Experience from the PhilIppin-- Frances F. Korten SWP528 WORLD BANK STAFF WORKING PAPERS Number 528 ~ ~~~~~~ ~~~~~~~~~~~~~ WORLD BANK STAFF WORKING PAPERS Number 528 Building National Capacity to Develop Water Users' Associations Experience from the Philippines Frances F. Korten (consultant) Agricultural and Rural Development Department Agricultural and Rural Development Department The World Bank Washington, D.C., U.S.A. Copyright © 1982 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 This is a working document published informally by The World Bank. To 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 publication is supplied at a token charge to defray part of the cost of manufacture and distribution. The views and interpretations in this document are those of the author(s) and should not be attributed to The World Bank, to its affiliated organizations, or to any individual acting on their behalf. Any maps used have been prepared solely for the convenience of the readers; the denominations used and the boundaries shown do not imply, on the part of The World Bank and its affiliates, any judgment on the legal status of any territory or any endorsement or acceptance of such boundaries. The full range of The World Bank publications is described in the Catalog of World Bank Publications; the continuing research program of the Bank is outlined in World Bank Research Program: Abstracts of Current Studies. Both booklets are updated annually; the most recent edition of each is available without charge from the Publications Distribution Unit of the Bank in Washington or from the European Office of the Bank, 66, avenue d'Iena, 75116 Paris, France. Library of Congress Cataloging in Publication Data Korten, Frances F. Building national capacity to develop water users' associations. (World Bank staff working papers ; no. 528) "July 1982." Bibliography: p. 1. Irrigation--Philippines--Citizen participation. 2. Water use--Philippines--Citizen participation. I. Title. II. Title: Water users' associations. III. Series: World Bank staff working paper .; no. 528, HD174I.0P5K67 1982 333,91'315'09599 82-11039 ISBN 0-8213 -005 1-2 ABSTRACT Over a five year period the National Irrigation Administration (NIA) of the Philippines has been building its capacity to develop water users' associations on small scale irrigation systems. The NIA approach is to develop a water users' asso- ciation prior to construction of the physical system and then to involve association members fully in the planning and con- struction of the system. Implementing this participatory approach has required a wide variety of changes in the agency's policies, procedures, and personnel. The paper details the changes that have been made within the agency, examines the nature of the learning process which has led to these changes, and discusses the implications for donor support of other small scale irrigation programs and more generally for programs involving village level work. - iii - ACKNOWLEDGEMENTS Many of the concepts of this paper are based on the continuing flow of creative ideas and actions of my colleagues on the Communal Irrigation Committee of the National Irrigation Administration including Felipe Alfonso, Romulo Coloma, Edilberto de Jesus, Romana de los Reyes, Alan Early, Carlos Isles, Faustino Inos, Jeanne Illo and Jess Ocampo. Special appreciation is due Committee Chairman Engr. Benjamin Bagadion whose vision, leadership, and dedication have been the primary stimulus of the work described in this paper. Many committee members have commented usefully on earlier drafts of this paper, although I alone must be held responsible for its flaws. I am also grateful to Dr. David C. Korten for both his comments on this paper and his clear articulation of the broader concepts which provide the framework for much of the work described here. I owe thanks to Dr. Michael Cernea who encouraged me to write this paper for presentation at a World Bank sociological workshop in July, 1981 and am grateful for the comments of the many World Bank staff members who took part in that workshop, particularly the panel discussants, Mr. John Blaxall, Mr. Kudlapur K.V. Krishna and Mr. Enzo Giglioli. Important to the work described have been the contributions of NIA's current adminsitrator Dr. Fiorello Estuar and NIA's past administrator, Minister Alfredo Juinio, who have provided the policy framework and personal support so important to the development of an innovative program. I would also like to thank my many colleagues at the Ford Founda- tion for their consistent support and encouragement for the program this paper represents. Peter F. Geithner, John C. Cool and Tom G. Kessinger deserve particular mention for their conceptual and organizational support. Finally, recognition must go to the hundreds of NIA personnel and thousands of farmers throughout the Philippines who have devoted so much energy to implementing this new approach to irrigation development. - iv - BUILDING NATIONAL CAPACITY TO DEVELOP WATER USERS' ASSOCIATIONS: EXPERIENCE FROM THE PHILIPPINES TABLE OF CONTENTS Page No. I. AN OVERVIEW .................................................. 1 II. INDIGENOUS IRRIGATION DEVELOPMENT ............................ 5 III. THE PHILIPPINE SMALL SCALE IRRIGATION PROGRAM .... ............ 10 IV. OPERATIONAL REQUIREMENTS OF THE PARTICIPATORY APPROACH ....... 17 A. Feasibility and Site Selection Stage ..................... 17 B. Preconstruction Stage .................................... 22 C. Construction Stage ....................................... 31 D. Operations and Maintenance Stage ......................... 35 E. Program Costs and Benefits ............................... 38 V. A LEARNING PROCESS APPROACH TO BUREAUCRATIC REORIENTATION ..... 44 A. A National Level Committee ................................ 44 B. Pilot Projects as Learning Laboratories .... .............. 48 C. Agency Based Implementation .............................. 48 D. Flexible Donor Support ................................... 49 E. Phased Expansion Accompanied by Agency Changes .... ....... 50 VI. IMPLICATIONS FOR DONOR PROGRAMMING ............................ 52 A. Irrigation Program Design and Implementation Issues ....... 53 B. Project Cycle Isues ....................................... 59 REFERENCES ........................... ......................... 64 - v - I. AN OVERVIEW 1. This paper describes the process by which the National Irrigation Administration (NIA) in the Philippines is building its capacity to develop water users associations on small scale irrigation systems. Four basic ideas have guided this work: - The farmer-users of small scale irrigation systems (under 1,000 hectares) are potentially capable of operating and maintaining their irrigation systems. - Involving the local water users association in the preconstruction and construction activities is an important means of developing more functional physical systems and developing the skills and structures of the associations. It is important also for insuring that farmers are willing to contribute towards the maintenance of the investments. - An irrigation agency which has been oriented nearly exclusively to the task of construction, will need to change many of its methods to implement an approach which involves participation of the local people in preconstruction and construction activities. - Understanding the nature of the changes needed in the agency s policies and procedures and implementing such changes on a nationwide basis requires building a learning process composed of an integrated set of action, research, and training activities. 2. The first two ideas listed are the premises which served as the stimulus for the NIA's initial pilot efforts to develop a participatory approach to irrigation development. Those premises were derived from exten- sive observation in the Philippines and elsewhere and were confirmed by the intensive documentation of the NIA pilot projects. The second two ideas are concepts which emerged from examination of the initial pilot project activity and have guided the efforts to implement the participatory approach more broadly. 3. The validity of the premise that a local association of water users can manage its irrigation system is amply evidenced by numerous indigenous systems in many nations including the Philippines. Section II briefly reviews some of the literature on such systems and notes a common historial pattern in the development and management of these indigenous systems. Usually the people who initiate the idea for an indigenous irrigation system are the same ones who plan its layout, construct the physical facilities and operate the system. Those people, who generally establish a formal or informal organization, provide important continuity between system development and operation. In contrast, in small scale systems developed by governments in many parts of the world there is often a sharp discontinuity between system development and operation. Outsiders plan and construct the system, leaving the local people as passive bystanders, yet then expect those local people to operate and main- tain the system once it is constructed. - 2 - 4. Philippine government assistance to small scale irrigation systems involved precisely this discontinuity for many years, as discussed in Section III. Government assistance focused nearly exclusively on the construction of physical facilities, although many officials within the NIA were concerned that more attention be given to the organizational aspects of irrigation development. This concern was translated into action in 1975 when a presiden- tial decree ruled that farmers would have to repay construction costs, thus necessitating the creation of local water users' associations that could agree to the terms of repayment. Because the NIA had no personnel for organizing farmers, it contracted with another government agency to do this work. Field experience quickly revealed that having one agency responsible for organizing the farmers and a different agency responsible for constructing the physical system led to confusion and friction and made involving the local people in the preconstruction and construction activities extremely difficult. 5. As an alternative the NIA in 1976, with Ford Foundation support, ini- tiated a pilot project in which NIA explicitly set out to involve the local people as fully as possible in project planning and construction. NIA hired its own organizers for this task who lived with the villagers in the project area. The positive response from the farmers convinced the NIA to pursue the approach further. Over a five year period the number of additional pilot projects was gradually expanded to 2, then 12 and then 24. The number is expected to continue to expand, eventually covering all of NIA's communal irrigation work. The pilot projects were intensively examined to clarify the methodology needed for developing and working with the water users' associations and to learn what changes in NIA's methods were needed for implementing the new approach. 6. Numerous changes were needed as detailed in Section IV. Key among these were: strengthening the site selection and planning process by adding social-organizational data and improving the technical data; fielding full-time community organizers (a minimum of one to a system) eight to nine months ahead of planned construction to develop the association and help it work directly with the engineers in planning the physical development of the system; devel- oping association committees for specific tasks such as mobilizing association labor for construction and monitoring project materials and costs; and offering financial and water management courses for water users' associations. At the level of the agency as a whole these changes have meant adding new personnel (primarily the community organizers and their supervisors), changing the plan- ning cycle to allow for improved feasibility work and to developp the asso- ciation prior to construction), and training both community organizers and the provincial and regional engineering staffs in the implications of the new approach for their work. 7. Five key elements supporting the reorientation of NIA's work are discussed in Section V. A national level committee helped stimulate and direct the process of change within the agency. This committee was chaired and managed by a high level official of the NIA and was comprised of both NIA personnel and academically based persons concerned with communal irrigation. Committee members viewed the pilot projects as "learning laboratories" which could reveal the nature of the changes needed throughout the agency to imple- ment improved projects in the future. The fact that it was the NIA implement- ing the pilot projects allowed the problems caused by the agency s own pro- cedures to be highlighted and stimulated necessary changes. Supporting this process were staff time and funds from the Ford Foundation (matched by NIA funds) programmed flexibly to allow activities to evolve directly from the needs evident from the field level experience. The fact that the pilot proj- ects were implemented only gradually meant that personnel experienced in the previous pilot projects could train and supervise additional personnel as the number of projects expanded. 8. The NIA experience may be useful to development agencies in two ways, as discussed in Section VI. One involves drawing on the Philippine program in the design and implementation of small scale irrigation projects in other countries. While different cultural, climatic, and governmental conditions will necessitate substantial variations in approach, four aspects of the Philippine program may be particularly important to consider for their relevance in other settings. One is the clear authority and responsibility of the water users associations. Associations are legal entities with water rights and total responsibility for operations and maintenance, thus leaving no gray areas of uncertainty regarding government versus farmer responsibility. A second is the use made of existing irrigation groups. When an informal or formal irrigation group already exists in an area where the government plans to develop an irrigation system, organizers generally use that group as the basis for building a more formal association, in order to take advantage of existing irrigation traditions. A third is the policy that the water users association must make substantial contributions to the cost of system develop- ment, including the provision of labor during construction and paying for the system over time. These local contributions help prevent systems from being built where farmers do not want them and provide checks and balances between the farmers and the engineers. A fourth is the policy of developing the water users association prior to construction and involving it in all the key decisions during system development. 9. Under this approach by the time the physical system is completed there is a water users association with membership, by-laws, clear expecta- tions regarding its responsibilities, and considerable experience in making decisions and implementing policies. Many of the numerous adaptations made by the NIA to implement this approach may be relevant in other settings. 10. A second way in which the NIA experience may be useful to donor and action agencies regards the lessons it reveals for the bureaucratic reorienta- tion process. The change process described in this paper involves a basic shift in the role of the action agency. Rather than being simply a provider of services (constructing irrigation systems), it becomes an enabler, enabling local communities to develop both the physical and organizational infrastruc- ture needed to manage their irrigation needs. This shift from providing some limited service to enabling communities to manage their own resources is a basic change in perspective relevant to a variety of development tasks. When - 4 - such reorientation is seen as needed by donor and action agency officials, the programming of donor support must be adapted to this need. Small scale support for an action/research process is needed long before major funds are committed to broader implementation. Such combined action and research must be programmed flexibly with intensive staff support and strong leardership from within the action agency. The appropriate research, training and changes in policy and procedures must flow from the examination of the field level action, rather than being rigidly preprogrammed before the needs are known. The focus of the work must be on building the agency's capacity eventually to execute the new approach on a program wide basis regardless of the source of funds for the individual village level projects, thus avoiding isolated "foreign funded" projects. While these requirements are difficult for both donor and action agencies, the NIA experience reveals their importance for building capacity for effective village level work. - 5 - II. INDIGENOUS IRRIGATION DEVELOPMENT In many parts of the world there is increasing interest in developing water users' associations to help manage and maintain irrigation systems. Concerns with the limited effectiveness of irrigation bureaucracies over- burdened by the rapid expansion of irrigated area has led many to conclude that the users of irrigation water must play a larger role in the operation and maintenance of both large and small irrigation systems 1/. On small systems the need is particularly acute since their scattered locations mean that government contact with them is limited and the effectiveness of irri- gation management depends primarily on the local people. The dilemma faced by irrigation bureaucracies throughout the world is how to help develop local water users' associations to insure effective system management. But while government officials ponder how to develop water users- associations, history shows that often the people themselves have done it on their own. Examina- tion of the experience of these "indigenous" irrigation systems is useful in understanding viable approaches government must take. The validity of the basic concept of a local association of water users' managing irrigation is amply confirmed by numerous indigenous systems in many parts of the world, some having existed for centures. The "subaks" of Bali, Indonesia, the "zanjeras" of the northern Philippines, and the "irrigation cooperatives" of Japan are particularly well known examples of such indigenous systems (Coward, 1980). Studies of these groups reveal that local associations are capable of: mobilizing significant amounts of 1/ The need for farmer organization on irrigation schemes is cited in numerous documents. A few recent examples include: Colombo, Johnson, and Shishido, 1978; Lowdermilk, Early, and Freeman, 1978; and FAO, 1980. - 6 - labor on a long term basis for maintaining the system; allocating water in close responsiveness to crop needs; resolving local conflicts over water; and coordinating cropping schedules to maximize the productivity of available water 1/. When effective, these local groups have significant advantages over bureaucratic management: their intimate knowledge of the local situation allows them to be more responsive to local needs for both water distribution and system maintenance 2/; they can use social pressure to enforce rules; and they place the cost of operation and maintenance directly on those who benefit from the system, avoiding a perennial drain on government resources 3/. A significant feature of indigenous systems is that the local people not only manage and maintain the system, but also plan and construct it. Examination of the history of many indigenous systems reveals a common pattern containing the following elements: Local farmers band together to put logs and stones in a river and dig channels to bring water to their fields; the same individuals that initiate the system's development serve as the irrigation leaders for operation and maintenance; during the physical development of the system, the farmers agree upon rules and obligations for operating and maintain- ing the system; later farmers farther from the water source join in, committing themselves to the same rules and obligations and plan and construct their own 1/ For example one description of a subak in Indonesia (Geertz, 1967) reveals intricate arrangements for assigning planting times in order to stagger the peak water demands. 2/ Within the North Indian context, Reidinger, 1974, has described how local water users associations could make water use more responsibe to crop needs. 3/ For example in the Philippines the routine operation and maintenance costs on the national systems managed by the National Irrigation Adminis- tration averaged $25 per hectare per year in 1980. At that rate, the half million hectares currently managed by community groups would add $12.5 million to the government s annual operation and maintenance costs. part of the system. Thus the social and physical systems develop simulta- neously--often over a period of many years 1/. In examining this process we might conclude that ideally small scale irrigation development should be left entirely to local initiative. But the pressure to increase food production to meet the needs of expanding popula- tions has meant that governments feel they cannot wait for this more leisurely pace of development. Consequently many governments have initiated programs to accelerate the pace of small scale irrigation development. Generally this has meant that initiative has been completely removed from the local people. Out- siders plan, design, and construct the physical system, and only when it is in place is there some (often feeble) attention to the local people who are expected to clean the channels and abide by water allocation rules. In con- trast to the experience of indigenous irrigation development, this government initiated process results in a sharp discontinuity between the planning and construction stage and the operation and maintenance stage, as diagrammed in Figure 1. Efforts to organize farmers into a water users association after the system is completed generally meet with apathy. During the planning and con- struction the local farmers, completely ignored, often assume that since the government is building the system, the government will also manage and maintain it. Once the water is flowing, those receiving it are likely to see little reason to spend their time in organizational meetings. Since no shared sense of commitment to managing and maintaining the system was developed during the I/ Histories of 51 community managed irrigation systems in the Philippines reveal this as a common pattern. (De los Reyes, 1980.) creation of the system, an "each man for himself" mentality is likely to reign 1/. Many of the farmers may in fact have been angered by construction activities which tore up their fields, and by canal layouts which may have taken little account of landholding patterns, soil conditions, or even topo- graphical irregularities. Figure 1 Process of Indigenous Irrigation Development Operation & Planning Design & Layout Construction Maintenance Local Water Users Group - J Process of Government Irrigation Development Operation & Planning Design & Layout Construction Maintenance Government Agency Local Water Users' Group To avoid this kind of discontinuity the National Irrigation Adminis- tration (NIA) in the Philippines has made an effort to simulate the indigenous experience under a government program by involving the local farmers at an early stage. Under this participatory program community organizers are 1/ Bromley, Taylor, and Parker, 1980 analyze the importance to production and equity of generating a "shared concept of justice" among water user association members during the formation of an irrigation system. - 9 - fielded to the area where construction is planned after preliminary feasibil- ity work has been done but prior to the design and layout of the system. They develop a water users' association which then works with the government engi- neers in planning the system development, determining the layout of the canals and constructing the system. Throughout these activities, the community orga- nizers work to insure that there is participation from farmers whose fields lie throughout the entire planned service area. Activities are designed to con- tribute to the development of responsible leadership and a well-structured association capable of moving smoothly into operating and maintaining the system. While conceptually this approach is simple, building a national capa- city to operate in this manner has turned out to be a major task. - 10 - III. THE PHILIPPINE SMALL SCALE IRRIGATION PROGRAM The Philippine terrain results in the nation's having many small (under 1,000 hectares) irrigable areas. The mountains in the center of the nation's many islands slope gently to the shore. The humid climate (average rainfall of 2,350 mm per year) produces myriad rivers and streams. The rainfall however, is uneven, with 70-75% falling between June and December 1/. Even in the wet season dry spells occur frequently, necessitating irrigation for flooded rice culture. For centuries farmers have built structures of logs and stones and dug channels to divert small streams to their fields. A 1978 NIA survey enumerated approximately 5,500 of these locally managed gravity irrigation systems, scattered throughout the nation. These are run-of-the- river systems which vary in size from a few hectares to as much as 4,000 hectares, averaging approximately 100 hectares each 2/. In the Philippines these locally managed systems are known as "com- munal" systems, as contrasted with "national" systems. The distinction rests on who owns, operates, and maintains the system. For national systems it is the government, through the NIA, while for communals it is exclusively the farmers. NIA data indicate that communals provide irrigation to over one-half million hectares, equalling in importance the national systems which cover approximately the same amount of hectarage. 1/ Rainfall data are based on 20 year records (1951-1970) for the entire country as reported in 1979 Philippine Statistical Yearbook, 1979. 2/ Data based on a survey by the NIA Communal Projects and Implementation Department. For more detailed presentations of these data see Bagadion and Korten, 1980. - 11 - Farms on communals are small. A sample survey of 51 communals (Illo, 1980) revealed that farm size averaged 1.1 hectares, with nearly half the farmers owner-operators, and the rest either lessees or share tenants. Generally the people who cultivate the land are active in their own irrigation management--though the way in which the systems are managed varies with the size of the system. Systems under 50 hectares tend to be managed more informally, often with a single village official playing the major role. Systems larger than 50 hectares more frequently have formal management struc- tures such as a registered association, paid water distributors, irrigation fees, and regularly scheduled maintenance (De los Reyes, 1980). The temporary nature of the farmer constructed diversion structures results in their frequent destruction by the heavy rains of the wet season. Not only is rebuilding these structures a significant drain on the farmers' time and energy, but also their destruction may mean crop loss. Further- more, these structures can capture only a fraction of the water in the stream, resulting in less irrigated land than would be possible with a more solid structure. Consequently the farmers on communal systems often seek government assistance to build more permanent irrigation structures. The government has responded to these requests since the early 1950's, but the scale of the response has increased substantially in recent years. In the last five years (1976-1980) the NIA completed construction assistance to approximately 1,000 systems. The scale of this assistance varies widely from major construction (such as building a concrete diversion structure, and extensive canals and canal structures) to more minor system improvements (such as extending and lining sections of a canal on an existing system). Consequently costs per - 12 - project vary substantially, and cost per hectare figures are difficult to interpret. Appropriations of Philippine government funds for communals have totalled $50 million during the last 5 years (Coloma, 1981). Until 1975 government assistance to communals was mostly confined to the construction of physical facilities, carried out at no cost to the farmers. Planning, design and construction was carried out by engineers with minimal attention to the local water users association. Subsequent to construction, some associations functioned effectively but many were weak. This resulted in poor maintenance and ineffective water distribution systems. Many NIA personnel observing these problems became convinced of the need to invest in the development of the water users associations. This conviction was strengthened by a presidential decree requiring NIA to collect repayment of irrigation construction costs. The cost of mate- rials, equipment and labor were to be financed on a loan basis to the farmer members of the association. Furthermore, farmers were to contribute a minimum of 10% of the cost of the construction through voluntary labor, materials, rights of way and/or money. This meant that work with the water users asso- ciation was needed not only for improved operation and maintenance, but also to insure that farmers agreed to repay the construction costs and had an asso- ciation strong enough to mobilize labor for the 10% counterpart contribution and collect fees to repay the construction costs. Because the NIA lacked personnel to carry out the work of developing water users associations, it turned to the Farm Systems Development Corpora- tion, a government corporation which had organized water users associations for pump irrigation. In 1975 NIA contracted this corporation to organize the - 13 - farmers on many of the gravity systems that NIA assisted. The contracting arrangement was based on the assumption that the organizing and construction work were separate tasks, appropriately carried out by different agencies. It was assumed that minimal coordination was needed at the field level, and that most of the organizing task could be done during or after construction. Subsequent experience showed that these assumptions were wrong. A major problem was that the issues that concerned the farmers were the technical ones such as the location of the diversion and the canals, the timing of con- struction, and the choice of hired laborers for the construction work. With the organizational tasks separated from the planning, design and construction tasks, the organizers could not deal with the issues of concern to ther farmers--they approached the farmers "empty handed." Organizing work was needed well before construction with close integration of the organizing and engineering work. In 1976 the NIA with the support of the Ford Foundation initiated a pilot project in Central Luzon to experiment with involving the farmers from the beginning of the project. The NIA recruited full-time community orga- nizers dedicated to the concept of "people's participation" and willing to live in the village and work on a daily basis with the farmers. Their task was to help the farmers develop an association which could then work with the engineers in planning and carrying out the construction assistance 1/. 1/ The variety of activities in which the association was involved are detailed in Isles and Collado, 1979. - 14 - While this initial project experienced numerous difficulties 1/, it produced three important lessons. One was that the local farmers wanted to participate in the development of the physical system and that their partici- pation strengthened their sense of ownership and concern with operations and maintenance. Their involvement in a wide array of preconstruction activities resulted in intense activity in the villages covered by the system, with high levels of participation from both the leaders and the members and widespread knowledge of the details of the project among the people of the area. A second lesson was that the farmers had knowledge that could con- tribute to system design and layout. The design of the dam provided a vivid example of this lesson. Because of expenditure limitations the engineers decided to build a semi-permanent type of dam (gabion) while the farmers insisted that the proposed structure would not withstand the force of local floods. The gabion dam was built and shortly after completion was washed out by a typhoon. The farmers knowledge was important on other items also. They knew the landholding patterns--and whether it would be possible to obtain rights of way in various areas. Often they pointed out land ownership bounda- ries proposing that where possible canals follow such boundaries to avoid taking too much land from any single farmer. They knew the rainy season con- ditions (the engineers' survey often being done in the dry season) such as 1/ In 1978 after a semi-permanent dam had been constructed, Typhoon Kading (internationally known as Typhoon Rita) struck directly on this area, washing out the dam, destroying the homes, and leaving many rice fields covered with sand and boulders. A number of farmers, unable to farm their land, left the area. Later the NIA built an intake which was able to irrigate about half of the area originally expected to be served by the dam. In a second communal, also part of the initial participatory project, construction was postponed until 1980 due to serious internal conflicts within the association. - 15 - which creeks would swell after heavy rains and what areas would become water logged. They also knew soil conditions, pointing out where sandy soil might lead to significant water loss from the canal. A third lesson was that it was extremely difficult for the agency to integrate the social and technical dimensions of its assistance. The engineers were not used to discussing plans with farmers or coordinating with community organizers. The community organizers were not, used to thinking in terms of the requirements of irrigation. Many of NIA's management systems and procedures presented obstacles to implementing a participatory approach. However, the promise that the participatory approach held for improved operation and maintenance, counterpart contributions, and repayment, encouraged NIA leaders to pursue the approach. By the end of 1978 the NIA was ready to move ahead in developing a small scale irrigation program based on the following components 1/: - Response to requests of indigenous irrigation groups for construction assistance when they exist in an area--or when no such group exists, assistance in forming the group. - Pre-construction development of the skills and viability of the water users association. - Involvement of the water users association in all key decisions and activities regarding the planning and con- struction of the system. - Construction cost financed on a loan basis with 10% of the cost of the system contributed during the construction. 1/ For further elaboration of these four components see Bagadion and Korten, 1980. - 16 - Independent ownership, operation, and maintenance of the system by the water users' association, with some external follow-up assistance. The logic and experience with the participatory approach had indi- cated it needed to be a key element in NIA irrigation development policy. The major question that remained in decision-makers' minds was not whether to use the participatory approach but how to build a national capacity to implement it. The personnel and management systems of the NIA were oriented primarily to the construction of physical facilities. To introduce a concern for the organizational aspects of irrigation and integrate that concern with the technical implementation process would require significant changes. A carefully phased series of follow-on pilot projects was carried out to develop and institutionalize the changes required. In 1979 two additional partici- patory projects were undertaken; in 1980 12 more; and in 1981 24 more. A national committee was formed to help examine the problems of these pilot projects and determine their implications for program change. The problems found were ones common to irrigation development around the world. The remarkable characteristic of the NIA was not that it had these problems, but rather that its officials were open and honest in examining their agency's work and searched systematically for improved implementation methods. Out of this process emerged a range of new operational procedures, changing significantly the agency's approach to communal irrigation assistance. - 17 - IV. OPERATIONAL REQUIREMENTS OF THE PARTICIPATORY APPROACH The changes in field operations needed to implement the participa- tory approach touched on every stage of assistance: feasibility and site selection; preconstruction; construction; and operation and maintenance. These field level changes summarized in Table 1, are discussed below, along with the needs which prompted them. The number and significance of the changes highlights why the carefully phased dissemination of the new approach was needed. The process is still underway. Thus, while some of the changes have become routine in the participatory projects, others are still experi- mental and are subject to further development. Additional changes are likely to be introduced as further experience is gained. A. Feasibility and Site Selection Stage A common occurence in NIA's communal assistance program is that shortly before construction is scheduled to start, a project is found to be un- suitable, resulting in a quick decision to substitute a different project so as not to lose the allocated funds. Reasons for cancellation vary substan- tially--from lack of water, inability to obtain critical rights of way, an unstable riverbed, higher than expected project costs, serious conflicts between community factions, and many others. The problem of cancellation of projects poses particular difficulties for the participatory approach. The investment of both the agency's and the farmers time in developing a water users' association prior to construction is only warranted if the project is highly likely to proceed to construction. Last minute substitutions mean - 18 - Table 1 Operational Changes Required to Implement the Participatory Approach Feasibility and Site Selection Stage Socio-technical profiles Improved estimation of area to be irrigated Change in evaluation criteria for engineers Improved technical planning Preconstruction Stage Specialized personnel for organizing Increased time to work with the association before construction Mechanisms for farmer participation in system layout Association working committees Mechanisms for coordinating technical and organizing work Construction Stage Association working committee Preconstruction conferences Periodic reconciliation of project accounts Two-week payment cycle of farmer construction contracts System testing and final adjustment Operations and Maintenance Stage Water management course designed for communals Financial management course designed for communals Continued organizer assistance during initial operations - 19 - not only that personnel and farmer time is wasted, but also that if construc- tion proceeds immediately on the substituted system, then preparatory work with the association cannot be done prior to construction. The experience of one early pilot project which was postponed due to serious conflicts within the association, suggested that both organizers and engineers could be more effective if they began with substantially greater knowledge of the setting. Issues such as who stood to gain or lose by the new system, whether there were important opposing factions involved, who were the key power brokers, and what interests they had in the project were ones that needed highlighting. These problems indicated a need for improved information on candidate systems before fielding community organizers. Another problem was that often the area to be irrigated by a system was overestimated. Under the participatory approach reasonably accurate esti- mates of the area to be irrigated were important since the area to be served determined who should be included in the water users association, which in turn determined who would bear the burden of repayment, elect officers and provide labor during construction. Farmers would be angry if they invested their time and energy in developing the system but then did not receive water. Four operational changes have been developed to deal with these needs: 1. Socio-technical profiles. While some technical and agronomic data had been routinely collected on candidate communal systems, few data were collected on the area's current irrigation arrangements or other relevant social characteristics. This made it difficult to anticipate problems such as in obtaining rights of way or water rights, generating farmer interests in the - 20 - project, converting non-irrigated cropland (such as coconut land) into irri- gated cropland, or developing a viable water users association. To provide a fuller profile of a candidate system, anthropologists worked with NIA engineers to develop guidelines to help NIA personnel gather, write-up, and analyze the needed data (De los Reyes and Borlagdan, 1981). For a 200 hectare system a trained person could gather the social data in about two weeks, then combine it with the engineers technical data to produce approximately a 15 page socio-technical profile. While the data lacked preci- sion, they provided a much improved basis for site selection and planning intervention strategies. 2. Improved water measurement. While a variety of factors determine the area ultimately irrigated, a key factor is the amount of water available. Because communal systems generally draw water from small rivers and streams, there are normally no data available on the stream's discharge. Consequently the NIA must develop these data as part of its feasibility work. Currently the NIA is striving to develop two to three years records on a stream's dis- charge before assisting a communal. Very rough estimates can also be derived by correlating the watershed area of the stream with a similar watershed where the discharge is known although this method is likely to be subject to impor- tant errors due to underestimates of upstream use of water. Currently in many pilot systems once community organizers are fielded to an area, a staff gauge is installed in the stream at the site of the proposed diversion and the farmers asked to make recordings of the water level. Later with the help of a hydro- logist these readings are converted into discharges. This not only provides some initial data, but also encourages the farmers to think in terms of the - 21 - water management needs of their system. In some projects farmers announce these readings at meetings of the water users' association. 3. Changed procedure for evaluating performance of engineers. The past practice has been for NIA records only to reflect "irrigable area"--the area theoretically capable of being irrigated by the canals constructed--not the area actually irrigated by them. Often there are substantial differences between the two. Since the engineers are credited with generating hectarage according to the "irrigable area" figures, there is a tendency to overestimate that area. Changes now being introduced will place greater emphasis on the area actually irrigated. For example farmer repayment schedules will be based on irrigated rather than "irrigable" area. This will call for a series of assessments following construction, which will be entered into NIA records. Engineers will also be evaluated on whether or not the farmers accept the system once construction is completed. Since one cause of farmer refusal to accept the system is that it irrigates less than the area promised, this measure should also contribute to more accurate estimates. Evaluation criteria will also give greater emphasis to farmer counterpart contributions and repayment rates, which are expected to encourage enginers to enter into genuine dialogue with the farmers before and during construction. 4. Improved technical planning. The last minute substitutions in con- struction sites and the inacurrate estimates of area to be irrigated are partly due to lack of personnel at the provincial irrigation offices. Once construction is due to start on a project, a full time construction engineer is assigned to it. But during the feasibility and planning stages, the work is left primarily to the provincial irrigation engineer, who is responsible for running all of the activities of the provincial irrigation office. - 22 - To fill this gap, NIA has now programmed a new position at the provincial level--a "planning engineer" who will be specifically responsible for the technical feasibility and planning work on communal projects. The NIA has also initiated a "retrospective study" examining, on a sample basis, the physical condition of the communal systems it has assisted over the past 15 years. It is expected that data from this study will help reveal key planning and design issues that these new planning engineers should take into account. Whether this one additional provincial level person will be able to improve the technical planning sufficiently, however, is still an open question. B. Preconstruction Stage The pilot projects revealed several major needs during the stage of planning the system and developing the water users' association. First was a need to make use of both the engineers' technical knowledge and the farmers' local knowledge in laying out the canal system. While the technical team provided basic topographical data, farmers had detailed knowledge of the physi- cal and social characteristics of the area and were particularly attuned to the effects of construction alternatives on operation and maintenance. Since in most areas there already existed at least some irrigated area, farmers were familiar with the requirements of irrigation and could provide relevant infor- mation. Second was a need to generate participation by a broad segment of farmers to insure that canal layouts were not biased to favor a few indi- viduals and to create the sense of shared commitment to the system so important - 23 - to operation and maintenance. Third was the need to develop association leadership gradually through activities that would reveal who was committed to the hard work required to make the association function. The customary governmental approach of developing a community organization by simply calling a meeting and asking for an election of officers often resulted in officers who were more interested in the prestige of the position than working to make the association function. Fourth was the need to develop clear understandings between the NIA and the water users association about the project--partic- ularly regarding what was to be constructed, when, by whom and at what cost to the farmers. In response to these needs, the NIA made five types of changes in its preconstruction work. 1. Specialized personnel for organizing. The NIA hires community orga- nizers (who generally have backgrounds in community development or social science) who become part of the provincial irrigation office team working on a project. These people live in the village and devote themselves full-time to developing the water users association--working on Saturdays and Sundays when necessary, since farmers often prefer to hold their meetings on weekends. Current assessments indicate that one community organizer is needed per 100- 150 hectares (with a minimum of one to a system) during the first 12 months, while later, after construction is well underway, a lower ratio is possible if a strong association has developed. Bulding a staff of community orga- nizers, trained and oriented to the needs of irrigation has been a major task for the agency--and remains one. While in early 1976 NIA had no community organizers for communal irrigation assistance, by March 1981 it had 86. Devel- oping the training and support for this new category of personnel represents one among many reasons for phased introduction of the participatory approach. - 24 - 2. Lead time. Experience indicates that community organizers should be fielded 8 to 9 months before construction for 100-400 hectare systems. This lead time has required important changes since NIA procedures and government appropriations have been based on a one-year planning and construction cycle. Since profile writing and analysis needs to be done before fielding the orga- nizers, these combined changes require substantially longer planning cycles from the provincial irrigation offices. 3. Farmer participation in system layout. A variety of mechanisms have been developed to insure broad based farmer participation in the layout of the system. a. Survey committee. On many projects a farmers' survey committee is developed before the NIA topographical survey team arrives in the area. This committee then encourages farmers to accompany the survey team to provide information about the area and discuss advantages and disadvantages of alternative canal locations. b. Paddy mapping. A recent innovation being tried in several projects is to have the farmers, assisted by the community organizers and technical staff, construct a map of all the bunded rice paddies in the area. This map serves a variety of functions. If each farmer mea- sures his own paddies, it insures early broad awareness and involvement in the project from all the potential members of the water users' association. It also identified exactly which farmers are eligible for membership and how much land they till. It can be used for discussions of alternative canal locations--and if elevations are taken of each paddy, can provide a better basis than the topographical map for deter- mining exactly which paddies can be reached by each canal (Moya and - 25 - Early, 1980). It also provides a basic tool for the association to use during the operation and maintenance stage in determining water allo- cation, collecting irrigation fees, and mobilizing for maintenance. c. Farmer-engineer meetings on system layout. To insure agree- ment between the farmers and the engineers on sysem layout a series of meetings are held. While the number and timing of such meetings vary from project to project, the basic flow is as follows: The NIA tech- nical personnel, taking account of both survey data and farmer supplied information, plan the canal locations and present these at a formal meeting of the water users' association. When possible this explanation is supplemented by a "walk-through" of the proposed canal locations, since farmers can generally relate to the on-the-ground descriptions better than to a map. These locations are then discussed by the farmers often over several weeks in small meetings in each of the residential clusters of the area. A second meeting is then held with the engineers where farmer suggestions are aired and any resurvey work agreed upon. When the resurvey work is completed, another meeting and walk-through are held to finalize the plans. The water users' association and the NIA are then supposed to sign a memo of agreement specifying what NIA will build and what the farmers will pay for, including the counterpart contribution. 4. Development of association preconstruction committees. The commu- nity organizers initially develop working committees to serve a variety of needs. The activities of these committees promote broad involvement of the members and allow leadership to develop. Some community organizers try to - 26 - delay the elections of association officers until after these committees have been functioning for some time. Members can then see who have shown respon- sible and committed leadership by the time elections are held. While the exact committees vary from project to project, the tasks performed include: a. Participating in the survey. (discussed above) b. Obtaining the water permit and registering the association. To receive NIA assistance a water users association in the Philippines must be officially registered with the government and hold a water rights permit. Association committees are formed to gather the needed information from the associations potential members and submit it through the provincial irrigation office. c. Obtaining right of way waivers. As preliminary canal locations are determined, farmers obtain right of way waivers from the landowners involved. This is an iterative process since if certain landowners refuse, then canal locations may have to be shifted. Because the NIA loan does not normally include right of way payments, obtaining the waivers is often difficult. It appears, however, that farmers can obtain rights of way that a government employee could not obtain, using techniques such as visiting a resistant landowner in large groups". However tactics for obtaining these rights of way often take a long time and make especially important the 8-9 month lead time before construc- tion. d. Developing the association's membership. Farmers identify who will be served by the system and encourage them to join the water users association. They also develop rules for the treatment of individuals who choose not to join the association. These usually involve higher irrigation fees than are charged to members. - 27 - e. Developing by-laws. Commonly in government programs local organizations are provided a set of standard by-laws. These, however, may be ill-suited to the particular needs of the individual local area, and are in actuality often ignored by the local people, resulting in the organization's not having a well defined framework 1/. To avoid these problems the community organizers encourage farmers to develop by-laws to fit their own association's needs. Proposed by-laws are discussed in small group meetings throughout the system s proposed service area as well as in general assemblies to insure there is wide involvement in the process and agreement on the rules all members will be expected to follow. While standard by-laws are avoided, experience has led community organizers to encourage a number of features in most of the water users associations. One is that members be composed of the people who actually use the water--the tillers of the soil--since it is these people whose behavior in operation and-maintenance will determine the effectiveness of the system. These might be owner-cultivators, lessees, or tenants, but would not include absentee landlords. A second is the creation of "sectors" based on the location of farms along each canal within the system. Experience in the pilot projects as well as other research on community irrigation systems indicates that farmers prefer to do most of the preconstruction, construction and oper- ation and maintenance functions in small groups, covering about 30-40 Lazaro, Taylor, and Wickham, 1979, report that experts at an international irrigation conference concluded that the structure of water users asso- ciations should be tailored to the individual setting rather than trying to implement uniform structures. - 28 - hectares 1/. With transportation and communication extremely difficult over large areas, these small groups become important working units of the association. Coordinating these sectors is also crucial and it appears functional for the board of directors of the association to be composed of representatives from each of these sectors. Third is the creation of structures that encourage careful regular reporting of use of association funds to the general membership. Suspicions and accusa- tions can be easily aroused regarding money, making these safeguards important. A fourth concern evident in some associations, regards women's participation in the association. Experience shows that while women participate actively in the meetings, generally they are not formal members because it is assumed that each household will have only one member, in which case it is normally the man. This means that generally women cannot be officers of the association--even though often they are well qualified. Mechanisms to allow joint memberships from each household have been proposed. 5. Close coordination of the technical and organizational work. All of the activities reviewed so far reveal the need for constant coordination between the technical staff and the community organizers. While in most proj- ects close coordination among the engineering and organizing staff has devel- oped as a natural part of the field level activities, there have also been projects in which tensions have existed between these two groups. Partly this is the result of introducing a new approach which requires substantial changes 1/ Coward, 1977, points out that mini-unit organization is a common charac- teristic of community run systems. These mini-units range from 15 to 70 hectares and result in a labor intensive style of irrigation management. - 29 - in the manner in which the technical staff approach their jobs. It is partic- ularly likely to occur when, for a variety of reasons, there exist distinct differences between the farmers' and the engineers' ideas about the project. Organizers who work and live with the farmers often develop a strong identi- fication with the farmers point of view, while the engineers' viewpoint may be shaped by technical considerations or agency constraints (such as funding, manpower or generated hectarage targets). The agency has developed a variety of mechanisms for encouraging good communication and problem resolution between the technical and organizing staff. In addition to frequent meetings between the two groups, five specific mechanisms developed so far include: a. The project profile. The data in the project profile (discussed under the feasibility stage) include both technical and social consider- ations. These profiles are analyzed jointly by the engineering and organizing staff and provide an initial basis for developing a common perspective on the nature of the project and the likely problems. b. Flow chart of preconstruction activities. To help the technical and organizing staff visualize the interrelationships of their work throughout the preconstruction stage a flow chart was developed (see Figure 2) depicting the parallel activities of the technical staff, the community organizers and the farmers. This provides an example from which each project team can develop its own flow chart to fit the needs of its particular project. - 30 - c. Planning and monitoring forms. Forms were recently developed for reporting the stage of accomplishment and planned activities on both the technical and organizing dimensions of each project. The expectation is that each month these will be filled out jointly by the project level technical and organizing staff, then discussed with the provincial irrigation engineer to help insure that both technical and organizational needs are considered in planning the next month's activities. d. Joint training. A variety of training courses and workshops are held involving both the technical and organizing staff which are meant to build a common perspective on irrigation development and help each type of specialist understand the viewpoints and constraints of the other. e. Regional review meetings. A recent innovation has been the conduct of monthly or bi-monthly review meetings at the regional level which provide a forum for airing the problems of the project and getting additional perspectives on the issues being confronted. C. Construction Stage The pilot projects revealed a variety of needs during the con- struction stage. One was for ways to insure that all farmers understood the expected construction schedule and planned their planting accordingly. Experience showed that if farmers planted where proposed canal lines would pass, they would want to delay construction until they had harvested their crop. A second need was for ways in which farmers could help keep down the costs of the project. The farmers took seriously the government's expectation - 31 - that they repay construction costs and were concerned about avoiding unneces- sary expenses. A third was for ways to respond to farmers desires to be hired as laborers in the project. They were willing to provide part of their labor free if they were paid for the rest. A final set of needs involved the process of project acceptance. Once construction was finished the water users' association and the NIA were supposed to test the system and sign documents signifying that the water users' association had officially accepted the system. In systems developed by NIA outside the paticipatory program farmers often refused to accept the system--not a surprising response after their having been involved only minimally during its planning and design. Within the participatory projects there have been problems in reaching mutual agree- ment on the project expenses, the farmers' counterpart contribution, and on the NIA's construction obligations. A variety of methods have been developed for dealing with these needs. 1. Association construction stage committees. The water users' associa- tion develops several committees to handle specific construction stage needs. While the number and names of these committees vary from project to project, the functions performed include: a. Participating in canvass and bidding. Representatives from the water users' association observe the opening of bids for any outside contracts and provide the NIA information on where the lowest prices are available for locally purchased goods. b. Checking on quantity and quality of materials. Association members check on the quantity and quality of goods delivered to the project and try to curtail misuse. For example commitees in some projects impound - 32 - government vehicles after work hours to make sure they are not used for personal purposes, since the fuel is charged to the association. c. Recruiting and placing laborers. The association determines the number of skilled and unskilled laborers available from among its member and from the local communities, and notifies those people as to when and where they need to appear for work. Generally farmers prefer work on the parts of the system nearest their own farms. d. Recording counterpart contributions and project costs. The association works out how much each association member needs to contribute, specifies the ways in which these contributions will be made and records the members' contributions. They also keep their own records of project costs which they then check against the NIA records. 2. Preconstruction conferences. Shortly before construction begins, several preconstruction conferences are held, attended by a general assembly of the water users' association and the NIA staff. These conferences are designed for reaching agreements regarding the construction schedule, the way in which the association will make its 10% contribution, the policies for hiring and contracting labor, and the means by which the NIA staff will coordinate with each of the association's committees. 3. Construction contracts with the association. To insure efficient work NIA prefers to contract construction work on a "work completed" basis rather than paying a daily wage. Since the association members generally want to work in the project, one mechanism to allow this has been to negotiate a contract with the water users' association for much of the work. The associa- tion contributes its part by having a percentage of the contract payment withheld by the NIA, resulting in lower than normal daily wages for - 33 - those who work. One problem, however, has been delays in paying the workers. A commercial contractor would normally have "front money" to pay laborers on a daily basis--later receiving payment when the contract is completed. The water users' association, however, has no such funds. Experience has shown that the farmers can work about two weeks without pay, but after that become so desperate for money to feed their families that they leave in search of other work. Consequently efforts have been made to develop small contracts that can be completed, inspected, and paid for within two week cycles. 4. Periodic reconciliation of accounts. Once construction is over the NIA presents to the water users' association a "final statement of project accounts" which the association is supposed to sign as part of the project acceptance process. This includes a list of all the expenses for the entire project. Experience has shown that farmers often refuse to sign--partly because they are unsure of whether the statements are correct. A new approach now being implemented in many of the participatory projects is for the NIA to reconcile their records with the water users' association on a monthly basis so that the final statement becomes simply an aggregation of these monthly statements. Results of this innovation indicate that it helps insure clear understandings about project costs on all sides. 5. Finalization of the construction. Once construction is completed the system is tested. No matter how good the design, this test is likely to reveal the need for additional work. A problem however has been that often there are no additional funds available for finishing the system. There is also not always agreement between the NIA and the association on exactly what the NIA agreed to build. To deal with this latter problem, it is now being recommended that a list of the exact structures to be built be part of the - 34 - preconstruction memo of agreement signed between the NIA and the association, which can serve as a reference document for finalizing the project. Insuring that funds are available so that the project can be finished, accepted, and moved directly into operation and maintenance is an issue currently under examination. D. Operations and Maintenance Stage During the operation and maintenance stage a water users association must carry out four key functions: water allocation, system maintenance, con- flict resolution, and fee collection 1/. At the present time there is only one pilot project fully into the operation and maintenane stage; therefore, pilot project data revealing the associations needs in carrying out these functions is limited. However, a survey of 51 existing communal systems in the Philip- pines was conducted which focused on the management characteristics of those local associations (De los Reyes, 1980). The characteristics of the more effective associations help reveal the key needs of a water users' association during operation and maintenance. To effectively control water allocation in systems over 100 hectares, associations generally hired water tenders to work full time in opening and closing outlets and repairing parts of the system. During times of water shortage they instituted water rotation and sometimes planting schedules. For routine canal cleaning they mobilized their members on a regular basis at the sector level (areas usually covering about 30-40 hectares). Maintenance of a given canal was normally done by the farmers using that canal, mobilized by a sectoral leader. 1/ This typology of the basic functions of an irrigation association follows closely that used by Coward, 1980a. - 3% - Conflict between neighboring farmers over water use was usually resolved at the sectoral level--by the water tender, sectoral leader or other community leader in the area, while conflict between sectors was brought to the association's board or to village leaders. Fee collection was generally done immediately following the harvest by the water tenders. Disputes about the use of association funds appeared to be easily generated and disruptive to the associations. This highlighted an area of particular need regarding the NIA assisted systems, since these would be making amortization payments to the NIA. On the NIA assisted systems the association's irrigation fees would generally amount to about $25 a hectare per year (150 kilos of unhusked rice)--half for the amortization of the construction costs and half for the association's own expenses. This represents substantial sums given the poor status of most association members. The NIA's basic strategy for developing the capability of the water users association to carry out these four key functions is the involvement of the farmers in the preconstruction and construction activities. In addition to contributing to the general development of the association, these activi- ties develop specific mechanisms directly applicable to operating and maintain- ing the system. One is the development of the sectors as working units. In systems over 100 hectares, many of the preconstruction and construction activities are done primarily on a sectoral basis. Thus the sectors--so important for system maintenance, water allocation, conflict resolution and fee collection--are already functioning units when the system begins operating. A second is the counterpart contributions. The experience of mobilizing labor and recording counterpart contributions are directly parallel to the needs for - 36 - mobilizing labor for maintaining the channels once the system is operating. A third is the paddy mapping. While there is still only a little experience with this approach, the hope is that the paddy map will provide a basic tool for the association in making water allocation decisions and in assessing and collecting irrigation fees. A fourth is the handling of money. As the association collects its membership fees, records equity contributions, and handles labor contracts, it gains experience in financial management useful for its later fee collection activities. Helping associations develop record keeping systems for these preconstruction and construction activities that establish appropriate habits for the operation and maintenance stage is an area still under development. In addition to these preconstruction and construction stage activi- ties the NIA is developing its capability to provide three kinds of assistance during the operations and maintenance stage. 1. Water management course. Research is currently underway to develop the prototype for such a course. The approach will emphasize providing tools and rules of thumb to the association by which it can make decisions according to the needs dictated by water availability, crops and soil. The paddy map is likely to be a key tool. With this map painted on boards, association water tenders can learn to chalk in the planting activities during each crop season, providing the association decision makers with basic information on the water needs of the system. 2. Financial management course. Research is also underway to develop the prototype for this course. Experience has indicated that appropriate financial systems must take into account a number of special characteristics -37 - of a water users association--such as the geographical dispersion of its members and the difficulty of measuring the association's primary commodity-- water. 3. Continued community organizer assistance during initial operations. While the association operates as an entity independent of government, some continued support from the community organizer is provided during the first year of operations. This person helps the association prepare for and make use of the courses in water and financial management, and helps the associa- tion make contact with other relevant government programs, such as the exten- sion service. This review of the needs and responses during the feasibility, pre- construction, construction and operation and maintenance stages shows the extent and variety of issues being addressed in the participatory program. The problems are not all solved nor are the operational changes described fully utilized in all pilot projects. More developmental work is needed. But substantial progress has been achieved through close examination of the field level activtity and creative responses to the problems identified. E. Program Costs and Benefits Assessing the economic costs and benefits associated with the operational changes reviewed above is difficult. The clearest additional cost is that of the community organizers. To give an idea of these costs in comparison to construction costs, the most complete source of data are the two pilot projects in Southern Luzon initiated in 1979--the Taisan and Aslong projects. For these projects a nine month lead time was used and construction is now nearly completed. - 38 - Taisan is a 200 hectare system with estimated total project costs of $163,000; Aslong is a 350 hectare system with construction costs of about $93,000. Taisan is unusually expensive and Aslong unusually inexpensive-- a consequence of their respective topographies. Combining the two systems, they cost $256,000 for 550 hectares or about $465 per hectare 1/. Two community organizers worked on each system during the first 12 months of these projects, while only one organizer worked on each during the next 12 months. Thus for the 24 months of preconstruction and construction activity, the two systems used 72 person months of community organizer time at a total cost of $18,000. This represented $33 per hectare or 7% of project costs. Offsetting this cost are the contributions of the farmers to con- structing the system. The farmers in Taisan and Aslong made substantial contributions of labor and materials--far higher than on most projects on which community organizers for the participatory approach were not employed. As of July, 1981 the farmer contributions totalled approximately $10,000 with additional contributions expected as construction continued. At the NIA labor wage rate of just under $2.00 a day this represented the equivalent of over 5,000 person days of labor. If these contributions are substracted from the costs of the organizers, their total costs are reduced to $14 per hectare, or 3% of project costs. Another cost of the participatory approach is the additional time which technical staff spend on the project. Quantifying this is difficult, as no extra technical personnel were added to implement the participatory projects. While some aspects of the participatory approach demand more time of technical staff, others demand less. The engineers have to spend more 1/ The conversion rate used is 8 pesos = US$1.00. - 39 - time in meetings with farmers, but the farmers supply information about the area, build the warehouse, obtain the rights of way, recruit the workers, etc.--which reduces the work load of the technical staff. The engineers who have worked on these projects are not sure themselves whether it takes more or less time. The general view appears to be that it takes more time in the beginning and possibly less time later on. Another cost is the training programs--both for community organizers and engineers. The training needed for organizers involves three modules: a 3-week initial course on community organization, irrigation and preconstruc- tion activities; a two week course on construction activities; and a 1 to 2 week course (still to be developed) on the operations and maintenance activi- ties. How large a proportion of total costs will be devoted to training in the longer run will, of course, depend on the eventual turnover rates of organizers and engineers. Of course, the participatory approach cannot automatically guarantee that the association will operate and maintain the system well after con- struction. But the development of active and committed associations during the preconstruction and construction stages appears to strongly increase the likelihood of good operation and maintenance of the physical system by the farmers themselves. This will add to the benefits resulting from the participatory approach by increasing irrigation effectiveness and preventing or reducing certain possible subsequent costs for system maintenance and/or rehabilitation. _ 40 - Indeed, the economic benefits of the participatory approach should be evaluated not just in the short run, but rather long-term. Ultimately, the purpose of developing an irrigators' association is to insure that organiza- tional capacity exists to manage the system once construction is completed. An indicator of the value of the participatory approach is the evidence on whether such organizational capacity is actually created. While, again, it is still early to give a final assessment, currently available data are presented below. Of the two systems that are being carefully documented only t:he Aslong system was largely operational during 1981. NIA community organizers report that the Aslong irrigators' association is reasonably representative of associations being developed under the participatory approach and is definitely not a "special case". In fact they predict that there will be many associations assisted more recently under the participatory program which will turn out to be stronger than the Aslong association. Consequently initial operations and maintenance activities of the Aslong association may be viewed as reasonably typical of an association developed under the participatory approach. In March, 1981, part of the Aslong system became operational. For the wet season, which started in June 1981, approximately 285 hectares were planted to rice and irrigated by the system. Process documentation of the activities of the association are available for April-December, 1981. Canal repair was a key concern of the association, because the system suffered from heavy rains and several typhoons. The association called a total of 29 different "group work days" in different sectors of the system. For these days, members of the association (from the area served by the system to be repaired or cleaned) voluntarily provided their labor, totalling 392 person - 41 - days. To manage the distribution of water and for routine canal cleaning, the association employed 10 water-tenders on a part-time basis whose contribution totalled 450 days. For the association's management needs it employed an operations manager, treasurer, bookkeeper and fee collectors, all on part-time bases, totalling an estimated 378 days. Thus over the nine-month period approximately 1,220 person days were contributed in addition to materials purchased by the association totalling $360.00. How does this contribution compare to the investment that was made in developing the association? To make such a comparison, the contributions of the association, which were based on 9 months, were annualized and work days multiplied by the government wage for unskilled labor of $2.00 per day. This value was then divided by the 285 hectares being irrigated by the system during this period to arrive at a per hectare value. The result as shown in Table 2, is that the association averaged a contribution of over $12.00 per hectare. This is a conservative estimate of the association's contributions since it does not include time consumed in board meetings, general assemblies, sector meetings or more individual and non-formal activities. If the value is compared to the investment of $14.00 per hectare for the community organizer it can be seen that in slightly over one year the investment is recovered. While one cannot be sure of what would have happened without the participatory approach, the process documentation research reveals a clear and steady development of the Aslong association's leadership, membership commit- ment, and organizational structure and skills through the preconstruction and construction periods, which led directly to these operation and maintenance activities. Thus, it appears that the development of an active and committed association during the preconstruction and construction stages strongly increases the probability of subsequent good operation and maintenance. Table 2 ESTIMATED VALUE OF OPERATIONS AND MAINTENANCE ACTIVITIES OF THE ASLONG IRRIGATORS' ASSOCIATION4/ No. of Person Annualized Annualized Annualized Per Days Worked, Rate of Value of Work. Hectare Value of Type of Contribution Apr. -Dec. 1981 Days Worked @ $2. 00/day Contribution Maintenance Activities (Voluntary work carried out by Association members) 392 523 $1,045 $ 3.67 Water Distribution Activities (10 water tenders working 5 days each permonth) 450 600 $1,200 $4.21 Management Activities (Operations manager, 1/2 time; treasurer, 1/4 time; book- keeper 1/3 time; 5 fee collectors 1/5 time each) 378 504 $1,008 $ 3.54 Cash outlays of Association for canal repair materials, office equipment and supplies, taxes and licenses for 9 mos. = $270. 00 3__ 60__ $ 1.26 Total 1,220 1,627 $3,613 $ 12.68 Data from Volante and Illo, 1981 and personal communications from Illo, 1982. - 43 - V. A LEARNING PROCESS APPROACH TO BUREAUCRATIC REORIENTATION In 1976 when the NIA started experimenting with a participatory approach, there was little sense that it would lead to the array of opera- tional changes reviewed above (and summarized in Table 1). That realization emerged over a five year period during which there was created what has been termed a "learning process approach" to building the agency's capacity to develop water users' associations 1/. The challenge was not only to identify the changes needed in the field operations, but also to institutionalize them throughout the NIA. The very nature of small scale irrigation defies central organiza- tional control. Projects are scattered throughout the countryside, often in remote areas implemented by the NIA's 11 regional and 67 provincial irrigation offices totalling over 1,000 individuals. To implement the new approach these people would need to understand the new needs and adjust their work accord- ingly. The learning process developed to meet this challenge has a number of distinctive characteristics worth highlighting since they reveal the way in which the needed bureaucratic reorientation is being achieved. A. A National Level Committee To direct the learning process a national level Communal Irrigation Committee was formed, including a variety of NIA officials concerned with communals, and chaired and managed by NIA Assistant Administrator, Benjamin U. 1/ David C. Korten, 1980, has reviewed a variety of village level programs throughout Asia concluding that the most effective programs have gone through a "learning process" in which they have first learned to be effective in a very few villages, then learned to be efficient and later learned to expand their operations to cover increased numbers of villages. - 44 - Bagadion. This high level agency leadership proved to be crucial to the effectiveness of the learning process, as it allowed appropriate changes in agency procedures, policies, personnel and training to be made in response to field level experience. However, internal NIA resources alone were not sufficient. Because the NIA was trying to build capabilities which it did not yet possess, it invited a variety of academically based people to join agency personnel on the Committee. This Committee, formed in 1979, was comprised of individuals from the Institute of Philippine Culture (and later the Ateneo de Naga and the Development Academy of the Philippines), the Asian Institute of Management, the International Rice Research Institute, the University of the Philippines at Los Banos, and the Ford Foundation. All of the committee members had already demonstrated their ability to contribute to understanding the initial pilot project. They brought different disciplinary perspectives but were united by their common approach to program improvement. This approach (see Figure 3) focused on the problems experienced in the pilot projects. For each problem the questions were posed--what was the cause and was there anything the agency could do to avoid or alleviate it in future projects. Frequently the answer to the latter questions was "yes" but it would involve developing new training, procedures or methodologies. Often a committee member, together with colleagues of his or her institution, developed the needed methodology or training. This resulted in a close fit between the need identified and the research or training undertaken. Social scientists from the Institute of Philippine Culture, the Ateneo de Naga and the Development Academy of the Philippines carried out the process - 45 - Figure 3 Process Utilized for Program Improvement Examination ot| Determination of Development of field experience to agency capacities -| required identify problems required to avoid capacities or alleviate each problem Application of the Application of the required +- required capacities on a capacities on a national scale limited scale - 46 - documentation 1/; helped develop the profiling technology (De los Reyes and Borlagdan, 1980); trained NIA personnel to use profiles; and studied existing communal irrigation associations (De los Reyes, 1980). The management special- ists from the Asian Institute of Management designed and moderated most of the workshops and training sessions in which NIA engineers have been oriented to the participatory approach; are writing case studies and conceptual papers highlighting key implementation issues (Alfonso, 1980); and have developed the planning and monitoring system used in coordinating organizational and tech- nical work at the field level. The agricultural engineers from the Interna- tional Rice Research Institute and the University of the Philippines at Los Banos together with NIA officials developed the paddy mapping methods (Early and Bagadion, 1981); are currently conducting cost/benefit analyses of some projects; and are developing water management training for communals. Finan- cial management experts from a private consulting firm together with NIA personnel helped develop bookkeeping systems for communals. The community organizers with the NIA have developed the community organizer training. The array of agency and academically based individuals meet at least once a month (and often more frequently) to discuss next steps in the leaning process, to advise on changes in NIA policy and procedures, and to develop and attend the workshops that help spread the new approaches throughout the agency. 1/ Process documentation includes monthly reports on the Bagting-Siclong Communal Irrigation Association, November 1978-June 1979, by Elena Chiong- Javier and from January 1980 to June 1980 by Elena Chiong-Javier and Hector Namay; on the Taisan Communal Irrigation System, March 1979 to June 1981 by Jesus R. Volante, Jeanne Frances Illo, Manuel Diaz, and Perla Makil; on the Aslong Communal Irrigation System from March 1979 to May 1980 by Nestor Felix and Jeanne frances Illo; on the Siwaragan Communal Irrigation System from May 1980 to the present by Sylvia Maria G. Jopillo, Elena Chiong-Javier, and Fely David; and on the Buhi-Lalo Irrigation Project from January 1981 to the present by Jeanne Frances Illo, Elena Chiong-Javier and Nestor Felix. All process documentation received consulting assistance from Romana de los Reyes. - 47 - These functions have cast the academically based members of the com- mittee into unaccustomed roles. While their work is research, it is so related to action that it is different from "normal" academic research. Its output generally is not the "policy recommendations" so common to applied academic research, but rather tools that NIA personnel and farmers can then use to make better decisions. It has taken unusually action oriented people with a systems perspective to carry out these roles. B. Pilot Projects as Learning Laboratories The first pilot project in Central Luzon convinced top NIA officials of the value of pursuing the participatory approach. But much remained to be learned to define more clearly the intervention methodology and needed support systems. To achieve this two new pilot projects in Southern Luzon were initi- ated on which intensive documentation was carried out. A full time social science researcher was placed in each of the two sites to provide monthly reports on the activities and concerns of the farmers, the community organizers, and the engineers. This "process documentation" research has provided the Communal Irrigation Committee, and other interested individuals access to detailed knowledge of the projects from which the implications for NIA's broader program can be derived. C. Agency Based Implementation An important characteristic of the pilot projects is that they are implemented by the very agency expected to learn from them--the NIA. This contrasts with many pilot projects that are done by universities, research organizations, or private voluntary groups with the idea that once an appro- priate model is developed, some large agency can implement it on a broad scale. - 48 - This approach often fails because the skills, attitudes, and procedures of the implementing agency are inappropriate to the new approach 1/. In the Philippine communal irrigation work having the pilot projects based in the NIA allows the needed changes to be institutionalized as the program grows. It also places leadership for the change process within the agency. The NIA Assistant Administrator who chairs the Communal Irrigation Committee devotes substantial attention to examining the details of the field level problems, developing strategies for dealing with them and leading the many workshops and training sessions that develop the organization's skill and commitment to pursuing these improved approaches. D. Flexible Donor Support Over a five year period the Ford Foundation has provided one-half million dollars to the NIA specifically to facilitate the learning process; all construction costs of the pilot projects are covered by normal government appropriations for communal irrigation. The Ford Foundation contribution, matched by NIA funds, creates a special fund under the control of the Chairman of the Communal Irrigation Committee which is used to respond to the needs identified from the pilot projects. These funds are not pre-programmed for specific activities, since those activities can only emerge from the very learning process the fund is meant to support. For example, once it became apparent that a socio-technical profile technology was needed for assessing candidate communal systems, NIA drew on this special fund to hire a committee member to develop such a technology. Only after a few trials had 1/ See D. Korten and Uphoff, 1981, for a more extensive discussion of the difficulties of relating pilot projects done outside of the action agency to a larger implementation effort. - 49 - indicated that NIA personnel could use that technology to write useful profiles, did NIA offer a contract for refining the approach and training NIA personnel in its use. In addition to providing grant funds the Ford Foundation has com- mitted substantial staff time to help develop and articulate the frameworks which have guided the learning process, structure the roles of the various assisting institutions, and facilitate inter-institutional linkages. One program officer has been assigned nearly full time to these activities. The Foundation's long history of support to each of the assisting institutions has also been important in developing the capability of these institutions to contribute to the learning process. Three of them have had active Ford Founda- tion grants supporting related work. E. Phased Expansion Accompanied by Agency Changes The learning process has so far taken five years and will likely take another three before the changes are institutionalized throughout the agency and applied in all its communal irrigation assistance work. Table 3 summarizes the gradual expansion of the program. Table 3 Phased Expansion of the Participatory Projects Number of Approximate Number of Approximate number participatory hectarage co- workshops of NIA participants projects added vered by the and training in the workshops and Year in each year projects courses training courses 1976-78 1 400 1 25 1979 2 550 2 75 1980 12 1,800 6 150 1981 24 3,500 36 600 Total 39 6,250 45 850 - 50 - The time span can be viewed as containing three major phases 1/. The first involves learning to be effective. This was the period from 1976-78 when those involved in the first pilot project were developing an approach that that made sense to the villagers. With the initiation of two new projects in 1976 the program entered its second phase--learning to be efficient. While there was still substantial attention to learning to be effective, it was during these projects that many problems encountered in the first project were anticipated and avoided and the lead time prior to construction was reduced. In 1980 the program entered its third phase--learning to expand. One pilot project was established in each of the nation's 12 administrative regions, thus providing some experience with the new approach throughout the agency. In 1981 each region added two more participatory projects, for a total of 24 new projects. Plans for 1982 involved participatory projects in 60 provinces and by 1984 the NIA expected to carry out all of its communal irrigation work using the participatory approach. Three types of agency changes have accom- panied this expansion: 1. Substantial staff changes. While in 1976 NIA had no community organizers, by 1981 it had over 80 and expected to add hundreds more as the program continued to expand. The community organizers on the early pilot projec became the trainors and supervisors of the new community organizers thus allowing personal transmittal of past lessons. 2. Extensive training activities. For each new set of pilot projects, workshops have been held at 2 to 4 month intervals to orient both engineering and organizing staff to the new approach, review lessons for past projects and discuss implications for the specific projects being implemented. In addition 1/ These phases are elaborated in D. Korten (1980). - 51 - 3 and 4 week formal training courses have been held for community organizers and engineering personnel based on materials developed specifically for the NIA participatory program. Over the five year period approximately 850 agency personnel have participated in these workshop and training sessions. 3. Numerous changes in policies and procedures. The changes detailed in Section IV have been gradually implemented throughout the regional and provincial offices. The careful phasing of the expansion of the program has been important to insure that the attitudes, skills and knowledge of personnel as well as appropriate agency policies and procedures were developed to allow implementation of the new approach. The learning process is far from over. Examination of the pilot projects continues to reveal both new and persisting problems, calling for additional changes in staffing patterns, training, policies and procedures as well as research to provide the methodologies and insights for improved approaches. VI. IMPLICATIONS FOR DONOR PROGRAMMING This paper has reviewed one national agency's efforts to develop strong water users' associations on small scale irrigation systems. What are the potential lessons from this experience for donor agencies such as the World Bank and for in-country action agencies? To begin with, it is certainly significant that the' approach des- cribed in this paper has been adopted in a recently appraised irrigation project assisted by the World Bank in the Philippines - the Communal Irrigation Development project" (1982). This project will assist in financing NIA's program of communal irrigation development. It would finance the - 52 - construction of about 25%, by area, of the expected NIA program for communal irrigation works during the project period, adding about 3% to the existing area, and would also rehabilitate 3% of the present communal irrigation area. The implementation strategy for this Bank assisted project was designed so as to help promote the participatory approach in many more small-scale irrigation systems in the Philippines, replicating on wider areas the pilot experience described in this paper. Thus, specific lessons are being already derived and implemented for the immediate development of irrigated farming in the country. There are also, however, certain lessons of more general relevance. Before searching for lessons of general relevance, it must be stressed that the approach described was developed to fit the Philippine context. It cannot be transferred directly. But it can be instructive in at least two respects. First it helps highlight some of the program design and implementation issues in a program intended to assist the development of water users' associations. Second it helps reveal lessons regarding the programming of donor support for building agency effectiveness for village level work. A. Irrigation Program Design and Implementation Issues A number of aspects of the Philippine program may be relevant to irrigation program design and implementation in other settings. 1. The clear authority of the water users' association. The authority given a water users' association can range from "cleaning the ditches" to having full responsibility for operating and maintaining the entire system. While irrigation agency officials are frequently eager to develop water users' associations responsible for cleaning the ditches, some are more skeptical of empowering local associations with control over water. It is often thought - 53 - that water control should be left to an impartial government authority in order to avoid inequities in water allocation. A problem, particularly acute on small scale systems, is that government contact is likely to be so limited that the concept of government control is only a myth 1/. The myth, however, officialy removes from the local people the authority to run the system, leaving a vacuum where individualism is likely to reign. The issue of the appropriate degree of authority to give the local water users' association is one that will need to vary according to each program setting. However, it should be recognized that the less the authority of the local association, the weaker it is likely to be. Farmers are not likely to be eager participants in an association that is simply expected to clean the ditches. Chambers, 1976, suggests a useful guideline for determining the appropriate degree of authority: "In general government should unambiguously avoid doing that which communities can do for themselves in their own interest, but should intervene when exceptional problems are beyond a community's power to overcome." 1/ A brief visit I made to a small system in Indonesia highlighted the mythical nature of the government management of that system. I walked through the rice fields with the farmer leaders of the water users' asso- ciation to reach the diversion which the government had built under their "Sedarhana" irrigation program. Under this program once government assistance is extended to a system, a Public Works official is supposed to be in charge of the diversion structure and the main canal, while the farmers are in charge of the remainder of the system. Noting the large chain and padlock on the wheels controlling the flood-gate and the gate to the main canal, I inquired as to who keeps the key to that padlock. The farmers dutifully replied that it was the Public Works official. I wondered aloud what they did when there were sudden heavy rains and they needed to close the canal gate and open the floodgate quickly. They assured me that this was no problem since the association also had two keys to the padlock kept by the leaders of the two subsectors closest to the diversion. They noted that since the Public Works man was eight kilometers away he couldn't possibly get to this area quickly enough to open the gate--especially in heavy rains. They then cited a variety of other occasions on which they opened and shut the gates, clearly indicat- ing it was the local association that managed the diversion weir. - 54 - One of the strengths of the Philippine communal irrigation program is its clear policy regarding the authority of the local associations. For each system under 1,000 hectares, there is one association which has respons- ibility for operating and maintaining the entire system. The associations are legal entities with water rights registered in the association's name. While the NIA might occasionally step in to help resolve disputes, it is the water users- association that runs the system. This provides the farmers a clear rationale for committing their time and energy to developing their association. 2. The use of existing irrigation groups. In developing a water users- association organizers can completely ignore existing groups or explicitly build from them. In some places, this is not an issue as there are no exist- ing irrigation groups. But in many places, particularly in the humid tropics, there are likely to be many such groups l/--even though irrigation agencies tend to be blind to them. A number of factors encourage this blindness. One is the target systems for generating irrigated hectarage common to many irrigation agencies. Since the agency prefers to assume it is generating new hectarage rather than improving on existing irrigation systems, it ignores existing irrigation groups. A second is the lack of data about these local groups. Even when local socio-economic data are collected, they are usually based on individual farmers, reporting farm size, crops grown, and yields--and hence do not reveal the local organizational arrangements for irrigation. Consequently existing irrigation groups may be overlooked. Ignoring existing l/ For example, in Indonesia 20% of the currently irrigated land is served by what are referred to as "village systems". Another 20% is irrigated by "sedarhana" (simple systems) which are also managed primarily by villages. - 55 - groups may entail considerable waste. While these groups sometimes do not operate as effectively as the government would like, they have stood the test of time and may have evolved useful traditions and leadership roles which would not easily be replaced. The participatory projects of the Philippine communal irrigation program have explicitly tried to build on existing irrigation groups. It is normally the local irrigation group that makes the request for assistance from the government in the first place. The socio-technical profile then highlights the current irrigation arrangements in the area, providing the NIA technical and organizational team the basis for making maximum use of the beneficial aspects of those arrangements. 3. The requirement of a contribution from the water users' association to the costs of the system. In many irrigation development programs farmers are not expected to contribute to the capital costs of the system. While generally it is not considered feasible or appropriate to require farmers to cover the full costs of the irrigation system, the Philippine experience reveals beneficial effects of requiring some contribution. The exact arrange- ments would need to vary from place to place. But the concept of farmers "buying" their own system, even at heavily subsidized rates helps reinforce the importance of the water users' association. In the Philippine program the fact that agency personnel are expected to elicit a 10% counterpart contribu- tion and subsequent repayment of construction costs encourages agency personnel to concern themselves with developing a water users' association. Similarily from the farmers' side, the requirement that they make a 10% contribution serves as an immediate and tangible organizational task for which the associa- tion must mobilize. The arrangement also bestows greater equality on the - 56 - relationship between the agency and the local people. The counterpart contri- bution requirement gives the local people the power to withhold that contribu- tion, providing them some leverage over agency personnel not available in a purely "dole out" program. 4. The involvement of the water users association in the planning, layout and construction of the system. A common assumption of many irrigation agency personnel is that there is no reason to develop a water users associa- tion until there is water to use. If this assumption is followed, then organizers are not fielded until after the system is built. Thus, the local people are denied the possibility of being involved in the system development. The Philippine experience reviewed in this paper, as well as the experience of numerous indigenous irrigation systems around the world, reveals how important such early involvement can be to developing the capacities of water users associations to manage and maintain the system 1/. 5. The gradual development of agency capacity to implement its strategy for developing water users associations. The above listed program design dimensions are likely to contribute to developing strong water users associations. But a program design is meaningless unless the implementing agency can carry it out. The Philippine experience helps reveal some key questions that might be raised regarding implementational capacity. (These do not deal with the purely technical issues, though, of course, capacities for engineering and construction are crucial.) 1/ The FAO (1980) has recently stresssed farmer involvement in planning and construction as a basic principle that should be followed for all irriga- tion development. - 57 - a. Does the agency have a "rapid assessment" technology by which it can learn about the current irrigation arrangements and other operationally relevant social characteristics of the people to be affected by the irrigation system? Does it have a way of using that knowledge in developing the irrigation system? Does it have a way of reasonably accurately estimating the area to be irrigated in a proposed project? b. Does the agency have specialized personnel who can develop the water users' association, working closely with the technical people? Is there a training program that develops such field workers' capabilities for the specific task of developing water users' associations? c. Is there a clear framework for relating the organizational and the technical work at each stage of a project so that the field level technical and organizational staff and the farmers understand their respective roles? d. Have the procedures that the agency requires of its field staff been closely examined to make sure they fit the operational requirements of the strategy for developing strong water users' associations? e. Are there training programs to help develop among technical staff the attitudes, skills and knowledge suited to the strategy? For most irrigation agencies the answers to most of these capacity questions would be "no". Some programs, facing this situation, have turned to a different agency to do the organizing work--an agency presumed to have greater knowledge of people than the irrigation agency. In some situations this may be the most appropriate answer to the problem. However, it has the inherent weakness that it divorces the organizational and technical work, making it difficult to involve the farmers in the key questions regarding the planning, design, construction and operation of the system-- and hence severely weakens the possibilites of developing strong water users' associations. - 58 - A different response to an initial lack of capacity to implement the preferred program design is the one described in this paper in which the NIA is developing the needed capacity by engaging in a learning process. This approach leads to the second set of issues for a donor agency. B. Project Cycle Issues The programmatic requirements of pursuing a learning process approach for improving village level work pose difficult issues for large donor agencies. 1. A focus on building program capability. A fundamental characteristic of village level work is that projects are dispersed over a wide geographical area with each having unique characteristics requiring on-the-spot judgements. Consequently efforts to improve the implementation must be thought of in terms of building the capacities of many local level offices charged with implementing those projects. The decentralized nature of the work makes particularly critical the careful development of policies, organizational structures, procedures, and training that can systematically affect the behavior of large numbers of agency staff in dispersed locations in ways consonant with the program strategy. Lack of such capacity cannot be overcome by bringing in a consulting firm or hiring a few high powered individuals to manage the "Project",--which is actually a program comprised of many individual projects. The time consuming task of developing agency-wide program capacity must be faced. 2. Five-year period prior to major external funding. The NIA's effort to build its own capacity to develop strong water users' associations began in 1976 and is only now approaching the point where it can absorb a substan- tial block of donor funding assistance. If large amounts of external funding - 59 - had been forced on the program too soon, the demands of normal project adminis- tration - procurement, funding accountability, meeting construction deadlines, etc. -- would likely have driven out learning. The systems and trained per- sonnel required to support the new approach would not have been sufficiently developed to allow its broader application. Long before a major loan is made, the learning process needs to be initiated within the implementing agency. The exact number of years required will depend on many factors, but the Philippine experience reveals that five years might be viewed as a minimum. 3. Combination of research and action projects. The learning process requires a combination of action and research aimed at building agency capaci- ties. Studies, apart from action, can make some contribution, particularly, studies of the operations of existing projects similar to the ones to be developed 1/. This can provide important background information for determin- ing what kind of new arrangements should be tried to improve the program. But a key part of the learning must come through action. Just as a person cannot learn to swim by making socio-economic studies of swimmers, an agency cannot learn to implement effective village level projects by making socio-economic studies of villagers. It must actually try to implement a strategy and examine its problem in doing this, drawing on the most creative people it can find to help understand the implications of these problems for its strategy, procedures, staffing, structure and policies. This requires funds for action as well as research -- action that gradually expands as the agency moves through the stages of learning to be effective, to be efficient, and to expand. 1/ In a World Bank supported tank irrigation project in the Indian state of Karnataka, studies of some existing tank systems are being carried out by the Indian Institute of Management at Bangalore to determine lessons appropriate to future tank irrigation development. - 60 - 4. Flexible funds with intensive staff support. The relatively modest funds allocated for this early work need to be programmed very loosely so that they can be used to support new needs as they become evident from the learning process. Rather than depending on preprogrammimg as the means for determining how the funds will be spent, the learning process requires committed, disciplined leadership within the agency and intensive donor staff support as the means for determining the use of these funds. These four characteristics suggest the need for an approach quite different from the way large donor agency loans are normally developed. To allow a five year lead time requires initiating a learning process at a stage when a program may have only been identified as a tentative candidate for a later loan. It requires initially specifying only the general problem to be addressed and the means for addressing it, rather than listing the particular activities to be carried out. It requires identifying agency leadership for the long term change effort and then providing funds and donor agency staff time to support the learning process. By the time of appraisal for the major loan a key task would be to determine how far the learning process has developed and to shape the loan in ways that promote the continued development of the agency's capacities to implement improved projects irrespective of the source of their funding. An important advantage of such an approach is that it means the loan not only supports some specific projects, but also leaves in place an agency capacity to go on developing additional projects as the country's need requires. While developing a loan in this manner would require substantial adjustments, they may be worth making. As Griffin and Ghose, 1979, concluded in their review of growth and impoverishment in the rural areas of Asia, - 61 - The centralized planning procedures existing at present are biased in favour of large projects designed to high engineering standards, amenable to sophisticated cost-benefit analysis and perhaps suitable for foreign financing. Anti-poverty focused rural development, however, requires a great many small, labour- intensive investment projects that are dispersed throughout the countryside. These projects cannot be identified, designed, and appraised at the centre; of necessity, projects of the type required must be identified locally and their desirability and feasibility determined there. This in turn implies that plans for rural development should be made locally, by those who will implement them, benefit from them and bear the major cost of them. Government, of course, has a major role to play in helping rural people to plan their own activities, to articu- late their demands and to become organized. The poor are unlikely to organize themselves spontaneously; some outside stimulus normally will be necessary and government can assist the process ... (viewing) local organizations as partners in rural development, sources of ideas, places where priorities can be hammered out, recruitment grounds for "barefoot" doctors and engineers, peasant leaders and local government personnel. In this way it should be possible to approach the ultimate objective in which planning for the poor becomes planning by the poor. - 62 - The development strategy implied in this statement involves a funda- mental shift in the role of many government bureaucracies. Rather than being a provider of services to people expected to play a passive receiver role, government becomes an enabler, creating the social and physical infrastructure that allows people actively to manage their own development 1/. Making this shift is a major challenge. The NIA experience helps reveal both the difficul- ties and the promise of the effort. 1/ D. Korten (1981) has discussed a variety of contexts in which such bureaucratic shifts are needed, drawing implications for both the economic planning process as well as implementational needs. A recent edited volume Bureaucracy and the Poor: Closing the Gap (D. Korten and Alfonso, 1981) also discusses issues involved in making such shifts in a variety of programs and countries. - 63 - REFERENCES Alfonso, Felipe B. 1981. "Assisting Farmer Controlled Development of Communal Irrigation Systems." Bureaucracy and the Poor: Closing the Gap. Singapore: McGraw-Hill. Bagadion, Benjamin and Frances F. Korten 1980. "Developing Viable Irrigators' Associations: Lessons from Small Scale Irrigation in the Philippines." Agricultural Administration. Vol. 7, pp. 273-287. Bromley, Daniel W., Donald C. Taylor, and Donald E. Parker 1980. "Water Reform and Economic Development: Institutional Aspects of Water Management in the Developing Countries." Economic Development and Cultural Change. Vol. 28, pp. 265-387. Chambers, Robert 1977. "Men and Water: The Organization and Operation of Irrigation." In B. H. Farmer, ed., Green Revolution?: Technology, and Change in Rice-growing Areas of Tamil Nadu and Sri Lanka. Boulder, Colorado: Westview Press, pp. 250-63. (Reprinted in Coward, 1980). Coloma, Romulo 1981. Personal communication. Colombo, U., D. G. Johnson, and T. Shishido 1978. Reducing Malnutrition in Developing Countries: Increasing Rice Production in South and Southeast Asia. The Trilateral Commission. Coward, E. Walter 1977. "Irrigation Management Alternatives: Themes from Indigenous Irrigation Systems." Agricultural Administration. Vol. 4, pp. 223-57. (Reprinted in Coward, 1980). Coward, E. Walter (ed.) 1980 Irrigation and Agricultural Development in Asia: Perspectives from the Social Sciences. Ithaca: Cornell University Press. Coward, E. Walter 1980a. "Irrigation Development: Institutional and Organizational Issues." In Irrigation and Agricultural Development in Asia: Perspectives from the Social Sciences. Ithaca: Cornell University Press. Pp. 15-27. De los Reyes, Romana 1980. Managing Communal Gravity Systems. Quezon City, Philippines: Insititute of Philippine Culture. De los Reyes, Romana and Salve Borlagdan 1981. "Formats for Interview, Write-up and Analysis of Communal Irrigation Project Profiles." Quezon City, Philipines: Institute of Philippine Culture. (To be reprinted in Irrigation Management Network Papers, London: Overseas Development Institute, November, 1981). Early, Alan C. and Benjamin U. Bagadion 1981. "Custom Fit Design of Farm Ditches: A Participatory Approach to Making Irrigation Systems Responsive to the Needs of the Farmers." Los Banos, Philippines: International Rice Research Institute, mimeo. - 64 - Food and Agriculture Organization 1980. Report on the Expert Consulation on Farm Water Management. Rome: Food and Agriculture Organization. Food and Agriculture Organization 1981. "Information Paper on the Interna- tional Support Programme for Farm Water Management." Rome: Food and Agriculture Organization, mimeo. Geertz, Clifford 1967. "Tihingan: A Balinese Village." In Koentijaraningrat, ed., Villages in Indonesia. Ithaca; Cornell University Press. (Reprinted in Coward, 1980). Griffin, Keith and Ajit Kumar Ghose 1979. "Growth and Improverishment in the Rural Areas of Asia." World Development. Vol. 7, pp. 261-383. Illo, Jeanne Frances 1980. The Farmers in Communal Gravity Systems: Rice Yields, Work, and Earnings. Quezon City, Philippines: Institute of Philippine Culture. Isles, Carlos D. and Manuel L. Collado 1979. " Farmer Participation in Communal Irrigation Development: Lessons from Laur." Philippine Agricultural Engineering Journal. Vol. 10, pp. 3-4, 9. Korten, David C. 1980. "Community Organization and Rural Development: A Learning Process Approach." Public Administration Review. Vol. 40, pp. 480-511. Korten, David C. 1981. "The Management of Social Transformation at National and Sub-national Levels." Public Administration Review. November- December. Korten, David C. and Felipe B. Alfonso (eds). 1981. Bureaucracy and the Poor: Closing the Gap. Singapore: McGraw-Hill. Korten, David C. and Norman Uphoff 1981. "Bureaucratic Reorientation for Rural Devlopment." Manila and Ithaca: Asian Institute of Management and Cornell University, mimeo. Lazaro, R. C., D. C. Taylor, and T. H. Wickham 1979. "Irrigation Policy and Management Issues: An Interpretive Seminar Summary." In D. C. Taylor and T. H. Wickham (eds.) Irrigation Policy and the Management of Irrigation Systems in Southeast Asia. Bangkok: Agricultural Development Council. pp. 2-12. Lowdermilk, M. K., A.C. Early, and D. M. Freemen 1978. Farm Irrigation Constraints and Farmers' Responses: Comprehensive Field Survey in Pakistan. Fort Collins: Colorado State University. Moya, T. B. and A. C. Early 1980. "Some Findings on Water Distribution with the Tertiaries of a Gravity Irrigation System." Los Banos, Philippines: International Rice Research Institute, mimeo. - 65 - Reidinger, Richard B. 1974. "Institutional Rationing of Canal Water in Northern India: Conflict Between Traditional Patterns and Modern Needs." Economic Development and Cultural Change. Vol. 23, pp. 79-104. Volante, J. R. and J. F. I. Illo, 1981, "RSC-NIA Documentation Research on the Operation and Maintenance of the Aslong and Taisan Communal Irrigation Systems in Camarines Sur," Naga City, Camarines Sur: Ateneo de Naga (mimeo). Wortd Bank World Bank Staff Working Faper No. and cooperation, the World Bank is 489. November 1981. About 50 pages. publishing three volumes on Niblicatlons Stock No. WP-0489. $5.00. appropriate technology for water sup- Stock No. | , ' ' 'ply and waste disposal systems in f Related Interrelationships In developing countries. Since the Energy Plannng: The Case technology for supplying water is Interest Energy Planning: The Case better understood, the emphasis in of the Tobacco-Curing these volumes is on sanitation and Industry in Thailand waste reclamation technologies, their Gunter Schramm and contributions to better health, and Mohan Munasinghe how they are affected by water ser- vice levels and the ability and willing- ness of communities to pay for the World Bank Reprint Series: Number sstems. 187. Reprinted from Energy Systems systems. and Policy, vol. 5, no. 2 (1981):117-39 Number 1: Appropriate Stock No. RP-0187. Free of charge. Sanitation Altenatives: A Technical and Economic Mobilizing Renewable Appraisal lectricity Pricing: Theory Energy Technology in John M. Kalbermatten, md Case Studies Developing Countries: DeAnne S: Julius, and 1ohan Munasinghe and Strengthening Local Charles G. Gunnerson eremy J. Warford Capabilities and Research This volume summarizes the techni- lescribes the underlying theory and Focuses on the research required to cal, economic, environmental, ractical application of power-pricing develop renewable energy resources health, and sociocultural findings of olicies that maximize the net in the developing countries and on the World Bank's research program conomic benefits to society of the need to strengthen the develop- on appropriate sanitation alternatives lectricity consumption. The ing countries' own technological and discusses the aspects of pro- iethodology provides an explicit capabilities for using renewable gram planning that are necessary to amework for analyzing system costs energy. (One of three publications implement these findings. It is nd setting tariffs, and it allows the dealing with renewable energy directed primarily toward planning iriff to be revised on a continual resources and issues in developing officials and sector policy advisers asis. Case studies of electricity pric- countries. See Alcohol Production for developing countries. ig exercises in Indonesia, Pakistan, from Biomass in the Developing The Johns Hopkins University Press, ie Philippines, Sri Lanka, and Countries and Renewable Energy March 1982. About 208 pages (includ- hailand describe the application of Resources in the Developing ing bibliography, index). ie methodology to real systems. Countries.) LC 80-8963. ISBN 0-8018-2578-4, he Johns Hopkins Uniuersity Press. July 1981. iv +52 pages. $12.95 (49.10) paperback. larch 1982. About 400 pages (includ- Stock No. EN-8101. $5.00. Number 2: Appropriate ig appendixes, index). Principles of Modern Sanitation Alternatives: C' 81-47613. ISBN 0-8018-2703-5, Elcrincipe ofModeng A Planning and Design 22.50 (,f15. 75) hardcover. ElecyMohan Munasinghe Manual John M. Kalbermatten, World Bank Reprint Series: Number DeAnne S. Julius, ilobal Energy Prospects 185. Reprinted from Proceedings of the Charles G. Gunnerson, and Woum Jone Choe, IEEE, vol. 69, no. 3 (March D. Duncan Mara ,drain Lambertini, and 1981):332-48. This manual presents the latest field 'eter K. Pollak Stock No. RP-0185. Free of charge. results of the research, summarizes background study for World World Bank Series in Water selected portions of other publica- 'evelopment Report, 1981. Examines Supply and Sanitation tions on sanitation program plan- djustments in the use of energy that ning, and describes the engineering ave taken place in five major country The United Nations has designated details of alternative sanitation roups since the 1973-74 oil price the 1980s as the International Drink- technologies and how they can be icrease. Based on these trends, ing Water Supply and Sanitation upgraded. nergy conservation and efficiency, Decade. Its goal is to provide two of The Johns Hopkins University Press, nd assumptions about income levels the most fundamental human March 1982. About 224 pages (includ- nd of a 3 percent annual increase in needs-safe water and sanitary Mac 1982.gabout 24pesiu !al energy prices, the paper provides disposal of human wastes-to all Ing bibliography, index). nergy projections by major fuels for people. To help usher in this impor- LC 80-8963. ISBN 0-8018-2584-9, iese country groups in the 1980s. tant period of international research $15.00 (£10.50) paperback. - 66 - Number 3: Sanitation and World Bank by The Johns Hopkins Economic Criteria for Disease: Health Aspects of University Press. Optimizing Power Systen Wastewater Management November 1980. Reliability Levels Richard Feachem and others Stock No. WS-8000. Free of charge. Mohan Munasinghe and Available in late 1982. Mark Gellerson World Bank Reprint Series: Numbe Low-Cost Technology Acohol Production 112. Reprinted from The Bell Journ Options for Sanitation- from Biomass in the Economics. vol. 10, no. 1 a State-of-the-Art Review Developing Countries (1979):353-365. and Annotated Bibliography Explains the techniques for manufac- Stock No. RP-0112. Free of charge. Witold Rybczynski, turing ethyl alcohol from biomass raw materials; analyzes the Chongrak Polprasert, and economics of and prospects for pro- The Economics of Power Michael McGarry duction and government policies System Reliability and A comprehensive bibliography that needed to accommodate conflicting Planning: Theory and Cai describes alternative approaches to needs of various sectors of the the collection, treatment, reuse, and economy in promoting production; Studies disposal of wastes. and discusses the role the World Mohan Munasinghe A joint World Bank/lInternational Bank can play in assisting developing A completely integrated treatmen Develpmen Resarch entr pubica- countries in designing national system reliability. Indicates that tiveon.m1978 Availablrom Ientrernatlional alcohol programs. (One of three application of the reliability optimi tion. 1978 Available fror International publications dealing with renewable tion methodology could help real Development Research Centre (IDRC), energy resources in developing coun- considerable savings in the electr P.O. Box 8500, Ottawa K1 G 3H9, tries. See Mobilizing Renewable power sector, which is especially Ontario (Canada). Energy Technology in Developing portant for developing countries i Countries: Strengthening Local limited foreign exchange reserves Capabilities and Research and Low-Cost Water Distribu- Renewable Energy Resources in the The Johns Hopkins University Pre& tion-a Field Manual Developing Countries.) 1980. 344 pages (including tables, Charles Spangler September 1980. ix + 69 pages maps, index). Provides nonengineers with the infor- (including 12 annex figures). LC 79-2182. ISBN 0-8018-2276-9, mation necessary to design and Egih rnh n pns.$75 X67)hrcvr implement simple water distribution Portngls (forth n Sn). ISB7 0-8018-2277-7, $12 50 (v6.25 systems for small communities. uguese coming B AvaIlable in 1982. Stock Nos. EN-8002-E, EN-8002-F, paperback. Aval able in 1982. EN-8002-S, EN-8002-P. $5.00. - Stock No. WS-8001. Free of charge. Electricity Economics: Essays and Case Studies Technical and Economic Costs Incurred by Residen Ralph Turvey and Options tial Electricity Consumers Dennis Anderson John M. Kalbermatten, Due to Power Failures Argues the merits of relating the DeAnne S. Julius, and ohan unasinghe price of electricity to the margina Charles G. Gunnerson World Bank Reprint Series: Number incremental cost of supply and di Reports technical, economic, health, 128. Reprinted from Journal of with interactions between pricing and social findings of the research Consumer Research, vol. 6 (March investment decisions, income project on "appropriate technology" 1980).361-369. distribution, and distortions in th and discusses the program planning pricing system of the economy necessary to implement technologies Stock No. RP-0128. Free of charge. The Johns Hopkins Universitj Pres available to provide socially and 1977. 382 pages (including tables, environmentally acceptable low-cost maps, index). water supply and waste disposal. The Economic Choice maps, index). December 1980. between Hlydroelectric LC 76-9031. ISBN 0-8018-1866-4, and Thermal Power $30.00 (,X13.50) hardcover; Stock No. WS-8002. Free of charge. Developments ISBN 0-8018-1867-2, $12.95 (,5.75 Herman G. van der Tak paperback. World Bank Research in A logically correct method for han- French: L'economie de l'electricite: Water Supply and Sanita- dling the economic comparison of essais et etudes de cas. Economica tion-Summary of Selected alternative systems. 1979. Publications The Johns Hopkins University Press, ISBN 2-71 78-0165-0, 58 francs. A bibliography summarizing the 1966; 4th printing, 1974. 80 pages Spanish: Electricidad y economia: papers in the Water Supply and Sanitation Series, as well as the (including 2 appendixes). ensayos y estudios de casos. Editon World Bank studies in Water Supply LC 66-28053. ISBN 0-8018-0646-1, Tecnos, 1979. and Sanitation published for the $5.00 (£3.00) paperback. ISBN 84-309-0822-6, 710 pesetas. - 67 - ;ctric Power Pricing Policy Integrated National Energy Renewable Energy han Munasinghe Planning in Developing Resources in the Id Bank Staff Working Paper No. Countries Developing Countries July 1979. 56 pages (including Mohan Munasinghe Examines the contributions that )pendixes, references). World Bank Reprint Series: Number renewable resources can make to 'k No. WP-0340. $3.00. 165. Reprintedfrom Natural Resources energy supplies in developing coun- Forum, vo. 4 (1980:359-73.tries and discusses the role of the Forum, vol. 4 (1980J:359-73. World Bank in renewable energy ergy in the Developing Stock No. RP-0165. Free of charge. development over the next five years. untries (One of three publications dealing usses the energy problemofthe Municipal Water Supply with renewable energy resources and :usses the energy problem of the Project Analysis: Case issues in developing countries. See t decade. States that by tapping Studies Alcohol Production from Biomass -rves of oil, gas, coal, and Frank H. Lamson-Scribner, Jr., in the Developing Countries and roeectic nd oret rsouces and John Hiuang, editors Mobilizing Renewable E~nergy roelectric and forest resources Technology in Developing Countries: biously considered uneconomical Eight case studies and fourteen exer- Strengthening Local Capabilities mporting countries could halve cises dealing with the water and and Research.) r oil-import bill by 1990. Outlines wastewater disposal sector. November 1980. iv + 33 pages isures for saving energy without World Bank (EDI), 1977. ix + 520 (including annex). ucing economic growth and pages. (Available from ILS, 1715 Con- Stock No. EN-8003. $5.00. orts industrialized and indus- necticut Avenue, N.W., Washington, lizing countries to adjust energy D.C. 20009, U.S.A.) Rural Electrification :es, incentives, and investments *mphasize domestic production. $8.50 paperback. Discusses the prospects for success- poses a World Bank program for A New Aproach to Power ful investment in rural electrification :rgy lending and explains the s P a and considers implications for World ,rational aspects of this program. ystem Plannsng Bank policies and procedures. rust 1980. vi + 92 pages (includ- Mohan Munasinghe A World Bank Paper. October 1975. 80 5 annexes). World Bank Reprint Series: Number pages (including annex). English, 147. Reprinted from IEEE Transactions French, and Spanish. -k No. EN-8001. Free of charge. on Power Apparatus and Systems, Stock Nos. PP-7505-E, PP-7505-F, PAS-99, no. 3 (1980):1198-1206. PP-7505-S. $5.00. ergy Options and Policy Stock No. RP-0147. Free of charge. ,ues in Developing The Oil Price Revolution of Village Water Supply ountries 1973-74 Describes technical aspects, costs, rrel G. Fallen-Bailey and 1973-74 and institutional problems related to wvor A. Byer Salah El Serafy supplying water for domestic use in rld Bank Staff Working Paper No. World Bank Reprint Series: Number rural areas and proposes guidelines rld ank taffWoring aperNo. 119. Reprinted from The Journal of for future World Bank lending in I. August 1979. vi + 107 pages. Energy and Development, vol. 4, no 2 this sector. ck No. WP-0350. $5.00. (Spring 1979):273-290. A World Bank Paper. March 1976. 98 Stock No. RP-0119. Free of charge. pages (including 4 annexes). English, dia: The Energy Sector French, and Spanish. D. Henderson Planning for Electrical Stock Nos. PP-7602-E, PP-7602-F, Power: Costs and n rncbn -nmary review of the sector, pro- Technologies PP-7602-S. $5.00. ing technical, historical, and tistical background information. Mohan Munasinghe Village Water Supply: Word Bnk eprntSeres:Numer Economics and Policy in the ord University Press, 1975. 210 World Bank Reprint Series: Number Developing World res (including map, 2 appendixes, 149. Reprinted from Modern Govern- Robert J. Saunders Wx). mentUN"(ational Development and Jeremy J. Warford N 019-560653-1, $12.95 (1980):75-84. .25; Rs3O) hardcover. Stock No. RP-0149. Free of charge. Addresses the problem of potable water supply and waste disposal in Prospects for Traditional rural areas of developing countries L Integrated Framework and Non-Conventional where the majority of the poor tend r Energy Pricing in Energy Sources in to be found. Emphasizes the Wveloping Countries Developing Countries economic, social, financial, and ,veloping Countries Developing Countries administrative issues that character- )han Mlunasinghe David r. Plughart ize village water supply and sanita- rld Bank Reprint Series: Number World Bank Staff Working Paper No. tion programs. B. Reprinted from The Energy Journal 346. July 1979. ii + 132 pages The Johns Hopkins University Press, io. 3 (1980):1-30. (including 2 annexes). 1976. 292 pages (including 4 appen- ck No. RP-0148. Free of charge. Stock No. WP-0346. $5.00. dixes, bibliography, index). - 68 - LC 76-11758. ISBN 0-8018-1876-1, $21.00 (f12.70) hardcover. French: L'alimentation en eau des communautes rurales: economie et politique generale dans le monde en developpement. Economica, 1978. ISBN 2-7178-0022-0, 45 francs. Spanish: Agua para zonas rurales y poblados: economia y poliftica en el mundo en desarrollo. Editorial Tecnos, 1977. ISBN 84-309-0708-4, 575 pesetas. - 69 - HG3881.5 .W57 W67 no.528 c.3 Korten, Frances F. Building national capacity to develop water users' associations : experience DATE NAME AND EXTENSION ROOM NUMBER The World Bank Headquarters: D 2 U 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. Telephone: (202) 477-1234 Telex: WUI 64145 WORLDBANK RCA 248423 WORLDBK Cable address: INTBAFRAD WASHINGTONDC European Office: 66, avenue d'Ilna 75116 Paris, France Telephone: 723.54.21 Telex: 842-620628 Tokyo Office: Kokusai Building 1-1, Marunouchi 3-chome Chiyoda-ku, Tokyo 100, Japan Telephone: 214-5001 Telex: 781-26838 ISSN 0253-2115/ISBN 0-8213-0051-2