Philippines - Communal Irrigation Development Project

Document of The World Bank L FOR OFFICIAL USE ONLY Report No. 3850-PH STAFF APPRAISAL REPORT PHILIPPINES COMMUNAL IRRIGATION DEVELOPMENT PROJECT May 12, 1982 Agriculture Division II Projects Department East Asia and Pacific Regional Office This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EQUIVALENTS Currency Unit - Philippine Pesos (P) P 1 = US$0.123 US$1 = P 8.1 SDR I = US$1.111 WEIGHTS AND MEASURES 1 ha = 2.47 acres I m = 3.28 ft 1 km = 0.62 miles I sq m = 10.76 sq ft 1 cu m = 35.31 cu ft 1 mm = 0.039 in 1 kg = 2.2 lb 1 cavan = 50 kg 20 cavans = 1 metric ton (referred to as ton in this report) ABBREVIATIONS AFT - Agriculture and Food Technician CIDP - Communal Irrigation Development Project CIS - Communal Irrigation System CO - Community Organizer COA - Commission of Audit FSDC - Farm Systems Development Corporation GDP - Gross Domestic Product IIYV - High Yielding Variety IA - Irrigators' Association IFAD - International Fund for Agricultural Development IRRI - International Rice Research Institute at Los Banos MAR - Ministry of Agrarian Reform M&E - Monitoring and Evaluation MOA - Ministry of Agriculture MPWH - Ministry of Public Works and Hlighways NEDA - National Economic and Development Authority NFA - National Food Authority NFAC - National Food and Agricultural Council NIA - National Irrigation Administration NWRC - National Water Resources Commission O&M - Operation and Maintenance OED - Operations Evaluation Department PDD - Project Development Department PIE - Provincial Irrigation Engineer  FOR OFFICIAL USE ONLY PIO - Provincial Irrigation Office PPAR - Project Performance Audit Report RID - Regional Irrigation Director RIO - Regional Irrigation Office SEC - Securities and Exchange Commission TSP - Triple Super-Phosphate GLOSSARY Palay = Threshed Unmilled Rice Barangay = Village GOVERNMENT OF THE PHILIPPINES FISCAL YEAR January 1 - December 31 |This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.   PHILIPPINES COMMUNAL IRRIGATION DEVELOPMENT PROJECT Table of Contents Page No. 1. BACKGROUND. . . . . . . . . . . . . . . . . . . . . . . . . . 1 General .*.*. ,.* 1 The Rural and Agricultural Sector . . .. . . . . . 1 Irrigation Subsector Institutions . . . . . . . . . . . . . 2 Agricultural Support Services . . . . . . . . . . . . . . . 5 Agricultural Inputs .... . . . . . . . . . . . . .... 6 Irrigation Development in the Philippines . . . . . . . . . 7 Communal Irrigation Subsector . . . . . . . . . . . . . . . 8 Future Development of the Communal Irrigation System. . . . 10 The Participatory Approach to Development of Communal Systems. . . . . . . . . . . . . . . . . . . . . . . . . . 11 Experience with Past Lending ... . ..... . . . . . . . 12 2. THE PROJECT AREA . . . . . . . . . . . . . . . . . . . . . . . 13 Natural Conditions .... . . . . . . . . . ...... . . 14 Socioeconomic Conditions . . . . . . . . . . . . . . . . . . 16 Existing Infrastructure. . . . . . . . . . . . . . . . . . . 18 3. PROJECT FORMULATION ................... 18 NIA Program ......... .......... 18 Major Components ....... . .......... 19 Other Components . . . . . . . . . . . . . . . . . . . . . . 20 4. THE PROJECT . . .. . . . . 21 Project Objectives ................... 21 Project Description . . . . . . . . . . . . . . . . . . . 22 Detailed Project Features . . . . . . . . . . . . 22 Status of Engineering. . . . . . . . . . . . . . . . . . . . 30 This report is based on the findings of an appraisal mission in September/ October 1981, consisting of Rafique Qureshi, Robert Hindle, John Macgregor, Abate Menkir, and Johannes ter Vrugt (Bank), and Jasper Ingersoll (Consultant).  - ii - Page No. 5. PROJECT COSTS AND FINANCING. . . . . . . . . . . . . . . . . . 31 Cost Estimates .31 Financing. . . . . . . . . . . . . . . . . . . . . . . . . . 32 Procurement .34 Disbursements. . . . . . . . . . . . . . . . . . . . . . . . 35 Accounts and Audit .... . . . . . . . . . . . . . . . . . 37 6. ORGANIZATION AND MANAGEMENT ................ . 37 Project Implementation ... . . . . . . . . . . . . . . . 38 Operation and Maintenance . . . . . . . . . . . . . . . . . 40 7. AGRICULTURAL PRODUCTION . . . . . . . . . . . . . . . . . . . 41 Present Cropping Pattern .41 Future Cropping Patterns .42 Yields and Production. . . . . . . . . . . . . . . . . . . 42 Harvest, Drying, Storage and Processing .45 8. MARKET PROSPECTS, PRICES, FARM INCOMES AND COST RECOVERY . . 45 Market Prospects .45 Prices .46 Farm Incomes . . . . . . . . . . . . . . . . . . . . . . . 47 Cost and Rent Recovery .48 9. BENEFITS, JUSTIFICATION AND RISKS. . . . . . . . . . . . . . 52 Benefits .52 Financial and Economic Analysis. . . . . . . . . . . . 54 Sensitivity Analysis .55 Environmental Effects. . . . . . . . . . . . . . . . . . . 56 Risks. . . . . . . . . . . . . . . . . . . . . . . . . . . 57 10. AGREEMENTS TO BE REACHED AND RECOMMENDATION. . . . . . . . . 58  - iii - Page No. ANNEXES 1. Climatic Characteristics by Region . . . . . . . . . . . . . 60 2. Table 1 - Strengthening of Provincial Irrigation Offices Classification Levels . . . . . . . . . . . . . 61 Table 2 - Office, Survey and Hydrologic Equipment. . . . . . 62 Table 3 - Construction Equipment, Machinery and Vehicles . . 63 Table 4 - Estimated Schedule of Expenditures . . . . . . . . 64 Table 5 - Estimated Schedule of Disbursements. . . . . . . . 65 Table 6 - Proposed Allocation of IBRD and IFAD Loan . . . . 66 3. Estimated Crop Labor Requirements . . . . . . . . . . . . . 67 4. Table 1 - Rice Farms: Physical Inputs, Yields and Unit Prices. . . . . . . . . . . . . . . . . . . . . . 68 Table 2 - Farm Budgets: Rainfed Rice Farms . . . . . . . . . 69 Table 3 - Farm Budgets: Irrigated Rice Farms . . . . . . . 70 Table 4 - Farm Budgets: Rainfed Corn Farms . . . . . . . . . 71 Table 5 - Farm Budgets: Economic Prices. . . . . . . . . . . 72 5. Annual Operation and Maintenance Cost of Typical Communal Irrigation System .... . . . . . . . . . . . . . . . . . 73 6. Accounting Procedures for Force Account Work and Minor Contracts .... . . . . . . . . . . . . . . . . . . . . . 74 7. Related Documents and Data Available in the Project File. . . 76 CHARTS: Figure 1.1 - NIA Organization Figure 1.2 - Regional and Provincial Offices' Organizations Figure 6.1 - Communal Irrigation Development Project Organizational Chart Figure 6.2 - Sample Flow Chart: Subproject Activities Figure 6.3 - Implementation Schedule MAP: IBRD #16197   PHILIPPINES COMHUNAL IRRIGATION DEVELOPMENT PROJECT 1. BACKGROUND General 1.01 The Government of the Philippines has requested the Bank and the International Fund for Agricultural Development (IFAD) to assist in financ- ing the National Irrigation Administration's (NIA) program of communal irri- gation development. The project would strengthen the Provincial Irrigation Offices (PIOs) which are responsible for implementing the program. It would finance the 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 Rural and Agricultural Sector 1.02 Agriculture is the predominant sector in the Philippine economy, accounting for about 30% of the gross domestic product (GDP), 60% of foreign exchange earnings and over half of total employment. About 70% of the population lives in rural areas, where nearly two-thirds of the households depend on small farms for their main source of income. About 85% of the 2.35 million farms in the Philippines are smaller than 5 ha; the average size is 3.6 ha, and rice farms average only 1.8 ha each. Between 50% and 60% of all farmer and farm-laborer households subsist below the absolute poverty income level of US$195 per capita (in 1980 prices). Population pressure has caused even the most marginal lands to be brought under culti- vation, and most future increases in production must come from higher yields and cropping intensities. Yet adverse price ratios retard the use of yield-enhancing fertilizers and crop protection chemicals; supporting ser- vices of research, extension, marketing and training are often inadequate; formal credit is severely limited and informal credit expensive; and insecure land tenure limits many farmers' willingness or ability to adopt innovations in production techniques. Hence higher yields do not come easily. Higher cropping intensities are also limited by existing crop varieties, inadequate extension support, and inadequacy of water through rainfall or irrigation. Rural social services are inadequate, nonagricul- tural production activities are limited, agricultural productivity is low and underemployment is high. 1.03 Although agriculture's share of GDP has been falling relative to other sectors, food production rose during the 1970s at an average of 7% p.a. This was facilitated by heavy government investments in high-yielding crop varieties and technical packages coupled with extension and farm credits and to a lesser extent irrigation. Most of these initiatives were  - 2 - directed at rice and corn, which together account for about 36% of the value of agricultural production, 57% of the area, and 70% of employment. 1.04 Rice is the primary staple of over 80% of the population, and accounts for about 30% of total harvested area. The Masagana 99 program of rice production support was instrumental in helping the country to achieve rice self-sufficiency in 1977, and since then the Philippines has exported small quantities of rice each year except 1981, when adverse weather and lower prices limited production to about the level of consumption. Corn is the primary staple of a sixth of the population and the main feed component of the livestock industry. Recent progress in developing varieties resistant to downy mildew offers the prospect that corn yields, now among the lowest in Asia at under one ton per ha, might rise enough to reduce or even eliminate the chronic need for corn imports, but this is only now being tested on a commercial scale. Coconuts and sugarcane are the third and fourth most important crops, both major foreign exchange earners whose profitability and production depend on unstable and currently depressed world prices. A combination of factors including relative prices, yields, soils and climate has limited the use of the expanding irrigation infrastructure largely to rice cultivation. 1.05 Government Objectives and Strategy. The Government's current (1978-82) five-year development plan sets general objectives of high growth, greater resource mobilization, reduced income disparities and improved environmental quality. All of these apply to rural sector development, but specific goals most relevant to the agricultural sector are poverty reduc- tion, continued food self-sufficiency and more diversified agricultural pro- duction. These goals are closely interlinked, since policies aimed at achieving one goal inevitably affect the others. Recognizing this, the Government has chosen to place most of its agricultural sector investments in infrastructure-building to support the broadest range of economic activities, while launching specific programs to address particular problems. The specific programs include land tenure reform, crop research, and a wide range of crop production programs combining credit, input supply and/or technical support. The infrastructure investments emphasize power, irrigation and highways. The problem of poor and often deteriorating ratios between prices of production inputs and farmgate prices of the major crops is a major con- straint on the poverty reduction and agricultural production goals, and has not yet been adequately addressed. The Government s draft development plan for 1983-87 states that performance during the current plan period has been satisfactory, and sets the attainment of sustainable economic growth, equi- table distribution of the fruits of development, and total human development as the major national goals for the next plan period. Irrigation Subsector Institutions (a) National Irrigation Administration 1.06 Organization. NIA, the central institution in irrigation development, was established in 1963 as the successor organization to the  - 3 - irrigation division of the Bureau of Public Works. NIA's mandate is a broad one. It is responsible for the planning, development, construction, and operation and maintenance (O&M) of nearly all irrigation works in the Philippines. Specifically, NIA is to (a) investigate and study all water resources for irrigation purposes; (b) construct, operate and maintain all national irrigation systems; (c) construct, temporarily administer and periodically repair communal and pump irrigation schemes; (d) collect water charges; (e) construct multipurpose water resource projects; and (f) construct drainage facilities in collaboration with the Ministry of Public Works and Highways (MPWH). In the past NIA has largely concentrated on construction of large new schemes and rehabilitation of existing ones, and O&M have received relatively lesser support. For both technical and financial reasons this historical concentration on large new capital works is shifting largely to smaller works with increasing attention to rehabilitation and O&M issues. The smaller works are largely located in the communal irrigation subsector. 1.07 Although as a semiautonomous agency NIA has a large degree of oper- ational freedom, it is attached to the MPWH for program and policy coordinat- ion. NIA is governed by a Board of Directors appointed by the President. At present Board members are the Minister of Public Works and Highways (chairman), the Administrator of NIA (vice-chairman), the Director General of the National Economic and Development Authority (NEDA), the Minister of Agriculture, the General Manager of the National Power Corporation and a Secretary. The Board takes an active role in setting guidelines and directions for the agency. The President appoints the Administrator and all Assistant Administrators of NIA. Within the headquarters at Manila, NIA's operations are divided among four Assistant Administrators: (a) Project Development and Implementation; (b) Operations; (c) Finance and Management; and (d) Personnel and Administrative Services. Most field activities are under the supervision and control of Regional Irrigation Directors (RIDs). (See Figure 1.1 for organization chart of NIA.) 1.08 NIA has a Regional Irrigation Office (RIO) in each of the 12 Regions as well as PIOs in 72 provinces. (See Figure 1.2 for the organizational structure of each Regional Office.) The identification, preparation and appraisal of communal irrigation systems (CISs) is initially handled by the PIOs. The RIOs review these proposals and assist in further preparation, particularly on agricultural and economic aspects. The proposals are then sent to the headquarters office where they are screened and annual operational programs are established. Detailed engineering and economic aspects are reviewed first by RIDs and then at headquarters by the Project Development Department (PDD). The approved program is then carried out by the PIOs. 1.09 Financial Situation. At the end of 1980 NIA had total assets of P 8.8 billion, the bulk of which (68%) represents civil works completed or in process. Equity capital from the Government amounted to P 1.0 billion. Total assets grew 27% over the previous year, although there is also an  -4- element of overstatement as NIA's investments are carried at original cost. NIA's current liabilities have been growing rapidly and at the end of 1980 amounted to f 2.2 billion. Government public works appropriations and equity contributions are by far the largest sources of NIA's income, but recent cutbacks have caused a constriction of liquidity which is slowing the implementation of some large projects. Increased irrigation fee collection (6% of total revenues in 1980) and reduced salary overheads (as completion of some major projects has allowed staff reductions) have not been sufficient to ease the tight liquidity. 1.10 Accounts. As regards individual foreign-funded projects, NIA's accounting procedures and staff are good. For the agency as a whole, how- ever, there are problems with the present system. First, the number of separately maintained accounts leads to errors. Second, NIA has some difficulty in ensuring accountability of cash disbursement officers in the field. Finally, and perhaps most importantly, the financial records system is too unwieldy to supply timely financial information to managers and to provide at any one point in time a complete picture of all NIA financial operations. Improving financial reporting and accounting for nonforeign- funded opeations is a high priority for NIA's management, and steps have been taken to strengthen reporting at the head office. In addition, a private auditor has been engaged to review selected accounting subsystems. Much remains to be done, however, to ensure development of a fully integrated and responsive accounting system and to train requisite numbers of staff. (b) Farm Systems Development Corporation (FSDC) 1.11 The FSDC, created in 1975, plays a minor role in the irrigation subsector. It has responsibility for farm technology development and extension, and entered the irrigation field largely through its work in establishing small pump irrigation systems and irrigators' associations (IAs, see para. 1.25). Until 1981 it had responsibility for organizing CIS farmers into IAs under an agreement with NIA in which NIA undertook all engineering tasks and FSDC all organizational tasks. This agreement also assigned FSDC the task of collecting on NIA's behalf the capital costs of CISs from the beneficiaries. In practice, there were many problems with this arrangement. NIA's PIOs were inadequately equipped to carry out their engineering tasks; FSDC's institutional organizers were too few, often inadequately trained in problems of irrigation, and isolated from the PIOs. Farmers did not get involved in the design and construction of irrigation systems; their involvement only began when the completed systems were turned over to them. Further, farmers' problems during O&M of the systems related mainly to technical matters which FSDC's institutional organizers were not trained to resolve. Disputes over design and costs have kept nearly half of the 350 CISs built in the last four years from being turned over to farmers. FSDC's organizers made little or no progress in improving the capital repayment record. In view of these problems, NIA and FSDC have agreed to phase out their joint operations by 1984, after which NIA will take full responsibil- ity - organizational and engineering - for all CISs of over 100 ha and also  - 5 - for many smaller systems, amounting to about 80% of the communal irrigation subsector; FSDC will assume responsibility for the small systems only, mainly in pump irrigation, constituting not more than 20% of the CISs. NIA has always had, and would continue to have, full responsibility for the national irrigation systems. Agricultural Support Services 1.12 Extension. A number of institutions provide extension and other forms of support to farmers in irrigated areas. The Ministry of Agriculture (MOA) is responsible for the extension service, providing on the average one agriculture and food technician (AFT) for every 104 farmers nationwide. The Bank's National Extension Project (Loan 1626-PH) is strengthening this service. Various credit institutions, rural banks, FSDC and the Ministry of Agrarian Reform (MAR) also provide some technical support to farmers. NIA itself trains water management technicians to help farmers use their irriga- tion systems efficiently. The National Food and Agricultural Council (NFAC) is responsible for coordinating the activities of all national agencies in food production while local municipal action committees and IAs coordinate at the level of individual irrigation systems. It is recognized that the extension and other support services provided by various agencies need better coordination, and that extension agents need broader training. AFTs are often ill-trained to advise farmers on irrigated rice production tech- niques, and water management technicians seldom receive much training in agronomy; the two sometimes provide conflicting advice to irrigators. The proposed project's training component addresses this need in relation to the communal irrigation subsector. 1.13 Rice Research. In addition to the central research facilities provided by the College of Agriculture of the University of the Philippines and the International Rice Research Institute (IRRI), both at Los Banos, there are six rice experiment stations run by MOA and a number of substations located all over the country. The main emphasis of the work on rice in these stations is on varietal screening, foundation seed production and certification, fertilizer trials and intensified land utilization through multiple cropping. High yielding varieties (HYVs) of rice, better farm management techniques and more appropriate farm machinery and tools are continually being developed at these facilities. Irrigated rice yields at these stations often exceed seven tons per ha per crop. 1.14 National agricultural research is coordinated and evaluated by the Philippine Council for Agriculture and Resources Research which assists in the formulation of relevant research programs and assists MOA in the packaging of technology for promotion by the extension service. Likewise, the National Nutrition Council, together with the concerned ministries and such agencies as the National Media Production Center, the Nutrition Center of the Philippines and the Food and Nutrition Research Institute are engaged in drawing up nutrition technology packages suitable for national programs.  -6- Agricultural Inputs 1.15 Seed. At present the average seeding rate is about 80 kg/ha, mainly because of poor germination, damage to seedlings during transplanting, too many seedlings per hill, improper spacing, and anticipation of drought or floods. Most of the seed is obtained by retention of on-farm production, though it is not uncommon for farmers to purchase seed from registered growers or from other farmers who grow the desired variety. In the future, given an assured supply of high quality seed and irrigation water, it is estimated that the average seeding rate will be about 50-60 kg/ha. No difficulties are foreseen in the production and distribution of registered seed. Most farmers are well informed about the new improved varieties and their characteristics and readily purchase the output of registered seed growers. 1.16 Fertilizers. The 1980 national annual consumption of fertilizer was about 820,000 tons, of which three fourths was imported. About half of the imported fertilizer was urea. Application rates on rice farms vary from 50 kg/ha or less in some rainfed areas subject to flooding to 250-300 kg on better farms with reliable irrigation. Farmers often use the wrong fertilizers or apply it at the wrong time, and tend to underrate the importance of a proper basal dressing. Sometimes, however, farmers are hesitant to apply the recommended amounts of inputs because of unassured water supply and unattractive price ratios between palay and fertilizer. The existing fertilizer distribution network is adequate and could easily be expanded to meet increased demand. 1.17 Agrochemicals. Agrochemicals are readily available in the Philippines and are widely used in small to modest quantities. Surveys show that most rainfed areas and almost all irrigated areas receive some pesticide application but these average only half or less of the recommended dosages. Consequently the applications are not always effective. Serious outbreaks of insect pests and diseases are, however, not common. Farmers are also well acquainted with herbicides, and about 50% of all rice farms receive some herbicide application, often at a lower dosage than recommended. Extensive rat damage has been reported for most irrigation systems at one time or another over the past six years. A rodent control component is included in the BPI-German Surveillance Program which, among other activities, trains plant pest control officers. For the future, the application rates of agrochemicals are not expected to increase dramatically, mainly because of increasing prices of these petrochemical products. 1.18 Credit. CIS farmers receive an estimated 35% of their seasonal and investment credit from formal credit institutions; the remainder comes from various informal sources. The major formal credit program affecting CIS farmers is the Masagana 99 program launched in 1973 to boost rice production by combining a noncollateralized supervised credit program with a production input package. This input package combines HYVs, fertilizers,  - 7 - biocides and technical advice on cultural practices. Masagana 99 credit is extended through Philippine National Bank branches and the rural banks under a rediscount arrangement with the Central Bank. Credit is for a term of about six months at an annual interest rate of 12%. All supervised credit is required to be insured to 100% of the value of the inputs. Repayment rates have been low, and a problem now facing many Masagana 99 beneficiaries is that, having defaulted on past Masagana 99 loans, they no longer qualify for credit on such attractive terms. Many rural banks, victims of widespread loan defaults have become ineligible to participate in the supervised credit programs, and are therefore precluded from access to the Central Bank's rediscount facilities. The Government has recently reviewed the arrears problem and has taken steps which will result in the requalification of a number of rural banks, and hence a corresponding increase in credit availa- bility. Loan defaults under Masagana 99 are also being rescheduled under some circumstances to allow a larger number of borrowers to participate. Irrigation Development in the Philippines 1.19 Farmer associations were building, operating and maintaining irri- gation systems in the Philippines before the Spanish era. Many of these systems are still in operation, as are about 180,000 ha of similar systems constructed in the Spanish times. Under American rule, the area covered by these simple systems is estimated to have doubled. Establishment of irriga- tion systems remained largely a local voluntary affair until the mid-twen- tieth century. The first government-financed irrigation project was comple- ted in 1923. By 1940, ten such irrigation systems covering 80,000 ha had been completed. By 1981, irrigated area totaled over 1.2 million ha out of about 3.5 million ha judged suitable for irrigation. The country's projects can be classified in three groups: national systems, communal systems and pump systems. 1.20 National irrigation systems are constructed, operated and main- tained by the Government through NIA (see paras. 1.06-1.10). Irrigators within the systems are expected to pay water fees sufficient to cover O&M, insurance and capital costs. These systems range in size from less than 200 to 78,000 ha, and average about 3,300 ha. Much of the post-independence national systems were financed with foreign assistance, including that of the Bank. Actual irrigated area under the existing 144 national systems now totals about 470,000 ha. Foreign-assisted national irrigation projects now under construction are expected to add about 300,000 ha of new area to the total of national irrigation systems by the end of 1985, while foreign- assisted rehabilitation efforts should improve the operation of a further 220,000 ha in existing systems. 1.21 Communal irrigation systems are the type constructed in the Philippines in the pre-colonial and Spanish periods. Unlike national systems, they may be built with or without government assistance. An association of irrigators rather than a government agency owns, operates and maintains the facilities of most systems. When Government assistance is  -8- involved in construction, IAs are expected to pay fees to cover capital costs, excluding the costs of roads and NIA's engineering and overhead costs. With the rising cost of investment, farmers are increasingly seeking government assistance to improve their systems as well as to construct new ones. Government assistance to communals rose very rapidly from about 30 systems in the late 1950s to over 900 systems in the late 1970s. At present, the communal irrigation development program has a budget appropriation of about P 150 million per year. Table 1.1 shows the growth of government assistance to communals. This assistance, mainly in design, construction and financing, was carried out with little local consultation or local contribution. Hence farmers were often reluctant to identify with the systems, and felt that the works were poorly adapted to local conditions and also that system maintenance was the Government's responsibility and not their own. Table 1.1: GOVERNMENT ASSISTANCE TO CISs, 1955-1980 No. of CISs Area (ha)/a Period completed During period Cumulative 1955-1959 51 7,500 7,500 1960-1964 109 12,000 19,500 1965-1969 135 17,000 36,500 1970-1974 340 61,800 98,300 1975-1978 887 149,100 247,400 1979 368 63,300 310,700 1980 241 47,400 358,100 /a Includes new as well as rehabilitation schemes. 1.22 Pump irrigation systems are nearly all privately-owned and opera- ted. Most of these draw water by diesel, gasoline or electric pumps from rivers or lakes and a few tap groundwater resources. The Irrigation Service Unit of NIA is responsible for the construction and supervision of pump schemes. About 190,000 ha are presently irrigated by pump systems. The high cost of fuel or electricity to operate the pumps reduces their financial attractiveness and constrains the further growth of pump irrigation. NIA is gradually phasing out its pump irrigation program. Communal Irrigation Subsector 1.23 Economic and Physical Aspects. The roughly 5,700 CISs account for about 45% (550,000 ha) of total irrigated area and 38% of irrigated rice production. A 1980 survey indicates that nearly 85% of CIS service area is  - 9 - irrigated (systems seldom being managed and maintained well enough to achieve 100%); virtually all of it is devoted to rice. About 13% of CIS farm area is devoted to coconuts, corn, sugarcane and vegetables. CISs comprise about 390,000 farm households with nearly 2.4 million people. They range in size from a few hectares to over 4,000 ha, but average slightly under 100 ha each, with half under 50 ha and over 70% smaller than 100 ha. Over one third of the systems, covering less than 20% of the total area, are in Region 1 (Northwestern Luzon), although they can be found in nearly all of the 73 provinces, along small rivers and tributaries of larger rivers as they flow out of the foothills into the river basins. The modest water supply and moderate topography in these areas have enabled local farmers to achieve limited water control with very simple temporary diversion structures often built of rocks or brush. Yields are generally low, about 2.8 tons per ha per crop of irrigated rice. 1.24 CIS Farmers. The farmers within CISs operate small farms, averag- ing 1.4 ha each. Farms are smallest in Region 1 (Northwestern Luzon - less than 0.9 ha) and largest in Region 11 (Southeastern Mindanao - about 1.8 ha). Farm households have an average of six members. With such small farms and low yields, farm income is low, and farm families earn about half of their total income from off-farm sources (nonfarm labor or labor on neighboring farms). Even with this supplementary income, the average per capita income of CIS households is estimated to be US$150-190 p.a., below the absolute poverty level. Contributing to their hardships is the fact that half the CIS farmers do not own their lands, but operate as share tenants (about 30%) or leaseholders (about 20%). Just under 50% are owners, ranging from 74% in Region 1 to 30% in most other regions. 1.25 Organizational Aspects. Communal irrigation systems vary widely in the way they are managed. Smaller systems tend to depend on informal leadership by village leaders or large landowners. These leaders call on irrigators to perform needed maintenance and may reprimand violators of the tacit water distribution rules. Management structure becomes more formal as system size increases. Over 90% of systems over 100 ha have IAs. 70% have people responsible for water distribution, 50% have paid personnel to operate control structures, perform maintenance and carry out administrative duties, and 40% have regular maintenance schedules and irrigation fees. Less than 25% of systems under 50 ha have any of these characteristics. A typical IA is registered with the Securities and Exchange Commission (SEC), and has an elected board of directors and a roster of officers, and has monthly meetings of members. In most cases, canals and farm ditches are cleaned, desilted, reshaped, repaired or cleared at least once per year, and usually twice per year. Most farmer-members contribute free labor and materials for system maintenance. 1.26 At present, irrigation service fees are collected by slightly less than half of all IAs. Fee collection arrangements are more common among NIA-assisted systems and systems in Mindanao and Southern Luzon. Most IAs  - 10 - collect fees in kind based on area irrigated. Fees range from 0.5 to 4.5 cavans of palay per ha per year, with a mode of about 2 cavans. About 80% of the members pay at least part of their dues. Delinquent farmers are three times as common among NIA-assisted systems as among non-NIA systems. This may be because NIA-irrigator consultations were inadequate during design and construction, leading irrigators to the conclusion that their CIS is really the Government's responsibility. It may also be connected with the fact that fee collection is more common in NIA-assisted CISs. Delinquent farmers in general report that they receive poor water service, get water from elsewhere, have crop failures or financial problems, or simply are never billed. Future Development of the Communal Irrigation System 1.27. Potential. NIA is placing increasing emphasis on the development of communal systems. These can be completed in as little as one year and give immediate benefits, and their future operating burden on NIA is far lower than that of national systems because O&M is largely performed by IAs. A preliminary NIA planning document envisages about 500,000 ha of communal irrigation development during the decade 1981-90; however, funding limita- tions and higher design standards could limit the area to under 250,000 ha. Additionally, small national systems covering an area of about 25,000 ha are planned for conversion to communal systems. 1.28 Problems. General agricultural sector problems such as adverse price ratios, inadequate support services, insecure land tenure and infrastructural inadequacies were noted in para. 1.02. Three other problems more specific to CISs must be resolved if the subsector is to be developed as envisioned: (a) Institutional Support. The PIOs of NIA are ill-equipped to carry out the type and scale of communal irrigation programs assigned to them. Training of PIO staff in organizational aspects of communal irrigation development is crucial, as is the provision of equipment. (b) Irrigation Association Development. In the past, most NIA- assisted systems were constructed with little or no effort to pre- pare the IAs which were expected to operate and maintain them. This has resulted in poor water distribution, water use conflicts and unsatisfactory conflict resolution, unfulfilled farmer expectations, poor maintenance, and low capital cost repayment. These problems are serious also on many non-NIA-assisted systems. Careful organization and training efforts are necessary to build strong IAs which can make the best use of available physical facilities. (c) Engineering Design. Most CISs depend on diversion weirs of rocks or brush which cannot withstand floods. Frequent  - 11 - rebuilding of these weirs is a major drain on labor which might be better employed in other tasks. Most systems also fail to divert as much water as farmers need to maximize cropping intensities, allow further losses through unnecessary seepage and poor design of distribution systems, and lack control structures and roads for maintenance. NIA-assisted CIS works are of a far superior design, and generally eliminate the drawbacks of traditional designs. The Participatory Approach to Development of Communal Systems 1.29 In order to deal with the above constraints, and at the same time realize the full potential of CISs, NIA has developed a new participatory approach to the planning, design, construction, and operation and maintenance of CISs. Under this concept the active collaboration among irrigation officials and farmers organized in IAs is an essential feature of some existing and all future CISs in the Philippines. The need for a new or improved irrigation system is the unifying element, and the strategy is to ensure full participation by officials and farmers from initial planning through construction, operation and maintenance. The technical goal is to construct properly designed physical systems; the institutional goal is to improve the capability of farmers to organize and sustain their association, improve their farm practices, and assume responsibiity and a sense of ownership of their system. 1.30 The organization of the IAs is built from the base upwards, rather than the conventional procedures of centralized, top-down design, construction and operation of most irrigation systems. The observed weaknesses of that conventional approach prompted NIA in 1976 to launch a special experimental communal project with Ford Foundation assistance to test new methods to integrate institutional and technical development by having farmers participate actively with NIA officials in both of these functions from the earlier stages of design through construction to operation. The encouraging results of this experimental project prompted NIA and the Ford Foundation to extend the program to 2 additional pilot projects in 1979, to 12 more in 1980, and to 24 more in 1981. 1.31 The participatory approach benefits NIA, its PIOs and the irri- gators. It enables NIA to reduce per hectare construction costs, transfer virtually all O&M costs to IAs, and improve the rate of cost recovery from the IAs. It enables the PIO to construct or rehabilitate more hectares of communal irrigation on a constant budget by shifting more costs to the direct beneficiaries. Finally it enables the farmers to get permanent well-designed and well-located systems requiring minimal maintenance and rehabilitation costs, achieve better responsiveness in their dealings with officials, and acquire their own modern systems. The result is a better-operated and -maintained set of systems, which ultimately means higher and more stable incomes for the farmers.  - 12 - 1.32 The participatory approach includes activities, both institutional and technical, for farmers at every stage of the project process. In the identification phase, the farmers organize themselves into an IA and request NIA assistance. In the design phase, they formulate by-laws, register their IA, obtain a water permit, establish IA committees as needed, conduct a man- power survey of IA members in preparation for construction, work with tech- nical staff on land surveys and negotiate over each version of the technical design, secure rights of way, and sign a pre-construction memo of agreement with NIA on the estimated cost of the proposed communal system. In the construction phase, they contribute labor, monitor construction materials, equipment, and costs, reconcile construction costs with NIA records, attend various training seminars, accept the completed system, and sign the loan with NIA. In the O&M phase, they take over full control of their system and continue attending training courses for water management, and financial control. The participatory approach thus combines the benefits of local initiative and self-reliance of the old indigenous communal systems with the benefits of modern technical design and construction. The historical context and details of the participatory approach to the implementation of CISs are available in the project file. Experience with Past Lending 1.33 NIA has extensive experience in the planning, design and construc- tion of irrigation projects. Sixteen projects have received Bank assistance, while thirteen others were assisted by the Asian Development Bank. Other main agencies involved in irrigation development include the United States Agency for International Development and Japan's Overseas Economic Coopera- tion Fund. NIA has established itself as a well managed and technically competent organization. Of the sixteen Bank-assisted projects, one was for watershed management and erosion control, and three for rural development. The remaining twelve Bank loans were mostly for large national irrigation projects providing irrigation and drainage facilities for new as well as rehabilitation areas, better road systems for efficiency of system operation and for marketing of farm products, and support services to assist farmers in adopting the new irrigation techniques needed to increase production. Implementation progress has been generally satisfactory on most projects. Some projects have experienced delays because of local funding constraints, design changes, contractor problems, or local order problems. Two projects have been completed. The first Bank-assisted project, Upper Pampanga River Irrigation Project (Loan 637-PH), was completed in December 1977. The Project Performance Audit Report (PPAR), prepared by the Bank's Operations Evaluation Department (OED) in June 1980, found that the dam was completed 10 months ahead of schedule, despite changes in design. However, there were delays of up to 30 months in construction of the irrigation facilities due to procurement problems, poor performance of contractors, and interruptions of work to allow for delivery of irrigation water to farmers. The cost overrun- was 105%, of which about 80% was due to inflation during the implementation period. While dry season cropping rose rapidly from less than 10,000 ha in 1974 to 55,000 ha in 1976 as a result of the project, the appraisal estimate of increase in annual cropping intensity proved overoptimistic. However, the paddy yields achieved in the project were higher than expected. The PPAR computed the revised economic rate of return of the project at 14% compared  - 13 - to 13% estimated at appraisal. The PPAR concluded that NIA's current excell- ent project preparation and farmer support capabilities were developed and nurtured within the organization which was created for the Upper Pampanga Project. The second project, Aurora-Penaranda Irrigation Project (Loan/Credit 984/472-PH), was completed in December 1981 and preparation of the Completion Report is underway. While work on the two dams and the transbasin canal was completed about one year ahead of schedule, difficulties in awarding con- struction contracts delayed the start of improvement works in the service areas which were further hampered by severe typhoon damages in 1978 and 1980. 1.34 Bank assistance to communal irrigation development began in 1975 with the Mindoro Rural Development Project (Loan 1102-PH), which included a small communal irrigation component of about 3,000 ha in conjunction with the development of about 12,000 ha ot national irrigation systems. These irriga- tion systems were completed about two years behind the appraisal schedule, mainly due to design changes and difficulties in right of way acquisition and shortages of construction materials and supplies. Under the Rural Infra- structure Project (Credit 790-PH), about 9,400 ha of CISs in six provinces were included which were subsequently expanded to 10,400 ha. Implementation progress is satisfactory in four provinces but is behind schedule in two pro- vinces. Delays are due to inadequate or late release of local funds, unusually severe weather, late design finalization for diversion weirs, and FSDC's ineffectiveness in organizing the farmers. The project is expected to be completed in 1983, one year behind the appraisal schedule. About 2,000 ha of CISs form part of the Samar Rural Development Project (Loan 1772-PH). Construction progress of one of the new subprojects is satisfactory, while the second was abandoned due to strong protests by the concerned farmers as they were not satisfied with the selection and preparation of the subproject which had been done without adequate prior consultation with them. Substitute subprojects are under investigation and evaluation. Rehabilitation ot the 23 existing CISs and improvement of buildings is progressing satisfactorily with 20 systems completed, one system under way, one system deferred and one system deleted. 1.35 The Philippines Agricultural Sector Operations Review by OED (Report No. 3796 dated February 10, 1982) concluded that while irrigation associations in the large national schemes have posed problems, communal associations have been relatively more successful. As communal schemes ultimately become the property of the associations, construction or rehab- ilitation is usually carried out using their ideas and suggestions. They are especially successful where local farmers have actively participated in the design and siting of structures and canals. 2. THE PROJECT AREA 2.01 The project would involve institutional strengthening in 67 provin- ces in eleven of the Philippines' twelve regions, and construction or reha- bilitation of 144 communal irrigation subprojects covering about 33,500 ha  - 14 - in 35 provinces of 11 regions. A summary of areas by region and province where irrigation facilities would be provided or upgraded is shown in Table 2.1. Natural Conditions 2.02 Climate. In general the climate in the Philippines is wet tropical with average temperatures around 270 C throughout the year in the whole country, allowing a twelve-month growing season with irrigation. The Philip- pines lie under the influence of the monsoons and in the path of tropical typhoons. The generally north-south alignment of mountain ranges leads to wide variations among regions in annual rainfall and its seasonal distribu- tion. Four main types of climate can, however, be distinguished. Type A and Type B climates are similar in that they both have annual rainfall of 1,500-2,000 mm and well defined dry seasons, although Type A has a more pronounced seven-month dry season beginning in October whereas Type B has a five-month dry season starting in December. The Type C climate has annual rainfall of 2,500-4,200 mm. Rainfall is heavy from October through March, and at its maximum from November to January. There is effectively no dry season since monthly rainfall between April and September is generally about 200 mm. In the Type D climate, annual rainfall of 2,000-3,500 mm is more or less evenly distributed over most of the year, with the wettest period from April to December, but rainfall is not as intense as in Type C; the dry season is short from January through March or April. The rainfall is gener- ally adequate in the country for a single rice crop in the wet season except in the southern part of Mindanao, although provision of irrigation would reduce risk and improve yields. Dry-season cropping entails, however, a greater risk, and irrigation is essential to ensure a crop. Typhoon frequen- cies generally increase from south to north with the largest incidence in Northern Luzon (averaging about seven a year in the 1948-77 period). Only Southern and Central Mindanao are relatively free from typhoons. Annex 1 gives details of climatic conditions for each of the Philippines 12 regions. 2.03 Topography. The Philippines has complex physiographic patterns. The entire country rests on a series of stress lines along which numerous faulting, folding and volcanic activities have taken place. The extensive earth movements have resulted in numerous mountain ranges which divide the islands into small watersheds with short and usually rapidly flowing rivers and isolated alluvial plains. The project areas are located along small rivers and tributaries of larger rivers as they flow out of the foothills into the river basins. 2.04 Soils. The dominant soil types are relatively young silty clay loams, clay loams and sandy clays. These soils are mostly derived from basalt andesites, volcanic rocks and tuffs, and limestones under intense and rapid natural weathering caused by high temperature, heavy rainfall and limited forest coverage. Most of the irrigable soils are in plains and valleys with deep alluvial deposits. Where volcanoes are present, as in  Table 2.1: SUMMARY OF PROJECT AREAS BY REGION AND PROVINCE I New Rehabilitation Total New Rehabilitation Total No. Area No. Area No. Area | No. Area No. Area No. Area Region/Province (ha) (ha) (ha) IRegion/Province (ha) (ha) (ha) Region I: ILOCOS |Region VII: CENTRAL VISAYAS La Union 1 100 2 330 3 430 | Cebu 1 250 2 250 3 500 Mt. Province 3 600 - - 3 600 | Negros Oriental and Pangasinan - - 1 100 1 100 I Siquijor 5 1,440 2 180 7 1,620 Subtotal 4 700 3 430 7 1,130 Subtotal 6 1,690 4 430 10 2,120 Region II: CAGAYAN VALLEY |Region VIII: EASTERN VISAYAS Ifugao 4 670 1 250 5 920 | Southern Leyte 3 395 1 480 4 875 Quirino 3 720 - - 3 720 | Northern Leyte 4 960 1 380 5 1,340 Cagayan - - 2 650 2 650 | Isabela 1 220 1 150 2 370 | Subtotal 7 1,355 2 860 9 2,215 Nueva Vizcaya 2 450 2 390 4 840 | IRegion IX: WESTERN MINDANAO Subtotal 10 2,060 6 1,440 16 3,500 | Zamboanga del Sur 3 690 1 120 4 810 I Zamboanga del Norte 4 1,080 1 280 5 1,360 Region IV: SOUTHERN TAGALOG Aurora 1 410 2 600 3 1,010 | Subtotal 7 1,770 2 400 9 2,170 Marinduque 1 120 1 200 2 320 Romblon 4 620 - - 4 620 |Region X: NORTHERN MINDANAO Batangas 3 340 1 80 4 420 | Surigao del Norte 2 1,220 - - 2 1,220 > Quezon 3 350 - - 3 350 | | Subtotal 2 1,220 0 0 2 1,220 Subtotal 12 1,840 4 880 16 2,720 I - Region XI: EASTERN MINDANAO Region V: BICOL REGION I Surigao del Sur 6 940 0 - 6 940 Catanduanes 2 300 3 230 5 530 | Davao del Norte 3 1,600 - - 3 1,600 Masbate 3 815 1 100 4 915 Davao del Sur 4 1,100 1 110 5 1,210 Sorsogon 1 160 5 810 6 970 | Davao Oriental 4 640 - - 4 640 Albay 5 1,580 2 370 7 1,950 | South Cotabato 2 670 2 640 4 1,310 Camarines Sur 2 750 3 1,130 5 1,880 | Camarines Norte 2 130 3 360 5 490 | Subtotal 19 4,950 3 750 22 5,700 Subtotal 15 3,735 17 3,000 32 6,735 |Region XII: CENTRAL MINDANAO | Sultan Kudarat 3 650 1 1,000 4 1,650 Region VI: WESTERN VISAYAS | Maguindanao 2 720 2 720 4 1,440 Negros Occidental 5 1,050 5 450 10 1,500 | North Cotabato 2 1,100 1 300 3 1,400 Subtotal 5 1,050 5 450 10 1,500 I Subtotal 7 2,470 4 2,020 11 4,490 I Total 94 22,840 50 10,660 144 33,500  - 16 - Bicol and Southern Mindanao, volcanic ash or tuff material is mixed with these deposits. The soils in the lowlands are usually heavy textured and well-suited for paddy production. On higher ground, soils are generally deep and loamy, with moderate to good internal drainage. There are no real problem soils in the Philippines, except for small spots of saline soils in Mindanao. Socioeconomic Conditions 2.05 Population. In 1980, when the population totalled 47.9 million, the population growth rate nationwide was about 2.4%, down from the average of 2.7% during the 1970s. However, population pressure on land is severe, and relatively little unexploited cultivable land remains. The project beneficiaries, an estimated 144,000 people in 24,000 families, would represent a cross section of the ethnic groups of the Philippines and the rural poor. A survey in 1980 of sample communal systems reported an average annual household income of P 7,140, a per capita income of P 1,190 or 80% of the rural poverty income level in 1980 prices. 2.06 Farm Size and Land Tenure. Philippine agriculture is characterized by the predominance of small farms (70% are smaller than 3 ha) and the per- vasiveness of tenancy, particularly among rice and corn farmers. The average farm holding in communal systems ranges from 0.9 ha in Regions I and II to 1.8 ha in Mindanao, with a national mean of 1.4 ha. In national irrigation systems the range is much wider and the national mean 70% larger. Average farm sizes by region are shown in Table 2.2. According to a detailed tech- nical and economic survey of 25 communal systems proposed for financing under the project, the average size farm was 1.6 ha, ranging from about 0.4 ha to 3.1 ha. About 54% of the farms in these systems range from 1.0 to 2.5 ha, and 29% are less than 1.0 ha. About 50% of the total area of the 25 sample systems consists of farms of 1.0-2.5 ha, 22% of 2.6-4.0 ha, 20% over 4.0 ha and about 8% less than 1.0 ha.  - 17 - Table 2.2: AVERAGE RICE FARM SIZES (ha) Region National systems Communal systems I Ilocos 0.56 0.89 II Cagayan Valley 2.81 0.89 III Central Luzon 2.59 1.54 IV Southern Tagalog 2.17 1.60 V Bicol 0.93 1.60 VI Western Visayas 1.56 0.95 VII Central Visayas 2.61 0.95 VIII Eastern Visayas 0.96 0.95 IX Western Mindanao 4.03 1.83 X Northern Mindanao 3.82 1.83 XI Southern Mindanao 2.79 1.83 XII Central Mindanao 3.65 1.83 National Average 2.37 1.39 2.07 About 50% of all farmers cultivate their own land, almost 30% are share tenants, paying an agreed percentage of their harvest; and the rest either rent extra land for their own use (11%) or are partowners (about 9%). In the 25 sample communal systems included in the project, about 46% are owner-operators, 25% are share tenants, 18% are lease holders (paying a fixed rent), and 11% are part-owners. The Government has pursued a program of agrarian reform aimed at the transfer of land ownership to tenant farmers on rice and corn lands since 1972, when the country's one million tenanted rice and corn farms covering some 1.5 million ha were declared land reform areas. The Land Reform Program is being implemented in two ways: "Operation Land Transfer," begun in 1972, applicable to all landowners who have more than 7 ha, whereby the tenants become owner-cultivators of rice and corn land in excess of 7 ha undertaking to pay for the land over a period of 15 years; and the Leasehold Enforcement Program for all farms up to 7 ha whereby the tenant receives security of tenure on the land being cultivated. Overall by October 1981, Certificates of Land Transfer had been issued to 416,000 tenants farming 645,000 ha, or about 88% of the area originally targetted for transfer. Leasehold contracts had been given to 550,000 tenants farming 615,000 ha, or 64% of the originally targetted area.  - 18 - Existing Infrastructure 2.08 The 35 provinces in which construction and rehabilitation under the project are planned include over 13,000 existing irrigation systems with a service area of about 744,000 ha, or about 61% of the total irrigated area in the country. By category, 68 are national systems irrigating about 289,000 ha, over 3,500 are communal gravity systems covering about 360,000 ha, and nearly 9,600 are small pump systems serving about 95,000 ha. The actual irrigated area in 1980 in the various regions of the country as a percentage of the technically potential irrigable area ranges from about 16% in Region X (Southern Mindanao) to about 60% in Region I (Ilocos). 2.09 The network of roads over most of the project area is sparse, averaging about 0.4 km per 100 ha. Road structures vary from paved national and provincial highways to gravel surfaced and unsurfaced barangay and private roads. Nationally about 80% of the freight and 60% of the passenger traffic move by road. Interisland traffic is largely by small vessels and to a limited extent by airplane. The Government and private sectors are cooper- ating in the development of port facilities. Postal and telecommunications services are available at major provincial cities and towns. Most of the communication facilities are still generally inadequate. Domestic water supply is also inadequate, although some of the areas included in the proposed project are provided with piped water systems. Most barangay families get their water supply from shallow wells. 3. PROJECT FORMULATION NIA Program 3.01 NIA's management recognizes that budgetary constraints and decreas- ing opportunities and need for large national projects will gradually change NIA's role. Inevitably, NIA's operations must shift from large new construc- tion toward the operation, maintenance and rehabilitation of existing sys- tems. Many of the remaining opportunities for new construction would tend to be smaller in scale, and well suited for communal system development. This suggests that future NIA programs will rely heavily on the PIOs. The PIOs are responsible for O&M and some rehabilitation of completed national sys- tems, and for the construction and rehabilitation of communal systems. As national projects now under implementation are completed, as the communal irrigation subsector expands, and as all systems age and begin to require rehabilitation, the demands on PIOs will grow. However, while the NIA cen- tral office, the RIOs, and the offices of project managers managing large foreign-assisted irrigation projects are well-equipped and competent organi- zations, the PIOs, who have not yet had the advantage of participating in foreign-assisted projects, are deficient in staff, buildings and equipment to carry out their tasks. The cornerstone of NIA's future program is,  - 19 - therefore, the strengthening of its PIOs. A concomitant element of the program is increased relative emphasis on communal irrigation development. And the most successful method of implementing communal irrigation projects has been found to be the participatory approach described in Chapter 1. The components of the proposed project reflect these conclusions. 3.02 In consultation with the Bank and IFAD, NIA prepared a proposal for Bank and IFAD assistance in financing the strengthening of its PIOs and the construction of a portion of its communal irrigation development program. The proposal was designed to be an integral part of NIA's overall irrigation development program, and to address directly the three specific CIS problems identified in para. 1.28. Major Components 3.03 The most important element of NIA's program, PIO strengthening, became the central element of the proposed project. Strengthening of PIOs involves training of staff and providing buildings and equipment. The specific support for each PIO was determined on the basis of its present resources and its potential role in communal subsector work. These were determined in a collaborative preparation effort by the PIOs, the RIDs, and the PDD, Equipment Management Department and Farmers' Assistance Department of NIA's headquarters. Sixty-seven PIOs were selected for strengthening; the other five PIOs are already being strengthened under other projects. 3.04 The preparation of the communal irrigation construction component followed similar lines. PIOs identified subprojects on the basis of present problems which could be addressed by irrigation works (low yields, low cropping intensity and poor land drainage), and potential for irrigation development (available water resources, extent of arable lands, potential irrigable area and possible alternative developments). PIOs are well suited to the task of identifying subprojects because they have grassroots contacts with the farming community and first-hand knowledge of the relative needs and potential of areas within their jurisdiction. Subproject proposals were reviewed by the RIDs, then forwarded to the NIA headquarters for final selec- tion. This is the normal procedure for approval of communal subprojects, and the PIOs identified a large number of subprojects for possible Bank financing. The list of 144 subprojects covering 33,500 ha, picked up for inclusion in the project, is indicative of the type and quality of subpro- jects to be financed. Specific subprojects included for Bank financing may change if unforeseen problems arise with some, while a few subprojects not currently included may qualify for inclusion in the project. Thirty-three sample subprojects, prepared by NIA in sufficient detail, were appraised by the Bank mission, and the financial and economic viability of the construc- tion component extrapolated from the sample results. This procedure is justifiable since the sample size is large, and the sample subprojects are fairly representative; the complete list of subprojects to be financed is not  - 20 - firmly determined, but NIA will subject it to the same criteria stipulated for the 33 subprojects noted above. 3.05 The innovative aspect of communal system financing in the proposed project was the decision that all subprojects would be prepared under the participatory approach described in paras. 1.29-1.32. As noted, this approach has been tested on about 40 subprojects, and found to strengthen the IAs. While the participatory approach itself is no longer innovative, its use in a Bank-assisted irrigation project in the country is. NIA was attracted to the approach after seeing many of its engineering efforts wasted because IAs were not prepared to operate and maintain communal systems properly. The partici- patory approach to be pursued under the project proceeds within the same institutional structure as is used in existing systems, but relies on improv- ing its operations. The only organizational changes have been the establish- ment of an advisory Communal Irrigation Committee under NIA's Assistant Administrator for Operations to set policies and goals and to monitor parti- cipatory developments, and the fielding of community organizers (COs) at the village level to work with farmers to build strong IAs in preparation for taking full responsibility for their part in designing, constructing and operating their own communal system. The participatory approach depends on farmers organizing themselves to play an active role in developing their communal system, and on PIO officials accepting farmers as partners in this development. The training of officials and farmers in joint, consultative planning was from the beginning recognized as a prerequisite to the success of the new approach. Other Components 3.06 NIA already has a training program but it is not large enough to meet all the needs of the communal subsector. Under the project, the ongoing training program would be expanded to include not only all PIO staff, but also farmers, NIA regional staff and technicians from the Ministry of Agriculture. The Asian Institute of Management and the Institute of Philippine Culture assisted in the design of the training component. To test the effectiveness of the participatory approach and the subproject design standards, a monitor- ing and evaluation component was also included in the project. It would be much simpler and more specific than similar components in past Bank-assisted irrigation projects in the Philippines, using only a sample of project and nonproject systems, and aimed at measuring differences in net benefits to irrigators on communal systems of varying types. Finally, the Bank appraisal team suggested, and NIA agreed, to include in the proposed project a study to prepare the way for an integrated financial management and reporting system (see para. 1.10). 3.07 Funds for detailed engineering studies for two multipurpose pro- jects (Lower Agno and Jalaur II) were included in NIA-s project preparation. Feasibility level studies for these two projects have been completed recently with Bank assistance. NIA had proposed, and the appraisal mission had  - 21 - agreed, to include financing for the second phase of the studies in the proposed project. Subsequent to the appraisal, however, it was concluded that these studies may be financed from the unutilized funds available under the Second National Irrigation Systems Improvement Project (Loan 1526-PH) and the Jalaur Irrigation Project (Loan 1367-PH). At negotiations the Bank would review justification, funding needs and sources for these studies. If they are to be included in the proposed project, total project cost would rise by about US$2.0 million. 3.08 In sum, a large share of the Philippines infrastructure development budget has been invested in irrigation, many future irrigation projects are planned, and further investments must be made to ensure that the full potential of past and future projects is realized. The inevitable shift in NIA's activities from new construction toward operation, maintenance and rehabilitation, and the expected shift from national to communal system development, creates a need for strong PIOs. The growing emphasis on communal irrigation development calls for more effective communal system institutions and better training of those involved in communal irrigation to ensure proper design, construction, and O&M of the new system. Efforts to improve communal system development need to be monitored and their effectiveness evaluated. Finally, the management of NIA must have easy access to accurate and complete financial information in order to plan and implement the entire irrigation development program efficiently. The proposed project was formulated to meet these needs. 4. THE PROJECT project Objectives 4.01 The project would have the following objectives: (a) supporting a national program under which farmers would effectively participate in planning and construction of CISs and, after completion, would assume full control of the O&M of the systems; (b) strengthening the ability of PIO's to select, design and implement efficient CISs, and to maintain and operate small national systems. The project would also enhance NIAs capability to support IAs in the repair of CISs in case of exceptional damage to the systems by typhoons, etc.; (c) increasing the incomes of about 12,000 farm families currently growing rainfed crops by providing irrigation and support services to produce rice, and improving the incomes of another 12,000 families by rehabilitating and upgrading their present run-down CISs; and  - 22 - (d) development of NIA's capability to undertake the selection of CIS subprojects according to criteria agreed with the Bank (para. 4.15), and to determine the phasing of their implementation. In this respect, the project places substantial responsibility on NIA to plan and execute nationwide CIS investments in the coming years, with regard to both new projects as well as the rehabilitation of existing projects. Project Description 4.02 Over a period of five years, the project would: (a) strengthen 67 PIOs through provision for: (i) construction and improvement of office buildings and workshops; (ii) equipment and vehicles for construction, improvement, and O&M of irrigation facilities; (iii) equipment for offices and workshops; and (iv) instruments for topographic surveying and hydrological observations; (b) support training of about 2,600 COs and technical staff of PIOs, MOA extension staff, and farmer leaders from about 1,900 IAs; (c) provide for the construction of about 94 new communal irrigation subprojects on about 22,800 ha, and for the rehabilitation and extension of about 50 communal irrigation subprojects covering about 10,700 ha; (d) establish a system for monitoring the implementation, and evaluating the impact, of the project; and (e) finance a study to improve NIA's internal financial reporting system. Detailed Project Features 4.03 Strengthening of Provincial Irrigation Offices. Under the project, the PIOs would be directly involved in, and be principally responsible for, the organization of IAs and the application of the participatory approach in the planning, design, and construction of CISs. In order to remove constraints which are inhibiting the efficient performance of PIOs and also to enable them to handle a considerable increase in their responsibilities, the project would include: construction of 23 new office buildings and work- shops; rehabilitation of 36 existing office buildings and workshops; and provision of office and workshop equipment, and instruments for topographic surveying and hydrologic observations. The project would also provide equipment, machinery and vehicles for the construction of minor and scattered works under force account and for O&M activities.  - 23 - 4.04 The 67 PIOs to be strengthened are in eleven regions. The follow- ing criteria were used to determine the appropriate level of investments: (a) Office buildings would be similar in all provinces. These would have a floor area of about 200 sq m and workshops of about 100 sq m; plot areas would be about 1,000 sq m. (b) Requirements of equipment for offices and workshops, instruments for topographic surveys, mapping and hydrologic observations, and construction machinery and vehicles have been based on a classi- fication of each province. For this purpose the provinces were graded by giving numerical weights to: (i) estimated annual poten- tial area for planning and investigation, (ii) average annual construction program of communal irrigation development, (iii) the O&M support to the existing CISs (pump and gravity), and (iv) the O&M requirements of the existing small national irrigation sys- tems. Level I would represent the basic operating office. Requirements of office, survey, hydrological, and construction equipment and vehicles would progressively increase to Level V. 4.05 The basic facilities to be provided in PIOs of differrent classifi- cation levels are summarized in Table 4.1. The classification level of each of the 67 provinces is given in Annex 2, Table 1. Total equipment require- ments for the project are listed in Annex 2, Tables 2 and 3. Incremental technical staffing requirements for each PIO would be one planning engineer, one construction foreman, one mechanic, on instrument man, one survey aide, one accounting clerk and one driver. Additional COs would be appointed in each PIO on the basis given in para. 6.04. Most of the incremental technical staff is available within NIA as some of the ongoing projects are nearing completion. Recruitment of the remaining staff is expected to be completed by end of July 1982. Recruitment of 46 COs for the project was completed in February 1982; they are undergoing training prior to their assignment to the field.  - 24 - Table 4.1: PIO CLASSIFICATION LEVELS AND BASIC FACILITIES Classification level I II III IV V Office machines 5 6 6 7 7 Drafting equipment 2 2 2 4 4 Surveying instruments 4 5 5 8 10 Hydrological instruments 1 1 1 2 2 Communications equipment 1 1 1 1 1 Vehicles and trucks 4 7 7 10 10 Track-type tractors 1 1 1 2 2 Excavators 1 1 1 1 1 Motor graders - - - 1 2 Compactors - 1 1 1 1 Concrete equipment 1 1 1 1 2 Workshop tools 1 1 1 1 1 4.06 Training. The project would support expansion of a program ini- tiated with assistance from the Ford Foundation to train officials and farmers to work together effectively to develop CISs. Training provided to PIO staff would also enhance their ability to operate and maintain small national systems. The program would extend to all provinces of the country, reaching some 2,600 NIA and MOA staff and some 1,900 IAs during the next five years. Training would be provided to a wide variety of staff including RIO and PIO economists and engineers, monitoring and evaluation teams, COs, MOA extension workers, technicians, construction workers and equipment operators. They would receive broad based training in their own functions and would become familiar with the functions of others with whom they must collaborate. 4.07 COs would get training in three modules of about two weeks each. The first module would be given immediately after hiring. It would include: orientation to NIA, particularly communal irrigation development using the participatory approach, subproject profile writing, integrating with the farming community, mobilizing farmers and assisting them in organizing them- selves, and matters related to subproject design. The second module would be given about two months before the start of construction. It would include: the technical and institutional process during the construction phase, coordinating IA organizational work with construction work, recording construction costs, organizing farmers counterpart participation, financial management for farmers, and cost reconciliation. The third module would be  - 25 - given about six months after the start of construction. It would include: assistance to farmers in O&M, water management, conflict resolution, fee collection, crop production, and communal monitoring procedures. 4.08 The technical staff of PIOs would receive training in three modules of about two weeks each at times appropriate to their working needs. It would include feasibility studies and project planning, the institutional process during the planning, design and construction phases, project program- ming and budgeting, and design of irrigation structures, canals and on-farm facilities, organizing and staffing, construction cost recording, project accounting, performance measures, potential problem analysis, assistance to IAs in the O&M phase, water management, communal system monitoring procedures, and follow-up activities with IAs. 4.09 Planning Engineers and CO supervisors of RIOs would be trained together in the review and assessment of socio-technical project profiles. They would also receive two weeks of training for trainers, with subsequent brief refresher courses. The training of irrigation technicians would be basically similar to that of CO supervisors, with additional emphasis on assisting farmers during the O&M phase in water management and crop produc- tion. To improve coordination they would receive water management and crop production training together with farm management technicians of MOA. Design engineers working on communal irrigation systems would further their skills through training with NIA engineers who have been trained abroad under the NISIP II project (Loan 1526-PH). Field engineers and construction foremen would receive an orientation on the participatory approach lasting about a week. This would be followed by a planning workshop held jointly with construction staff, COs and IA leaders before starting construction of any communal system. The NIA central office monitoring and evaluation super- visory team and nine regional teams would be given about two weeks of train- ing in monitoring and evaluation of CISs. It would include an orientation to the organization and work of the regional offices, PIOs and IAs, the partici- patory approach, and the main tasks and problems of O&M. 4.10 IA leaders and selected members would be given training in sessions of one to three days at different times through the course of development of their system. The first module would be a one-day session for the entire membership and would include a presentation and discussion of feasibility designs and an introduction of the technical and institutional processes to be followed during the detailed design phase. In the second module, about one month before construction, training would be provided to IA leaders on keeping records of construction costs, cost reconciliation procedures, and of counterpart participation by the IA. For the farmers convenience, this two-day module would be divided into shorter one-day or half-day sessions, as would the later modules. About three months after the beginning of con- struction, a two-day training module would be held for IA leaders in the management of the financial affairs of the IA, stressing bookkeeping proce- dures for all expenditures and receipts. Before completion of construction  - 26 - but at least a month before the IA begins to operate the system, the first training for O&M would be given. It would include: O&M procedures, conflict resolution, management of fee collection, follow-up on record keeping, and the procedures for subproject monitoring (for those systems to be monitored). This module would last one day for the entire membership and about two days for selected IA leaders. Shortly after the IA begins operating the system a second training session of about three days would be provided for the O&M phase, stressing water management and crop production for all members of the IA. 4.11 A team drawn from three of the 40 pilot communal systems composed of about nine IA farmer leaders and about six NIA personnel (engineers, COs, CO supervisors) would be sent overseas in 1982 to study various aspects of irrigation practices, water management, O&M and crop production. A second team from three other systems would receive similar training in 1983. By 1984 six model communal systems would be established, two each in Luzon, the Visayas and Mindanao. These would serve as demonstration and training cen- ters for farmers from other communal systems in the Philippines. 4.12 At the beginning of each year PI0 technical and institutional staffs would meet for a three-day regional workshop to review their exper- ience in coordination, study cases, receive additional training in coordina- tion and conflict resolution skills, and set their general coordination plans for the year. The technical and institutional staff would then meet at least once each month with their PIE for a review session. 4.13 Communal Irrigation Rehabilitation and Development. The project would provide for the improvement of irrigation facilities on about 10,700 ha in 50 existing CISs and the construction of new irrigation works on about 22,800 ha, consisting of 94 new communal irrigation subprojects. The sub- projects range in size from 50-1,000 ha and would be based on run-of-the- river supply, served by gravity irrigation. Typical subprojects would be served by diversion weirs; a few would need intake structures only. The diversion weirs would be of concrete ogee type with a crest height of 1-2 m above the stream bed. The length of each weir would be determined by the 25-year frequency flood. Intake structures would be steel-gated with constant head orifice turnouts for discharges up to 1 cu m/sec. Water flow measuring devices would be included in structures discharging more than 1 cu m/sec. Generally, the main canals would be unlined: concrete lining would be used only where soils are susceptible to erosion and where possibility of excessive seepage is identified. Check structures and gated turnouts would be provided as necessary to ensure effective water control. 4.14 For areas smaller than 100 ha or where the service area is elongated, secondary canals would be eliminated and the main canal would serve the distribution system. Secondary unlined canals would be provided for larger subprojects, as necessary. Service roads of 4.0 m width and 15 cm thick gravel surface would be built along the main canals, and access roads  - 27 - would connect service roads to the nearest barangay, provincial or national road. Drainage would be provided through outlets to secondary drains and natural channels. Farm drainage, in most cases, would be from one field to the adjacent lower field. 4.15 Subproject Selection Criteria. All NIA communal irrigation sub- projects would meet following minimum criteria: (a) the expected cropping intensity should be at least 130%; (b) the proposed subproject area should be suitable for irrigated crop production; (c) there should be minimal conversion of land use from productive permanent crops like coconuts and orchards; (d) the area should not be less than 50 ha or more than 1,000 ha; (e) the average farm size should not be more than 2 ha; (f) the internal economic rate of return of the subproject should be not less than 10%; and /1 (g) farmers should be actively involved in preparing, and should concur with, the initial feasibility design of the subproject. During negotiations agreement was reached with the Government that NIA would by December 31, 1983, begin to implement a phased program satisfactory to the Bank and IFAD with a view to achieve by December 31, 1985, a uniform set of selection criteria and design standards satisfactory to the Bank and IFAD to be applied to all its CISs, regardless of financing source. 4.16 Subproject Preparation and Evaluation. Feasibility studies of 33 sample subprojects were prepared before appraisal. These were evaluated in detail by PDD and appraised by the Bank mission. Detailed engineering design of these subprojects would be undertaken by the RIDs of NIA and reviewed by the Project Manager. Construction is scheduled to start by January 1983. The preparation of the remaining subprojects including topo- graphic mapping, soil and water resources studies, layout and cost estimates would be done by the respective PIOs. Short gestation subprojects would be given priority. Completed feasibility studies and subproject profiles would be evaluated by a multidisciplinary team from the PDD. Detailed engineering /1 The rate of return of the 33 sample subprojects evaluated by PDD varies from 10% to 36%.  - 28 - design would then be undertaken by the RIDs and would be reviewed by the Project Manager against standard design criteria. The technical and institutional staff of the PIOs would be responsible for finalizing the design and undertaking requisite preconstruction and construction activities in consultation with the IAs. Supervision of field work would be the responsibility of the RIDs who would regularly report their findings to the Project Manager. During negotiations agreement was reached with the Government that NIA would submit for review by the Bank prior to implementation full evaluation reports for subprojects costing more than P 5.0 million each and summary data forms for subprojects costing less than P 5.0 million. 4.17 Monitoring and Evaluation. In order to learn from experience gained under the project, NIA would improve its existing monitoring and eval- uation (M&E) system with the intention of improving CIS policy and adminis- trative procedures and developing more effective training programs for the staff and farmers involved. This would be done in two ways: NIAs existing reporting requirements for communal works would be expanded and a new M&E scheme would be implemented. 4.18 NIA has an established program of monthly reporting of CIS con- struction progress from the PIO to the RIO and then to the NIA central office. This system is being computerized to increase its usefulness to NIA's senior management. Under the project, reporting would be extended to include progress during the design phase of a CIS including establishment of the IA, fielding of the CO, obtaining of water permit, completion of design, and selection of farmer committees. 4.19 NIA would also establish a system to monitor the O&M of a sample of communal subprojects by lAs during the first three years after the completion of construction and to evaluate the impact of these subprojects after the first and third years of cropping. The sample would comprise 40 Bank-assisted subprojects and 40 locally-funded subprojects in order to compare the benefits achieved in the two sets of CISs with regard to types, amount and distribution of benefits to members. A supervisory team in NIA's Farmers' Assistance Department would design and oversee the M&E work. This national team would comprise a statistician, an economist, an agriculturalist and a sociologist. Nine comparable teams at the regional level would conduct studies, supervise local data collection, and supervise and train one farmer- member from each of the 80 IAs who would be employed at a modest wage to conduct periodic data collection. Regional teams would comprise an agri- cultural economist, an agriculturalist and a sociologist. Work would proceed in three phases: benchmark survey, O&M monitoring, and O&M evaluation. During negotiations agreement was reached with the Government that NIA would submit for review by the Bank and IFAD before December 31, 1982, the indicators that would be monitored and other details of the M&E system.  - 29 - 4.20 A benchmark survey would be carried out during construction of each of the selected systems. This would measure level of farmer organization, water control and cropping practices, use of inputs, level of outputs and level of agricultural incomes. Regional teams would verify data before analyzing and summarizing them for the IA, the PIO, the RIO and the NIA central office team. 4.21 During the wet and dry cropping seasons of the first three years of system operation, simple surveys would be conducted of a sample of IA mem- bers twice each cropping season regarding input availability and distribution system, input utilization and farm management practices, adequacy of irriga- tion water and water management practices, O&M of irrigation facilities, and crop production levels and farm income. The regional teams would also gather data once each cropping season regarding IA organization, management and performance. The regional teams would verify data and analyze and present it for review by the IAs, PIOs, and RIOs. Results of each periodic survey would then be presented and discussed in a provincial meeting of IA leaders and the PIO. This joint IA-PIO review of progress and problems arising from monitor- ing reports would be a critical channel for improving the performance of some CISs, providing useful recommendations for training and helping PIO staff improve their contribution to the large number of CISs not included in the M&E sample. 4.22 After the second year of monitoring, survey forms and system-level monitoring studies would be modified to prepare for evaluation. Provisions would be added that would help compare third year system outputs more com- pletely with the benchmark data and would help compare system benefits more completely with project objectives. The third year of monitoring would thus be somewhat more extensive in that it would enable NIA to make a thorough evaluation of impacts and benefits from the sample systems. After completion of the third year of monitoring (in the fourth and fifth years of the project), the regional teams and central M&E teams would produce a general evaluation of accomplishments and recommended redirection of NIA strategies. A copy of this report would be forwarded for review by the Bank and IFAD. 4.23 Review of NIA's Financial Reporting System. The project would support a consultant study of NIA's financial reporting system. The purpose of the study would be to review the present system and prepare a five-year work program to develop an integrated financial management and reporting system. Work undertaken for this component would be coordinated with NIA's ongoing efforts to improve its electronic data processing and management information sytem. It is estimated that about 35 man-months of local consulting services would be required at about US$4,500 per man-month inclusive of overheads, office costs, and local travel. Assurance was obtained during negotiations that NIA would prepare by December 31, 1982, for review by the Bank, terms of reference for consulting services for the financial reporting system study.  - 30 - Status of Engineering 4.24 Feasibility grade designs, layouts, and drawings have been prepared by NIA for 33 sample subprojects (22 new areas and 11 rehabilitation areas) covering 6,700 ha in 24 provinces or about 20% of the total project area. This would serve as the basis for the entire first year construction program. Construction quantities have been estimated for the diversion works, main and secondary canals, tertiary canals and drains, access and service roads and all appurtenant structures, from plans and profiles supported by field surveys and site investigations. Adequate investigations have also been conducted to ensure the availability of ample supply of construction mater- ials. Designs and drawings for buildings in all provinces have been prepared. Preparation of bid documents for equipment and vehicles is underway. Prepa- ration of detailed designs and tender documents is thus sufficiently advanced for procurement and construction to start soon after loan effectiveness (see para. 6.06). 4.25 Hydrological and soil studies, and determination of water require- ments for the 33 sample subprojects have been completed. Studies for the remaining subprojects are underway. Hydrological studies were carried out by the Research and Development Section of the Agricultural Division in each regional NIA office. Data on temperature, relative humidity, wind velocity and rainfall were collected from the various ordinary rainfall and synoptic stations operated and maintained by the Ministry of Public Works and Highways (MPWH), and the Philippine Atmospheric Geophysical and Astronomical Service Administration. Evaporation data for the study were taken from eleven stations located near the project areas. In areas where no evaporation data were available, Penman's formula was used to estimate the monthly evaporation rate. Estimates of monthly stream flows for ungauged rivers and creeks and for areas with incomplete data on river discharges were correlated with nearby gauged streams, using concurrent periods of operation studies using synthesized flow records and estimated water requirements and assuming no return flows. Available water supply was assumed to be about 90% of river flow for subprojects with diversion dams and about 70% for subprojects pro- vided with intakes only. Water quality is suitable for agriculture. Only 5 rivers out of the 33 sample project rivers have prior approved water rights to users other than the proposed communal subprojects. Previously appropri- ated waters are small compared to the total discharges of the rivers and would not affect the irrigation supplies for the project. 4.26 For each of the sample subprojects, available supply of water was matched against the projected water requirements for an annual cropping intensity of 165%. The proposed cropping calendars for the sample projects would maximize the use of rainfall for both wet- and dry-season crops without exceeding local manpower constraints. Field water requirements have been estimated for most provinces and differ from each other mainly because of varying evapotranspiration according to different climatic conditions, growth stage of the crop, and varying percolation losses depending on the permeability of the soils. The annual diversion requirements were estimated  - 31 - by adjusting the field water requirements with the effective rainfall in the area and the efficiency of the irrigation delivery system. 4.27 Studies done by NIA show that all sample subprojects have soils suitable for rice cultivation. Soil maps and land-use maps from the Bureau of Soils were used in the identification of subproject areas. Composite soil samples (four samples/ha) were processed either in the regional soil laboratories of the Bureau of Soils or in one of NIA's own soil and water laboratories. 5. PROJECT COSTS AND FINANCING Cost Estimates 5.01 Detailed cost estimates for 33 sample subprojects based on feasi- bility grade designs (para. 4.24) were examined at appraisal. Cost estimates for the remaining 111 subprojects, for which surveys and designs are being prepared, have been derived from the detailed estimates of the sample sub- projects by a process of comparator extrapolation. For the canal systems, the sample subprojects were grouped into new and rehabilitation schemes. Each of the two categories were further divided into three subgroups accord- ing to size: below 250 ha, 251-500 ha, and 501-1,000 ha. Based on the sample subproject estimates, the average costs per ha for each subgroup were determined. The costs of the canal systems for the remaining 111 subprojects were then determined using the appropriate unit cost per ha corresponding to the size and category of each subproject. For the diversion works, the estimated cost per meter of weir length for various heights of weirs were determined on the basis of actual detailed engineering designs and layouts of the sample subprojects. These unit costs were then applied to determine the costs of diversion weirs for those of the remaining subprojects wherein the approximate heights and lengths of diversion weirs had been decided by actual field investigations. For rehabilitation of diversion works, the average cost of the sample rehabilitation subprojects was used to determine the cost of the remaining rehabilitation subprojects. 5.02 Total project costs are estimated at US$120.8 million, of which US$61.5 million or about 51% is foreign exchange. Including a capitalized front-end fee of US$1.05 million on the proposed Bank loan, the total financing required for the project is estimated at US$121.85 (rounded to US$121.8). Taxes equivalent to 3% of the costs of contracts for civil works are included in the cost estimate. The Government exempts NIA from paying taxes or duties for equipment, materials and consulting services procured under foreign-assisted projects. NIA's costs for design and construction supervision are included. Unit prices are based on recent quotations for similar items received by NIA and are indexed to mid-1982. Physical contingencies of 20% for irrigation works, 10% for buildings, training and monitoring and evaluation, and 5% for equipment have been applied. Price contingencies have been applied at the following rates:  - 32 - Annual inflation rate (%) 1982 1983 1984 1985 1986 1987 Equipment & Civil Works Local 10.0 8.0 8.0 8.0 8.0 8.0 Foreign 8.0 8.0 7.5 7.0 6.0 6.0 Over the implementation period these amount to about 24% of the base cost plus physical contingencies. Cost estimates are presented in Table 5.1. Financing 5.03 The proposed Bank loan of US$71.1 million, including capitalized front-end fee, and the IFAD loan of SDR 10.8 million (US$12.0 million equivalent) would together finance about 71% of total project costs excluding taxes and the cost of land acquisition, or about 68% of the gross total costs. The loan would cover 100% of the foreign exchange requirement (US$62.5 million) and also about 35% (US$20.6 million) of the local costs. Financing of local costs is justified by the shortage of counterpart funds in the Philippines. Of the remaining project cost of about US$38.7 million, Government would finance about US$33.7 million and the rest would be contributed by farmer beneficiaries through their equity contributions to communal irrigation system construction through land, labor, materials or cash. The full amounts of the loans would be channeled through Government directly to NIA. In the past, the Goverment has treated transfer of these resources as direct public works appropriations or as an equity contribution to NIA and no onlending arrangement between the Borrower and NIA has been necessary. At present Government is considering changes in its financial treatment of NIA and other autonomous institutions. At negotiations Government agreed that in the event NIA would be responsible to service this debt obligation, the terms and conditions of the onlending agreement would be discussed with the Bank and IFAD. 5.04 In order to begin construction of project works in 1983, the first year of project implementation, certain preparatory works would have to be undertaken. PIO buildings would need to be constructed and office equipment and construction machinery purchased for the provinces where the first sub- projects, selected from among the sample communal systems for which feasibi- lity studies and preliminary cost estimates have been prepared, are to be implemented. Staff would need to be trained and community organizers mobi- lized to help form IAs to review and concur with the preliminary design and cost proposals. Detailed topographic surveys would be carried out and the detailed design prepared in advance. To provide for these preconstruction activities during the course of 1982, expenditures of up to $1.5 million, made after January 1, 1982, would be considered for retroactive financing.  - 33 - Table 5.1: COST ESTIMATE /a Foreign Percentage Local Foreign Total Local Foreign Total exchange of base cost ----- (P000) -- -- ---- (US$000) ------- (%) (%) Strengthening of Provincial Irrigation Offices (PIO) Office buildings 11,275 7,517 18,792 1,392 928 2,320 40 Workshops, sheds and site works 7,484 3,208 10,692 924 396 1,320 30 Office equipment 2,025 8,100 10,125 250 1,000 1,250 80 Survey and hydrological equipment 527 4,738 5,265 65 585 650 90 Construction machinery and vehicles /b 5,988 113,659 119,637 738 14,032 14,770 95 Workshop equipment and tools 97 1,847 1,944 12 228 240 95 Subtotal 27,386 139,069 166,455 3,381 17,169 20,550 84 24 Training Local programs 14,783 - 14,783 1,825 - 1,825 - Foreign programs 405 3,645 4,050 50 450 500 90 Consulting services 1,620 2,430 4,050 200 300 500 60 Subtotal 16,808 6,075 22,883 2,075 750 2,825 27 3 Irrigation Systems Diversion and intake works /c 80,190 80,190 160,380 9,900 9,900 19,800 50 Main canals /c 40,873 33,444 74,317 5,046 4,129 9,175 45 Service and access roads /c 30,051 45,077 75,128 3,710 5,565 9,275 60 Distribution canals and drains 14,029 11,486 25,515 1,732 1,418 3,150 45 On-farm ditches and drains 34,968 28,617 63,585 4,317 3,533 7,850 45 Topographic survey, mapping and services 8,505 5,670 14,175 1,050 700 1,750 40 Subtotal 208,616 204,484 413,100 25,755 25,245 51,000 50 60 Land Acquisition Strengthening of PIOs 3,022 - 3,022 373 - 373 - Irrigation systems 26,017 - 26,017 3,212 - 3,212 - Subtotal 29,038 - 29,038 3,585 - 3,585 - 4 Monitoring and Evaluation Personnel and training 8,586 - 8,586 1,060 - 1,060 - Equipment and vehicles 146 584 730 18 72 90 80 Materials and supplies 680 1,020 1,700 84 126 210 60 Subtotal 9,412 1,604 11,016 1,162 198 1,360 15 2 Engineering, Supervision & Administration /d Strengthening of PIOs 4,374 486 4,860 540 60 600 10 Irrigation works 37,908 4,212 42,120 4,680 520 5,200 10 Subtotal 42,282 4,689 46,980 5,220 580 5,800 10 7 Financial Reporting System Study 324 1,296 1,620 40 160 200 80 - Base Cost 333,866 357,226 691,092 41,218 44,102 85,320 52 100 Physical contingencies /e 45,878 49,702 95,580 5,664 6,136 11,800 52 Expected price increases 100,716 91,133 191,849 12,434 11,251 23,685 48 Total Project Cost 480,460 498,061 978,521 59,316 61,489 120,805 51 Front-end Fee on Bank Loan /f - 8,511 8,511 - 1,051 1,051 100 Total Financing Required 480,460 506,572 987,032 59,316 62,540 121,856 51 /a Indexed to mid-1982. 7T Includes 10% of cost for spare parts. 7T Includes 3% tax on civil works contracts. Th About 10% of irrigation works, 5% of buildings and 2.5% of equipment costs. /e About 20% of irrigation works, 10% of buildings, training, and monitoring and evaluation, and 5% of equipment. /f Based on Bank loan of US$70.0 million.  - 34 - 5.05 Special Account. Government's time-consuming reimbursement pro- cedures and sometimes untimely budgetary allocations during each fiscal year have presented difficulties for smooth project implementation. At the level of project implementing agencies, such as NIA, liquidity is frequently constrained by the agency's inability to plan and manage its own working capital resources. This situation would be exacerbated in the communal irrigation project because its geographically dispersed nature and the large number of small individual disbursement requests to be submitted for each subproject extend further the processing time between agency expenditure and final reimbursement. Moreover, the whole approach to participatory communal development assumes continuity of work and timely responsiveness to farmer inputs and places a premium on smooth implementation. And finally, because the small contractors and suppliers who will be servicing the project have very limited financial resources, prompt payment of their claims is vital for ensuring continuity of their services which would otherwise be subject to frequent and serious interruption. Because of these considerations, and at the specific request of NIA and the Government, it is proposed to establish a special account to accelerate disbursements. 5.06 A peso account (the Special Account) would be established in the Philippines National Bank with advances from the proposed loans. The initial advance from the Bank loan account would amount to $250,000 rising to a maximum of $850,000, as needed. IFAD would advance up to $150,000 pari passu with the Bank advance. Disbursements from the Account would cover the Bank-s and IFAD's shares of expenditures for force account civil works, small civil works contracts (less than P 1,000,000), and small local purchases of equip- ment and materials (items costing less than P 50,000 equivalent). Replenishment of the Special Account would take place periodically on the authorization of requests by NIA-s Assistant Administrator for Finance. Justification for disbursements made out of the Account would be submitted by NIA prior to replenishment, and a separate annual audit by an external auditor would be provided. It is NIA's intention, with which the Bank agrees, to extend gradually the purposes for which funds in the Account are used and to ensure that the Account continues to provide liquidity after loan closing. Therefore, NIA would fund its liquidity by opening and maintaining a counter- part account (the Operating Account) during project years 2-6 with the deposit of the equivalent of US$250,000 per year from its own resources. This would result in an established fund of over US$1.5 million equivalent for general corporate purposes after the end of the Bank replenishment of the Special Account. The creation of the Special Account would constitute, therefore, an important institution-building element, the significance of which extends well beyond the proposed project. During negotiations, agreement was reached with the Government that NIA would establish the Special Account and the counterpart Operating Account and maintain and operate the accounts in a manner satisfactory to the Bank and IFAD. Procurement 5.07 Goods. Construction machinery, vehicles and spare parts, equipment for offices and workshops, surveying and hydrologic instruments and con- struction materials and supplies, totalling about US$19 million equivalent,  - 35 - including contingencies, would be grouped in appropriate bidding packages and procured by international competitive bidding (ICB) in accordance with Bank Guidelines. Qualifying domestic manufacturers would receive a preference in bid evaluation of 15% of the c.i.f. price of imported goods, or the import duty, whichever is lower. Miscellaneous items of equipment and machinery, costing less than US$20,000 each and totalling not more than US$3 million, including contingencies, would be procured under local competitive bidding procedures, which are satisfactory to the Bank and IFAD. 5.08 Works. Office buildings and workshops totalling about US$5 mil- lion, and the diversion works, main canals and roads totalling about US$56 million, including contingencies, are scattered over 67 provinces and 145 locations. The small size and wide geographical dispersion of these works are unlikely to attract foreign contractors. To the extent possible, contracts would therefore be let by local competitive bidding procedures which are satisfactory to the Bank and IFAD and in which foreign bidders have the opportunity to participate. In cases, where the small size or remoteness of works discourage even local bidders, works would be done by force account, up to a maximum of US$14 million. The balance of civil works, totalling about US$18 million, including contingencies, consist of minor works related to distribution systems, onfarm facilities and appurtenant structures and are not likely to interest local contractors. Therefore these works would be carried out under force account. Assurance was obtained during negotiations that the aggregate cost of civil works done by force account (including US$18 million for minor works and up to US$14 million of the diversion works, main canals and roads) would not exceed 40% of the total cost of civil works, or US$32 million. 5.09 Contract Review. Between 300 and 400 bidding packages are expected to make up the estimated US$61 million in domestic civil works contracts and US$22 million in the procurement of goods, including US$19 million under ICB procedures. All bidding packages for works estimated to cost over US$400,000 equivalent and bidding packages for goods over US$200,000 equivalent would be subject to the Bank's prior review of procurement documentation. This would result in a coverage of about 40% of the total estimated value of civil works contracts and about 75% of goods contracts. About 25% of the total number of packages or between 75 and 100 contracts amounting to about US$41 million (US$17 million in goods and US$24 million in works) would be subject to prior review over the five-year period, which represents a reasonable work load for Bank staff and is consistent with NIA-s record of good performance in the preparation of procurement documents. The balance of contracts would be subject to random review by the Bank after contract award. Disbursements 5.10 Disbursements would be made as follows: (a) 100% of foreign expend- itures for directly imported equipment and materials, 100% of the ex-factory cost of such items manufactured locally, and 80% of the cost of imported items procured locally; (b) 75% of expenditures for civil works; and (c) 100% of expenditures for training, consultants' services, and the monitoring and evaluation of component. Disbursements for force account civil works would be made against statements of expenditure based on agreed unit rates. NIA  - 36 - would submit a detailed unit rate analysis with supporting documents for Bank review before September 30, 1982, or before claiming the first reimbursement for force account work, whichever is earlier. These unit rates would exclude depreciation of the equipment directly financed under the project. Disburse- ments for minor civil works contracts (less than P 1,000,000) would also be made against statements of expenditure based on actual progress payments made to contractors. In addition, statements of expenditure would be used for minor local expenditures for equipment and materials (items costing up to P 50,000 equivalent), for local training programs, and for expenditures under the monitoring and evaluation component. The use of statements of expenditure for these items is justified because of the small and frequent payments involved. NIA has the institutional capability to maintain related records in a satisfactory manner, and supporting documentation would be made available for review by visiting the Bank missions. All disbursements shall be apportioned by the Bank, as nearly as practicable in the circumstances, between the Bank loan and the IFAD loan in the ratio of 6:1. 5.11 It is expected that disbursements would be completed by June 30, 1988, approximately one year after the end of construction, due mainly to required lagtime for releasing contractors' retention monies. An estimated schedule of disbursements is given in Annex 2, Table 5, which also includes a comparison with regional and country disbursement profiles for the irrigation subsector. The proposed allocation of IBRD and IFAD loan proceeds is given in Annex 2, Table 6. The country disbursement profile shows higher disbursement percentages than the regional profile during the first four years of implementation but both profiles indicate that full loan disbursement is accomplished in eight years. Despite these indications, it is expected that the proposed loan would be disbursed in six years owing to the following factors: (a) the project provides for retroactive financing from January 1, 1982. Therefore, the effective disbursement period is 6-1/2 years; (b) the proposed project is essentially different from other projects in the disbursement profile insofar as it would pick up a time-slice of ongoing activity in the subsector. CISs covering an area of over 300,000 ha have been implemented during the decade 1971-80; (c) the project comprises a large number of small subprojects which are easy to design and construct under the authority delegated to a large number of regional and provincial offices; (d) a large number of subprojects would be implemented simultaneously by small contractors and independent force account crews in different provinces; (e) the status of engineering on 33 subprojects is sufficiently advanced to permit an early start on project implementation, and the engineering of the remaining subprojects is expected to be completed well ahead of implementation;  - 37 - (f) the project would not suffer from land acquisition delays (see para. 6.07). Accounts and Audit 5.12 The NIA is a government agency and its accounts are audited annually by the Government's Commission on Audit (COA). Assurances would be obtained during negotiations that NIA would cause all its regional and provincial offices to maintain separate accounts for the project. The project accounts would be collated by the Project Manager. After audit by the COA, the pro- ject accounts, together with the auditor's comments, as well as his opinion on the certification of expenditure on force account works, would be sent to the Bank within six months of the close of each fiscal year. Accounting proce- dures for force account work and minor contracts are given in Annex 6. 6. ORGANIZATION AND MANAGEMENT 6.01 NIA would be responsible for all components of the project, which would be under the Assistant Administrator for Operations. An ad-hoc advisory committee, named the Communal Irrigation Committee, consisting of specialists from within and outside NIA, has been in existence for some years and would continue to advise the Assistant Administrator for Operations on policy issues and provide feed-back on the state-of-the-art in communal irrigation develop- ment. A Project Manager has been appointed to handle the day-to-day imple- mentation of the project and supervision of the work of concerned Provincial and Regional offices, and of the project-connected functions of other agencies in NIA central office. He would also assist the Assistant Administrator for Operations to oversee the entire communal irrigation subsector and partic- ularly to monitor the adherence to the subproject selection criteria (para. 4.15). He would be responsible for compiling and submitting to the Bank quarterly progress reports covering all project activities. The organization of the Project Manager-s office and its interrelationship with other agencies concerned with the implementation of the project are shown in Figure 6.1. The Project Development Department of NIA central office would have responsibility for support functions related to the overall preparation and economic viability of the subprojects; Design and Specifications Department, for the engineering design of civil works; Equipment Management Department, for equipment specifications and procurement; Farmers' Assistance Department, for institutional development and agricultural coordination, and for the M&E component; and Training and Development Division, for the training component of the project. The Project Manager took up the assignment end-October 1981 and started preparatory activites for project implementation.  - 38 - 6.02 The Project Manager, under the guidance of the Assistant Adminis- trator for Operations of NIA, would plan, schedule and coordinate the various activities of the project. He would arrange for release of funds, maintain records of financial transactions and exercise overall control of funds for the project. He would conduct inspections of project works to ensure compli- ance with standards of quality, examine progress of work in all regions to ensure that status of construction and institutional development is on target, coordinate with other departments to facilitate extension of agricultural support services to the project areas, and arrange for the training of project personnel and farmers. He would prepare required reports on the project for submission to the NIA Board and the Bank. 6.03 The regional directors would review the plan of work prepared by the PIOs and request the Project Manager for release of funds. They would also review subproject designs prepared by the PIOs and prepare the design of structures beyond the technical capability or delegated authority of PIOs. They would supervise construction, conduct inspections, watch progress of work, supervise formation and organization of IAs and conduct training of project personnel and farmers. One CO supervisor in each RIO would assist the regional directors in the functional supervision of the work of institutional development under the participatory approach. RIOs have adequate staff and do not need further strengthening. 6.04 The PIOs would be the principal functionaries of the project. They would maintain lists of promising communal irrigation sites and undertake sociotechnical profile studies of potential subprojects as a basis for initial selection. A feasibility study would be done of selected subprojects to ascertain that necessary conditions exist for development. The PIOs would prepare plans of work for subprojects included in the project, prepare designs of canals and structures, assist farmers in formation and organization of IAs, and arrange and supervise construction of subprojects. They would process disbursement vouchers, maintain accurate records of financial transactions and control budget. Aside from the technical staff, a number of COs would assist each PIO in the work of institutional development under the participatory approach. Normally, one CO would be assigned in each subproject less than 200 ha. Additional COs would be assigned in larger subprojects or in sub- projects with unusually small land holdings. NIA already has about 150 COs and there is a continuing program of further recruitment and training. Project Implementation 6.05 Project implementation would follow established NIA procedures with some adaptations for the participatory approach in which NIA staff and farmers would together build viable IAs and work together through all phases of designing and building the system. Farmers would be organized into an IA prior to final design, and the IA would agree to the construction of the subproject. The coordinated technical and institutional process is derived from NIA's experience with about 40 communal projects undertaken in the past five years with assistance from the Ford Foundation.  - 39 - 6.06 Implementation of a typical subproject is depicted in the flow chart Figure 6.2 which describes the sequential steps from feasibility study of the subproject to completion of construction and turnover of the system to the IA for O&M. COs were fielded beginning January 1982 in about 20 sub- projects, where IAs were already sufficiently organized, to review and concur with feasibility design and cost proposals. While the technical staff would conduct detailed surveys and investigations, and prepare final designs and cost estimates, the CO would assist the farmers to adopt a set of by-laws, register with the SEC, obtain a water permit, and acquire land for construc- tion. Continuing close coordination between officials and IAs would enable farmers to understand and jointly review the construction designs and cost estimates, so that tender and contract documents could be prepared to permit an early start of the tendering process and the award of construction con- tracts for diversion works, main canals and service and access roads soon after loan effectiveness. The cycle would be repeated and the number of subprojects increased in subsequent years of the project. Land acquisition and procurement of materials for construction of buildings and workshops for 22 PIOs was started in November 1981 and construction would be undertaken in 1982. Land for the remaining 45 PIO buildings and workshops would be acquired, and tender documents prepared, so as to commence construction early in 1983. Preparation of tender documents and technical specifications for construction machinery, office and workshop equipment, surveying and hydro- logical instruments was started in November 1981 and is making good progress. Bid invitations would be issued soon after loan effectiveness so that award of supply contracts can be made in early 1983. To enable an early start of noncontract activities and force account works, a limited quantity of con- struction machinery and equipment for offices and workshops would be procured locally. Field investigation, preliminary design, consultation with IAs, design finalization, and preparation of tender and contract documents would be carried out for groups of subprojects covering about 5,500 ha in 1982, 7,500 ha 1983, 9,500 ha in 1984, and 11,000 ha in 1985 for implementation in the respective subsequent years. An estimated schedule for project implementation is given in Figure 6.3. 6.07 In the past, some projects for the construction of national irri- gation systems have experienced land acquisition delays. The proposed project design specially addresses this problem. Firstly, since the farmers would themselves own the irrigation system after completion of construction, the Government or NIA would not be "acquiring" land in the usual sense, but the farmers would be "contributing" land to the IA, of which they would be members. So the farmers would in a sense continue to own the "acquired" land which would vest in the community. Secondly, the farmers would fully parti- cipate in all stages of designing and constructing the irrigation system. Various alternative layouts of the system would be fully discussed in farmers' meetings, and the layout finally adopted would be fully acceptable to the farmers. By the same token, once a layout is adopted, it would be very difficult for any member of the community to refuse to provide his land to the IA. Finally, if any subproject has land acquisition problems which cannot be satisfactorily resolved, it would be dropped in favor of another as the size of the subsector affords a large field of selection.  - 40 - 6.08 Overall supervision of the project's training component would be the functional responsibility of the NIA Training and Development Division. The Division-s functions include preparing long-range manpower development pro- grams, establishing training needs and course curricula, evaluating training effectiveness and administering NIA's external and overseas training program. NIA's training capability has been strengthened through implementation of previous Bank-assisted projects, particularly the substantial training compo- nent under the Second National Irrigation Systems Improvement Project (Loan 1526-PH). Staffing is adequate both in the central Training and Development Division and in the regional Farmer's Assistance Divisions. Only eight addi- tional positions of trainers would be required in NIA. MOA's Bureau of Agri- cultural Extension would train the agricultural extension workers. Outside specialist assistance would consist of three consultants from the Asian Institute of Management, two from IRRI, and one from the Development Academy of the Philippines. The Asian Institute of Technology, Bangkok, would assist NIA in the formulation of a training program for supervisors on subproject design and management. Training would be provided at the 11 regional centers and at NIA's national training center at San Raphael. 6.09 NIA's Farmers' Asistance Department would have overall functional responsibility for the M&E component of the project. Two previous Bank- assisted projects, Chico River Irrigation Project (Loan 1227-PH) and the Second National Irrigation Systems Improvement Project (Loan 1526-PH), helped in the establishment of a large scale input-output monitoring system in NIA. These projects have helped to enlarge NIA's responsibilities and shift NIA's concern just with irrigation to concern for the broader problems of produc- tion. NIA no longer concentrates solely on how to design dams or align canals, but is now also interested in securing adequate inputs for producers and in arranging for well-organized disposition of their products. NIA has acquired substantial experience and expertise in the M&E of large irrigation systems which would be modified and simplified for the small-scale, widely scattered systems under the project. This process of adaptation and applica- tion has already started. Incremental staff required for the M&E component of the project is given in para. 4.19. Operation and Maintenance 6.10 Upon completion of construction and the computation and final recon- ciliation of costs, the PIO would turn over the system to the IA. The IA would accept this responsibilty by contracting with NIA to pay back the construction costs. Before operations begin, IA leaders and members would receive training in O&M, including record keeping, fee collection and system supervision. After operations have begun they would have another training workshop in water management and crop production. With this training and continuing help from various officials, the IA would take charge of all irrigation functions: water allocation, system maintenance, conflict reso- lution, and fee collection. The CO would remain with the IA during the initial settling-in phase and facilitate assistance from MOA extension agents, suppliers of inputs and sources of credit. The technical and institutional  - 41 - processes would become integrated into a joint process of continuing efforts by the IA to deal with both the technical and organizational problems of sustaining the irrigation system in order to improve agricultural production. During negotiations agreement would be reached with the Government on a typical proforma of the memorandum of agreement that would be signed by NIA and the concerned IA at the time of turnover of a completed system, setting out the rights and obligations of both parties, and the terms of repayment by the IA of the construction costs of the system. 6.11 The IA would pay water tenders to inspect structures, canals and ditches regularly and to do minor maintenance work, such as clearing debris from the water and from the screens of intakes, and patching minor cases of erosion. IA members would form work parties to repair eroded places along canals and remove silt from them. The IA capacity to maintain the physical system and collect water fees regularly is essential to keeping the system solvent. Further, capacity to resolve disputes, deliver water to members reliably, and to increase the availability of agricultural inputs and marketing facilities to members is crucial to lAs benefitting from the system. 6.12 For purposes of analysis, the annual O&M cost has been estimated at about P 125 per ha (see Annex 5). This would include materials and supplies (P 20); salaries of gatekeeper/water tender, ditch tenders and honorarium and traveling expenses for officers of the IA (P 80); equipment rental and operation (P 15); and miscellaneous expenses (P 10). Free labor would be contributed by the IA members. Fee collections by the IAs from their members would cover the O&M expenses - either in cash or in kind - as well as the loan repayments to NIA. Major repairs beyond their technical capability would be referred by the IAs to NIA for execution assistance as well as funding. The amount incurred by NIA for such major repairs would be charged to the IA concerned, and would be capitalized as an addition to the loan for construction of the system. 7. AGRICULTURAL PRODUCTION Present Cropping Pattern 7.01 In most of the project systems land preparation for the first or wet-season crop begins in May or June and transplanting is usually completed by July/August, while the harvesting is done between November and January. The second or dry season crop usually extends from December to April/May. In general, cultivation is by animal-drawn implements. The cropping calendar for rainfed paddy is dependent on the arrival of the southwest monsoon which can vary by a month or more from year to year. According to a national survey, about 75% of all rice paddies (irrigated and rainfed) are planted with high yielding varieties (HYVs). These HYVs require high levels  - 42 - of physical inputs and management which are often lacking. Practically all rice is transplanted from field nurseries onto previously puddled land. In general, farmers use low levels of fertilizer and only small amounts of insecticides and herbicides. Crops are usually hand-weeded twice during a season and are manually harvested and threshed. Grain is sun-dried on mats, roads, drying floors or other suitable surfaces, and stored in the house for home consumption, seed or sale. Future Cropping Patterns 7.02 In the absence of the project, there would be no significant change in cropping patterns. The majority of the potential subproject areas would continue to have one wet season rainfed rice crop or an inadequately irrigated crop and only a limited area would have a second, rather risky, dry season crop depending on water availability, rains and timely planting. However, with the project, farmers would grow only irrigated rice in the proposed CISs. Cropping intensities for the individual systems would depend on the available water supplies. However, no subproject with a realistic potential for future cropping intensity of less than 130% would be accepted in the project, and the average cropping intensity of the systems is expected to be 165% at full development, four years after completion of each subproject. Yields would vary depending on the degree of water control, the availability of extension services, credit and input supplies. Yields and Production 7.03 Without the project, yields in the rainfed areas are estimated to increase at 1.25% per year, and at 1.4% per year in the presently irrigated areas needing rehabilitation. This means a yield increase by 1990 of about 0.4 tons/ha over the present level of 1.7 tons/ha for rainfed areas, and of about 0.5 tons/ha over the present level of 2.2 tons/ha for irrigated areas in need of rehabilitation. These yield increases would be induced by improved infrastructure, better availability of inputs and more effective extension services, but offset somewhat by the continued deterioration of these systems. 7.04 Yields with the project are estimated at 3.3 tons/ha four years after implementation of each subproject. Increases would be due mainly to the availability of a reliable supply of water in the project areas, together with better access to institutional credit and physical inputs, and improved rice extension service support through adequate hiring and training of rice production technicians from NIA and MOA. Present and projected palay (threshed unmilled rice) yields are shown in Table 7.1. With the project, palay production would increase from the present 61,000 tons to about 186,000 tons at full development in 1990. Phasing of production from start of irrigation facilities under the project to full development is given in Table 7.2.  - 43 - Table 7.1: PRESENT AND FUTURE PALAY PRODUCTION, 1990 /b /c /d Present Present Future without project Future with project Phase status /a Hectares Yield/ha Production Yield/ha Production Yield/ha Production