role is less apparent because of a lack of technology for the service to extend, extension has a key role in ensuring that the research system is responsive to farmers and develops technology to meet their needs. Quantifying such benefits is naturally very elusive and, in any event, they usually take longer to emerge than the investment period of most projects. The World Bank has, however, accepted the general case for extension investment not least because it has recognized the probable presence of significant positive extemalities (see Chapter 6). This acceptance, however, has been accompanied by a growing interest in improving the management and organization of extension in order to improve the cost effectiveness of public extension systems. CHAPTlER 2 THE ORGANIZATION OF AGRICULTURAL EXTENSION The ways in which agricultaral extension services are organized are as diverse as the socio-economic and administrative environments in which they operate. There is, however, widespread agreement on the main principles that should underpin extension services.15 Nonetheless, although there may be agreement on principles of extension there can be disagreement on the principles of extension management (for example the degree of supervision) or over extension methods (for example, the use of farmer groups).16 There are four basic principles of effective agricultural extension: 1. Extension must provide agricultural knowledge and skills to farmers which will enable them to cope better with any difficulties they have and improve the returns to their farm investments. For low-technology agriculture this knowledge and skill is likely to be related to specific farm practices and technologies., whereas for farmers already using a range of new technologies and impiroved inputs the emphasis is more likely to be upon improving farm management and diagnostic skills. But the general principle holds that effective extension must offer agricultural advice which is useful to farmers. 'Agricultural advice' normally covers advice on crop and livestock production, although it also includes advice on access to inputs and in some circumstances it can include advice on how farmers might organize collectively to advance their farm interests. 2. Extension staff must understand the production systems within which they are offering advice, the farm businesses or household economies with which they are dealing, and the scientific basis of the knowledge they are advancing. Their training and methods of work must be organized to ensure that they can perform this enabling role both for farmers and for farmer sernices, including research. 3. Extension must be supported by a research system which produces properly tested and fully appropriate new practices and varieties which extension agents can recommend to farmers. The nature of the research system and its relationship to extension staff can vary from one based heavily on research station-based technology transfer to one that relies most 15 For a further account of 'fundamentals', see A.T. Mosher, An Introduction to AEricultural Extension. Agricultural Development Council, New York, 1978. 16 See John Howell, 'Making Agricultural Extension Effective', in Howell (ed.)., op.cit.. - 10- heavily on on-farm trials, but the principle of an effective research base remains. 4. Extension work may involve a range of methods but frequent visits to individual farms are particularly important. This direct personal contact part of extension work is essential to ensure that extension staff themselves fully appreciate the possibilities and constraints of the farm community in which they work and that advice is adapted to specific farm contexts. There is clearly enormous scope for enhancing extension and farmers' education more generally through mass communication technologies but as a component of effective extension, direct individual contact is irreplaceable. Making these principles operational needs a clear sense of purpose and a manageable set of tasks (principle 1); a properly trained work-force (principle 2); a supportive research base (principle 3); and methods of work which encourage direct contact with farmers and relay their demands to other agricultural services (principle 4). But, there is often a large gap between what is expected of extension and what, in practice, prevails in many agricultural ministries and agencies employing large numbers of extensionists. And, such gaps in performance are not necessarily related to the particular extension approach employed - as the following sections indicate. Following the categorization suggested in Chapter 1, we now examine in turn rural extension, commodity-specific extension, university-based extension, and Ministry of Agriculture field service extension. Rural Extension A common extension approach in many developing countries, at least until relatively recently, has been one based on the multi-purpose extension agent. This is primarily associated with the community development approach to rural development which was particularly well developed in India as a means of providing government support across a range of activities. Village-level workers were directed particularly to assist poorer rural groups: those generally disadvantaged in their access to land and other resources.17 Animation rurale as promoted in some Francophone countries has much in common with community development, with its aim of generally elevating rural living standards.18 The term 'rural extension' is employed for this form of extension as it gives administrative expression to the notion that advice on increasing agricultural output should not be divorced from improvements in public health, adult education, community organized labour for public works or off-farm employment and so on. Similarly, within the narrower field of 17 See Guy Hunter and Anthony F. Bottrall (eds.), Serving the Small Farmer: Policy Choices in Indian Agriculture. Croom Helm, London, 1974. 18 On the approach generally, see contributions in Bruce R. Crouch and Shankariah Chamala (eds.), Extension Education and RuMa Development. 2 Vols., John Wiley, New York, 1981. - 11 - agricultural production, the emphasiis is upon addressing multiple constraints so that the administration of input supply and credit in particular become the direct responsibility of extension workers. In most cases, the rural extension approach requires the extension agent to be responsible for the village-based programs of several, often most, government departments with rural service programs and thus the technical supervision of some of the agent's work is likely to come from officers outside the agricultural sector. As a rule, the technical content of agricultural extension under a rural extension system is modest although sometimes a mechanism for obtaining expertise elsewhere is built in. For example, in the animation rurale approach, the role of the multi- function village level animateurs is to encourage farmers to make use of the representatives of specialized extension services, which are thus one step removed from the village. The multi-function extension approach (multi-function in the sense of handling either services in addition to agriculture or services related to agricultural production in addition to extension) has advantages, especially where rural welfare services are poorly developed and there is little reliable private provision of agricultural services. The system offers governments a relatively lowv-cost and flexible administrative instrument and, in agricultural terms, there are benefits from having an extension agent whose acceptability in the community is enhanced by performing useful non-agricultural roles. Yet there are also limitations to the rural extension approach. Above all, there is generally a low and irregular level of technical (agricultural) direction given to extension agents, who often work in the poorest, most challenging and resource constrained environments, yet are expected to develop recommendations appropriate to such farming systems.19 This lack of direction can normally be attributed to the range of tasks expected of the extension agent and the (lifficulty of incorporating agents with such broad responsibilities into the work of the research departments and specialized services of the agriculture ministries. Commodity Extension Commodity-based extension approaches are generally based on the technical, administrative and commercial requirements of a specific crop. In most cases, extension is directed by a commodity authority which manages all activities from research and extension through to input (including credit) supply, and product marketing and processing. In developing countries, such authorities are generally confined to export crops, though they 19 Geoffrey Lamb and Linda Muller, _Cgtrol. Accountability and Incentives in a Successful Development Institution: The Kenya Tea Development Authority. World Bank Staff Working Paper No. 550, Washington D.C., 1982. - 12- may also be found in domestically oriented activities such as dairy production. Well- known commodity-based extension systems include the FELDA rubber schemes in Malaysia, the CFDT cotton companies in West Africa, the Kenya Tea Development Authority, and the dairy cooperatives of India. There are also commodity-based extension systems operating in the private sector such as those run by the British American Tobacco Company in Kenya.20 In both the private and public sectors the customary pattem is for crop authorities to operate on a commercial basis whereby the cost of providing extension and other services is recovered by a management charge or crop cess at the point of sale. This cost-recovery mechanism is largely assured by the monopsony powers of the authority or the company. Most commodity-based field services are successful in attaining their objective of producing and marketing a commodity effectively and efficiently. Field operations are facilitated by integration of the various elements of technology transfer, a usually limited geographic scale of operation, and the specific technical and managerial requirements of the crop concerned. Another potential strength of commodity-based extension is the degree to which growers can hold extension agents responsible for the services they provide. In KTDA for example, grower committees and individual growers have a substantial influence over the way the Authority manages its day-to-day operations.21 Yet this is not the general rule. Most commodity-based extension systems are designed in part to ensure quality control and, in some cases, sufficient throughput for the efficient utilization of processing factories. As a consequence, extension tends to be authoritarian in character and limited to the technology of a single crop even where farmers have a spread of crop and animal production activities on which their livelihoods depend.22 Furthermore, in these circumstances commodity-based extension tends to contribute to a fragmented series of farmer support services. This is not necessarily against the interests of farmers but it is frequently the case that the staffing and other resource requirements of an efficient commodity approach can detract from the general extension service for crops and regions where the commodity is not produced. 20 For an account of BAT, see Jon Moris, Extension Alternatives in Tropical African Development. ODI, London, (forthcoming). 21 Guy Mahdavi 'A Commodity-Driven Approach: the Experience of the Compagnie Francaise pour le Developpment de Textiles', in Nigel Roberts (ed.)., Agricultural Extension in Africa. The World Bank, Washington DC., (forthcoming). 22 See H. C. Saunders, (ed.)., The Cooperative Extension Service, Prentice Hall, New York, 1986. - 13- University-based Extension In the United States, where U niversity-based extension originated, it took the form of a cooperative extension service based on Land Grant colleges. Under this system, extension staff are employed by the university, and the service is normally funded under a cooperative arrangement between federal, state and local agencies. One important aspect of this approach is the close link between the research, extension and training functions; another is the general educational function of extension including, community development and youth work. The approach is a product of US agricultural cooperative history dating from the establishment of the land grant universities in 1862 which effectively distanced extension services from direct government control. These special circumstances have limited the adoption elsewhere of university- based extension on the US pattern, particularly the cooperative aspects. Nonetheless, agricultural universities have been promoted by USAID in Latin America and there is some evidence of long-term success in improving the effectiveness of the links between research and extension. In India, there has been a partial adoption of university-based extension. There, staff of the agricultural universities do not hold joint extension/research appointments, as in the US model, and field extension operations remain the responsibility of State Departments of Agriculture. Nevertheless, research is generally sensitive to extension possibilities because of its strong university focus. While components of the :Land Grant model have been established in various countries, nowhere has the model been fully adopted. Moreover, key elements in the evolution of the US extension system were not 'exported'; there has been much less farmer demand for extension (and research) in most countries compared to the historical experience of the United States. Thus extension clients have played a weaker role in extension program design; mass mnedia support for field extension operations has been lacking; technical specialist expertise has been generally deficient; and the close links between state extension operations and those of cooperatives and private firms has generally been absent Ministry of Agriculture Extension Services The bulk of agricultural extension work in developing countries is undertaken by Ministries of Agriculture. Unlike the commodity approach, the Ministry approach is designed to meet the needs of all farmers throughout the country whatever their production system, and this role is particularly strong where there is an absence of private sector involvement in agricultural inputs and advice or where farmer associations are weak. Strictly speaking it is inaccurate to describe the wide range of Ministry field services as a distinctive extension 'approach' because, unlike the other three approaches described - 14- above, there is an absence of a recognizable strategy. However, there are now a number of countries which have adopted measures to reorganize their field services based on 'Training and Visit' (T&V) principles and, taking these countries together, it is possible to distinguish an extension approach which differs from the rural extension, university extension, and commodity extension approaches. The basic principles of this approach are that the extension service should deal only with agriculture (in the sense of including all productive activities carried out on a farm); that staff at each level should have monitorable tasks and responsibilities; that farmers deal mainly with one field level agricultural extension agent; that there are effective two way links between extension and agricultural research; and that the capabilities of extension staff should be constantly upgraded.23 The management principles of training and visit extension can be applied to other extension approaches and most commodity-based systems stress such features as work scheduling, regular visiting and staff training. However, T&V extension is distinguishable in so far as it has been adopted by Ministries of Agriculture who have used the method to improve the work of large numbers of often poorly-trained staff working with farmers in very diverse production systems. Different Organisational Methods: Strengths and Weaknesses Most countries simultaneously support a mix of extension approaches and this partly reflects diverse farmer requirements and agricultural policy objectives. Large and heterogeneous countries such as India or Brazil, for example, maintain some diversity in their approaches to extension. Nonetheless, there is some utility in examining the relative performance of different approaches, as they must all be judged against the general principles of effective agricultural extension described earlier. The commodity-based approach is normally the most purposeful in conveying knowledge as there are relatively strong pressures from both farmers and the industry itself to ensure that extension advice is practical. This approach also normally has a strong research base and endeavours to maintain the calibre of its staff through training. Its weakness lies in the lack of concern with the production of crops other than the defining commodity and, as a consequence, there is a lack of understanding of farm-level constraints and opportunities which often results in pressures on farmers to misallocate land or labor resources. The rural extension approach often suffers from having too many functions. This contributes to a lack of clear purpose and tends to reduce the relevance and usefulness of 23 See Benor, Daniel, J.Q. Harrison and M. Baxter, Agricultural Extension: The Training and Visit Svstu, The World Bank, Washington D.C. 1984, and Daniel Benor and Micheal Baxter, Training and Visit Extension, The World Bank, Washington D.C., 1984. - 15- the agricultural advice it offers. T he responsiveness to village-level demand in such extension can be strong but this asset is often undermined by the poor level of technical understanding of field staff and weak links to research. The university-based approach is generally stronger on research than extension and fanner contact. There is nothing intrinsically non-consultative about such an approach although in practice outside the USA it tends to be dominated by an emphasis upon research station-based technology transfer. Ministry of Agriculture extension using T&V methods is similarly prone to a transfer of technology bias which reduces the scope for consultative extension and may weaken its responsiveness to farmers. On other counts, however, T&V extension within Ministries has the merits of simplifEying field tasks, of making provision for regular staff instruction and of reinforcing research inputs into extension work. CHAPT'ER 3 THE WORLD BANK AND AGRICULTURAL EXTENSION From modest beginnings in the late 1960s, the World Bank has become a significant supporter of agricultural extension in developing countries. Key features of this involvement in extension are presented in this chapter. The first part of the chapter describes the scope of Bank extension support in terms of both project investments and non-project activities. The second part identifies some major trends in this support. The Dimensions of Bank Support for Agricultural Extension Between 1965 and 1986, the World Bank funded 460 projects in 79 countries that supported agricultural extension mainly for crop production.24 The extension components of these projects were estimated at appraisal25 to cost $4,027 million, of which $1,807 million was covered by Bank loans (Tables 1 and 2).26 During the same period, in addition to these projects the Bank supported extension components in 25 livestock, 12 forestry and 3 fisheries projects; the cost of the extension components in these projects was $66.4 million (Table 3). On June 30, 1986, there were 244 projects with crop focused agricultural extension components underway in 70 countries: their cost, at appraisal, was $2,904 million, of which $1,260 million was covered by Bank loans (Table 4). Bank support for agricultural extension should, however, be seen in the context of all its lending for agricultural and rural development (Table 5). For example, between 1965 and 1986 slightly more than 40% of the 1,146 Bank supported agricultural and rural development projects included some funding for extension but the share of extension in total agricultural and rural development lending was only about 4.5%. Clearly, although widespread, the financing of extension constitutes a minor part of Bank lending for agriculture. 24 See the 'Note on Data' in the Statistical Annex. 25 Appraisal is that stage in the 'project cycle' when the Bank determines whether a project sufficiently satisfies the technical and economic criteria necessary to justify a loan. Hence, the costs of a project at appraisal are estimated costs. In general, actual total project costs do not vary greatly but the actual costs of individual components of projects may differ substantially from appraisal estimates. 26 The term 'Bank' includes IDA and all cost figures are in current US$ unless otherwise noted. Tables 1 and 2 and all other tables cited below are to be found in the Statistical Annex. - 17 - There are significant regional and country variations in the Bank's lending for agricultural extension (Tables 1,2,4, and 5). Although Western Africa has had the greatest number of projects with extension activities, Bank investment for agrcultural extension has been greatest in the Latin America aLnd Caribbean and South Asia Regions; together they account for 52% of all Bank lending for extension. Not only has investment in extension been skewed to particular Regions, but agricultural extension lending has been dominated by a relatively small number of borrowers. The ten countries (Table 14) with the largest Bank involvement in extension account for nearly 70% of Bank extension funds, and 34% of Bank extension projects. These countries are: Brazil ($328M of Bank investment in extension), India ($279M), Nigeria ($138M), Mexico ($133M), Indonesia ($94M), Turkey ($76M), Thailand ($65M), Bangladesh ($57M), Philippines ($50M) and Malaysia ($30M). Like agriculture and rural development projects, the number of projects supporting extension has increased with time (Table 6). From an annual average of 10 new projects with extension components in the period 1970-74 (out of 39 agricultural and rural development projects), there were more than 35 a year during the following decade (out of an annual average of 75 for agriculture and rural development). For the two years 1985 and 1986 the number of new extension projects averaged 24 per year, compared to 67 agriculture and rural development projects. Extension projects suppoited by the Bank incorporate a variety of extension methodologies. Following the classification in Chapter 2, two are identifiable as discrete methods: commodity specific and training and visit extension. In some projects, a 'modified T&V' method is used, the modification being either implicit or explicit.27 A variety of other extension methods have been utilized in Bank-supported projects, but none is as readily identifiable as these ltwo methods, and many in fact are not comprehensive systems. Some 99 Bank projects hiave supported agricultural extension without specifying in appraisal documents or working papers the extension methodology to be used. Bank lending for different extension methods is summarized in Tables 7 through 10. Of the 361 projects for which an extension method is specified, commodity approaches are used in 15%, T&V in 23%, modified T&V in 15% and 'other methods' in 47%. In terms of the cost of these extension components, 48.4% of Bank extension funds have gone to 'other methods,' 31.7% to T&V, 13.4% to modified T&V and 6.5% to commodity systems. These figures suggest that introducing the T&V method is relatively more expensive than other methods. As indicated in Table 8, ithe number of projects initiated with T&V or modified T&V extension methods is greatest in the 1975-79 and 1980-84 quinquennia. In both these 27 We have classified projects as 'modified T&V' where the method is so identified in the Staff Appraisal Report (though rarely are precise details of the modification given) or where the method is not specifically mentioned but the extension system as described follows some key features of the T&V approach. - 18- periods, however, more projects used 'other methods' or an unspecified method than used T&V methods. Commodity-based extension systems have not been as common in Bank projects started in the 1980s as in earlier years. Agricultural extension projects have been supported by the Bank in 79 of its 93 borrowing member countries. Training and visit or modified T&V extension methods were being implemented in 38 of the 70 countries that had an active extension project as of June 30, 1986, and commodity specific approaches in 15. Including those projects that were completed before June 30, 1986, the Bank has supported commodity specific approaches in 26 countries, T&V in 30, modified T&V in 29 and T&V and modified T&V together in 45 countries.Extension has generally been supported by the Bank as a relatively small component of an agricultural project (Table 1 1). In 60% of 'extension projects' extension has accounted for less than 10% of total project costs. On average, these projects had extension components that cost $4.3 million. These components accounted for 26% of all Bank funds that have been allocated to extension. In contrast, the 13 projects that have had 90% or more of total costs allocated to extension accounted for almost one-third (31%) of Bank investment in extension but only 4% of extension projects. The World Bank's involvement in agricultural extension goes beyond the preparation and implementation oversight of the individual projects that comprise the extension portfolio. This further involvement has taken two main forms: conferences designed to raise borrowers' awareness of extension issues and publications on aspects of extension methodology and evaluation. The Bank has supported, with co-sponsors, four international conferences on agricultural extension. These were at Chiangmai, Thailand (1982, on T&V extension); Denpasar, Indonesia (1984, on research-extension linkages); Eldoret, Kenya (1984, on extension methodology); and Yamoussoukro, Cote d'Ivoire (1985, on extension methodology). Proceedings of the first two have been published by the Bank.28 The Bank has also published several technical papers on extension including the seminal Agricultural Extension: The Training and Visit System (1977).29 There has been, however, relatively little analysis during the course of either project preparation or sector work that focuses on extension. Aside from postfacto examinations of project performance by the Bank's Operations Evaluation Department,30 internal analysis has been largely confined to surveys of extension strategies in India, Zambia and 28 Michael Cernea, John Coulter and John Russell (eds.), Agricultural Extension by Training and visit. The Asian Experience. 1983, and Research-Farner-Extension: A Two-way Continuum for Agricultural DevelQment. 1985. 29 Daniel Benor and James Q. Harrison, 1977; revised and reissued under the same title by Benor, Harrison and Baxter, 1984. See also D. Benor and M. Baxter, Training and Visit Extension, op. cit.. 30 Operations Evaluation Department, Agricultural Research and Extension: An Evaluation of the World Bank's Experince. The World Bank, Washington D.C., 1985. - 19- Mexico. As the number of projects without an identifiable extension methodology suggests, the attention given to extension even during project preparation can be modest. In general, there has been little systematic evaluation by the Bank of extension operations even in countries with major Bank investments in extension. A notable exception is the research study on the impact of extension in northwestern India undertaken jointly by the Bank and researchers in India (see Chapter VI). Despite its substantial support of agricultural extension via lending operations and other means, the World Bank has no official policy on agricultural extension apart, perhaps, from insistence that an extension system should be in place if the potential benefits of agricultural projects are to be attained. There is not, for example, an agricultural extension policy paper, as there is f:or agricultural research, forestry, rural credit and other agricultural sub-sectors. The absence of a comprehensive policy and a standard approach to extension investments no doubt contributes to the variety of extension projects supported by the Bank and to the differences in levels of input into, and quality of, project preparation and supervision. In some projects, extension principles, objectives and methodologies are spelled out in detail, and Bank staff give considerable attention to reviewing these aspects of project implementation. It is nolt unusual for there to be significant changes in extension methodology or project scope during project implementation: in some respects this reflects the importance of flexibility in implementing extension projects, but also, it often derives from the lack of attention paid to extension during project preparation, particularly when the investment in extension is a small part of total project costs. Trends in Bank Support for Agricultural Extension A number of distinct trends in Bank support for agricultural extension can be discerned from the experience of the past two decades. One is the increasing level of support for extension: from an annual average of less than six new projects incorporating agricultural extension during 1965-1974 to more than 33 annually over the following twelve years. Some evidence of this build-up was given in the preceding section and further details are to be found in the statistical annex. It is enough to note here that while there may not have been an explicit Bank extension policy, there has been sufficient consensus on the need to eliminate the chronic inefficiencies present in many extension services, Ministries of Agriculture and agricultural research institutions, to contribute to a steady expansion of support for extension. This has allowed the Bank to learn continuously and to improve its ability to appraise interventions and to be innovative. Another trend has been the change in the nature of the Bank's support for extension. Extension in early projects was usually a component of commodity development or region-specific agricultural or integrated rural development projects. In many cases these extension investments were not linked, in design or implementation, to national agricultural extension policies or strategies; often there were parallel agricultural - 20 - research components with overlapping objectives and design. More recently, however, it has become common to integrate support for extension into the national research and extension frameworks and to support extension development on a national scale. Where sub-national components are supported, there is now recognition that extension organizations or methodologies should not be changed if the changes cannot be replicated nationally for financial or administrative reasons. Another change in the Bank's support for extension has been the increase in the number of projects introducing the training and visit extension method. Since the mid- 1970s when this method was first articulated, the Bank has supported 84 projects with this approach and another 54 that use it in a modified form. A significant concomitant of the Bank's growing interest in agricultural extension methodology and systems is an increased awareness of the need for a sound administrative and technical basis for extension. The principle that an extension system must be administratively sustainable and technologically innovative seems better understood. Interest in the administrative structures supporting extension has also focused more attention on the capital and recurrent costs of extension. There has been an increasing acceptance of the need for agricultural extension project preparation to include detailed assessment of technological constraints to farming and the availability and development of technology. This, in turn, has led to more careful assessment of the costs involved and their justification. Another trend in Bank support for agricultural extension has been the growing acknowledgement that such support is part of a broader process of institutional and educational development. Extension projects with which the Bank has been involved have increasingly recognized the links between research, education and agricultural extension. Similarly, more attention is now paid to ways of improving the planning and implementation activities of the agricultural administrations within which extension services operate. Lastly, Bank involvement in agricultural extension now tends to take a long-tern view. This derives from the recognition that meaningful institutional development is usually a slow process. The usual five-year project period is often to short to complete extension improvements because of the time normally taken to institute the reforms associated with establishing an effective extension system and the need to allow time for the system itself to evolve. Recognition of the long time it takes to build up effective extension services has led to longer project investment periods, follow-on projects and extension 'sector' programmes. There have also been changes in the way extension projects are supported during implementation. For example, intensive field-based support teams have been established in the New Delhi, Nairobi and Abidjan World Bank Resident Missions, each staffed in part by experienced local officers. These teams assist local administrations - 21 - implement extension projects ancd, in some cases, projects concerned with related agricultural support services. CHAPTER 4 OTHER DONORS AND AGRICULTURAL EXTENSION There is no satisfactory method of providing useful comparative data on levels and forms of donor investment in agricultural extension. There is no standard definition of expenditure on extension and, in any event, few donors isolate extension as a discrete item of investment. Some comparisons are made later in the chapter but they relate largely to estimates of expenditure on extension gathered from a small number of donors and accept spending on such items as 'technology transfer' as a proxy for extension. Types of Support A wide range of international agencies are involved in agricultural extension, and they are involved in a variety of ways. In spending terms, the main form of support is financial aid, on grant or concessionary loan terms, for agricultural projects which contain extension components. These often support the development of a particular technological package (as in the case of British aid to the Hindustan Fertilizer Corporation in Northeastern India) or production system (as in the multi-donor Gezira Rehabilitation Project in the Sudan). It is rare for donors, other than the World Bank and USAID (see below), to be involved in substantial projects with extension as the primary component. Like research projects with an extension component, donor-assisted 'extension projects' tend to be modest in size and in most cases targeted on parts of the extension service rather than its overall operation. For example, UK aid for extension in East and Central Africa has been largely for the establishment of extension communication aids (e.g., the Agricultural Information Centre in Kenya and the Extension Aids Branch in Malawi). German and Canadian aid for extension in developing countries has often been concentrated on building Farmer Training Centres, encouraging demonstration farms and training 'progressive' farmers. Particular technical services within extension have also been subject to donor support. For example, several bilateral donors have collaborated with FAO in extension projects in East Africa to promote disease and pest control measures (for East Coast Fever and Armyworm, particularly). Another example is USAID assistance for farm-level production and storage of improved seed for local distribution in the hill farming areas of Nepal. - 23 - In most of these extension projects, technical assistance as well as financial aid has been provided. Manpower aid for extension provided by bilateral donors tends to concentrate upon specialist technical or training inputs rather than the supply of extension supervisors or managers. The major exceptions are the various commodity schemes for African export crops (see below) where some expatriate management continues. In the past USAID provided extension management in its Latin American programmes. The most common form of technical assistance in extension is in training. The UK, Holland and USA provide awards for extension training in their own countries, as they each have internationally-recognized extension training and research centers. Such training is for extension managers and trainers rather than extension agents, although the UK and Holland, among others, also providle some in-country extension training directed at staff with field-level responsibilities. Apart from official aid agencies, local and external non-governmental agencies (private voluntary organisations: PVOs) are also involved in extension. There are a number of large PVOs which operate like official donors and play an important role in extension work. For example, the Aga Kharn Foundation's Rural Support Programme in Northern Pakistan has effectively substituted for the government's own extension service in the remoter districts. In Zaire, much of the agricultural extension effort is externally financed and implemented through PVOs. In general, however, PVOs narrowly target their interventions which are almost always supplementary to the main govemment extension effort. International (as opposed to indigenous) PVOs are normally engaged in financing local initiatives of a community development nature rather than agricultural extension in the technical sense; but there are instances where PVOs (such as Oxfam) provide staff for extension work. Such direct intervention tends to occur in areas of relative government neglect. Forestry and soil conservation extension in the drier parts of Kenya and Burkina Faso, and livestock extension in the Sudan, for example, have a substantial international PVO involvement. Another important category of donor investment in agricultural extension is where government funds are provided for overseas commercial investment (not necessarily on concessional terms and thus not official development assistance) by publicly-owned companies or corporations. The UK's Commonwealth Development Corporation (CDC) and France's Compagnie Francaise pour le Developpement des Fibres Textiles (CFDT) are cases where investment in smallholder export crop production has been made conditional - in effect, on the establishment of commodity extension systems which integrate technical advice with input supply and regulation functions at the field level.31 There are 'aid' 31 Guy Mahdavi, "A Commodity Driven Approach: the Experience of the Compagnie Francaise pour le Developpment de Textiles", in Roberts (ed.)., op cit.. - 24- elements in such investments, including the provision of technical assistance and, in Africa at least, it would be accurate to categorize the main British and French impact on extension as commodity-specific. Considering this range of external aid for agricultural extension as a whole, a number of points emerge. First, the forms of extension which receive support, and the mechanisms of support, are often different in character even within particular countries. Second, it is very difficult to isolate 'extension' as a discrete component in most projects, and donors do not normally attempt to do this. There have been very few attempts to formally evaluate extension projects (exceptions include the Indo-German and Indo-British Fertilizer Projects mentioned above which attribute higher levels of fertilizer use in north- eastern India to enhanced extension coverage).32 Similarly, projects (or parts of projects) aimed at improving training have rarely been evaluated: donors do not think it helpful to isolate 'extension' training aid from wider training support to Ministries of Agriculture to assist in the development of the technical capabilities needed to support a field extension service. Third, the issue of 'donor co-ordination' within extension does not appear to be as contentious an issue to governments as it sometimes is in other areas of agricultural aid. Governments appear to accept a varied pattern of extension assistance involving the private and voluntary sectors as well as the public sector. They also seem to allow some degree of local experimentation often based upon externally-assisted training and research projects within their own extension schemes. In the rest of this chapter we examine available evidence on flows of external financial resources for extension investment from multilateral donors other than the the World Bank. We then consider the record of the largest bilateral donor for agricultural extension - USAID - which is also the donor with the most explicit set of policies towards extension. Finally, we examine the resources and objectives of FAO which is probably the major single source of technical assistance for extension projects and which maintains the largest extension advisory establishment of any donor or international development agency. IFAD and the Regional Development Banks33 The emphasis within IFAD's mandate upon services to small farmers has meant that the bulk of its project loans include some extension or extension-related component. Up to September 1986, the IFAD Executive Board had approved 188 projects; of these, 150 had components of agricultural extension, research or farmer training, or combinations of two or more of these activities. These 150 projects entailed loans of $1,740M out of $2,149M 32 National Council of Applied Economic Research., Impt of Indo-German Fertilizer Education Project, New Delhi, 1979. 33 IFAD: The International Fund for Agricultural Development. - 25 - in total loans approved by IFAD. Figures on commitments to agricultural extension within the projects are not available. As Table 4.1 indicates, Africa is particularly important in the IFAD portfolio of extension lending.. The size of individual [FAD project investments for extension, research and training ranges from $7,000 in Dominica to $203M in Turkey. IFAD's data on the number of intended beneficiaries in its extension projects shows a range from 425 in Djibouti and 1,000 in the Gambia to 1.2M and 1.7M in Turkey and Kenya respectively. IFAD extension, research or training projects are usually financed in conjunction with other donors. Only about 25% of such projects is financed exclusively by IFAD. Table 4.1 IFAD Projects Approved to September 1986 Number of Projects Latin Near East/ Africa ,4sia America North Africa Total AR projects 67 50 37 34 188 Projects with extension 57 37 27 29 150 components [FAD has not taken a position in favour of any particular extension system or methodology. Provided that the FLnd's basic target of serving smaller farmers is met, any system is supported. The Asian Development Bank (ADB) and the Inter-American Development Bank (IDB) treat agricultural extension in dissimilar ways. ADB does not give explicit attention to extension and is not active in lending for agricultural extension. It believes that the World Bank is adequately handling the subject in Asia and so does not see a need to make additional investments. IDB, on the other hand, provides funds for extension although it does not make extensive professional inputs into country strategy, or project design and evaluation. Between 1980 and 1986, IDB funds were used in 65 'integrated agricultural' projects that involved some agricultural extension. Five were primarily extension projects with total project cost averaging about $30M. United States Agency for International Development (USAID) Of the development agencies involved with agricultural extension in developing countries, USAID has probably given the most thought to the objectives and strategy of its - 26- involvement. This introspection and the agency's long period of involvement with extension in a wide range of countries, make the USAID experience particularly relevant to this review. Through its predecessor the International Co-operation Administration, USAID became involved in extension support in the early 1950s. One of its early activities was to place extension agents from the United States in developing countries where they worked as local extension agents. While sometimes successful, this approach generally revealed that US agricultural technology was not directly transferable and showed that to be effective extension agents require a close familiarity with the local culture and community and a strong institutional base. In response to these lessons, USAID altered its strategy and began to support the formation of ministries of agriculture and national extension systems. To this end, from the early 1960s attention turned toward promoting the US Land Grant model of agricultural extension, research and education. The Land Grant focus lasted into the 1970s but its limited impact (see Chapter 2) and criticism of USAID's support of national extension systems34 resulted in the early 1970s in a shift of USAID attention towards the support of extension within area-based integrated agricultural development projects. At the same time, extension support (often in conjunction with agricultural research) was also directed to commodity-specific projects. In these projects, extension' activities were normally set up parallel to the national extension system for a single commodity or group of related commodities. As with area-based projects, these commodity projects were successful in the short-term due to the increased resources and attention they provided but little survived their eventual reamalgamation into national programmes. Over the last decade, USAID involvement with agricultural extension has been characterized by an increased questioning of the priority to be accorded to strengthening extension delivery systems compared to developing farmer-relevant technology. USAID evaluation studies have also raised questions concerning the adequacy of farm policy and rural infrastructure. These findings stimulated discussion of the merits of strengthening extension and adaptive research programmes along the lines of farming systems research - which, expanded to farming systems research/extension, was seen by some as a substitute for extension. Adaptive research and extension projects based on a farming systems model are now a major component of USAID support to extension. USAID has a much more explicit extension policy than other donors, and it has also issued detailed guidelines to its field offices. In 1985 a circular on extension noted that the 34 E.B.Rice, Extension in the Andes: An Evaluation of Official US Assistance to Agricultural Extension Services in Central and South America. USAID, Washington D.C., 1971. - 27 - results of the Agency's efforts in technology transfer had been disappointing35. The circular implied that this was attributable to a dependence on approaches that tended to rely on extensive cadres and a hierarchy cf public sector extension personnel. Another was that 'a simple transfer of the U.S. public sector extension model, or other developed country model, has serious limitations in many developing country situations.' To supplement and improve the work of public sector extension operations, USAID proposed a strategy to develop private sector extension, to intensify the use of mass media communications, and to selectively increase the effectiveness of public sector extension. This was to be done by taking some or all of the following steps. (a) Relieving extension agents of non-production responsibilities (e.g., credit administration, marketing, tax collection and regulatory enforcement). (b) Strengthening training, technical backup and mobility of extension workers; (c) Focusing research and extension more sharply on improved technologies for priority crops and animals; (d) Increasing interaction between extension workers and researchers, including joint participation in on-farm research and demonstration and feedback to research (this is often accomplished under the aegis of farming systems research projects); (e) Supporting farmer training programmes (including literacy, numeracy, basic science and farm management training) and vocational agriculture programmes; (f) Closely coordinating public sector extension with mass media approaches and private sector activities; (g) Improving attention to the needs of low-resource farmers and women by increasing the number of female extension agents and developing appropriate techniques to reach these farrners; and (h) Development of mcore rigorous planning and setting of objectives and methodology for the public extension service. The circular concludes by saying that 'The net effect of implementing these reforms could well be a smaller (emphasis) but better qualified (emphasis) public sector extension service.' Despite these shifts in focus, there has been much consistency in USAID's interest in extension. It has, for example, been continually concerned with increasing the output and incomes of small farmers. It has tried to avoid building up large central extension 35 USAID, -AID Support for Agricultural Extension " (unclassified), Department of State, Washington D.C., 1985. - 28 - organizations that would have little impact on production. USAID has also consistently emphasized the pitfalls in a bureaucratic, top-down approach to extension, the need for close links between research and extension, and the need to tailor support to local conditions. However, during the 1970s and 1980s, the scale of USAID support for extension has decreased. This is partly in response to the general difficulty of assessing extension impact, but it is also a response to the increasing involvement of the World Bank in supporting national agricultural extension systems. USAID initiated 1,065 projects between 1975 and 1984 which involved agricultural extension, though in only 266 of these was agricultural extension a major component. These 266 projects accounted for about $50M p.a. or 7% of the Agriculture, Rural Development and Nutrition budget between 1980 and 1985. However, the share going to agricultural extension is impossible to determine.36 These projects were largely implemented in single countries and 64% were with national government organizations. A further 14% were with PVOs. Over half of the projects involved technical training of extension staff and farmers, and a quarter contained technology demonstrations at universities and institutes. Sixty-five of the 266 projects incorporated what USAID has termed 'innovative' approaches, primarily involving improved research/extension interaction (including farming systems research). There have also been women farmers' programmes, and extension through private companies has been supported in a few cases. The Food and Agriculture Organisation (FAO) The impact of FAO on agricultural extension has been primarily through its technical assistance programme rather than direct financial support. This programme is fmanced in two ways. The 'Regular Programme' activities of FAO in extension are planned and organized by FAO Headquarters and Regional Office staff. They concentrate on providing technical advisory assistance to member countries, formulating project proposals, developing publications and training materials, and conducting workshops and consultations on extension. A number of technical guides have been produced for extension staff and trainers. There are twelve headquarters-based staff responsible for agricultural education and extension and training (in FAO's Agricultural Education and Extension Service), and four in the regional offices; other headquarters divisions are also involved in training and communications support related to extension activities. The latter formulated 34 projects (total cost $43.8M) in 1986 that mainly provided technical assistance.37 36 In comparison, the World Bank supported 356 agricultural extension projects between 1975 and 1984, with an average annual commitment of new funds for extension of $135.4M. 37 This and subsequent information on FAO has been drawn from "Notes on FAO's Involvement and Experience in Agricultural Extension", Agricultural Education and Extension Service, Human Resources, Institutions and Agrarian Reform Division, FAO, March 1987 (processed). - 29 - The largest part of FAO direct support for agricultural extension comes through its Field Programme. Extra-budgetary funds (from the UNDP and country Trust Funds in particular) finance the programme for agricultural extension which is reported for several years in Table 4.2. As that Table shows, 1092 (44%) of the 2,506 FAO field projects in 1986 had an extension and/or training component. FAO also provides limited funds for field projects through its Technical Cooperation Programme. This funding does not exceed $0.4M per project and is confined to selected, small-scale extension and/or training initiatives. Table 4.2: FAO Extension Field Projects, 1981-86 FAO Field Projects FAO Field Projects with 25% of Costs Going to Extension Extension With With Total Experts & Total Extension Training Project Extension Year Consultants Projects Component Component Budget Budget No. No. No. (%) No. (%) $M $M (%) 1986 178 2,506 171 (6.8) 921 (36.8) 143.9 57.9 (40.3) 1985 163 2,604 167 (6.4) 810 (31.1) 146.4 57.4 (39.2) 1984 159 2,534 138 (5.4) 801 (31.6) 140.8 54.6 (38.8) 1983 138 2,450 135 (5.5) 796 (32.5) 136.0 51.5 (37.8) 1982 122 2,429 131 (5.4) n.a. .. 84.3 36.7 (43.5) 1981 144 n.a. 126 *- n.a. .. 75.1 32.9 (43.8) Agricultural extension as defined by FAO is a 'service or system which assists farm people, through educational processes or procedures, to improve farm methods and techniques, increase production efficiency and income, and improve the quality of life by lifting the social and economic standards of rural populations.' Thus, has FAO given a high priority to promoting a comprehensive, educational co.icept of extension. This emphasis on a broad-based approach to extension also reflects the particular long- term concerns of FAO in land and tenancy reform and rural development generally which were reiterated at the 1979 World Conference on Agrarian Reform and Rural Development. Within its broad extension mandate, three specific aspects are currently stressed by FAO. The first is that extension should include problem identification and analysis of technology adoption constraints rather than simply disseminate technology or train farmers. The second is institution-building designed to improve the formal and non-formal education of trainers (mainly extension stafiF) and farmers. The third aspect is the links between agricultural research and extension. Given these concerns, FAO's emphasis on the inter- - 30- relations between agricultural education and training, extension and research is understandable. FAO also emphasizes its catalytic role in extension support, especially in terms of its ability to train trainers. In 1984 its 801 field projects with an extension or training component provided training for 65,000 participants at the professional, technical, vocational and producer levels. Between 1979 and 1984, at least 350,000 people were trained by FAO field projects. Beyond training, FAO feels it can best serve member countries by strengthening their national capabilities in extension planning, manpower development and programme design. Thus, assistance is aimed at establishing national and unified agricultural extension systems, often with an extension programme planning unit (responsible for planning, coordinating and monitoring extension activities, and for training staff in these fields) as an integral and vital component of the system. FAO believes that its extension and training staff and consultants should always work with national counterparts - supported by formal training opportunities - in order to develop a trained, national extension cadre. Another feature of FAO extension policy is that national extension systems should deliberately seek to work with smaller and 'harder-to-reach' farmers. There is also a strong emphasis upon working with women and youth. In terms of the extension methods it supports, FAO is catholic. It has provided technical assistance to 82 countries and has supported a range of extension approaches, which it categorizes as T&V, commodity-specific, animation rurale, Land Grant/university, participatory, general agricultural advisory, area development, cooperative (cost-sharing) and audio-visual communication orientation. FAO's position on the promotion of particular extension methods 'is flexible and as a technical agency of the UN prefers to present alternative extension approaches to member countries and assist them in adapting a chosen approach and set of methodologies to the specific local conditions and capabilities of each country.' FAO does, however, believe that 'the participatory extension approach' has the greatest potential in reaching large numbers of male and female farmers, and favors the 'cooperative funding' of extension with central, regional and local participation. In reviewing its experience with agricultural extension, FAO has identified nine requirements for sound and effective extension. They can be summarized as follows. (i) Policy Commitment. A strong and specific policy is required providing a clear direction to, and authority for, the establishment and operation of agricultural extension and training activities as an integral part of institutionalized agricultural development programmes. (ii) Adeguate Infrastructure. There should be adequate logistical and management support for extension staff, who should also have clear and realistic work programmes and supervision. - 31 - (iii) Appropriate Extension Approaches. The extension approach used should be responsive to local country and client needs, and may be a combination of approaches. (iv) Extension Programme and Messaae Relevance. Extension pro- grammes and messages must be relevant to the problems and needs of farmers. Functional and operational links between research, extension and farmers are needed for this end, as is attention to farming systems research- related practices. (v) Media Support for Exte nsion. Multi-media extension methods should be an integral part of the extension programme. (vi) Appropriate Extension and Training Materials. These are required by field staff and trainers. (vii) Trained Agricultural Personnel. Formal and informal training programmes are needed to upgrade the quality of extension staff. (viii) Strategies for Reaching Women Farmers. Specific extension and training programmes are needed for women farmers, even where they are covered by extension programunes more generally. (ix) Evaluation for Extension Programme Improvement. Changes in farmers' knowledge, attitudes and practices as a result of extension should be measured and the results utilized for programme planning. As regards its future work, FAO has a number of new priorities. One is the strengthening of national extension services by increasing the attention given to assisting educational institutes to provide pre-service extension training, and to improving the management, monitoring and supervision of extension programmes. The development of extension methods that take advantage of progress in communications technology is another area of priority, and one that FAO believes closely fits its brief as a technical assistance agency. Finally, FAO hopes to give priority to improving its own ability to act as an informnation clearing house on extension approaches and methodologies and to increasing coordination among international agencies involved with agricultural extension in developing countries. CHAPTER 5 THE DESIGN OF AGRICULTURAL EXTENSION SYSTEMS The preceding two chapters describe the evolution and nature of World Bank support for extension and the approaches of other agencies that support agricultural extension. Although there are differences in approach, there are also similarities; together they suggest a number of important issues that require careful reflection during the planning of investments in extension. The Policy Framework38 In developing effective research and extension systems, a basic concern is the national policy framework in which they operate. Policy should indicate national agricultural development priorities; outline the organizational structures necessary to implement these priorities and the corresponding institutional linkages; and the extent and nature of the commitment to encouraging farmers to act in a manner supportive of national policy. An important issue in national agricultural development policy is the priority given to agricultural research and extension. The need to support research, particularly basic and applied research, is usually not disputed. However, the case for agricultural extension is not always so easily made. Reasons for this include a belief that researchers are responsible for disseminating their results or, more commonly, that applicable, useful research results will be disseminated independently of institutional support because of their innate attractiveness to farmers. This notion also reflects the difficulty of effectively organizing and managing an extension system and attributing to extension an impact on production (see Chapter 6), as well as confusion over different approaches to extension. Coupled with limited resource availability, these factors help explain why the role of extension in agricultural development is not usually made explicit in national agricultural policy statements. National agricultural development policy should indicate the contribution of research and extension to agricultural development and to a country's particular agricultural development strategy, and should show how the two activities are interrelated in implementation and institutional arrangements. It is not for national policy statements to detail the modes of extension or research, but basic parameters should be established. For 38 This section draws extensively on M. Baxter and W. Thalwitz, in Cemea, Coulter and Russell (eds.)., 1985., op.cit.. - 33 - extension, these might include its professional and technical orientation, basic work responsibilities and management principles, criteria by which its effectiveness should be monitored, ways in which extension contributes to national policy formulation and implementation, and the basic institutional arrangements linking it with other developmental services, such as agricultural research. Even where a comprehensive policy for agricultural research and extension exists, its translation from the national level to the field is not easy. Not only are there many institutions and levels of administration, but farmers frequently have priorities different from those of national governments. Moreover, in contrast to policy at the national level, research and extension activities become increasingly locally oriented as they move into the field, particularly in their response to farmers' production conditions and needs. Indeed, a national policy is likely to be only as successful as the degree to which it coincides with the interests of (and incentives received by) farmers. A national policy is unlikely to find acceptance with producers unless t]here are appropriate economic incentives for them to follow it. For example, without adequate prices for their crops farmers are unlikely to support a national policy of self-sufficiency in food production. Although policy makers must take a longer and broader view of priorities than farmers, national policy cannot be oblivious to farming conditions and farmers' practices and priorities; ways must be developed to take account of them - including the use of farmer-oriented research and extension services. Just as national policy on agricultural development should be supported by realistic incentives to farmers, so should adequate resources be deployed for its implementation. The absolute level of resource allocation is significant, but so also are the timing of investments (expenditure) and regional priorities. Investment in extension rarely has an immediate, directly attributable impact, a factor that can lead to frequent changes in organization and strategies and to uncertain financial support. To tailor and adapt an extension system to a specific set oif environmental, economic, or administrative conditions often requires many years. Consistent long-term policy and financial support is an imperative. These considerations highlight the need for any investment in extension to be framed in the light of particular national agricultural policy and economic circumstances. Although seemingly obvious, this injunction is all too often ignored. Systems of extension are introduced without adequate attention being given to the policy and resource environment with the consequence that adequate commitment remains elusive. Equally serious, however, is the unquestioning acceptance of a given policy environment. This can lead to extension services being created or expanded and becoming desirably efficient and professional yet left to operate in a policy environment which meets few of the conditions noted above. The reform and development of extension brings with it an obligation not only to address the technical issues of good extension practice and organization but also a responsibility to address complementary policy and institutional issues. - 34 - The Technological Base The World Bank's experience with extension shows that effective extension must have a solid technological basis. The basic functions of agricultural extension are to advise farmers on how to increase their production and income and, to advise those who operate other agricultural support services how to improve the support they offer farmers. To do this, extension agents must offer advice on crops and activities that are relevant to farmers. If such information is not readily available, there is little justification for significant expenditures on extension. This does not mean, however, that there should be extension only where a significant stock of technology is available. There is sufficient variation in farming practices and standards in most farming communities for extension to work even when significant 'improved' technology is not available. Moreover, in such situations, extension has a major responsibility to ensure that research services, as well as other agricultural services, remain aware of farmers' problems and work towards relevant solutions. Just as extension cannot operate for long without strong agricultural research, neither can farmer-related research survive for long without extension. While the need for a technological base for extension might seem a truism, it is more frequently than not poorly handled in the design of extension systems. It is often assumed that there is a backlog of 'messages' that most farmers do not know (or that they do not understand them); that the adoption of information is relatively simple and largely cost-free; and that an extension service is required to simply diffuse messages. There is rarely a detailed analysis of current farmer practices (particularly analyses specific to different resource conditions) and of the suitability of messages. Where such analyses have been done, it has been shown that few 'messages' of immediate relevance to farmers are available. As much as other organizations, the Bank's experience with extension is marked by projects that have had limited success because of a poor technological base, for example, the Sine Saloum Project in Senegal and the Thaba Bosiu Rural Development Project in Lesotho. Often this is less the result of poor applied research in areas such as plant breeding and more a result of what Byerlee calls the poverty of adaptive research.39 The latter is normally considered to be the final stage (before farm level demonstrations) of the research continuum - aiming to take the fruits of applied research and evolve good or better ways of managing new or changed technologies at the farm level. Adaptive research, partly because responsibility for it is often diffused, is frequently poorly managed and is generally the weakest, most neglected and most confused aspect of national research systems.40 The consequence is that extension is required to disseminate information that is often irrelevant 39 See Derek Byerlee in John Howell (ed.)., forthcoming, op. cit.. 40 World Bank, AgdcultuW Research, Sector Policy Paper, World Bank, Washington, D.C., 1981. - 35 - (e.g., in Nigeria during the 1970s) or uneconomic, excessively packaged (that is, farmers are encouraged to adopt a complete 'package' of interrelated technologies even though farmers rarely adopt changes in this way)41 or is insufficiently adapted to local circumstances. This last point is a weakness commonly found in India. These weaknesses are a main factor behind the development of farming systems research and the growing recognition that farmers themselves are also a good source of invention and problem solving. However, as yet research systems have not been modified to create the localized structures in which farming systems research can flourish. While this is now a growing focus of World Bank efforts, progress has been and is likely to be slow. Plans and projects for extension, must explicitly recognize these problems and avoid over-ambitious expectations. This done, attention can turn to the matter of the appropriate organizational form of extension. The Organization of Extension The World Bank's involvement with agricultural extension, like that of other organizations, suggests that there is no one extension system suited to all conditions. Rather there is a standard set of procedures that should be used to determine the type of extension service for a particular location. There are two main steps to be undertaken to this end, in each of which there must be a high level of participation by staff of the existing extension service and department of agriculture. The first step is to understand the current system of extension (since there are few places without one) and the local administrative, cultural and agricultural conditions since they have an important bearing on the objectives and implementation of extension operations. Rural settlement patterns and cultural dynamics (whether, for example, there are traditional groups that may be utilized for farrner training and information diffusion), the range and seasonality of agricultural activity, the distribution of factors of production, the orientation and effectiveness of agricultural research and other agricultural services, the availability of useful technical messages for farmers, the numbers and skills oi staff available for extension, as well as government and farmers' financial resources, can all have significant impact on the design of an effective extension organization. If such factors are properly taken into account, extension activities are likely to be consistent with the technical and operational realities of the countries in which they take place. Once the organizational structure has been settled, the design of an extension project should turn to objectives and operating principles. Here, extension's role relative to other agricultural services should be established and its basic orientation determined. From these objectives, relevant principles of extension organization and operation should be 41 See Bruce R. Crouch and Shankariah Chamela (eds.)., 1981., op. cit.. - 36- elaborated. For example, the basic principles of an effective technology - based extension service normally suggest that extension staff of all levels should have relevant, feasible and monitorable work programs and, in particular, the work programs of field staff should be known to farmers. Their work should be focused almost entirely on information dissemination and improved understanding of farm practices and problems in their territory. Furthermore, the product of extension is knowledge and its focus should be a concern for the way in which agricultural services can enhance farmers' understanding of new and relevant ideas. Thus, extension staff should receive frequent and regular training, be supported by technical specialists, and be in frequent two-way contact with agricultural research and other farmer services. Finally, as extension's client is the farmer, extension activities should take place primarily at the farmers' place of work and should cover all aspects of a farmer's production activities. Farm visits should be the chief extension method, supplemented where possible by other means of communication. Once the technical and organizational aspects of extension operations have been determined and the objectives and principles of extension established, the design of an extension system is relatively straightforward. Ignoring the basic principles of extension operations can, however, be perilous. For example, an extension service burdened with the responsibility for input distribution and sales, general data collection or regulatory programs, clearly has limited time to devote to its two-way knowledge transfer and education function. Similarly, an extension service that rarely meets farmers or operates to a schedule or with a purpose unknown to farmers has little chance of understanding what farmers really need from agricultural services, and so cannot usefully advise farmers, let alone advise other services on how to improve their support to farmers. It was mainly this concern with objectives and organizational principles that led the World Bank to advocate the training and visit method for organizing and managing extension. As noted elsewhere in this paper, the T&V method is an approach in which basic principles of management are applied to agricultural extension. Thus, the methods of extension used by a "T&V system" (none of which are new or unique to T&V) are not as important as the way in which they are managed. The T&V method focusses on management because management is often the least satisfactory aspect of an agricultural extension system. The simplicity of the management principles embodied in T&V are appealing but widely assumed to be rigid and inflexible. This is incorrect. Although the basic principles of T&V have remained essentially unchanged, there have been numerous changes in its method of operation. For example, in some countries (e.g., Turkey, Nepal), T&V was first applied using paid farmers as the final stage in formal technology diffusion - they were paid to teach other farmers what they had learned. This system proved to be unworkable even in the short-term. Moreover, where appropriate technology is available, this practice has been found to be unnecessary since extension workers in regular contact with many farmers are able to ensure the widespread dissemination of extension's recommendations. - 37 - Experience has also shown the importance of frequent working sessions involving both agricultural research and extension staff in order to discuss farmers' technical requirements and to develop suitable extension messages. In the early years of T&V such meetings were biannual or seasonal but this proved to be too infrequent to ensure that productive pressure was maintained on both the research and extension services; a monthly interval has proven the most effectiive. Another change in T&V has been the emphasis given to contact farmers as the link between the extension worker and farmers at large. In that an extension worker cannot, and need not anyway, have direct contact with all farmers, some system must be developed to ensure that he does meet farmers representing the main local economic and production conditions in a regular and monitorable fashion. In communities with strong individualistic traditions, individuals may be used; where there is a tradition of group activity, formal or informal farmers groups may be a more appropriate contact medium. Another way in which T&V operations have evolved is the increasing attention given to the training of staff and to developing the role of mass media to complement field extension activities. The Bank's experience with T&V extension methods highlights a number of design issues requiring careful attention. One is the need to isolate in extension project design and implementation the principles of extension operation from the means by which the principles are implemented. For example, a principle of T&V is that farmers should know the extension worker responsible to them and his work program. Whether extension agents' visits are weekly or to some other schedule is not as important as the fact that they take place in accordance with a regular schedule appropriate to the farming conditions of the area and which is known to all iarmers. Similarly, farmer:extension agent ratios in themselves are not immutable - they should represent a balance between the needs of farmers, communications, staff availability (including cost considerations), and the possibility of mass media complementing extension field activities; clearly, they should change with time to reflect changes in these considerations. A second difficult design issue is the establishment of the two- way function of extension. On the one hand, this means that farmers must be actively involved in the design of agricultural extension activities and that their preferences must be a major consideration in designing the topics to be covered by extension. On the other hand, it means that extension must take an active stance in advising all agricultural services of farmers' conditions and needs, and in helping these services design programs that will meet those needs. This role is increasingly recognized in relation to agricultural research, but extension should also work with services such as input suppliers (including banks and irrigation systems) and marketing organizations to help improve their responsiveness to farmers. Through its provision of regular staff training and work review sessions, T&V attempts to create a venue where such contact can take place on a regular, sustained basis. A third issue for design is to accept that strengthening extension on any scale and especially on the scale often implied by T&V entails significant institutional reform and - 38 - manpower development. The World Bank's experience with T&V has shown that this can be costly. Hence the Bank now focuses considerable attention on the resource implications of extension development at an early stage in the design process. From the foregoing it follows that the design of effective systems of extension is not simply a narrow matter of ensuring a simple one way flow of technology to farmers. It is much more than that. In essence it requires first that there be a clear articulation of national agricultural policy from which the basic strategy of extension and its relationship to agricultural research can be deduced. Once this has been done a formal commitment by government to efficient extension must follow and, with that, will come the requisite resources. These resources must then be deployed and managed in a manner which ensures that extension not only delivers in a systematic and monitorable way the fruits of research to farmers but also, by accurately reflecting farmers concerns and difficulties, helps to ensure the relevance of new agricultural research. That the resulting system may be expensive (need extensive resources - both human and material) is obvious but this is of little account if the return to such expenditures is, as it should be, substantially greater than the cost. CHAPT ER 6 ECONOMIC ASPECTS OF AGRICULTURAL EXTENSION This chapter first considers rnethods of economic analysis and their utility in both extension project appraisal and evaluation studies of the productivity effects of agricultural extension. It then discusses the role of monitoring and evaluation work in extension projects. Finally, the chapter briefly discusses the separate issue of cost recovery and the transfer of the costs of extension to private sector alternatives. The World Bank's concern with the economics of extension began, in effect, with the advent of projects that were devoted exclusively or almost so to extension development. Hitherto, because extension was merely a component in larger and usually wider-ranging projects (e.g., rural development) the economics of extension were ignored or the costs and benefits were subsumed within the calculations of rates of return for the larger project. The basic economic issue can be put very simply, but hypothetically. If in a country without an extension service of any kind a decision is to be made on whether such a service is worthwhile it would be necessary to first identify the full cost of establishing such a service and then to identify (estimate) the benefits that are expected to result. Once these two streams of costs and benefits have been defined using the conventional rubric of benefit-cost analysis42 it is a simple matter to calculate, for some accounting rate of interest (opportunity cost of capital), the economic or internal rate of return. In principle this would allow decision-makers to assess the utility (value to society) of the investment and, comparing it with alternatives (other projects), make a judgement as to whether it is worthwhile. Such a situation is never found in practice. All countries with a significant agricultural sector have an extension service; and most extension projects as defined and financed by the World Bank involve modifications (usually additions) to existing publicly- financed services. Hence, the basic issue in practice involves clearly defining the real choices at appraisal. If (as is typically the case) the only decision to be made is whether to 42 No attempt is made here to define this term further, to examine the underlying economic theory or to describe the derivation of the efficiency prices used in such analyses. For such information the reader is referred to J. Price Gittinger, Economic Analysis of Agricultural Projects, Johns Hopkins, Baltimore, 1982, for the simplest exposition and to Ian Little and James Mirlees, Proiect Appraisal and Planning for Developing Countries, Basic Books, New York, 1984, Edward J. Mishan, Cost Benefit Analysis, Allen and Unwin, London, 1975, or Lyn Squire and Herman G. van der Tak, Economic Analysis of Projects. Johns Hopkins, Baltimore, 1975, for more comrplex explanations. - 40 - approve or deny an expansion, reorganization or intensification of an existing extension organization, then the concept of benefits must refer to the additions (with time) to the net value of farm output over and above those expected in the absence of the project. The relevant costs are those required to bring about the reorganization or intensification (i.e., incremental costs). To the extent that the project calls for government employees not presently involved in extension, to be reassigned to extension work, the cost of these employees should be attributed to the project. Thus, the primary concern is the estimation of incremental costs and benefits. Only in the very unlikely situation (owing to bureaucratic rigidities) of a government considering the abolition of its existing structure would it be necessary for a cost-benefit analysis to take into account both the costs of the existing extension system and the incremental costs and the benefits due to that system plus those from the proposed project.43 Bank Practice The essentially conventional approaches outline above are not often followed in practice. Among recent pure extension projects, most (thirteen of seventeen) do not calculate a rate of return in this way. Five do not calculate one at all. Why is this so? The answer is that although costs can be readily identified, calculating benefits presents a serious problem. Some projects simply regard the benefits as being too intangible to permit reasonable and realistic quantification while others (the majority) argue as follows: "Economic Benefits. Attributing a precise level of benefits to this type of project is difficult since it is not possible to determine what proportion of benefits are due to extension alone and what are due to additional purchased inputs and other factors. In practice, it is the combination of a number of factors, with extension playing the role of catalyst, that brings desired benefits. It is also difficult to estimate acceptance rates of recommended practices. However, since the project relies primarily on reorganization and strengthening of an existing extension structure the incremental cost is low per hectare and per farm family. Hence, even very small and slow production increases in the project area would generate a high rate of return. This project would be able to generate a 50% rate of return if, by 1988, yields of foodgrains (which cover 75% of cropped land area) increase by only 1.3%. Per hectare yields of paddy and millets would have to increase by 65 kg/ha and 22 kg/ha, respectively." (World Bank Staff Appraisal Report, Tamil Nadu (India) Extension Project. March 1981). 43 In this situation it is theoretically possible for the incremental project to be viable but the overall benefits to extension to be insufficient to justify the overall costs. If so, the 'first best' solution would be to discontinue all extension activities, while the 'second best' solution, if scrapping extension altogether is not feasible, would be to approve the incremental project. - 41 - Such arguments amount to an appeal to plausibility and, by taking a minimalist approach, suggest that extension may have modest effects. Their main weakness, however, is that they fail to fully think through the chain of cause and effect as it operates in extension and hence find a practical method for estimating the benefits or show more strongly than they do why positive and sufficient economic benefits are plausible.44 Cause and Effect in Extension Project Analysis Ex-post, in contrast to the ex-ante situation of an appraisal, demonstrating that extension has a positive effect on output requires that a clear cause and effect relationship be demonstrated. The following five conditions must be satisfied. First, it must be shown that the extension service is well organized, adequately staffed, mobile and properly trained. Second, that the extension agents reach the majority of farmers. Third, that farmers learn of innovations from contact with extension agents and that farmers understand these innovations and find them useful.45 Fourth, the cause and effect relationship requires that farmers apply the knowledge they have acquired. And, fifth, as a result, output must be demonstrably higher. It is in establishing these last two steps that severe identification problems arise owing to the presence of many other variables which add to, or subtract from, output. It is not possible to establish such cause and effect relations ex-ante, of course, but the argument still holds. By aclequately providing human and other resources the possibility of delivering information is reasonably assured. At appraisal, it must then be assumed that knowledge is effectively transferred and freely disseminated. Next, however, explicit attention should be given to estimating realistic trajectories for the incremental adoption profiles. That is, the rate that improved husbandry practices are likely to be taken up.46 Such a process makes transparent the thinking underlying any attempt to attribute production increases ex-ante. The latter, however, may still not be possible owing to the reasons outlined above. Even giving explicit attention to adoption profiles is complex and precise quantification difficult. Consider the following. 44 In this process more general problems of the correct efficiency prices to use in profect appraisal may arise, as well as the way in which benefits are distributed between consumers and producers. These problems, however, are not uniquely fouind in extension projects and are extensively discussed in the more general texts on agricultural project analysis. 45 Up to this point the chain of cause and effect is uncomplicated by large numbers of exogenous influences, except the presence of other knowledge disseminating media. 46 This necessarily requires a rigorous examination of the existing and foreseeable technologies that are likely to suit the present farming system. The World Bank SAR for The Zimambwe Extension and Research Project provides an example of where this was done. - 42 - When there are only one or two major crops and very few elements in the technology to be diffused, (i) the relation between adoption and increases in output, and (ii) the calculation of the incremental adoption profiles which might justify the project, may both be discernable and feasible. But when there are several crops and many technological elements (not all of which need to be adopted simultaneously), the calculations become intractable. In a project financed by the World Bank in Haryana, India, for example, wheat, rice, sugarcane, cotton, pearl millet, sorghum and pulses are all major crops. The technical package for each one of these includes more than a dozen recommendations, some of which are composite recommendations involving more than one element. For most of the crops some of the recommendations were known to at least some farmers before the project through the pre-project extension system. Moreover, some diffusion would probably have taken place even if the extension system had not been intensified. In such a case it is not feasible to estimate incremental adoption profiles for each practice for each crop. Nor would it be possible to relate them to production increases in the absence of a detailed production function specifying the relevant interactions. Hence, unless the crop - technology situation is very simple, the estimation of incremental adoption profiles and hence production increases is infeasible.47 Therefore, an appraisal should contain a detailed discussion of the technology which extension is expected to promote, as well as an indication of the technologies currently under development and likely to be available for diffusion in the short and medium run. The incremental output increases for major crops which would be sufficient for project justification can then be specified and an assessment made of the feasibility of realizing such gains considering the available and expected technology. However, a target minimum incremental adoption profile can be specified only if a major component of a technology package for a major crop can be isolated. Of course, the amount of guesswork involved in defining the "without project" diffusion path could be reduced if experience in similar regions elsewhere has been documented. But this is almost never so. In any case, the importance of such a "minimum incremental adoption profile" should not be over-emphasized, since it ignores possible gains due to other components of the technology. In many cases, the adoption of more than one element of the technology may need to be speeded up sufficiently to justify the investment, but the calculations to show this may not be tractable. If, then, these problems are, to a large degree, intractable ex-ante are they less so ex-post? Experience of evaluating extension both within and outside the Bank in the 1970s suggests that they are. 47 Nevertheless, some indication of the effects of adoption on farm production and profitability can be derived from farm budget analysis, or better still, linear programming models, for representative farms. - 43 - Studies in the Ex-Post Evaluation of Extension Extension is heavily concerned with the adoption of technology on which there is a voluminous literature. From a survey of this literature48 it is possible to conclude that studies dealing with the evaluation of extension are a distinct sub-set. Many of these studies, also extensive, have recently been summarized.49 Those who have attempted to evaluate extension have done so from a variety of perspectives with mixed results. Some, usually aggregative studies50, recognizing the dependence of extension on agricultural research, combine the two and have shown a positive relationship between them and agricultural productivity. Other studies have concentrated on estimating the effects of extension by comparing farmers who have contact with extension agents with those who do not.51 These evaluations underestimate the effects by not taking into account the way in which information, once acquired by one farmer, can be passed on to others. Other studies have concentrated on assessing extension performance by measuring the extent of farmer-extension agent interactions.52 Other studies have examined the extent to which extension agents visits to farmers are biased in favor of the rich and influential.53 Some evaluators see the internal efficiency of the extension system as the crucial parameter and study farmer to agent ratios and ithe quality and motivation of extension agents,54 or the 48 Gershon Feder, Richard E. Just and David Zilberman, Adoption of Agricultural Innovation in Developing Countries: A Survey, Staff Working Paper No. 542, The World Bank, Washington D.C., 1982. 49 Francois Orivel, 'The Impact of Agricultural Extension: A Review of the Literature' in Hilary Perraton, D. Jamison, J. Jenkins, F. Orivel and L. Wolff (eds.)., Basic Education and Agricultural Extension, Staff Working Paper No. 564, The World Bank, Washington D.C., 1983. 50 See for example, Robert E. Evenson and D. Jha, 'The Contribution of the Agricultural Research System to Agricultural Production in India', Indian Joumal of Anricultural Economics. Vol. 28, No. 4, 1973. 51 For example, B.R. Harker, 'The Contribution of Schooling to Agriculture and Modernization: An Empirical Analysis', in P. Foster and RJ. Sheffield (eds.)., Education and Rural Development, Evans Bros, London, 1973. 52 For example, D. Giltrow and J. Potts, 'Agricultural Communication: The Role of the Media in Extension Financing', The British Council, London, 1978, and Robert Chambers and M. Wickremanayake, 'Agricultural Extension: Myth, Reality and Challenge', in B.M.Farmer (ed.)., Green Revolution? Technology Change in Rice Growing Areas of Tamil Nadu and Sri Lanka. MacMillan, London, 1977. 53 See D. Leonard, 'Why do Kenya's Agricultural Extension Services Favor the Rich Farmer'. Paper read at 16th Annual Meeting of the African Studies Association, Syracuse, New York, 1977. 54 Bernhard Hoeper, 'Selected Results of an ADO and VEW Survey in Jind, Karnal and Mahendragarh, Districts of Haryana, India', Institute of Socio-Economics and Agricultural Development, Working Note No. 1, Berlin, 1983, (processed). - 44 - location and mobility of extension workers.55 In the context of exploring the relationship between education and farmer productivity, some investigators have sought to establish whether extension is a substitute for, or a complement to, education.56 Even when they yield unambiguous results, such studies shed little quantitative light on the net benefits of extension. To do so requires not only rigorous comparative analysis but a formal means of establishing whether, if positive effects are observed, the effects are commensurate with the costs incurred to produce them. Such studies have been undertaken in developed countries, of which Griliches' (1958) study of hybrid corn in the USA is perhaps the most well known.57 Studies in developing countries have usually been less rigorous, evaluating extension through a simple before and after comparison of crop yields.58 Such studies, however, do not identify and hence evaluate, the contribution of extension to increases in output because they do not separate out the contributions of factors such as material inputs, soil quality, supply constraints and other variables likely to influence output. Nor do they take into account the possibility that, with time, there may be autonomous growth in productivity. These weaknesses, together with the difficulties noted in earlier sections, led the World Bank to fund a substantial research project designed to estimate directly the productivity effects of incremental investment in extension. This project was undertaken between 1981 and 1984 in Haryana and Uttar Pradesh, India, in conjunction with the Haryana Agricultural University. The Productivity Effects of Extension The research project noted above followed closely and in a comparative manner the chain of cause and effect mentioned earlier. It was conducted in two agro-climatically similar and geographically adjacent areas -- Karnal district and Kairana tehsil in north west India -- with different extension systems.59 55 S.A. Rahim, 'The Comilla Programme in East Pakistan', in C.M. Wharton (ed.)., Subsistence Agriculture and Economic Development Aldene, Chicago, 1966. 56 M. Lockheed, D. Jamison and L. Lau., 'Farmer Education and Farm Efficiency: A. Survey', Economic Development and Cultural Change, Vol 29, No. 1,1980. 57 Zvi Griliches 'Research Costs and Social Returns: Hybrid Corn and Related Innovations', Journal of Political Economy. Vol. 66, 1958. 58 See for example Daniel Benor and James.Q. Harrison, op cit., and U. Lele, The Design of Rural Development Lessons from Africa. Johns Hopkins, Baltimore, 1975. 59 For a full account see Gershon Feder, L. Lau and R. Slade., The Imnact of Agricultural Extension. Staff Working Paper No. 756, The World Bank, Washington D.C., 1986. The rest of this section draws heavily on this work. - 45 - In the state of Uttar Pradesh, in which Kairana is located, the extension system at the time of the study consisted of the traditional network of village level workers (VLW). These workers provided extension advice, regulated the supply of inputs and credit, and provided links to the rural population for several government agencies. In the adjacent state of Haryana, of which Karnal is a part, the extension system was changed to the T&V system from late 1979. The reorganization reduced the ratio of villages to VLW and created higher level extension positions (e.g., supervisors and subject matter specialists). Village level workers were relieved of non-extension duties and renamed village extension workers (VEW). The main thrust of the study was an analysis of productivity differentials between the two areas for two crops - HYV wheat and rice - and the extent to which any estimated differences could be attributed to the introduction of T&V extension. A more precise definition of a productivity differential will be helpful. To compare the output obtained by a farmer in Karnal with that of a farmer in Kairana, any differences in soil types, farmer characteristics, irrigation variables and the production environment must be controlled. Any remaining difference in output between the two farmers when they apply the same quantities of physical production inputs is called the disembodied productivity differential. It does not depend on the level of physical production inputs; rather, it involves better utilization and/or timing of inputs, the adoption of better practices and more timely responses to production problems. Thus, if extension increases farmers' knowledge about improved cropping methods, it will increase the disembodied productivity differential. In contrast, extension information can increase crop yields by expanding the use of physical production inputs. The resulting change in output is called the embodied productivity differential. Using farm level survey data and econometric estimation procedures, production and supply functions were fitted which explicitly incorporated, and hence controlled for, a number of variables which might cause productivity differences. The resulting estimates showed that in 1982/83, after three years of T&V extension, HYV wheat yields in Karnal were about 9% higher than in Kairana. This estimate of the disembodied differential was compared to a combined, but separate, estimate of the disembodied and embodied differential plus price effects of about 13%. Owing, however, to an inability to separately estimate the price effects the additional difference of about 4% was ignored. However, sufficient additional analysis was performed to suggest that the embodied differential was small. The results for rice although similar were not statistically significant. The rest of the analysis, therefore, focused only on wheat. The disembodied differential of 9% was, however, gross of any differential between the two areas that existed before the more intensive T&V system was introduced. Thus the next step in the analysis was to estimate the initial or baseline differential. In the - 46 - absence of comparable farm level survey data for the immediate pre-project period, secondary data were utilized to estimate the productivity differential in HYV wheat yields between the two areas in the baseline year (1979/80). This established that there was a baseline productivity differential of between 2 and 3% in favor of Karnal which should be subtracted from the post-project productivity differential. Thus, the end result was that after three years of T&V extension, there was a gain in productivity for HYV wheat of between six and seven percent which could be attributed to T&V extension. The final step in the evaluation required that the value of the 6.3% increase in farm output be set against the additional costs incurred to make the additional output possible. The stream of incremental extension costs was constructed using data on the actual costs of the first four years of T&V extension in Karnal and projections made at the time of project appraisal. The increase in yield attributable to T&V extension was estimated for the third year of the project. In the absence of data with which to estimate future extension-induced gains, a dynamic model to simulate the future evolution of the change in productivity, both with and without T&V extension was constructed. The model assumed that in the absence of T&V extension, the average yield grows at a constant rate and that after a few years the initial impetus to productivity resulting from the introduction of T&V slows down. The final results showed that there was a 90% probability that the actual rate of return to the incremental investment in T&V extension was at least 15% over the project life60, and at least 18% over an efficient project life. This account of the research shows clearly that even the ex-post evaluation of extension, although possible, is difficult and requires special conditions. It also indicates that a full and rigorous evaluation of extension's impact on productivity would require a comparative approach along two dimensions of 'before and after' the project and 'with and without' the project. Rarely is this possible for either practical or administrative reasons. Sometimes, however, it is possible to work along only one dimension, most commonly studies of the situation in the area where the project is undertaken both before and after (as well as during) implementation. Even this somewhat inadequate approach is not possible in areas where monitoring and evaluation work is not done prior to the project and subsequent implementation covers the entire country or state. In these situations evaluation must remain, as it were, within a single cell of the matrix - the situation in the area of the project during and after implementation. Such a restricted form of evaluation, common in many project situations, is unable to yield definite answers about the effects of extension. 60 Calculations were performed for hree assumed lives: 7,10 and 17 years. - 47 - Monitoring and Evaluation in Extension Parallel with the emergence of free standing extension projects and the concern with how to evaluate them (both ex-ante and ex-post) has come the realization that monitoring and evaluation work could also play a useful role. At an early stage, and based on anticipated Indian experience with the implementation of T&V extension, Cernea and Tepping6l developed a system for monitoring and evaluating agricultural extension projects. This system, novel in many ways, was not seriously implemented because it was later found to be too complex and did not offer sufficiently detailed guidance to practitioners on the ground. Subsequently, and as part of the research into the productivity effects of extension (see above) Slade and Feder62 prepared a manual on the monitoring and evaluation of T&V extension. This document defined more clearly than its predecessor the nature of M&E in extension and the role that it should play. It also provides detailed step-by-step guidelines on how to plan and conduct monitoring and evaluation work. More recently, in the context of introducing T&V extension into Kenya another system was developed by Marchant.63 Although similar in style and objectives to the work of Slade and Feder, Marchant's approach places the study of adoption rates at center stage. There are also less formally defined methods that have been introduced elsewhere such as in Thailand.64 These various systems, now being applied in countries as diverse as Somalia and Sri Lanka, all tend to emphasize monitoring rather than evaluation. Some Governments, most notably India, and the World Bank have recognized the importance of reliable information in implementing effective extension and have increased the priority accorded to M&E. There are now active and reasonably well staffed and equipped Monitoring and Evaluation Units in fourteen of the seventeen states in India where extension has been reformed along T&V lines. These units have produced over 300 reports and added greatly to the evidence of what extension is achieving and how it does so. There are indications from Pakistan, Sri Lanka, Kenya, Somalia and Zambia that a similar process is under way. Nevertheless, there are still considerable shortcomings in many places. 61 Michael Cemea and B. J. Tepping, A System for Monitoring and Evaluating Agricultural Extension Projecs, Staff Working Paper No. 272, The World Bank, Washington D.C., 1977. 62 Roger Slade and Gershon Feder., 'The Monitoring and Evaluation of Training and Visit Extension: A Manual of Instruction', The World Bank, Washington, D.C., 1981, (processed). 63 Josette Murphy and Tim Marchant, Monitoring and Evaluation in Extension Agencies, Technical Paper No. 79., The World Bank, Washington D.C., 1988. 64 See for example Adisak Sreensunpagit, 'Monitoring and Evaluation of Extension: Experience in Thailand', in M. Cernea, J. Coulter and J. Russell (eds.)., op.cit.. - 48 - Monitoring and evaluation is a technical activity requiring highly developed and specialized skills in the social sciences and statistics, combined with substantial knowledge of the extension system and the constraints surrounding its development. However, the most serious and frequently encountered weaknesses are in staffing, and particularly staffing in the senior ranks. In some situations, insufficient professional and supervisory posts have been provided, while in others posts have been created but remain unfilled owing to a lack of qualified staff or grading difficulties. This can have severe consequences on both data quality and work output. In one large state in India, extensive and apparently acceptable data have been collected and satisfactorily tabulated but remain unreported owing to a shortage (unfilled positions) of economists to interpret and summarize these results. While M&E is making substantial progress and its technical proficiency is improving overall performance remains handicapped by two weaknesses: (i) a lack of understanding by extension managers of M&E; and (ii) inadequate and sometimes very slow analysis and report production. Managers of extension are not yet sufficiently familiar with M&E and tend to either regard its products as uncalled for criticism or a questioning of objectives or try to use it as an internal police force to identify and report wrongdoing. Management in extension is not yet committed to the notion of objective reporting of progress and unprepared to request and use imaginatively the information it provides. Although, the evaluation of extension and the attribution of productivity changes is difficult, a more rigorous approach to the analysis of monitoring and evaluation data is required. There is a growing realization that this objective can be greatly aided by introducing more sophisticated data processing facilities. As a result, micro-computers are beginning to be deployed in extension projects and eventually these will help to improve analysis and report-writing. Nevertheless, they are not a substitute for improved human input. M&E staff, now that basic systems are in place and functioning, still need to improve their intellectual grasp of the complexities of extension and farmer response and ally this understanding to the creative analysis of the data they collect. Although monitoring and evaluation work is now quite widespread in extension projects, much of it is too recent for its results to have become widely available. An exception is India, where as earlier noted some 300 monitoring and evaluation studies have been completed. Although some of these deal with special issues affecting particular state extension services many contain the results of sample surveys conducted routinely each season on the basis of virtually identical questionnaires and sampling designs. Their sheer - 49 - volume precludes individual summary. Fortunately, however, researchers have analyzed and published some of the data in many of these reports.65 Cost Recovery and the Role of the Private Sector As this paper has made clear, extension services, which are mainly public services, are intensive users of manpower. Hence, the costs of running an extension service are dominated by wage costs and other expenditures directly related to the number of staff. Unfortunately, few data to demonstrate this point are available but World Bank appraisal reports show that about 75-85% of the recurrent costs of running an extension service should be devoted to wages and non--discretionary allowances, that is, allowances that must be paid as part of the explicit contract governing employee terms of service. The remaining 15-25% of recurrent costs consists of expenditures such as travel allowances, petrol, oil and repairs for vehicles and other miscellaneous operating costs which are discretionary; that is, payment or non-payment will not affect the incomes of staff. Inadequate provision for such items, however, has a disproportionate effect on the efficiency of extension operations which obviously depend on well trained and mobile staff. There is much evidence that the provision for such items in some projects is inadequate.66 Such events typically reflect more widespread budgetary problems - depressed government revenues and weak national and sectoral planning - which result in a failure to properly prioritize government investments and service expenditures. These problems contribute to the need for economy in resource use and a questioning of whether services such as extension should continue to be provided free of charge. Of equal importance, however, is the growing realization worldwide that the public sector is not always the best and certainly not the only, actual or potential, supplier of extension services. As a result, there is a search for ways to recover the costs of public services and to explore ways of increasing private sector involvement. Extension is no exception. To what extent, therefore, is it possible to consider user charges or other forms of cost recovery in extension? Information on new technology, including some farmer-specific advice, is typically thought to be a public good: that is the provision of information to one user does not exclude other users from obtaining the information. Indeed, effective extension, because not all farmers can be provided with information individually, depends on the free 65 See Gershon Feder, R. Slade and A. Sunderam, 'The Training and Visit Extension System: An Analysis of Operations and Effects', AgrcultyS Administration, Vol. 21, 1986, and Roger Slade, G. Feder, and R. Chikkara 'Reforming Agricultural Extension: The Training and Visit System in India', Ouarterly Journal ot Intmmpional Agjjcult=., (forthcoming). 66 See for example John Howell, ecrent Costs and Agricultural Development, ODI, London, 1984. - 50 - dissemination of information amongst farmers. This feature provides a powerful argument for free public provision of agricultural information implying that cost recovery should be made only indirectly through general taxation or a user tax or fee on all actual or potential beneficiaries. However, as a given level of agricultural technology becomes well known, extension agents are likely to be most valuable when they provide a more individual advisory service - solving particular problems for particular farmers. Such services are less of a public good providing there is, with time, a steady sophistication in the nature of the problems and the quality of the advice. Under these circumstances, the intensity and mechanisms with which more generalized technology is disseminated may change or be reduced - implying, of course, some cost savings. However, a 'market' may also begin to open for skilled and specific agricultural advice. At this juncture, the possibility of introducing specific user charges arises. At the same time, however, government must reconsider its role in this market and carefully evaluate its comparative advantage. If it is a monopoly purveyor of such advice it can exploit its position. This is unlikely, however, as the private sector, especially manufacturers and distributors of agricultural inputs, may have equally good information. Thus, it is sensible for government to create conditions in which private suppliers of advice can emerge and flourish. If demand for such services is strong enough, governments may have only to ensure a free but also a well regulated environment to prevent abuse and ensure quality control. If initial demand conditions are less certain but the prospects appear promising in the longer run, government may wish to use short-term tax or other incentives to encourage response in the private sector. This may be a particularly valid approach for specialized agricultural advice to poor and small farmers as such advice has many of the characteristics of what economists call merit goods. These goods are a form of public good in as far as they may bestow externalities - in this case positive ones - but their distinguishing feature "is not that third parties benefit from the provision of such goods, but that the recipients themselves benefit to a greater extent that they themselves believe".67 It follows that they would tend to under-purchase such goods if no incentive were provided. In agriculture these goods may emerge in the form of specialized farrn advisory companies or inputs suppliers providing information as part of the price competition in the market they serve. This is now a common feature of developed agricultural economies and developing countries are likely to follow suit. Thus, commercial agents specializing in certain types of information will operate side by side with a more generalized public sector extension service. 67 Gabriel Roth, The Private Provision of Public Services in Developing Countries, Oxford, New York, 1987. CHAPTER 7 CONCLUSIONS It is evident that much remains to be learned about the process of transmitting agricultural technical knowledge and the role of extension services in the process. But the apparently more manageable task of assessing the impact of public investment in extension services raises formidable problems. For example, it is not possible on the basis of current evidence to isolate the effects of 'extension' as a component of public agricultural investment. No government or public extension service is readily able to indicate the total recurrent and capital cost of its extension operations. Even where approximations can be made, there remain legitimate questions about which parts of an agricultural service system as a whole, and its administration, constitute 'extension' expenditure. Without such information, it is difficult to justify unequivocally different levels of investment in extension or to present definitive statements on the cost of one extension approach in comparison to others, even assuming that the different objectives of, and approaches to, extension would allow valid comparisons. While there is inevitable ijmprecision over the costs and benefits of agricultural extension, two general points enjoy wide acceptance. The first is that the advance of technical knowledge is a prerequisiite of agricultural development, and without some system of field-based involvement with farmers, improved agricultural technology cannot be diffused effectively and agricultural research (and other farmer services) will not focus sufficiently on farmer concerns and priorities. There are, of course, major questions of the particular form that this field-based involvement and support should take; the relative role of public extension operations in relation to private systems, and the relative roles of public extension and educational development of the rural population are cases in point. The appropriate mix of direct contact between extension agents and farmers, and the use of mass media is another fundamental issue. But these are not questions that impinge on acceptance of the importance of extension in agricultural development. The second point is that while extension operations can and should be organized in a variety of ways in response to local conditions and objectives, there are a number of conditions that are necessary for successful extension investment, and these are constant for all agricultural extension operations. One is the need to acknowledge and facilitate the dual function of extension - to advise farmers on how to increase their productivity and incomes, and to learn from farmers their production conditions and priorities in order to be able to advise and guide agricultural research. Another fundamental condition is that extension must have a deliberate technological basis. The prime function of extension staff is to help farmers increase the product of their farm activities. Unless extension staff have - 52 - means to do this, they can serve little purpose. A third condition is that just as extension must have a close two-way relationship with agricultural research, so must it be with all other agricultural support services - such as input supply, rural credit, marketing and irrigation system management. A further requirement of successful extension is effective management. Given the geographic dispersal of extension staff and their clients, the contributory role of extension to agricultural production, and the reliance of extension on other agricultural support services, there are few other activities that make a greater demand on management than agricultural extension. The best extension methods, staff, research and input support can have little impact on farmers unless they are effectively managed. It is evident that there is a wide range of contexts in which extension operates: different production systems, client diversity, changing requirements of clients and technology with time, different support from the private sector and different policy objectives are ready examples of such contexts. Agricultural extension must be organized differently to reflect these variations. Similarly, there are different entry points for external agencies wishing to support agricultural extension. Depending on the needs and priorities of the extension service and government on the one hand, and on the skills, experience and resources of the external agency on the other, such support could focus on overall system management, specialized training needs, or agricultural research development and co- ordination with extension. Some external agencies may be better suited to provide intensive specialist support for extension staff in a specific area than others. Whatever the form of support, it will only be effective if it is responsive to the priorities of the extension service it is designed to assist and then, only if it is in a form that is sustainable managerially and financially once the external support is terminated. Despite this diversity in extension approaches and in environments in which extension takes place, the experience of extension - and not simply the World Bank's experience - is that principles of effective organization and field management relevant to all extension operations can be identified and should be stressed in programme design. These normally include the following. First, extension staff of all levels should have feasible and monitorable work programmes, and the work programmes of field staff should be known to farmers. Second, the product of extension should be knowledge that enables farmers to increase the productivity and profit of their work and assists other agricultural services to improve their support to farmers. To enhance this knowledge, extension staff should receive frequent and regular training, support from technical specialists, and have frequent contact with agricultural research and other services for farmers. Third, extension activities should take place primarily at a farmer's place of work and should cover all aspects of a farmer's production activities. Much has been learned over recent decades which is relevant to the establishment of effective agricultural extension systems and considerable experience in programme design - 53 - for a range of approaches to the organization of extension in different administrative and agro-ecological environments has been gained. Nevertheless,there are a number of important areas where much more must be learned. In part, this is because any extension organization must be continuously adapted to serve the changing needs of its clients, and because there are a number of other iissues which donor experience in extension has shown to require a further period of experimentation and learning. Three broad categories of issues can be distinguished. The first is operational and design issues within public extension services and their relationship with other extension services. The second, those issues which relate to the relationships between extension and other components of agricultural knowledge systems. Thirdly there are issues concerning the strategic nature for extension services and agricultural development policy more generally. In this last category one of the most important issues appears to be the problems of access to resources and services which specific groups encounter. These, for example, may be women farmers in some areas, or very poorly endowed farmers unable to take advantage of technical opportunities. In both cases, the possibility of bias in extension services towards relatively more privileged farmers needs to be scrutinized. Such issues cannot be treated as exclusively 'extension' questions, however, as they are also issues related to national agricultural priorities. The operational and management issues are more various. They involve the development of approaches to the design, monitoring and evaluation of national extension programmes. This is closely linked to extension financing, particularly the possibilities of cost recovery, cost reduction, or even cost transfer by giving extension responsibilities to the private sector or farmers' organizations. But it is also linked to questions about extension methods as there are major operational implications for extension in the growing use of improved communications technology and ways in which they can be coordinated with field operations. The linking of extension to research and improving extension's accountability to farmers is the other main area requiring development and experimentation. While these issues may appear daunting there are encouraging signs that they can be addressed. The foundation for effective agricultural extension operations has now been established in many countries. In most cases, the basic administrative structure of a field oriented extension service is in plaLce; field staff, supervisory and specialist functions have been defined and the requisite staff appointed. A system for work programme formulation and for the development of technical recommnendations is also present Links in some form generally exist between agricultural extension and research, as well as with other agricultural services and the agricultural planning process. There is widespread recognition that extension agents should be trained to listen to, and learn from, farmers; though achievements in this field are everywhere inadequate. In most countries, a staff training programme operates; some use is already made of media support for field extension activities; and some form of monitoring and evaluation is in operation. The challenge is to - 54- improve on this beginning while recognizing that there is no single 'best' extension system suited to all circumstances. It is difficult to imagine a situation where the quality of extension services cannot be improved. REFERENCES Baxter, M. and W. Thalwitz, 'National Policies and Institutional Constraints to Linking Research with Extension in Asia', in Cernea, Coulter and Russell (eds.), 1985. Benor, D., and M. Baxter. Training and Visit Extension, The World Bank, Washington D.C., 1984. Benor , D., and J. Q. Harrison. Agricultural Extension: The Training and Visit System, The World Bank, Washington D.C., 1977. Benor., D., J. Q. Harrison and M. Baxter. Agicultural Extension: The Training and Visit System, The World Bank, Washington D.C., 1984. Blankenburg, P. von. Agricultural Extension Systems in Some African and Asian Countries, FAO Economic and Social Paper 46, Rome, 1984. Boyce, J. K., and R. E. Evenson. National and International Agricultural Research and Extension Programs, Agricultural Development Council, New York, 1975. Byerlee, D., 'Extension in Post Green Revolution Agriculture', in Howell (ed.), (forthcoming). Cernea, M., and B. J. Tepping. 'A System for Monitoring and Evaluating Agricultural Extension Projects', Staff Working Paper No. 272, The World Bank, Washington D.C., 1977. Cernea, M., J. Coulter and J. Russell (eds.). Agricultural Extension by Training and Visit: The Asian Exerience. The World Bank, Washington DC., 1983. Cernea., Michael., J. Coulter and J. Russell (eds.). Research-Farmer-Extension: A Two- way Continuum for Agricultural Development. The Worll Bank, Washington D.C., 1985. Chambers, R. and M. Wickremanayake. 'Agricultural Extension: Myth, Reality and Challenge' in B.M.Farmer (ed.) Green Revolution? Technologv Change in Rice Growing Areas of Tamil Nadu and Sri Lanka. MacMillan, London, 1977. Chambers, R. Settlement Schemes in Tropical Africa. R.K.P., London, 1969. Collinson, Michael. Farn Management in Peasant Agriculture. Westview, Boulder, 1983. Crouch, B. R. and S. Chamala (eds.). Extension Education and Rural Development. 2 Vols, John Wiley, Chichester, 1981. Evenson, R. E. and D. Jha. 'The Contribution of the Agricultural Research System to Agricultural Production in India', Indian Journal of Agricultural Economics. Vol. 28, No. 4, 1973. - 56 - Evenson, R. E. 'IARC Investment, National Research and Extension Investment and Field Crop Productivity.' Yale University, New Haven, Connecticut, undated, (processed). FAO. Agricultural Extension: A Reference Manual. Rome, 1972 (first edition prepared by D. Bradfield) and 1984 (much revised second edition prepared by Interpaks, University of Illinois). FAO. "Notes on FAO's Involvement and Experience in Agricultural Extension", Agricultural Education and Extension Service, Human Resources, Institutions and Agrarian Reform Division, FAO, March 1987 (processed). Feder, G., L. Lau and R. Slade. The Impact of Agricultural Extension. Staff Working Paper No. 756, The World Bank, Washington D.C., 1985. Feder, G., R. E. Just and D. Zilberman. Adoption of Agricultural Innovation in Developing Countries: A Survey, Staff Working Paper No. 542, The World Bank, Washington D.C., 1982. Feder, G., R. Slade and A. Sundaram. 'The Training and Visit Extension System: An Analysis of Operations and Effects' Agricultural Administration, Vol. 21, 1986. Giltrow, D., and J. Potts. 'Agricultural Communication: The Role of the Media in Extension Financing', The British Council, London, 1978. Gittinger, J. Price. Economic Analysis of Agricultural Projects, Johns Hopkins, Baltimore, 1982. Griliches, Z. 'Research Costs and Social Retums: Hybrid Corn and Related Innovations', Journal of Political Economy. Vol. 66, 1958. Harker, B.R. 'The Contribution of Schooling to Agriculture and Modernization: An Empirical Analysis' in P. Foster and R.J. Sheffield (eds.) Education and Rural Development, Evans Bros, London, 1973. Howell, J. (ed.). Agricultural Extension in Practice. ODI, London, (forthcoming). Howell, J. 'Making Agricultural Extension Effective', in Howell (ed.), (forthcoming). Howell, J. Recurrent Costs and Agrcultural Development, ODI, London, 1984. H6eper, B. 'Selected Results of an ADO and VEW Survey in Jind, Karnal and Mahendragarh Districts of Haryana, India', Institute of Socio-Economics and Agricultural Development, Technical University, Working Note No. 1, Berlin, 1983 (processed). Hunter, Guy and A. F. Bottrall (eds.). Serving the Small Farmer: Policy Choices in Indian Agriculture, Croom Helm, London, 1974. Jones, G. E.(ed.). Investing in Rural Extension. Elsevier, London and New York, 1986. Jones, G. E., and M. Rolls. Progress in Rural Extension and Comrnunity Development, Vol.2, John Wiley, Chichester, 1982. - 57 - Judd, A. M. et al., Investing in Agricultural Suppl. Economic Growth Centre, Yale University, New Haven, Connecticut, 1983. Lamb, G. and L. Muller. Control. Accountability and Incentives in a Successful Development Institution: The Kenya Tea Development Authority. World Bank Staff Working Paper No.550., Washington DC., 1982. Lele, U. The Design of Rural Development: Lessons from Africa. Johns Hopkins, Baltimore, 1975. Leonard, D. 'Why do Kenya's Agricultural Extension Services Favor the Rich Farmer'. Paper read at 16th Annual Meeting of the African Studies Association, Syracuse, New York, 1977. Little, I. and J. Mirlees. Project Appraisal and Planning for Developing Countries, Basic Books, New York, 1984. Lockheed, M., D. Jamison and L. Lau. 'Farmer Education and Farm Efficiency: A. Survey'. Economic Development and Cultural Change, Vol 29, No. 1,1980. Mahdavi, G. 'A Commodity-Driven Approach: the Experience of the Compagnie Francaise pour le Developpnient de Textiles', in Nigel Roberts (ed.) Agricultural Extension in Africa. The World Bank, Washington D.C., (forthcoming). Mishan, E. J. Cost Benefit Analysis, Allen and Unwin, London, 1975. Moris, J. Extension Alternatives in Tropical African Development. ODI, London, (forthcoming). Moris, J. Managing Induced Rural Development. International Development Institute, Bloomington, 1981. Mosher, A.T. An Introduction to' Aericultural Extension. Agricultural Development Council, New York, 1978. Murphy, J. and T. Marchant. Monitoring and Evaluation in Extension Agencies, Technical Paper No. 79., The World Bank, Washington D.C., 1988. National Council of Applied Economic Research. Impact of Indo-German Fertilizer Education Project, New Delhi, 1979. Orivel, F. 'The Impact of Agricultural Extension: A Review of the Literature' in H. Perraton, D. Jamison, J. Jenkins, F. Orivel and L. Wolff (eds.)., Basic Education and Agricultural Extension, Staff Working Paper No. 564, The World Bank, Washington D.C., 1983. Rahim, S. A. 'The Comilla Programme in East Pakistan' in C.M. Wharton (ed.) Subsistence Agriculture and Economic Development, Aldene, Chicago, 1966. Rice, E. B. Extension in the Andes: An Evaluation of Official US Assistance to Agricultural Extension Services in Central and South America. USAID, Washington D.C., 1971. - 58 - Rivera, W. M. and S. G. Schram. Agriculture Extension Worldwide, Croom Helm, London, 1987 Roling, N. G. Extension Science: Agricultural Information Systems in Agricultural Development. Cambridge University Press (forthcoming). Rolls, M. J., G. E.Jones and C. Garforth. 'The Dimensions of Rural Extension', in Jones (ed.), 1986. Roth, G. The Private Provision of Public Services in Developing Countries. Oxford, New York, 1987. Saunders, H.C. (ed.). The Cooperative Extension Service, Prentice Hall, New York, 1986. Slade, R. H. and G. Feder. 'The Monitoring and Evaluation of Training and Visit Extension: A Manual of Instruction', India Department, The World Bank, Washington, D.C., 1981 (processed). Slade, R. H., G. Feder and R. K. Chikkara. 'Reforming Agricultural Extension: The Training and Visit System in India', Ouarterlv Journal of International Agriculture., (forthcoming). Squire, L. and H. G. van der Tak. Economic Analysis of Projects. Johns Hopkins, Baltimore, 1975. Sreensunpagit, A. 'Monitoring and Evaluation of Extension: Experience in Thailand' in Cernea, Coulter and Russell (eds.)., 1983. True, A.C. A History of Agricultural Extension Work in the United States 1785-1923. US Dept. of Agriculture, Washington D.C., 1929. USAID, "AID Support for Agricultural Extension " (unclassified), Department of State, Washington D C., 1985. World Bank, Agricultural Research and Extension: An Evaluation of the World Bank's Experience. Operations Evaluation Department, The Wor1d Bank, Washington D.C., 1985. World Bank. Agricultural Research. Sector Policy Paper, World Bank, Washington D.C., 1981. STATISTICAL ANNEX WORLD BANK LENDING F'OR AGRICULTURAL EXTENSION Tables 1. World Bank Lending for Agricultural Extension in Current Dollars. 2. World Bank Lending for Agricultural Extension in Constant 1985 Dollars. 3. Extension Components in World Bank Livestock, Forestry and Fisheries Projects, 1975-1986. 4. World Bank Agricultural Extension Projects Currently Being Inplemented. 5. World Bank Agricultural Extension Projects Compared with Agricultural and Rural Development Projects Pre-1965 to 1986. 6. World Bank Lending for Agricultural Extension by Quinquennia 1965 - 1986. 7. World Bank Lending for Agricultural Extension by Different Extension Methods. 8. World Bank Agricultural Extension Projects: Number of Projects with Different Extension Methods. 9. World Bank Lending for Agricultural Extension: Number of Countries using Different Extension Methods. 10. World Bank Lending for Agricultural Extension: Number of Countries with Current Projects using I)ifferent Extension Methods. 11. World Bank Agricultural Extension Projects: Extension Cost Relative to Project Cost. 12. World Bank Agricultural Extension Projects: Families Affected and Extension Cost per Family for Different Extension Methods. 13. World Bank Agricultural Extension Projects: Total Extension Cost and World Bank Share. 14. World Bank Lending: Ten Largest Borrowers for Agricultural Extension. -60 - Note on Data The source of all data- unless otherwise indicated is a survey of the documentation for all World Bank-supported68 agriculture and rural development projects presented to the Bank's Board on or before June 30, 1986. An agricultural 'extension project' was defined to be any project having some agricultural extension activity. 'Agricultural extension' is extension for general field agriculture; it excludes livestock, farm forestry and farm fisheries extension activities. Because the primary data source for the survey was the Staff Appraisal Report (SAR) for each project data refer to planned actions rather than actual experience. The conventions used in the tables follow: 'n.a.' indicates data are not available. All years are World Bank financial years (1975, for example, is July 1974 - June 1975). The year of a project is that in which it was approved by the World Bank's Board. All costs are in current US dollars unless otherwise noted. 'Project cost' is the total estimated cost of a project, at the time of project appraisal. 'Extension cost' is the estimated cost of extension activities in a project, at the time of project appraisal. 'Bank Loan/Credit amount' is the amount of the Bank Loan/Credit at the time of Board approval. 'Bank extension support' is the estimated amount of Bank disbursement against project extension cost, at the time of project appraisal, based on the disbursement rates estimated in the SAR. The extension method of a project was determined from the SAR and its related annexes and working papers. Five categories are defined. 'Commodity Specific,' for commodity based extension systems (i.e., multi- function operations for a commercialized crop). 68 Unless otherwise stated the tenn "World Bank' includes IDA. -61 - Training and Visit' (T&V) extension when this was explicitly noted as such in the SAR. 'Modified T&V,' where the method is specifically identified in the SAR as a 'modified' T&V system (though rarely with precise details of modification) or where the methodology is not specifically mentioned but the extension system as described incorporates key features of the T&V method. 'Other,' where Commodity Specific, T&V or Modified T&V methods are not indicated but reasonable detail is presented of the proposed extension methodology. 'No Methodology,' for cases where there is no reference to extension method or activities in the SAR or related documents. Since the SAR is the source of extension method information, changes in extension methodology during project implementation are not reflected in the tables. Table 1: WORLD BANK LENDING FOR AGRICULTURAL EXTENSION IN CURRENT DOLLARS (June 30th 1986) Project Costs World Bank Share No. of Total Extension Total Extension Countries with a Bank Share of Extension Region Projects Cost ($m) Cost ($m) Cost ($m) Cost ($m) Cost greater than $25.0 million EASTERN AND SOUTHERN AFRICA 75 2207 418 1035 174 WEST AFRICA 97 4880 666 2072 273 Nigeria EUROPE. MIDDLE EAST AND NORTH AFRICA 70 6224 469 2308 173 Turkey LATIN AMERICA AND CARIBBEAN 80 13854 1349 5153 524 Brazil, Mexico EAST ASIA AND PACIFIC 64 5012 476 2324 265 Indonesia, Thailand, Philippines, Malaysia SOUTH ASIA 74 4622 649 2357 397 India, Bangladeshl Pakistan TOTAL 460 36798 4027 15249 1807 Table 2: WORLD BANK LENDING FOR AGRICULTURAL EXTENSION IN CONSTANT 1985 DOLLARS (June 30th 1986) Project Costs World Bank Share No. of Total Extension Total Extension Countries with a Bank Share of Extension Region Projects Cost ($m) Cost ($m) Cost ($m) Cost ($m) Cost greater than $25.0 million EASTERN AND SOUTHERN AFRICA 75 2564 475 1275 211 Malawi WEST AFRICA 97 5356 758 2352 324 Nigeria, Cote d'lvoire, Burkina Faso, Senegal EUROPE, MIDDLE EAST AND NORTH AFRICA 70 7214 479 2662 180 Turkey LATN AMERICA AND CARIBBEAN 80 14681 1416 5477 551 Brazil, Mexico EAST ASIA AND PACIFIC 64 5675 531 2619 290 L-donesia, Thailand. Philippines, Malaysia SOUTH ASIA 74 5753 693 2881 421 India, Bangladesh, Pakistan TOTAL 460 41243 4351 17267 1978 Table 3: EXTENSION COMPONENTS IN WORLD BANK LIVESTOCK, FORESTRY AND FISHERIES PROJECTS, 1975-1986 EASTERN & EUROPE MIDDLE LATIN AMERICA EAST ASIA SOUTHERN WESTERN EAST & NORTH AND SOUTH AND ALL AFRICA () AFRICA AFRICA CARIBBEAN ASIA (5*) PACIFIC REGIONS LIVESTOCK Extension Cost (Sm) 16.5 15.2 0.9 2.5 11.9 1.6 48.6 No. Couttries 8 5 2 2 6 2 25 No. Projects 8 5 2 2 4 2 23 FORESTRY Extension Cost ($m) 2.7 0.3 0.9 0.1 1.6 5.9 11.5 No. Coautries 3 2 1 1 4 1 12 No. Projeet 3 2 1 1 1 1 9 FISHERIES Extension Cost (Sm) 0.2 0 0.2 0 5.9 0 6.3 No. Countries 1 0 1 0 1 0 3 No. Projects I 0 1 0 1 0 3 TOTAL Extension Cost (Sm) 19.4 15.5 2 2.6 19.4 7.5 66.4 No. Countries 12 7 4 3 11 3 40 No. Projects 12 7 4 3 6 3 35 NOTE: (*) Includes one Livestock Project in one country with an Extension Cost of Sl.lm that is included in all other tables. (**) Includes four livestock Projects in two countries with Extension Costs totalling S8.3m that are included in all other tables. SOURCE: World Bank Agricultural Projecta Database - 65 - Table 4: WORLD BANK AGRICULTURAL EXTENSION PROJECTS CURRENTLY BEING IMPLEMENTED (June 30th 1986) PROJECTS WITH EXTENSION COMPONENTS NO. OF NO. OF PROJECT EXTN BANK SHARE REGION PRO- COUNTRIES COST COST PROJECT EXTENSION JECTS W/PROJECTS (Smn) (Sm) COST ($m) COST (Sm) EASTERN & SOUTHERN AFRICA 35 15 1416.90 287.57 617.23 101.67 WESTERN AFRICA 41 16 3303.40 449.60 1339.70 166.83 EMENA 44 10 4385.82 395.20 1643.10 142.58 LATIN AMERICA AND CARTBBEAN 49 13 7856.00 699.95 2882.40 249.10 EAST ASIA AND PACIFIC 33 10 3151.57 131.65 1485.30 85.86 SOUTH ASIA 29 6 2134.96 83.36 1149.10 64.73 TOTAL 231 70 22248.65 2047.33 9116.83 810.77 'PURE EXTENSION PROJECTS NO. OF NO. OF PROJECT EXTN BANK SHARE PRO- COUNTRlES COST COST PROJECT EXTENSION IECTS W/PROJECTS (Sm) (Sm) COST (Sm) COST (Sm) EASTERN & SOUTLIERN AFRICA I 1 28.60 28.60 15.00 15.00 WESTERN AFRICA 0 0 0.00 0.00 0.00 0.00 EbENA I 1 33.80 33.80 7.50 7.50 LATINAMERICAANDCARIBBEAN I 1 349.10 315.65 155.00 140.64 EAST ASIA AND PACIFIC 2 2 152.00 152.00 87.00 87.00 SOUTH ASIA 8 2 326.50 326.50 199.30 199.30 TOTAL 13 7 890.00 856.55 463.80 449.44 TOTAL (COMPONENT AND 'PURE" PROJECTS) NO. Ol NO.OF PROJECT EXTIN BANK SHARE PRO- COUNTRIES COST COST PROJECT EXTENSION JECTS WiPROJECTS (Sm) (Sm) COST (Sm) COST (Sm) EASTERN& SOUTHERN AFRICA 35 15 1445.50 316.17 632.23 116.67 WESTERN AFEUCA 41 16 3303.40 449.60 1339.70 166.83 EMENA 45 10 4419.62 429.00 1650.60 150.08 LATINAMERICAANDCARIBBEAN 50 13 8205.10 1015.60 3037.40 389.74 EAST ASIA AND PACL:IC 3 10 3303.57 2S3.65 1572.30 172.86 SOUTH ASIA 37 6 2461.46 409.86 1348.40 264.03 TOTAL 244 70 23138.65 2903.88 9580.63 1260.21 NOTE: Projects with extision components ae projecl in which exteison oomprises 90% or more of total project costs. - 66 - Table 5: WORLD BANK AGRICULTURAL EXTENSION PROJECTS COMPARED WITH AGRICULTURAL AND RURAL DEVELOPMENT PROJECTS PRE - 1965 TO 1916. Anic and Rural Dev Proiecta Extension Proiects Tota WoeAJ Wadd Bank No. of Project Banu No. of Extension Extension REGION PERIOD Projeps Coat Sm Share Sm Projects Cost Sm Share Sm EASTERN Total (al year) 192 na 2885 75 418 174 AND Total (1975-1986) 135 5137 2391 59 376 149 SOUTHERN Pre-1965 6 n.a. 75 0 0 0 AFRICA 1965-69 15 na. 2 3 6 4 1970-74 36 n.a. 368 13 37 21 1975-79 55 1913 818 26 91 41 1980.84 54 2363 1157 23 219 73 1985-86 26 860 416 10 65 35 WESTERN Total (all years) 180 na. 3112 97 666 273 AFRICA Total (1975-1986) 136 7198 2860 83 617 249 Pre-1965 0 n.a. 0 0 0 0 1965-69 5 n.a. 57 1 3 0 1970-74 39 n.a. 195 13 47 24 1975-79 69 2044 947 43 170 85 1980-84 48 4236 1545 36 410 150 1985-86 19 919 368 4 36 14 EUROPE, Total (all years) 171 na. 7137 70 469 173 .IDDLE EAST Total (1975-1986) 142 22608 6562 65 462 170 AND Pre-1965 4 n.a. 30 0 0 0 NORTH AFRICA 1965-69 5 na. 73 1 0 0 1970-74 29 niL 473 4 6 3 1975-79 63 8247 2207 32 54 39 1980-84 60 9096 2999 33 321 117 1985.86 19 5265 1357 8 87 23 LATh4 AMERICA Total (all years) 209 n.a 9597 so 1349 524 AND Total (1975-1986) 129 20691 8524 74 1343 521 CARIBBEAN Pre-1965 25 na. 127 0 0 0 1965-69 20 n.a. 311 0 0 0 1970-74 35 na. 635 6 7 3 1975-79 58 5976 2214 29 441 175 1980-84 52 10859 3912 35 442 160 1985-86 19 3856 2398 10 460 187 EAST ASIA Toal (all years) 201 n.a. 8010 64 476 265 AND Total (1975-1986) 153 15850 4685 55 458 258 PACIFIC Pre-1965 3 na. 27 0 0 0 1965-69 11 na. 202 1 1 0 1970-74 34 n.L. 550 S 17 7 1975-79 64 5290 2546 23 183 90 1980-84 64 6980 3116 22 249 145 1985-86 25 3580 1568 9 D7 23 SOUTH ASIA Total (all years) 247 n.a. 10816 74 649 397 Total (1975-1986) 194 29832 9799 67 641 392 Pre-1965 16 n.a. 212 0 0 0 1965-69 8 n.a. 185 0 0 0 1970-74 i9 n.a. 620 7 8 6 1975-79 89 5729 2638 35 247 140 1980-84 88 11584 4742 26 224 149 1985-86 26 12518 2419 6 169 103 GRAND TOTALS Totl (all years) 1200 nma 41558 460 4027 1807 Total (1975-1986) 889 101314 34821 403 3896 1738 Pre-1965 54 na 471 0 0 0 1965-69 64 n.a 879 6 9 5 1970-74 193 n.a 2842 51 122 63 1975-79 389 29199 11369 181 1187 562 1980.84 366 45118 17471 175 1865 792 198546 134 26998 8527 47 844 385 NOTh Data for Agric. and R.D. Proje' are from the Banks Agriculural Project Data Base - 67 - Table 6: WORLD BANK LENDING FOR AGRICULTURAL EXTENSION BY QUINQUENNIA 1965-86 1965-69 1970-74 No. of Project Extension Bank Ext No. of Project Extension Bank Ext REGION Projects Cost Sm Cost $m Share Sm Projecs Cost $m Cost Sm Share Sm EASTERN AND 3 13.7 5.6 4.4 13 198.2 37.1 20.9 SOUTHERN AFRICA WESTERN AFRICA 1 29.1 2.6 0.3 13 162.1 46.6 24.2 EUROPE, MIDDLE EAST 1 63.0 0.3 0.3 4 414.4 6.4 2.5 AND NORTH AFRICA LATIN AMERICA 0 0.0 0.0 0.0 6 417.3 6.9 29 AND CARIBBEAN EAST ASIA 1 3.3 0.5 0.3 8 242.9 16.9 7.0 AND PACIFIC SOUTH ASIA 0 0.0 0.0 0.0 7 565.2 7.8 5.6 TOTAL 6 109.1 9.0 5.3 51 2000.2 121.7 63.1 1975-79 19s0-84 No. of Project Extension Bank Ext No, of Project Extension Bank Ext Projects Cost Sm Cost $m Share Sm Projects Cost Sm Cost Sm Share Sm EASTERNAND 25 655.5 91.4 41.4 23 1011.1 219.0 72.5 SOUTHERN AFRICA WESTERN AFRICA 43 1403.1 170.3 85.2 36 2948.9 410.4 149.8 EUROPE, MIDDLE EAST 24 1789.7 54.4 30.3 33 3361.7 320.7 116.9 AND NORTH AlRICA LATINAMERICA 2 3919.2 440.9 174.7 35 7452.0 4420 159.6 AND CARIBBEAN EAST ASIA 24 1706.5 182.8 90.2 22 2144.3 248.6 144.6 AND PACIFIC SOUTH ASIA 35 17655.7 247.5 139.9 26 1922.5 224.2 148.6 MOTAL 181 11244.6 1187.2 561.7 175 18840.5 1864.9 79Z0 1985-86 TOTAL No. of Project Extension Bank Ext No. of Project Extension Bank Ext Projects Cost $m Cost Sm Share Sm Projects Cost Sm Cost Sm Share Sm EASTERN AND 2L0 328.0 65.2 34.7 75 2206.5 418.3 174.0 SOUTHERN AFRICA WESTERN AFRICA 4 331.3 36.1 14.0 97 4879.5 666.0 273.4 EUROPE, MIDDLE EAST 8 S95.5 87.0 22.7 70 6224.3 468.8 17Z8 AND NORTH AFRICA LATINAMERICA 10 2065.1 459.6 187.0 80 13853.6 1349.4 524.2 AND CARIBBEAN EAST ASIA 9 914.8 26.8 22.8 64 5011.8 475.6 265.0 AND PACIFIC SOUTH ASIA 6 369.0 169.1 103.4 74 4622.5 648.5 397.4 TOTAL 47 4603.7 843.9 384.7 460 36798.1 4026.6 1806.9 Table 7: WORLD BANK LENDING FOR AGRICULTURAL EXTENSION BY DIFFERENT EXTENSION METHODS COMMODITY SPECIFIC T & V MODIFIED T & V No.of No.of No.of Pro- Extension Cost $m Pro- Extension Cost Sm Pro- Extension Cost $m REGION jects Total Mean Min Max jects Total Mean Min Max jects Total Mean Min Max EASTERN AND 13 46.0 3.5 0.1 8.8 5 63.0 12.6 2.3 27.6 12 82.2 6.8 0.9 27.6 SOUTHERN AFRICA WESTERN 22 145.5 6.6 0.4 22.67 14 92.6 6.6 2.8 41.1 9 101.0 11.2 1.4 79.0 AFRICA EUROPE, MIDDLE EAST 1 1.7 1.7 - - 9 215.5 23.9 1.6 176.8 10 86.6 8.7 0.2 22.5 AND NORTH AFRICA LATIN AMERICA 2 3.3 1.6 1.2 2.1 3 37.2 12.4 2.5 29.5 6 124.3 20.7 7.5 54.3 AND CARIBBEAN EASTASIA AND B 41.7 3.2 0.6 12.3 9 259.2 28.8 0.4 71.2 11 141.2 12.8 0.2 61.6 PACIFIC SOUTH ASIA 3 26.4 8.8 0.1 21.6 44 621.1 14.1 0.1 83.6 6 7.4 1.2 34.0 3.6 TOTAL 54 264.6 4.9 - - 84 1288.5 15.3 - - 54 542.6 10.0 - - ON 00 OTHER METHOD NO METHOD SPECIFIED ALL METHODS No.of No.of No.of Pro- Extension Cost $m Pro- Extension Cost $m Pro- Extension Cost $m jects Total Mean Min Max jects Total Mean Min Max jects Total Mean Min Max EASTERNAND 34 259.0 7.6 0.2 117.3 11 24.4 2.2 0.5 7.5 75 474.6 6.3 0.1 117.3 SOUTHERN AFRICA WESTERN 34 339.3 10.0 0.4 79.0 18 79.7 4.4 0.1 16.2 97 758.0 7.8 0.1 79.0 AFRICA EUROPE, MIDDLE EAST 23 110.2 4.8 0.2 33.8 27 64.8 2.4 0.1 33.8 70 478.7 6.8 0.1 176.8 AND NORTH AFRICA LATIN AMERICA 52 1176.8 22.6 0.4 302.6 17 74.7 4.4 0.5 23.8 SD 1416.2 17.7 0.1 302.6 AND CARIBBEAN EASTASIA AND 13 49.6 3.8 0.1 12.8 18 39.1 2.2 0.3 8.9 64 530.8 8.3 0.1 71.2 PACIFIC SOUTH ASIA 13 31.5 2.4 0.4 2.6 8 6.3 0.8 0.0 2.6 74 692.7 9.4 0.1 83.6 TOTAL 169 1966.3 11.6 - - 99 288.9 2.9 - - 460 4351.0 9.5 - - Table 8: WORLD BANK AGRICULTURAL EXTENSION PROJECTS: NO. OF PROJECTS WITH DIFFERENT EXTENSION METHODS 1965 - 69 1970 - 74 1975 - 79 Exeznsion Meahod = (1) (2) (3) (2)4(3) (4) (5) (1) (2) (3) (2)+(3) (4) (5) (1) (2) (3) (2)4(3) (4) (5) EASTERN AND I 0 0 0 2 0 3 0 0 0 7 3 5 2 3 S 13 3 SOUTHERN AFRICA WESTERNAFRICA I 0 0 0 0 0 5 0 0 0 3 5 13 0 3 3 18 9 EUROPE, MIDDLE EAST 0 0 0 0 0 1 1 0 2 2 0 1 0 2 1 3 6 15 AND NORTH AFRICA LATIN AMERICA 0 0 0 0 0 0 0 0 0 0 2 4 0 0 3 3 16 10 AND CARIBBEAN EAST ASIA I 0 0 0 0 0 3 0 0 0 2 3 3 3 6 9 3 9 AND PACIFIC SOUTH ASIA 0 0 0 0 0 0 0 0 0 0 5 2 2 20 5 25 4 4 TOTAL 3 0 0 0 2 1 12 0 2 2 19 18 23 27 21 48 60 5) 19U0 - 84 1985 - 86 TOTAL ExtensionMethod= (1) (2) (3) (2)4(3) (4) (5) (1) (2) (3) (2)(3) (4) (5 (1) (2) (3) (2)4(3) (4) (5 EASTERN AND 4 2 4 6 10 3 0 1 5 6 2 2 13 5 12 17 34 11 SOUTHERN AFRICA WESTERN AFRICA 3 14 6 20 10 3 0 0 0 0 3 1 22 14 9 23 34 IS EUROPE, MIDDLE EAST 0 4 5 9 14 10 0 3 2 5 3 0 1 9 10 19 23 27 AND NORTH AFRICA LATIN AMERICA 2 3 3 6 24 3 0 0 0 0 10 0 2 3 6 9 52 17 AND CARIBBEAN EAST ASIA 4 6 4 10 4 4 2 0 1 1 4 2 13 9 11 2 13 18 AND PACIFIC SOUTH ASIA 1 19 1 20 3 2 0 5 0 5 1 0 3 44 6 50 13 8 TOTAL 14 48 23 71 65 25 2 9 8 17 23 5 54 84 54 138 169 99 NOTE: Extensbnmethods are: (1) Comnodity Specific (2) T&V (3) ModifliedT&V (4) Oth:rMedhod (5) No Method Specified Table 9: WORLD BANK LENDING FOR AGRICULTURAL EXTENSION: NUMBER OF COUNTRIES USING DIFFERENT EXTENSION METHODS 1965 - 69 1970 - 74 1975 - 79 Euaesio-Method= (1) (2) (3) (2)"(3) (4) (5) (1) (2) (3) (2)4(3) (4) (5) (1) (2) (3) (2)+(3) (4) (5) EASTERN AND I 0 0 0 1 0 1 0 0 0 4 2 5 2 2 4 8 3 SOUTHERN AFRICA WESTERN AFRICA I 0 0 0 0 0 4 0 0 0 2 3 7 0 3 3 9 7 BUROPE, MIDDLE EAST 0 0 0 0 0 1 1 0 1 1 0 1 0 1 1 1 3 9 AND NORTH AFRICA LATINAMERICA 0 0 0 0 0 0 0 0 0 0 2 3 0 0 2 2 5 9 AND CARIBBEAN EASTASIA I 0 0 0 0 0 2 0 0 0 2 2 2 3 1 4 2 6 AND PACIFIC SOUTHASIA 0 0 0 0 0 0 0 0 0 0 2 2 2 6 3 6 4 3 TOTAL 3 0 0 0 1 1 8 0 1 1 12 13 16 12 12 20 31 37 1980 - 84 1985 - 86 TOTAL O Eimuin Method= (1) (2) (3) (2)+(3) (4) (5) (1) (2) (3) (2)(,3) (4) (5) (1) (2) (3) (2)4(3) (4) (5) EASTERN AND 4 2 3 5 7 3 0 1 4 5 2 2 8 4 7 10 10 7 SOUTHERN AFRICA WESTERNAFRICA 3 10 6 11 5 2 0 0 0 0 3 1 9 10 8 13 10 10 EUROPE, MIDDLE EAST 0 3 4 5 6 7 0 3 2 3 3 0 1 6 5 8 8 11 AND NORTH AFRICA LATINAMERICA 2 1 2 3 8 2 0 0 0 0 4 0 2 1 2 3 11 9 ANDCARIBBEAN EASTASIA 1 2 3 5 3 2 2 0 1 1 2 2 3 3 4 5 5 7 AND PACIFIC SOUTH ASIA 1 6 1 6 2 2 0 3 0 3 1 0 3 6 3 6 4 5 TOTAL 11 24 19 35 31 18 2 7 7 12 15 5 25 30 29 45 48 49 NOTE: Exiension mediods are: (1) Commodity Specific (2) T & V (3) Modified T & V (4) Ow Method (5) No Method Specified Table 10: WORLD BANK LENDING FOR AGRICULTURAL EXTENSION NUMBER OF COUNTRIES WITH CURRENT PROJECTS USING DIFFERENT EXTENSION METHODS (June 30th 1986) PROJECTS WITH EXTENSION COMPONENTS "PURE" EXTENSION PROJECTS COMPONENT AND "PURE" PROJECTS NO. OF COUNTRIES NUMBER OF COUNTRIES NO. OF COUNTRIES NUMBER OF COUNTRIES NO. OF COUNTRIES NUMBER OF COUNTRIES PROJ- WnH USING MEJHOD:- PROJ- WITH USING ME'HOD:- PROJ- WITH USING MElHOD:- REGION ECTS PROJECTS (1) (2) (3) (2+3) (4) (5) ECTS PROJECTS (1) (2) (3) (2+3) (4) (5) ECTS PROJECTS (1) (2) (3) (2+3) (4) (5) EAS'ERNAND 35 15 5 1 6 7 7 5 1 1 0 1 0 1 0 0 36 15 5 2 6 7 7 5 SOUrHERNAFRICA WESTERNA HRA 41 16 4 9 6 10 6 2 0 0 0 0 0 0 0 0 41 16 4 9 6 10 6 2 EMENA 44 10 0 5 5 7 8 4 1 1 0 0 0 0 1 0 45 10 0n5 5 7 8 4 LATINAMERICA 49 13 2 1 2 3 10 5 1 1 0 0 0 0 1 0 50 13 2 1 2 3 10 5 AND CARIBBEAN EASTASIA 33 10 3 2 4 5 3 5 2 2 0 1 1 2 0 0 35 10 3 2 4 5 3 5 AND PACIFIC SOUTHASIA 29 6 1 6 2 6 3 2 8 2 0 2 0 2 0 0 37 6 1 6 2 6 3 2 TOTAL 231 70 15 24 25 38 37 23 13 7 0 4 1 5 2 0 244 70 15 25 25 38 37 23 NOTES: Pure extension projects are projects in which extension comprises 90% or more of total project costs. Extension methods are: (1) Commodity Specific (2) T&V (3) Modified T & V (4) Other Method (5) No Method Specified Table 11: WORLD BANK AGRICULTURAL EXTENSION PROJECTS: EXTENSION COST RELATIVE TO PROJECT COST 1-9% OF PROJECT COST FOR EXTENSION 10-29% OF PROJECT COST FOR EXTENSION 30-69% OF PROJECT COST FOR EXTENSION No.of Bank Funnce for Total Extension No.of Bank Fmane for Toat Extension No.of Bak Finae for Total Extensica Pw- Extension $m Cost Sm Pn- Extension Sm Cost Sm Pro- Extension $m Cost Sm jeds Total Mean Tol Mean jes Total Mean Total Mean jecs Total Mean Tol Mean EASERN AND 34 34.6 1.0 61.9 1.8 26 49.0 1.9 77. 3.0 12 58.9 4.9 206.4 13.8 SOUTHERN AFRICA WESTERN 35 33.7 1.0 89.6 26 53 210.4 4.0 51Z9 9.7 9 29.4 3.3 63.4 4.2 AFRICA EUROPE MIDDLE EAST 54 63.2 1.2 141.8 Z6 9 23.7 2.6 82.6 9.2 5 17.9 3.6 31.0 2Z1 AND NORTH AFRICA LATIN AMERICA SD 178.5 3.0 481.3 8.2 18 94A 5.2 275.6 15.3 1 13.0 13.0 324 2.2 AND CARIBBEAN EASTASIAAND 48 60.8 1.3 100.7 21 8 15.1 1.9 26.6 3.3 2 4.2 2.1 7.0 0.5 PACIFIC SOUTH ASIA 47 35.8 0.8 56.4 1.2 6 22.7 3.8 31.8 5.3 5 58.9 11.8 81.7 5.4 TOTAL 277 406.6 1.5 931.7 3A 120 415.3 3.5 1007.0 8A 34 182.2 5A 498.6 33.2 70-89% OF PROJECT COST FOR EXTENSION 70-89% OF PROJECT COST FOR EXTENSION TOTAL No.of Bank Fnance for Total Extension Noof Bank Fnane for Total Extension Noof Bank Fnane for Total Extension Pr- Extension Sm Cost $S Pro- Extension $m Cost Sm Pro. Extension Sm Cost Sm jeas Total Mean Total Mean jecs Total Mea Total Mean jecs Total Mean Total Mea EASTERN AND 2 16.5 8.2 43.8 21.9 1 15.0 15.0 28.6 28.6 75 174.0 2.3 418.3 5.6 SOUTHERN AFRICA WESTERN 0 0.0 0.0 0.0 0.0 0 0,0 0.0 0.0 0.0 97 273A 2.8 666.0 6.9 AFRICA EUROPE, MIDDLE EAST 1 60.6 60.6 179.6 179.6 1 7.5 7.5 33.8 33.8 70 172.8 2.5 468.8 6.7 AND NORTH AFRICA LATIN AMERICA 0 0.0 0.0 0.0 0.0 2 238.3 119.1 560.2 280.1 SD 524.2 6.6 1349.4 16.9 AND CARIBBEAN EASTASIAAND 2 60.3 30.2 103A 51.7 4 124.6 31.1 237.9 59.5 64 265.0 4.1 475.6 7A PACIFIC SOUTH ASIA 7 55.7 8.0 98.7 14.1 9 224.3 24.9 380.0 42.2 74 397A 5.4 648.5 8.8 TOTAL 12 193.1 16.1 425.5 35.5 17 609.7 35.9 1240.5 73.0 460 1806.8 3.9 4026.6 8.8 Table 12: WORLD BANK AGRICULTURAL EXTENSION PROJECTS: FAMILIES AFFECTED AND EXTENSION COST PER FAMILY FOR DIFFERENT EXTENSION METHODS COMMODITY SPECIFIC T & V MODIFIED T & V No.of No of Farm Average No. Extn Avoage No.of No of Farm Average No. Extn Avenge No-of No of Farm Average No. Extn Avrage Pro- Families FFAffected Cost EXL Cost Pro- Families FFAffected Cost Ext. Cost Pr- Families FFAffected Cost EXL Cost REGION jects Affected '00 per Proj '000 $ m per FF in S jects Affected '000 per Proj '000 $ m per FF in $ jects Affected 000 per Proj '000 $ m per FF in $ EASTERN AND 13 1033 79 45.9 44 5 2270 454 63.0 2S 17 2925 172 145.2 5D SOUTHERN AFRICA WESTERN 22 285 13 145.5 510 4 909 227 92.6 102 23 1672 73 193.6 116 ARI.CA EUROPE, MIDLE EAST 1 5 5 1.7 378 9 1985 221 215.5 109 19 2241 118 302.0 135 AND NORTH AFRICA LAT-NAMERICA 2 !! 5 3.3 313 3 431 144 37.2 86 9 611 6S 161.5 264 AND CARIBBEAN EAST ASIA AND 13 231 18 41.7 181 9 25742 2860 259.2 10 20 31701 1585 400.4 13 PACIFIC SOUTH ASIA 3 1123 374 26.4 24 44 208896 4748 621.1 3 50 209057 4181 628A 3 TOTALS 54 2688 5) 264.5 98 74 240234 3246 1288.5 5 138 248206 1799 1831.1 7 OTIIER METIIOD NO METIIOD SPECIFIED ALL METIIODS No.of No of Faun Average No. Extn Average No.of No of Farm Average No. Extn Avesge No.of No of Farm Average No. Extn Aveage Pro- Families FF Affected Cost EXL Cost Pro- Families FFAffected Cost Ext. Cost Ptn- Families FF Affeted Cost EXL Cost jects Affected'000 per Proj '000 $ m per FF in $ jects Affected '00 per Proj '000 S m per FF in $ jects Affected '00 per Proj '000 $ m per FF in S EASTERN AND 34 3133 92 259.0 83 11 71S7 653 24A 3 75 16548 221 537.5 32 SOUTHERN AFRICA WESTERN 34 3022 8S 339.3 112 18 346 19 79.7 230 97 6234 64 850.6 136 AFRICA EUROPE,MIDDLEEAST 23 1237 54 110.2 89 27 583 22 64.8 111 70 6051 86 694.2 115 AND NORTH AFRICA LATIN AMERICA 52 4387 84 1176.8 268 17 237 14 74.7 316 80 5675 71 1453.4 256 AND CARIBBEAN EASTASIAAND 13 15280 1175 49.6 3 18 1141 63 39.1 34 64 74095 1158 790.0 11 PACIFIC SOUTH ASIA 13 2644 203 31.5 12 8 1127 141 4.3 4 74 422847 5714 1311.7 3 TOTALS 169 29703 176 1966.3 66 99 10621 107 286.9 27 460 531451 1155 5637.4 11 Tabl 13: WORLD BANK AGRICtJLTURAL EXTENSION PROJECTS: TOTAL EXTENSION COST AND WORLD SANM SlAREUnI,NE 36TH IoU (Cotmt 19S5 dgener) 1985 TOTAL EXENSON COST EUNK SHARE EXT COST PPLATION COST PER C5PNTA COST PER CAP1TA REGION AND COUNTRY IN MIlONS USS m USS US S m USS EASTERN AND SOUTHERN AFRICA: 205.0 474.3 2.32 211.5 1.03 BUUNDI 47 12.4 2.64 10.1 2.15 COMOROS 14 0o9 132 a9 2.32 EIHI8'A 42.3 38.7 (192 23.4 as5 KENYA 20.4 44.4 21 23.5 1.16 LEsalo L. 4.9 3.22 1.1 Q73 MADAGASCAR 10.2 7.5 074 62 (161 MALAWI 7.0 56.0 0.06 38.4 5.45 MAURITIUS L0 (8 (178 4 037 RWANDA 6.0 26.1 4.32 20.3 3.36 SOMALIA 14 34.4 6.39 11.3 2 10 SUDAN 21.9 51.7 236 22.1 1.01 SWAZILAND (s 7.5 9.89 3.2 4.25 TANZANIA 22.2 32.4 1.46 16.4 0.74 UGANDA 15.5 3.0 0.19 1.5 a1o ZARE 30.6 10.5 0.60 11.1 0.36 ZAMBI 6.6 17.2 250 13.5 2.03 ThIABWE 0.4 117.3 13.95 &01 096 WESTERN AFRICA: 190.1 7S8.0 3.99 325.9 1.72 BENIN 4.0 27.4 6.77 &1 2.00 BURIONAPASO 7.9 44.8 5.69 27.8 3.52 CAMEROON 10.2 34.3 3.37 17.2 1.69 CL.A. z6 OA 3.27 (10 o.0 CHAD 20 1.4 (128 1.4 0.23 OTmEDIVO1RE 10.1 71.0 7.03 28.6 234 TIEMGAMBIA a7 5A 7.33 4.1 552 GHANA 12.7 25.0 2.03 11.4 (0s9 GUINEA 6.0 1.3 21 (4 0.06 LMEIA 212 19.5 &86 0.1 3.71 MAU 7.5 31.3 4.17 13.0 1.99 MAURITANIA 1.7 10.2 6.01 23 1.30 NI.ER 64 11.5 1.00 4.5 (171 N.IA 99.7 390.s 3.92 147.9 1.48 SENBGAL 66 36.6 5.57 26.0 3.97 SIERRA LEONE 2.7 10.4 2.70 5.5 1.47 TOGO 3.0 27.9 9.17 17.8 2.4 EUROPE, MIDDLE EAST AND NORTH AFRICA: 215.4 478.7 2.20 130.1 0.82 AFGHANISTAN M. 4.8 M. 3.0 M. ALGEUIA 21.9 6.9 0131 23 0.24 CYPRUS (17 0.4 12.69 5.5 030 BG`fPT 47.1 22.9 0.49 23 0.11 GREPECE 9.9 3.3 033 L (11s JORDAN 2.5 04 all a( (103 MOROCCO 21.9 40.6 1.65 20.5 094 POKIXAL 10.2 10.0 090 4.3 0.42 SYRIA 10.5 41.6 3.97 9.7 (192 TUNISIA 7.1 31.4 4.39 15.1 2.11 TURKEY 49.4 232.6 4.71 81.1 1.64 YEMEN, ARAB REPLUC &O 54.7 6.07 16.3 2.03 YEMEN, PEOPLES DEMOCRAnC REP. I 1 7.0 3.37 4.2 2.03 YUGOSLAVIA 23.1 13.9 0.60 0.O 0.34 LATIN AMERICA AND CARIBBEAN: 312.2 1416.2 4.54 551.5 1.77 BOLIVIA 64 2.7 0.42 2.0 32 BRAZIL 1325 869.4 6.41 345.0 2.55 CHn0E 12.0 16.5 1.38 0.3 069 CttOM 23.4 53.9 1.90 20.4 (172 COSTA RICA 26 2.5 as97 (17 (126 DOMINICAN REPUEBrC 03 .6 0.58 1.3 (21 ECUADOR 94 11.7 1.25 3.4 0.36 CRENADA (l 1.0 10.31 19 0.96 GuYANA as 0. 1.04 a( 0.06 HAm 25 3.4 (62 as 0.1 JAMAICA 22 5.1 2.29 3.0 1.3L MEICO 78.8 401.5 5.09 143.7 1.82 PANAMA Z2 (17 0.33 (14 (117 PARAGUAY 24 2.8 (0.2 1.0 (130 PERUi 18.7 40.5 z217 19.7 1.06 EAST ASIA AND PACIFIC: 1332.3 530.6 0.40 289.5 0.22 CHNA 1041.1 12.2 (1o0 10.9 (101 INMD E4EA 162.2 211.2 1.30 105.5 (65 LAOS 326 1.0 0.27 (1s (123 MALAYSrA 15.6 86.1 5.52 37.3 2.39 PAPUANEWGUINEA 3.5 19.5 258 13.9 3.99 PHILIPPINES 34.7 69.9 1.28 44.9 (162 SOLOMONISLANDS (3 2.8 1(164 2.8 10.34 IHaILAND 51.0 125.0 2.45 70.7 1.39 VA.NUATU (11 1.5 11.46 1.5 11.39 WESTERN SAMOA (12 1.5 9.37 1.0 6.30 SOTITH ASIA: 1030.2 692.7 0.67 421.4 0.41 BA3OLADESH 10(16 S7.0 (16 67.4 (167 BURMA 36S. 0.6 (23 2.2 0.09 INDIA 7621 301.9 0.60 29(12 0130 NEPAL 16.5 18.9 1.14 15.6 (194 PAISTAN 94.9 52.3 a55 30.1 (132 UILANXA 16.1 23.9 1.4S 14.9 (192 TOTAL 3285.0 4350.9 1.32 1980.0 0.60 NOM1R Popalio&nu mfm Is WVdd Bwk AIa 1987. - 75 - Table 14: WORLD BANK LENDING: TEN LARGEST BORROWERS FOR AGRICULTURAL EXTENSION Total Borrowings for Agricultural Extension Current Constant Dollars 1985 Dollars $m Sm Brazil 328.39 345.85 India 279.17 290.18 Nigeria 137.64 147.85 Mexico 132.92 143.74 Indonesia 93.34 105.51 Turkey 76.32 81.06 Thailand 65.39 70.03 Bangladesh 57.05 67.43 Philippines 49.6 44.94 Malaysia 29.9 37.32 TOTAL 1249.72 1333.91 Distributors of World Bank Publications ARGENTINA FINLAND MALAYSIA SRI LANKA AND THE MALDIVES Caios Hinch, SRL Akateedine, Kirjakauppa Un verdity of Madya Cooperative Lake Houe Boohkhop GabenaG uere P.O. Box 128 Bookahop, Liited P.O Bo 244 Flolida 165,4th Floorik. 453/465 SFcmOI P.O Bo. 1127, Jalan Pantai Baru 100, Sir CttPamd A. Garder 1333 Bunio Airot Helsinki10 Kuoa liupur Mawabha Cudomibo 2 AUSTRALIA, PAPUA NEW GUINEA, FRANCE MEXICO FIJI SOLOMON ISLANDS, Wod Sank Publication INFOTC SWEDEN VANUATU, AND WESTERN SAMOA 66,avonued'1ns Apartado Potal 22-60 Fornrgtitlt DA. Book Journals 75116Padc 14060Tlalp n,MtiDonD. Fitz- Fadibokaforetaget 11-13 Stabon Street Regaiogogatan 12, Boo 16356 Mitcha 3132 GERMANY. FEDERAL REPUBLIC OF MOROCCO S10327 cttdnhol Vidoria ULNO-Ve&ag Sodete d'Etuder Marketing Marxaine Poppedorfer Allo 5 12 rue Moart, Bd. d'Anf FPaeasii4ln nfeT AUSTRIA D-5300 Bonn I Casablnca Wemoirersn-Wiffaon AB Ge,old and Co. Boo 3004 G-abo 31 GREECE NETHERLANDS S-10425 Stockbid A-1011 Wien KEME InOe-Pubhikatiinbv 24, Ippod StnouEtoretPSlmt MPlataa P.O. 3. 14 SWITZERLAND BAHRAIN Athn-11635 7240 BA Lchem ornirgerrJ Baheahi Reoeatch and Connultany LUbedde Payot Asodalte lAd. HONG KONG MACAO NEW ZEALAND 4 rue Grnur P.O Bo. 22103 Aia 200 Ltd. Hill Library and ffrmaion Service Cae potal 381 M-nama Town 317 6 FL 146 Pnc Edward Road, W. PvateBag CH 1211 Gena 11 Kowloo New Marke BANGLADESH Hong Kong Aukdand Foroeece*t-inms ion: MiRr findu-tni Devdopment Libraiie Payt Asitance Sodety (MIDAS) HUNGARY NIGERIA Servie denAbonn-n--t. House 56, Road 7A tuluac Uai..ity P- Liid C.- po4 3312 Dhanoondi R/Aea P.O. Box 139 Th Crwn-BuidingJUridh CaaHes2sLa3 e Dhaka 120° 1389 Budapert 62 Private Mail Bag 5095 Ibadan TANZANIA BELGIUM INDIA O-ford Unversity Pron Pubcahti do Nation Union Allied Pubhhen Pvahte Ltd. NORWAY P.O. Box 1299 Av du Rht 202 751 Mount Road Narveren Itfoemation Cinh rDar On B Saam 1060 Bru"ss Madra - 600 002 Bert-and Narreenr vei 2 P.O. Boo 6125 Enerstad THAILIAND BRAZIL Boch jfa N-0602 Oslo 6 Cenht-I Departmnnt Store Publicco Te-ica Int-cadonai- 1IS.N. Heredia Mag 3D65ilm Road Ltda. Balard Ertte OMAN Bangkok Rua Peixota Gomide, 279 Bombay -400 038 MEMRB infoeraton Ser-i-c 01 409 San Paulo, SP P.O. Boo 1613, Seb Airport TRIIDAD & TOBAGO, ANTICiUA 13/14 Aaf A13 Road Mut BARBUDA, BARBADOS, CANADA New Dlhi - 110 02 DOMINCA, GRENADA, GUYANA. LeDiffuaseu PAKISTAN JAMAICA, MONTSERRAT, ST. C.P. 55, 11B-reAmpbre 17 Chitaranjan Avoue Mia Book Ageny XITTS & NEVIS, ST. LUCiA Boudh-rviDe, QOebc Calcutta- 700072 63, Shiaah-e-Quaid-Aza ST. VINCENT tkGRENADINES J4B 5E6 P.O. BSo No. 729 Sypteonatic Studieo Unit Iyadera Honth Bulditg Lahore 3 5S Estern Main Road CHINA 5th Main Road CGodhinagar Carepe (lies F.oandal & Eonomic Publishing Bngalore- 560009 PERU Trinidad, Wont Indiec Hous Sdio,al Donarll 5,A H, Da PoSi OgJie 3-5-1129 Kaciguda Croo Road Apartado 3824 TURKEY Beijr,.g Hyd-abad -103027 r Haet KVtapei. AS. DavotpasaCaddoni COLOMBIA Prathana Flab 2nd Floor PHILIPPINES SergekanSokak,115 EnlceLtd.. Nr ThakoreiB Navragpura National BokStore Topkapi ApartadoAereo34270 Ahrnedabad-313009 701 RizadAvenue Ibtul Bogota D.E. P.O. BSo 1934 Pati.a Houre Metro Manila UGANDA COSTA RICA 16-A Arhok Mag Uganda Bookohop Liberia Tqr.os Lucknow - 226 001 POLAND P.O. So 7145 Cale111-13 ORPAN Kampala Av. FernandezGull INDONESIA PatacKlthryyiNauki SanJone Pt Indra limitcd 002-1 Warcawa UNIfED ARAB EMLRATES 31. Sow Ratolangi 37 MEMRB Gulf Co. COTE a'IVOIRE Jakarta Puast PORTUGAL P.O. Bo 6097 Cotte dEdiin et de Diffiaon P.O. Box 181 Lvria Portugal Sharjah Afhicainot (CEDA) Ra Don Carmo 70l-74 04 B.P. 541 IELAND 120 Lisbon UNITED KINGDOM Abidjan 04 Plateau TDC Publishers Mi t Ltd. 12 North FPdick Sheet SAUDI ARABIA, QATAR PO Boo 3 CYPRUS DNbu, I Jrir Book Store Alton, Hanphire GU34 2PG MEMRB Information Services P.O Bon 3196 England P.O. Bon 2099 ISRAEL %ydhll471 Nico-ia Thejeru-aodt Pont LRUGUAY TheJetnales Pont Building SINGAPORE, TAIWAN, BURMA, Instituto Mactonal dd Libro DENMARK PO BoxSI BRUNEI Sanjoaell16 Sa-frndLittmarahr Rome jerusalem91000 hIormation Publicadons Mesvigyideo Roatouea AIR I I Private. Ltd. DK-I970 PFrdemikabex C ITALY 02-06 1st Ft Pt-Fu InduUtrial VENEZUELA liuts Comumisiomaia Sairnoni SPA Bldg. lb-ria de Eon DOMINICANREPUBLIC Via Bndetto Foemi 120/10 24NewIndutrialRoad Aptdo. 60337 Editors Taler, C. pe A. Caella Pstale 552 Sungj- 1953 Caaca 106-A Rotta-rn -on e LNabel I Catoica 309 50125 Horner Aparlado Pontal 2190 SOurrH AFRICA YUGOSLAVIA S-to Dom-g. JAPAN Forli,,gkldW: Jugonlovenik.nKpga Eariern Book S-ior Onford Univerdty P-ert Soathern YU-11000 Blgrade Trg Republike EL SALVADOR 37-3, Hongo 3-Chome Bunkyoku 113 Africa Faades Tokyo P.O, B.o 1141 ZIMBABWE Edifi La Centro Americana 6. Pisw Cape TownM00 Long-n Zimbabwe Apartado PontaI01-278 KENYA P.O Box SST 125, Soutwerton Sa Sa3adr 01I Afica Book Svie (EA) Ltd. Fr omib n,ss: P.O. Boo 45245 Int-tna lnal Sobsription Servie EGYPT, ARAD REPUBLIC OF Naibi P.O, Boo 41095 Al Abeam Craighall Al Gala Strt KOREA, REPUBLIC OP Johanmesburg 2024 Catr Pa Knots Book Corporathin P.O. Box 101, Kw-mgwh-aun SPAIN The Middle Knot Observer Seou1 Mondi-Prsa LiUroc, SA. 8 Chaw-rbi Streer Ctello 37 Cairo IKUWAIT 28001 Madrid MEMRB PO. Boo 5465 RECENT WORLD BANK TECHNICAL FAPERS (continued) No. 77. Gunnerson, Wastewater Management for Coastal Cities: The Ocean Disposal Option No. 78. Heyneman and Fagerlind, University Examinations and Standardized Testing: Principles, Experience, and Policy Options No. 79. Murphy and Marchant, Monitoring and Evaluation in Extension Agencies No. 80. Cernea, Involuntary Resettlement in Development Projects: Policy Guidelines in World Bank- Financed Projects (also in Spanish, 80S) No. 81. Barrett, Urban Transport in West Africa No. 82. Vogel, Cost Recovery in the Health Care Sector: Selected Country Studies in West Africa No. 83. Ewing and Chalk, The Forest Industries Sector: An Operational Strategy for Developing Countries No. 84. Vergara and Brown, The New Face of the World Petrochemical Sector: Implications for Developing Countries No. 85. Ernst & Whinney, Proposals for Monitoring the Performance of Electric Utilities No. 86. Munasinghe, Integrated National Energy Planning and Management: Methodology and Application to Sri Lank-a The World Bank Headquarters European Office Tokyo Office U I818 H Street, N.W 66, avenue d'lena Kokusai Building Washington, D.C. 20433, U.S.A. 75116 Paris, France 1-1 Marunouchi 3-chome Telephone: (202) 477-1234 Telephone: (1) 40.69.30.00 Chiyoda-ku, Tokyo I00, Japan Facsimile: (202) 477-6391 Facsimile: (1) 47.20.19.66 Telephone: (3) 214-5001 Telex: WUI 64145 WORLDBANK Telex: 842-620628 Facsimile: (3) 214-3657 RCA 248423 WORLDBK Telex: 781-26838 Cable Address: INTBAFRAD WASHINGTONDC ISSN 0253-7494 Cover design by Bill Fraser ISBN O-8213-1193-X