Philippines - Irrigation program review

Report No. 3545-PH FILLE COPY The Philippines: Irrigation Program Review December 15, 1982 Country Projects Department East Asia and Pacific Regional Office FOR OFFICIAL USE ONLY U Document of the World Bank 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   FOR OFFICIAL USE ONLY PHILIPPINES IRRIGATION PROGRAM REVIEW Preface i. This report originated in a mission /1 to the Philippines in 1980, and Chapters 1 through 4 are essentially as that mission's yellow cover draft, issued in June 1981. That version drew attention to the likelihood of dramatic rice surpluses and led to a more detailed examination of the poten- tial rice balance. The results were incorporated in the green cover draft, issued in January 1982. Based on the data available, very large rice surpluses were projected through the 1990s, leading to questions on the advisability of the Government continuing with a major program of investment in new irrigation capacity. ii. The draft report was discussed with the Government in February and May 1982, and, as a result, discrepancies between data on irrigated areas and cropping intensity produced by the National Irrigation Administration and the Ministry of Agriculture became apparent. The data proved to be reconcilable, however, and in this light a new set of projections were made. These are included in Chapters 5 and Annex 2 of this report. They suggest that while the Philippine rice supply is still likely to be considerably in excess of domestic demand through the 1990s, provided real prices to farmers are maintained, this will be by a smaller amount than earlier projected. The broad conclusion remains - that unless the Philippines can establish itself in the rice export market on a regular basis (a possibility which should be studied urgently), generate greater domestic demand, or manage a large-scale conversion of land use from rice to more profitable crops, the current plans for irrigation development should be scaled down - but with the proviso that the need for further new investment in the late 1990s should be carefully monitored. /1 The mission, which visited the Philippines in December 1980, comprised Mr. M. Cox (Bank) and Sir Kenelm Guiness and Mr. A. Yambao (consultants). Messrs. W.T. Smith and E.G. Giglioli assisted with the report, and Messrs. J. H. Cleave and J. Macgregor undertook later discussions and drafting. 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 IRRIGATION PROGRAM REVIEW Table of Contents Page No. SUMMARY AND CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . i - iv 1. BACKGROUND .... . . . . . . . . . . . . . . . . . . . . . . 1 Regions and River Basins . . . . . . . . . . . . . . . . . . 1 Scale of Irrigation Expenditures by the Philippine Government 2 External Assistance for Irrigation Development . . . . . . . 3 Purpose and Objectives of the Irrigation Program Review . . . 4 2. IRRIGATED AGRICULTURE IN THE PHILIPPINES. . . . . . . . . . . . 5 Background. 5 Farm Size, Land Tenure and Land Reform. 5 Topography, Climate and Soils. 8 Crop Production. 8 Prospects for Irrigation of Crops other than Rice . . . . . . 12 Agricultural Supporting Services . . . . . . . . . . . . . . 13 3. INSTITUTIONAL ORGANIZATION OF THE IRRIGATION SECTOR . . . . . . 17 National Irrigation Administration . . . . . . . . . . . . . 17 Other Institutions Involved in Irrigation . . . . . . . . . . 23 4. IRRIGATION DEVELOPMENT: PRESENT SITUATION AND POTENTIALS . . . . . . . . . . . . . . . . . . . . . . . . . 25 Background . . . . . . . . . . . . . . . . . . . . . . . . . 25 Present Situation and Potential by Regions . . . . . . . . . 28 5. A PROGRAM FOR THE FUTURE . . . . . . . . . . . . . . . . . . . 33 Government Objectives .33 Rice Production and Consumption Projections . . . . . . . . . 33 Supporting Policies .39 Alternatives to New Investments in Surface Irrigation . . . . 41 Implications for Future Irrigation Development. . . . . . . . 44  -2- Page No. ANNEXES 1. External Financing for Irrigation Development ... . . . . . . . 48 A. World Bank ......................... . 48 B. Asian Development Bank ... . . . . . . . . . . . . . . . . 53 C. Other Financing Agencies .................. . 56 2. Rice Demand and Supply Projections . . . . . . . . . . . . . . . 57 Table 1 - Supply and Disappearance, Crop Years 1970-82 .60 Table 2 - Harvested Areas, Yield and Production of Paddy for Selected Years .61 Table 3 - Rice: Supply and Demand Projections. . . . . . . . . . . . 62 Table 4 - Sensitivity Analysis of Rice Supply-Demand Balance. 63 Table 5 - Rice Yield Growth Assumptions .64 Table 6 - Irrigated Rice Cropped Area Growth .65 Table 7 - Rice Demand Projection Assumptions. . . . . . . . . . . . . 66 FIGURES 1 Past and Projected Palay Yields . . . . . . . . . . . . . . . . 67 MAP - Location of Projects (IBRD 15720 R) Tables in Text 2.1 - Distribution of Rice Farm Holdings in the Philippines . . . . . 6 2.2 - Targets of Agrarian Reform Program. 7 2.3 - Climatic Characteristics by Region. 9 2.4 - Area, Yield and Production of Three Major Crops .10 3.1 - Collection of Water Charges on National Irrigation Systems . . 20 4.1 - Areas Served by Irrigation Systems, 1980 . . . . . . . . . . . 27 5.1 - Summary of Rice Supply and Demand Projections . . . . . . . . . 36 5.2 - Sensitivity Analysis of Rice Supply-Demand Projections . . . . 38  PHILIPPINES IRRIGATION PROGRAM REVIEW SUMMARY AND CONCLUSIONS 1. In 1981 about 950,000 ha of crops in a service area/l of 1,230,000 ha benefited from irrigation in the Philippines. About 480,000 ha of service area are served by national systems under the control of the National Irrigation Administration (NIA), while some 600,000 ha are communal systems owned and operated by the communities they serve. About 150,000 ha are served by miscellaneous other schemes, of which many are pump irrigation projects originally built by the Bureau of Public Works but now under the general control of NIA. The service area has grown to its present level from about 80,000 ha in 1950. 2. Virtually all of the irrigated land in the Philippines is devoted to rice cultivation; this is due to a combination of climate, topography, soils and prices. Annual rainfall varies considerably from place to place in the thousands of islands forming the Philippines but is seldom less than 1,800 mm and in some places exceeds 4,000 mm. Thus, most crops other than rice can be grown without irrigation. Topography and soils favor rice irrigation because the rivers flow through the broad flat flood plains where rainfed rice has been grown for centuries and the soils are generally ill-suited for other crops. Irrigated cropping accounted for about 44% of the 3.6 million ha of rice harvested in 1979/80. 3. The creation of NIA in 1964 presaged a major government program, which began in 1968 and is still in progress, to improve the standard of existing projects and to extend irrigation to large areas of rainfed lowlands. Government support for NIA-s program was stimulated by the Philippines failure to meet its own rice requirements. Imports of milled rice reached a peak of over 600,000 tons in 1972 following a series of poor harvests caused by typhoons and droughts. In parallel with the irrigation program, the Government also launched a major credit program, Masagana 99, to help rice farmers finance the inputs needed to exploit the new high yielding varieties (HYVs) which had become available. As a result of NIA's irrigation program and the technology ushered in by Masagana 99, the Philippines has had a surplus of rice in five of the six years since 1976. Gains in production will continue through the 1980s and 1990s as benefits are realized from the /1 Service area includes service roads canals, farmhouses and land not receiving irrigation water but within the intended irrigated area of a project, as well as actual irrigated cropped area.  - ii - strengthening of agricultural support services and from the many large irrigation projects now under construction with assistance from the Bank and the Asian Development Bank. 4. Only about 40% of the technically irrigable area of the Philippines has been developed. The potential exists for developing a further 1.0-1.2 million ha of service area mainly in Central Luzon, already the most developed area and the "rice bowl" of the Philippines, and Mindanao, where a major impediment to smooth development is the security problem present in much of the island. NIA has prepared a corporate plan which in its latest (September 1982) version envisages the development of some 800,000 ha of new irrigation and rehabilitation of a further 400,000 ha over the period 1982-2000. These figures include some 350,000 ha on which design or implementation of schemes has already started. The planned development, although less than earlier envisaged, is on a scale which will produce rice considerably in excess of domestic demand. On the basis of an anticipated continuing decline in population growth and a moderate growth in average per capita income, the domesic effective demand for Philippine rice is pro- jected to increase by about 70% over 1980 levels by the year 2000. On the supply side, assuming only completion of schemes on which a start had already been made by 1982 but no starts on new irrigation other than three multipurpose projects, and allowing for moderate yield increases on rainfed and irrigated rice and a modest growth in cropping intensity in irrigated areas, milled rice production is projected to increase by about 65% in the same period. On this basis, the rice surplus over the domestic market demand would rise from nearly 600,000 tons a year in 1985 to nearly 800,000 tons by 1990 as projects currently under implementation come on stream, and would still be somewhat more than 300,000 tons a year by the year 2000. These projections presuppose a positive pricing policy for rice producers, and assume adequate maintenance and rehabilitation of irrigated areas, the perpetuation of research and extension services, and the availability of credit. Clearly, on the basis of this analysis, NIA's program of new development -- certainly until the late 1990s -- should be considerably delayed. 5. However, the demand for rice would be considerably increased if the Philippines were able successfully to enter the export market on a regular basis (rather than merely exporting whatever is surplus to domestic requirements in any year) or if the Government were to mount a substantial subsidized rice distribution program aimed at the under-nourished segments of the population. The surplus of rice would also be reduced if it were possible to find alternative crops which could be produced on irrigated land in significant quantity on an economic basis. 6. Production for export would require production of quality rice acceptable on world markets: present Philippine production is of poor quality with mixed varieties and a high percentage of brokens. The switch would necessitate investments to improve production and postharvest handling  - iii - and probably would require designation of export areas from which an even flow of uniform varieties could be directed through modern facilities to export centers. Two studies are needed before a decision could be made to move into production for export, the first to assess the market for export rice and to examine the structure of the distribution system to ensure that the Philippines could enter without impediment, and the second to identify a least-cost export area and assess the investment and support services required. 7. One of the thrusts of a Food and Nutrition Plan published by the Ministry of Agriculture early in 1981 is direct intervention to reduce under- nutrition by subsidizing incremental consumption of rice and vegetable oil by targeted beneficiaries. Various scales of intervention have been postulated, calling for from 90,000 tons to over 450,000 tons of rice a year. Clearly an intervention on the scale of the latter scenario would absorb much of the projected rice surplus, but the FNP proposals have not been accepted by the Government, partly at least because of the subsidy and administrative cost, and a greatly scaled-down program is probably the most which can be anticipated. 8. The scope for diversifying into crops other than rice needs close examination and quantification but appears limited by physical and market factors. Because of their scale, obvious crops for consideration as alternatives are corn and sugar. Two major factors limit the scale of diversification into corn. First, most soils suitable for lowland rice are too heavy or flood-prone for corn to be grown more profitably than rice. Second, market factors are likely to limit the expansion of corn area: the Ministry of Agriculture (MOA) estimates that the introduction of high-yielding varieties of seed combined with fertilizer and biocides will raise average corn yields by about 50% by 1987, while area rises by one-eighth. If the technical package is indeed sound and more profitable than traditional methods, then the increase in yields alone would produce large corn surpluses, depress corn prices and limit planted area. Although some sugar areas could be irrigated economically, these are largely in the hands of estates and large farmers which have low priority in NIA's plans. Irrigation of specialized crops is also possible, but the potential scale in these and in sugar would probably be insignificant compared to the area under rice. 9. Even if an export policy and adoption of a food-subsidy program were so to increase the projected demand for rice, and competing crops so reduce the availability of irrigated land for rice, that additional investment were needed in the near term, alternatives other than expanded surface irrigation should be reviewed. Such alternatives include greater efforts in the support areas underlying the projections - more investment in family planning to reduce demand; stepped up research, extension and credit ser- vices, and positive pricing policies to increase intensification of produc- tion; and investment to improve milling out-turn rates and reduce postharvest  - iv - losses. In the area of irrigation, even greater emphasis on rehabilitation is an economic alternative to extending the irrigated area; groundwater development, for which there is a reported potential to irrigate 1.5 milLion ha of dry season cropping, may be a lower-cost alternative to surface irrigation in many areas; and there may be some opportunities for grafting irrigation systems onto existing hydro-power schemes. 10. The Government should be encouraged to focus on the crucial policy questions concerning its export strategy and its position on nutrition support, and should be prepared to undertake policy studies necessary to facilitate early decisions. NIA should be encouraged to keep its medium-term investment program under review, relating this directly to the likely demand for irrigated production, and addressing the feasibility of alternatives to new surface irrigation. NIA should carefully examine the implications for its structure, staffing and operations of a new emphasis on rehabilitation and maintenance instead of new investment. Subject always to making an appropriate response to the Government's policy decisions, the thrust of future Bank lending to irrigation should be to support and strengthen NIA in this new direction, while emphasizing diversification through rainfed agriculture in other areas of the lending program.  1. BACKGROUND 1.01 The Philippine's territory extends over 300,000 sq km and includes some 7,100 islands. The population is estimated at 49 million (mid-1980), with a growth rate declining from 3% in 1970 to roughly 2.6% in 1980. The Philippine economy grew at an average annual rate of 6.3% between 1971 and 1981 when per capita GNP reached US$715. Details on the economic situation are contained in the Bank's latest economic report./l Regions and River Basins 1.02 There are considerable differences in resources, socioeconomic conditions and production potential between the country's 12 administrative regions. Historically, economic development has tended to concentrate around Central Luzon, the "rice bowl" of the country, and near Manila, where most of the infrastructure development has taken place. The less developed areas are Northern Mindanao in Region X, the Visayas in Regions VI, VII and VIII and Bicol in Region V. The major river basins are briefly described below. Climate, soils, and topography, in relation to irrigated agriculture, are discussed in Chapter 2. 1.03 Five major river basins, three in Luzon and two in Mindanao, contain the largest irrigable areas in the Philippines. They are the Cagayan, Agno and Pampanga basins in Luzon, and the Cotabato and Agusan Basins in Mindanao. 1.04 The Cagayan River Basin in Northern Luzon covers some 28,000 sq km and is bounded on the north by the Babuyan Channel, on the south by the Caraballo mountains, on the east by the Sierra Madre mountains and on the west by the Cordillera Central range. The Cagayan River traverses the entire length of the valley, from its source near the Nueva Vizcaya-Quezon bound- ary to its mouth on the Babuyan Channel. Its major tributaries are the Magat River in the southwestern part of the basin where a major multipurpose dam is being constructed to provide year-round irrigation to some 100,000 ha, and the Chico River in the northwestern part where several irrigation systems covering about 20,000 ha are being upgraded. 1.05 The Agno and the Pampanga are the principal rivers of Central Luzon. The Agno River rises in the mountainous area at the north of the Central Luzon Plain and has a length of about 270 km from its source to its mouth in the Lingayen Gulf. Principal tributaries include the Pilar, Ambayoan, and the O'Donnell, Camiling, and Tarlac Rivers where several large irrigation projects are being upgraded. The Pampanga River drains the mountain ranges to the south and east of the Plain and is formed by the Carranglan and Pantabangan /1 The Philippines: Selected Issues for the 1983-87 Plan Period, June 1, 1982 (World Bank Report No. 3861-PH).  - 2 - Rivers which join at the northern end of Nueva Ecija Province. From there it flows south to its mouth in Manila Bay. Eastern tributaries of the Pampanga River include the Angat, Bulu, Coronel, Penaranda and the Digmala Rivers. On the west there are the Bamban, Rio Chico, Talavera and Carranglan Rivers. Two major multipurpose dams have been constructed, one on the Pampanga River which irrigates 78,000 ha and another on the Angat River, which irrigates 30,000 ha. 1.06 The Agusan River Basin, of about 11,500 sq km, is located in the eastern portion of the island of Mindanao. It includes a major part of Agusan Province, a substantial part of Davao Province, and small areas of Bukidnon, Misamis Oriental, and Surigao del Sur Provinces. The Agusan River originates in Davao Province, traverses the entire length of the basin in a northerly direction and discharges into Butuan Bay. The river channel is wide and navigable by motor boats and launches up to about 100 km from the mouth. About 60,000 ha in the Agusan Basin are being developed through projects financed by the Asian Development Bank (ADB). 1.07 The Cotabato Basin, covering an area of about 20,000 sq km, is located in the central part of Mindanao. Over 70% of the basin area is located in Cotabato Province and the rest in Bukidnon, Agusan del Sur, Davao del Sur and Lanao Provinces. The Mindanao River, the principal river in the basin, is formed by the Pulangi and Allah Rivers, and from the confluence of these two rivers it flows in a northwesterly direction into the Illona Bay near Cotabato City. The Liguasan and Libungan marshes cover extensive areas in the lower portion of the basin. They are fed by overflow from the rivers during periods of heavy rainfall and runoff. Also located in the basin's southern portion are two major lakes, the Buluan and Sebu. The Bank-financed NISIP II Project includes upgrading of about 24,000 ha of existing systems in the basin, and ADB has some 40,000 ha of irrigation projects in the basin. 1.08 Other smaller river basins include the Bicol River Basin located in the southern portion of Luzon and covering an area of about 3,100 sq km, the Ilog-Ililabangan River Basin covering about 2,100 sq km in the southwest of the island of Negros, and the Panay-Jalaur River Basin, located in the island of Panay and covering about 3,700 sq km. Scale of Irrigation Expenditures by the Philippine Government 1.09 Government irrigation expenditures grew rapidly during the 1970s. Using the narrow measure of infrastructure expenditures in the national budgets, and excluding from the category of irrigation multipurpose projects providing power, flood control and drainage, irrigation-s share of infra- structure was about 8% in 1970, rose to 11.6% in 1978, and then declined to 9.2% in 1980. For the 1981-87 period, the Government plans to allocate about 11% of its infrastructure budget to irrigation. The capital expenditure  - 3 - budget of the National Irrigation Administration (NIA) grew (in P mil- lions) from 972 in 1978 to 1,748 in 1981, a nominal growth rate of 22% p.a. NIA has recently prepared a Corporate Plan, scheduled to be updated annu- ally, which projects a 1981-90 national and communal irrigation development program costing some P 15.5 billion (1981 terms) in total, with annual costs declining from P 2.9 billion in 1982 to P 1.0 billion in 1990. This program would be heavily dependent on foreign assistance, and is currently under review. Some reduction in the one million ha national and communal systems construction and rehabilitation program is anticipated (para. 5.25). External Assistance for Irrigation Development 1.10 Since 1969 the Bank has financed twelve irrigation projects, one watershed project, and three rural development projects with irrigation components. These are all being implemented by the National Irrigation Administration. Loans and credits in the period 1969-1982 amount to US$686 million to irrigate some 464,000 ha. Details are given in Annex 1. 1.11 In the Cagayan River Basin, Bank-financed projects include the three stages of the Magat River Irrigation Project (1975 and 1978) and the Chico River Irrigation Project (1976). The Magat Project consists of a multipurpose dam on the Magat River with a storage capacity of 830 MCM, and canals and drainage systems to rehabilitate and upgrade some 52,000 ha of irrigated lands and provide new irrigation to some 50,000 ha. The Chico River Irrigation Project, located just north of Magat, involves construction of a diversion dam on the Chico River and irrigation and drainage systems to rehabilitate and upgrade some 7,800 ha of existing irrigation systems and to extend irrigation to 11,900 ha of currently rainfed areas. The ADB is also involved in the Cagayan Basin through the Angat-Magat Integrated Agricultural Development Project. 1.12 In Central Luzon, Bank-financed projects include: the completed Upper Pampanga River Project (UPRP) (1969) comprising a dam and reservoir at Pantabangan on the Pampanga River, rehabilitation and upgrading of 46,000 ha, and construction of new irrigation facilities for 32,400 ha; the Aurora Penaranda Irrigation Project (1974) which utilizes the storage reservoir and main canal systems of the UPRP; and the Tarlac Irrigation Systems Improvement Project (1974). 1.13 The Bank is also financing two National Irrigation System Improvement Projects (NISIP I and NISIP II) which will together lead to the upgrading of irrigation facilities on about 87,000 ha, and the extension of irrigation to about 38,000 ha of rainfed lands in areas of smaller national systems handled by NIA in Northern Luzon, the Visayas and Mindanao. The Communal Irrigation Development Project (1982) will finance the construction of 23,000 ha and the rehabilitation of 11,000 ha while strengthening the Provincial Irrigation Offices which do the operation and maintenance of national irrigation systems and the construction of communal systems. Other Bank-financed irrigation projects are located in the Visayas  4 (Jalaur, 1977) and in the Islands of Mindoro and Palawan (Medium Scale Irrigation, 1980). There are irrigation components Ln the Mindoro Rural Development Project (1975) involving rehabilitation of 13,000 ha of existing irrigation systems and extension of irrigation to 2,000 ha of rainfed lands; in the Rural Infrastructure Project (1978), involving construction of new irrigation systems serving 9,400 ha in Ilocos, Cagayan Valley and the Visayas; and in the Samar Island Rural Development Project (1980) which includes 2,000 ha of new irrigation systems. 1.14 ADB-s main activities have been in the island of Mindanao, where it has financed nine projects to construct new irrigation facilities on 95,000 ha of formerly rainfed lands and to upgrade and rehabilitate about 21,000 ha of existing irrigation systems. Outside of Mindanao, ADB projects include the recently completed Angat-Magat project in the Pampanga and Cagayan River Basins covering about 70,000 ha, the two stages of the Laguna de Bay Development Project (1975 and 1980) which will upgrade 24,000 ha of existing systems and bring irrigation to 5,300 ha of rainfed areas, and the Bicol River Basin Irrigation Development Project (1979) which will upgrade 5,200 ha of existing systems and irrigate 17,000 ha of rainfed areas in Southern Luzon. Other agencies involved in irrigation development in the Philippines include the United States Agency for International Development (USAID), Japan's Overseas Economic Cooperation Fund (OECF) and the Australian Development Assistance Bureau (ADAB). USAID financed the studies of the major river basins in the country, the feasibility study for the Magat multipurpose project, and is currently financing irrigation development in Bicol. OECF is currently financing an integrated agricultural development project in the Cagayan Valley and an irrigation project in Ilocos, as well as equipment for the rehabilitation of the smaller national systems. ADAB is financing irrigation development in Mindanao. Purpose and Objectives of the Irrigation Program Review 1.15 In line with the Government's drive towards self-sufficiency in rice in the early 1970s, NIA embarked on a major irrigation program to increase the irrigated area by 30% and upgrade 80% of the existing national systems. Through increases in the irrigated area and the wider use of new rice varieties and fertilizer stemming from the Masagana 99 credit program, the Philippines is now self-sufficient in rice. The full benefits of the large investment in irrigation during the 1970s are yet to be realized because many major projects are still under construction, but as these come on stream rice production is likely to stay comfortably ahead of domestic demand through the 1980s. Given the long lead time to bring projects into production, decisions will have to be made soon on the scale and timing of irrigation investments required to meet the needs of the 1990s. With this in mind, a Bank mission visited the Philippines in December 1980 to review with NIA an appropriate approach to future development in the irrigation subsector. The findings of the mission were reviewed and modified in the light of new information and later discussions with NIA and Ministry of Agriculture officials. The revised findings are presented in this report.  1.16 The report begins with an overview of irrigated agriculture in the Philippines with reference to climate, soils, topography, land tenure, crop production and supporting services, and looks at the prospects for irrigation of crops other than rice. This is followed by a discussion of the present status and future potential for irrigation in the major geographic units of the country, and then by a review of NIA and other agencies involved in irrigation. The report concludes with a presentation of rice supply and demand projections through the 1990s and outlines the implications for govern- ment policies and irrigation investment. 2. IRRIGATED AGRICULTURE IN THE PHILIPPINES Background 2.01 During the 1970s, agriculture (including forestry) grew steadily at an annual rate close to 5%, while agricultural terms of trade improved by 21% up to 1975 and declined by 23% during the next five years. It is still the predominant sector in the Philippine economy, accounting for about 27% of GDP, over one half of total employment and slightly less than 60% of export earnings. Nearly three quarters of the cultivated land is devoted to grain production, notably rice and corn; the other major crops are sugar and coconuts. Together these four crops account for more than 80% of the area planted. Foodgrain production is mainly oriented to domestic consumption, while coconuts (copra) and sugar are the main export crops. 2.02 During the 1970s the growth in agricultural production came from both expansion of cultivated area and improvement in yields. The performance of different crops has varied considerably with rice, sugar, pineapple and banana yields growing significantly, corn and tobacco yields growing slowly, and coconut and abaca yields declining. The area under rice has remained relatively stable at around 3.5 million ha, area under sugarcane has fluctuated severely in response to unstable prices, and areas under corn, coconut, root crops and other minor crops, such as banana, abaca and pineapples, have increased considerably. Since there is little land left for further expansion of cultivation, future production growth will have to come mainly from higher yields and more intensive use of presently cultivated lands. Farm Size, Land Tenure and Land Reform 2.03 Characteristics of Philippine agriculture are the predominance of small farms and the pervasiveness of tenancy, particularly among rice and corn farmers. The average rice farm is 1.8 ha, ranging from 0.8 ha in the densely populated region of Ilocos to 2.4 ha in Southern Mindanao. According to the 1971 census figures, the farm size distribution for rice farms in the Philippines was as follows:  - 6 - Table 2.1: DISTRIBUTION OF RICE FARM HOLDINGS IN THE PHILIPPINES Size of Farm Percent of Farms Percent of Area (ha) t%) (%) Less than 1.0 14.6 2.9 1-3 54.1 33.6 3-5 21.5 28.3 5-10 7.3 16.6 10-25 2.2 10.8 25.0 and above 0.3 7.8 Total 100.0 100.0 Although there is no up-to-date information for the Philippines as a whole, the size of farms has probably decreased as a result of populationi pressures. This is confirmed by information on land tenure recently collected in the preparation and appraisal of Bank-financed projects. Almost 30% of all farmers are tenants with an additional 11% renting extra land. About 60% of these tenants were found to be in absolute poverty in 1971, with the highest incidence occuring in Northern Mindanao, Bicol and the Visayas (Central and Eastern). While more recent figures are not available at the regional level, it is likely that this proportion has decreased in the latter half of the 1970s as a result of the Government's Land Reform program and through gains in productivity. Since 1972 the Government has pursued a program of agrarian reform aimed at the transfer of land ownership to tenant farmers on rice and corn lands. All tenanted rice and corn farms were declared land reform areas under Presidential Decree No. 27 of October 1972. The reform is being implemented in two ways: "Operation Land Transfer" applicable to all landowners above 7 ha, whereby the tenants become owner-cultivators 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. Table 2.2 shows the intended targets of the reform program.  Table 2.2: TARGETS OF AGRARIAN REFORM PROGRAM Size of tenanted Total Average holding estates area Tenants Landlords per tenant (ha) ('000 ha) ( 000) ( 000) (ha) Operation Land Transfer 100 and over 205.0 95.0 2.0 2.2 50 - 99.9 83.0 5.0 2.0 1.8 24 - 49.9 95.0 57.0 4.0 7 - 23.9 348.0 200.0 41.0 1.7 Subtotal 731.0 397.0 49.0 1.8 Leasehold 7 and less 730.0 609.0 435.0 1.2 Total 1,461.0 1,006.0 484.0 1.45 Source: Ministry of Agrarian Reform. 2.04 The Ministry of Agrarian Reform (MAR) is responsible for the over- all administration and implementation of the program, while the Land Bank of the Philippines (LBP) is concerned with the financing of land acquisition under Operation Land Transfer. After a slow start due to poor land title documentation, lengthy negotiations with landlords over land valuation, and the need to develop the institutional framework and resources within MAR, program implementation has recently accelerated. As a result of improved procedures on the part of MAR, by the beginning of 1982 certificates of land transfer (CLTs) had been issued to over 105% of the number of tenants targeted by the program covering over 90% of the area to be reformed. Leasehold contracts had been issued to over 90% of the potential beneficiaries covering over 85% of the land subjected to the program. The entire program accelerated since whole areas were subjected to the agrarian reform program instead of proceeding on a case by case basis as was done before. In irrigated lands the progress has been uneven. The process has been completed in Upper Pampanga, Magat, and Tarlac and is close to completion in Jalaur. However, in Chico 20% of the farmers are still share tenants, and in the irrigation systems being upgraded under the National Irrigation Systems Improvement Project about 38% of the farmers are share tenants. Land reform has moved fastest in Central Luzon and areas near Manila and slowest in Mindanao and other regions with security problems.  Topography, Climate and Soils 2.05 The Philippines has complex physiographic and climatic patterns. The 45 largest islands account for 98% of the land area, leaving 2% for roughly 7,000 islets. The entire country sits on a series of stress lines along which numerous faulting, folding, and volcanic activities have taken place. These extensive earth movements have resulted in numerous mountain ranges, which divide the islands further into small watersheds with short and usually rapidly flowing rivers and isolated alluvial plains bounded by mountains, hills or the sea. 2.06 The islands lie under the influence of the monsoon and in the path of tropical typhoons. The generally north-south alignment of mountain ranges protects part of all islands from either the north-east or south-west monsoon, and this leads to wide variations between regions in annual rainfall and its seasonal distribution. Annual rainfall ranges from 1,500 mm to over 4,000 mm and four main types of climate can be distinguished. Type A and Type B cli- mates are similar in that they both have well-defined dry seasons, but Type A has a seven-month dry season beginning in October whereas Type B has a five- month dry season starting in December. Annual rainfall for Types A and B ranges between 1,500 and 2,000 mm. The Type C climate has a clearly defined wet season with heavy rain throughout the period October through March, and there is effectively no dry season since between April and September, monthly rainfall is generally on the order of 200 mm. Annual rainfall for Type C varies between 2,500 and 4,000 mm. In the Type D climate, the wettest period is April to December, but rainfall is not as intense as in Type C; the dry season is short (January-March) and months with less than 50 mn of rain can occur. Annual rainfall for Type D climates is between 2,000 and 3,500 mm. Typhoon frequencies increase generally from south to north with the largest incidence in Northern luzon (averaging about seven a year in the 1948-77 period). Only southern Mindanao is relatively free from typhoons. A common element is temperature, which provides the whole country with a year-round growing season for both crops and forests. The climatic characteristics of each region are given in Table 2.3 below. 2.07 Frequent structural movements and volcanic activities have given most of the Philippines a young topography. Natural weathering is intense and rapid under high temperature, heavy rainfall and limited forest coverage. Most of the irrigable soils are in plains and valleys with deep alluvial deposits and, where volcanoes are present, 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, 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. Crop Production 2.08 Four crops, rice, corn, coconuts and sugarcane, take up almost 90% of the area cultivated in the Philippines. Trends in the cropped areas, yields and production of the three main field crops - rice, sugarcane and corn - which have to be considered in irrigation planning are shown in Table 2.4 and discussed below.  Table 2.3: CLIMATIC CHARACTERISTICS BY REGION Total rainfall Average Wet Dry Monthly rainfall No. of Type of Annual season season Wet Season Dry Season typhoons Region Climate Months Max. Min. Months Max. Min. per year Ilocos } 7.2 Central Luzon I A 2,,000 1,700 300 Jun-Sep >500 275 Oct-May 100 5 3.2 Southwest Tagalog } (25)/a (17) (15) (2) } West Cagayan B 2,500 1,900 300 Jun-Nov 300 175 Dec-May 150 20 } (28) (20) (20) (5) } } 7.2 Southeast Tagalog } D 3,500 2,800 700 May-Dec 450 250 Jan-Apr 200 150 } East Cagayan } (20) (18) (18) (15) } West Visayas A 1,800 1,400 400 Jun-Nov 350 200 Dec-May 125 20 3.8 (20) (15) (15) (5) Central Visayas B 2,000 1,600 400 Jun-Dec 210 135 Jan-May 115 55 1.4 (18) (15) (13) (11) Bicol and C 4,000 2,600 1,400 Oct-Mar >500 250 Apr-Sep 250 150 3.8 East Visayas (27) (14) (20) (15) West and B 1,500 1,000 500 Jun-Nov 300 175 Dec-May 125 40 1.4 North Mindanao (19) (12) (10) (6) Southwest Mindanao D 2,000 1,650 350 Apr-Dec 250 175 Jan-Mar 125 50 0.2 (19) (15) (10) (6) East Mindanao C 4,200 2,800 1,400 Oct-Mar >500 250 Apr-Sep 300 170 0.2 (26) (18) (21) (15) /a Figures in parentheses are number of rainy days in a month.  - 10 - Table 2.4: AREA, YIELD AND PRODUCTION OF iHREE tAJUR CROPS 1970 1975 1981 Area Yield Prod. Area Yield Prod. Area Yield Prod. ('000 (ton/ ('000 ('000 (ton/ ('000 ('000 (ton/ ('000 ha) ha) tons) ha) ha) tons) ha) ha) tons) Rice /a 3,115 1.68 5,233 3,540 1.60 5,660 3,330 2.27 7,572 Corn /b 2,390 0.84 2,008 3,060 0.84 2,570 3,560 0.98 3,489 Sugar /c 425 5.03 2,127 480 5.14 2,474 414 5.79 2,395 /a Paddy /b Shelled /c Processed sugar. Source: Bureau of Agricultural Economics and Policy Analysis Staff of Ministry of Agriculture. 2.09 Rice is the main irrigated crop in the Philippines. Of the nearly 3.5 million ha of rice harvested in crop year 1981, over 1.6 million ha or 47% were from irrigated land. This proportion has been rising steadily since the late 1950s when it accounted for less than 30% of the area. By 1985, when most ongoing irrigation projects are completed, almost 55% of the harvested area will be irrigated. The total physical area under rice has probably reached a peak but the harvested area will increase as higher cropping intensities become possible with the expansion of irrigation. 2.10 Total production reached 8.0 million tons of paddy or 4.6 million tons of milled rice in crop year 1982. Production grew at 3.6% p.a. during the 1970s with irrigated production growing at 4.3% p.a. Most of this growth was due to yield increases (2.3% p.a. overall) brought about by the widespread use of HYVs which in 1979 covered about 72% of the area and accounted for about 82% of total paddy production. In irrigated areas, HYVs covered 90% of the area. Yields in irrigated areas have been growing at 3.3% p.a., slightly faster than in the lowland rainfed areas which have been growing at a rate of 2.9% p.a. Yields in upland areas have also been improving but at the slower rate of 2.1% p.a. Despite this progress, yields still remain relatively low and the national average paddy yield passed the 2 ton/ha figure only in 1978. Lowland rainfed yields reached 1.9 ton/ha in 1982. The yield potential for irrigated paddy under good water control and management is well over 6 ton/ha per crop. Average yields in irrigated  - 11 - areas reached over 2.8 ton/ha in 1982 and in the national systems they are reported to have averaged over 3.7 ton/ha in 1981. 2.11 The quality of Filipino rice has usually been low, and the most common grade marketed in the Philippines contains 30% to 35% brokens. So long as the rice was consumed only domestically, the quality situation did not cause major problems. However, now that the country has reached self- sufficiency and an increasing exportable surplus is being produced, quality becomes a major issue. The existing milling industry in the Philippines is capable of providing the higher qualities for export. However, the absence of quality standardization and quality incentives in rice pricing is an impediment to the production for export of higher quality rice. An addi- tional difficulty is the practice of mixing several rice varieties of local origin at the milling stage, which prevents the achievement of uniform rice qualities. 2.12 Corn surpassed rice in terms of harvested area for the first time in 1981, although the gap may be closed again in 1982. It is a rainfed crop grown by small farmers in gently rolling and upland areas. Production is estimated to be 3.7 million tons in 1982, having increased steadily at a rate close to 6% p.a. during the 1970s. Most of this growth is, however, a result of expansion in area; yields have only increased by 16% in the last five years and at an estimated 1.04 ton/ha in 1982 are among the lowest in Asia. Mindanao is the major corn growing area with almost 50% of the area, followed by the Visayas with 25% of the area. Productivity is still low despite Government efforts through the Masagana Maisan and Masagana 77 supervised credit programs. The absence, until recently, of improved varieties with sustained resistance to downy mildew, and slow adoption of existing suitable technological packages, are the main reasons for the low yields. While the Philippines is self-sufficient in white corn, largely used for human consumption, it has been importing over 200,000 tons of yellow corn a year since 1980 for the livestock feed industry. Demand for all corn is expected to rise at about 3.8% p.a., but with rising incomes and declining population growth the livestock feed component will rise faster than the human consumption component of demand. The Government's new Maisagana program, launched in 1981, is intended to expand the use of new varieties in conjunction with fertilizer and biocides by over 500,000 ha, to achieve corn self-sufficiency by 1985. Although the new varieties seem better protected against downy mildew, they still have not been tested on a commercial scale and caution on how far the technological problem has been solved would still be appropriate. Until proven, uptake of improved packages will probably be slow, and it is still open to question that the Government objective of achieving self-sufficiency in corn and even eventually becoming an exporter will be realized soon. Neither is a shift of irrigated areas out of rice and into corn likely to take place until agreed improvements reach a level that makes corn as profitable as rice. 2.13 Sugarcane production in the Philippines is mostly concentrated in the island of Negros in the Visayas and in certain areas of Central Luzon and Mindanao. Harvested area is estimated at 425,000 ha in 1982, the same area as in 1970. Sugar has been one of the major export crops of the Philippines. Because of sharp fluctuations in world market prices, the contribution of  sugar to total export proceeds has varied widely, ranging from a low of 6% in 1978 to a high of almost 30% in 1974. Production, however, has increased slightly from 2.1 million tons in 1970 to about 2.5 million tons in 1982. This has been the result of an increase in yields from 5.03 tons/ha of sugar in 1970 to an estimated 5.95 ton/ha in 1982. While presently there is little irr.i;ated sugarcane, it is the only major crop other than rice that could benefit from large-scale irrigation especially in areas with distinct wet and dry seasons. 2.14 Sugarcane is grown mostly on estates and large farms. The average size of holding is 18 ha, and about 80% of the sugarcane area is owned by farmers with 10 ha or more. Farms below 5 ha, which account for 55% of the total number of sugar farms, have only 9% of the area. This concentration of ownership is especially prevalent in the main sugar-growing areas of the Visayas and Mindanao. In these regions, farms bigger than 10 ha account for 84% of the area and one third of the farms. 2.15 Other Crops. Other major crops in the Philippines include coconuts on about 2.8 million ha, banana (320,000 ha), abaca (230,000 ha), cassava (210,000 ha), coffee (170,000 ha), tobacco (60,000 ha) and other minor crops such as groundnuts, beans, pineapple, and vegetables with less than 100,000 ha each. Dry-season irrigation of vegetables and some upland crops such as groundnuts and mungbeans is being practiced in small areas of the national irrigation systems. Sprinkler irrigation is being used on some of the foreign-owned banana and pineapple plantations in Mindanao. Prospects for Irrigation of Crops other than Rice 2.16 Corn and sugarcane, because of their importance in terms of area, are obvious crops for consideration as alternatives to rice in irrigation developments. A switch from rice to sugarcane in most of the presently irrigated areas is, however, not feasible because the soils are too heavy and the risk of flooding is too great for sugarcane cultivation, but in parts of Central Luzon, especially the Tarlac system, irrigated land covering about 15,000 ha is suitable for either rice or sugarcane. Nine potential irrigation projects covering some 56,000 ha in Western Negros have been identified by NIA, and would probably involve some sugarcane-growing areas. In such areas, sugarcane is grown mainly by the estates and large farmers; the small farmers prefer rice to sugarcane since it is a crop with an assured market and more stable prices. In the traditional sugarcane-growing areas in the Visayas and Mindanao there is potential for irrigation, but NIA has given such developments low priority in order to concentrate on the rice lands where most of the country-s poorest farmers live. 2,17 Corn can be grown as an irrigated crop in the dry season after rice on some irrigated lands. A number of factors, however, explain why corn has not been widely grown as an irrigated crop. First, only about 25% of the existing irrigated lands have soils suited to corn production. Second, many of the areas in Luzon where corn can be grown have sufficient  - 13 - water for irrigation of a dry-season rice crop which has been much more profitable to the farmer. Third, only areas with pronounced dry seasons are suited to dry-season irrigation of corn; in the wetter climates the risk of flooding is too great on the low-lying rice lands and in any case, the rainfall is more than adequate to meet crop requirements. Finally, yields have been such that construction of new schemes to irrigate corn would have been justified only if costs could have been kept to between US$1,100 and US$1,400/ha, whereas for new developments, costs are about US$2,500/ha. Nevertheless, MOA has assumed that the bulk of dry season cropped area to come under new irrigation schemes in the 1980s will not be in rice as in the past, but apparently in corn as a result of the Government-s Maisagana program (para. 2.12). Overall, the harvested area of corn is projected to rise by 450,000 ha (nearly 13% from its 1981 level to 4.01 million ha by 1987. With MOA's projected increase in corn yields of some 57%, these projections imply an annual corn surplus in excess of 1.5 million tons by 1987. In fact this is unlikely to occur unless there are large corn exports because of the probable effect on corn prices and, therefore, area planted. Moreover, at MOA's projected average yield for 1987, even the 1981 area of corn harvest would produce an exportable surplus of over 970,000 tons (22% of domestic consumption). To export such a quantity (or to export the livestock or poultry products which it could produce) would require concerted marketing efforts beginning immediately, and would neither require nor stimulate substantial diversion of irrigated rice area into corn. 2.18 There is potential for irrigation of a wide variety of specialized crops in existing irrigated systems where soils and climate are suitable. As noted in para. 2.15, bananas, pineapple, and vegetables are irrigated in some areas at present. Further gains in the irrigation of such crops will largely be determined by market prospects, but the area of land taken up will be insignificant compared to the area of irrigated rice. 2.19 In summary, the scope for large-scale irrigation of sugarcane or corn on existing lands is limited by price relations, soils, climate, hydrology and technology. Irrigation of new lands may be economically justified for sugarcane in the rather limited areas with suitable soils and topography, but from the social standpoint sugarcane irrigation has fairly low priority. Irrigation of corn is not likely to be economically justified except in special cases where development costs can be kept very low. Some growth in the irrigation of various specialized crops is possible, but the scale would be very small compared to the area under rice. Quantification of the technical and economic scope for crop diversification in the irrigated areas of the Philippines should be given high priority, but at present it would appear not to have significant bearing on the formulation of a long-term strategy for irrigation. Agricultural Supporting Services 2.20 Extension. About 15 government agencies provide varying forms of agricultural extension. The Ministry of Agriculture (MOA), however, is primarily responsible, through its regional offices, for the dissemination of technical information for crop and livestock production, domestic  - 14 - science, nutrition and rural youth development. Where staff of other agen- cies are involved in similar activities, some overlap does occur. In most cases, however, informal agreements exist concerning their respective roles. Extension services for coconut, sugar, and tobacco are provided by special commodity-specific agencies most of which are supervised by MOA. 2.21 Within MOA there are about 8,600 technical staff, 90% of whom have university-level degrees or certificates in agriculture or related subjects. Both the ratio of extension workers to farm families and the level of basin academic training of extension workers in the Philippines compare favorably to other countries in Asia. However, the effectiveness of the extension service has been held back by: (a) lack of training in extension methodo- logy; (b) a shortage of equipment and facilities for field-level training and demonstration; (c) inadequate mobility of extension staff; (d) inadequate links between research and extension; (e) unclear lines of command between the national, regional and field levels; and (f) low salaries and poor allowances for field staff. A major effort has been initiated by the Minis- try, under the Bank-financed National Extension Project (Loan 1626-PH), to reorganize and strengthen the extension service. In particular this includes: (a) the establishment of a more unified extension service and closer coordination between research and extension; (b) provision of vehi- cles, communication, and demonstration and training equipment; and (c) implementation of a staffing and training program to increase the number of trained extension staff, including the provision of additional subject matter specialist positions and livestock extension technicians. In national irri- gation systems, NIA's water management technicians cooperate with extension officers in introducing farmers to improved techniques in irrigated agriculture. 2.22 Research. The standard of agricultural research in the Philippines is relatively high. For the past 40 years, the University of the Philippines in Los Banos (UPLB) has been internationally recognized as a center for training and research in tropical agriculture. Its capacity was further strengthened by the establishment of the International Rice Research Institute (IRRI), adjacent to the university complex. It has also been strengthened through three Bank-financed education projects. A major portion of the Bank-financed Agricultural Support Services Project is also devoted to strengthening agricultural research. The agricultural research done at UPLB, IRRI and the various MOA and private research stations is coordinated and evaluated by the Philippine Council for Agricultural and Resources Research (PCARR), and the packaging of technology for promotion by the extension service is carried out by the National Food and Agriculture Council of the Ministry of Agriculture. The National Nutrition Council, together with ministries and such agencies as the National Media Production Center, the Nutrition Center of the Philippines and the Food and Nutrition Research Institute (FNRI) are engaged in drawing up nutrition technology packages suitable for national programs.  - 1 5 - 2.23 Training. The agricultural education and training system, at the higher level, consists of about 46 colleges giving degrees in general agri- culture, 18 in forestry, 21 in animal science, and 4 in veterinary medicine. Within this system, the University of the Philippines offers the best quality of instruction. More than 60 technical schools graduate about 600 technicians each year in poultry or pig husbandry, farm equipment and forestry. Training in animal disease and protection has been started only very recently. Systematic and coordinated programs for effective training of extension workers and farmers also had not been developed in the Philippines until recently. Beyond this, however, training tends to be seen as a classroom activity and not as an essential part of, or supplement to, experience gained on the job. Particularly at the local level this limits the effectiveness of the existing training effort. 2.24 The Government, assisted by the Bank-financed Fourth Education Project, is in the process of establishing an improved agricultural educa- tion system comprising (a) lead institutions at the national level; (b) one high-level agricultural college in each of three major regions, surrounded by selected colleges and agricultural high schools serving as vocational/tech- nical training centers; (c) experimental agricultural high schools; (d) post-secondary two-year technician training programs; and (e) systems of training extension workers and farm leaders. The latter is to be accom- plished through the establishment of the Philippine Training Center for Rural Development (PTC-RD). 2.25 Training of its personnel is a major NIA activity. In 1979 around 7,000 employees received training of different kinds ranging from short orientation courses to university training. Prominent among these courses is the training program for water management technicians (see para. 3.10) Additionally, every year between 20 and 30 NIA employees undergo overseas training. 2.26 Agricultural Credit. The severe typhoons and floods that struck most of the Philippines in 1972 and 1973 caused a decline in rice production of almost 20%. As a means of promoting domestic self-sufficiency in rice by minimizing the adverse factors, the Government introduced, among other measures, supervised credit programs under the generic name of Masagana 99 that would help bring to the rice farmer a new technological package to increase rice production. The program reached 610,000 farmers cultivating an area of 928,000 ha in the first year of operation and peaked in 1974, reaching 787,000 farmers on 1.1 million ha. Production recovered and in 1976 self-sufficiency in rice was achieved. The use of HYVs now covers about three quarters of the rice area and most of the technological advances introduced in the early years of Masagana 99 have remained despite the decline of program coverage brought about by low repayment ratios. The explosive growth of the program in the presence of a weak organizational  - 16 - base in the local and rural banks, as well as climatic problems, are the major reasons explaining the arrears. While these developments have weakened the Philippines agricultural credit system, the Government is currently implementing measures to strengthen and improve its functioning. In accordance with a general approach towards a less specialized financial system, rural banks are being allowed to diversify their lending portfolio. Loan restructuring is taking place beyond the previous maximum term of a year, especially in the case of crops affected by natural calamities. Bank capitalization is being encouraged through the provision of additional rediscounting facilities tied to capital increases. These measures, coupled with more selective lending based on the repayment capacity of the would-be borrowers and with more rigorous technical and managerial support to be provided under the National Extension Project, are likely to improve the functioning of the agricultural credit system. 2.27 Agricultural Marketing and Storage. Traditionally, the marketing and storage of food crops and livestock were functions of the private trad- ing sector. This is still largely the case, although an increasing share of the marketing of foodgrains, feedgrains and pulses has been assumed by the National Food Authority (NFA), formerly National Grains Authority (NGA). The private trading sector is a complex system of market intermediaries who interact in the buying, transport, storage, processing and financing of agricultural commodities. Storage facilities are owned primarily by millers as independent warehousing activities have become less profitable in recent years due to government control of marketing margins. 2.28 NGA was originally established to implement the Government-s grain policies, both through industry regulations and through market intervention, the basic premise being that NGA would not displace the private sector. NGA was granted monopoly powers in international grain transactions and acquired storage, milling and transport facilities as well as sufficient working capital to operate a sizeable grain purchasing program covering about 15% of the total market. Its successor, NFA, has had its authority extended to deal with all other foodstuffs in addition to grain. 2.29 Farmer Organizations in Irrigated Areas. As part of the agrarian reform program, the Government, through the Ministry of Agriculture, is pro- moting the formation of pre-cooperative farmer organizations, called Samahang Nayon, based on a barrio or hamlet. The organizations would serve as channels for the inflow of technical services, credit and inputs and as assembly points for produce intended for markets. It is expected that these organizations will eventually grow into full-fledged, formally constituted cooperatives. This approach to cooperative organization fits in well with the NIA effort to set up irrigators associations based initially on the 50 ha irrigation service units. Currently about 2,300 such associations have been formed with a membership of about 60,000 farmers. NIA-s aim is to cover all the national systems and gradually hand over the operation and maintenance of the system below the service units to such associations.  - 17 - 3. INSTITUTIONAL ORGANIZATION OF TdE IRRIGATION SECTOR National Irrigation Administration 3.01 Republic Act No. 3601 of 1964 established the National Irrigation Administration (NIA) to investigate, study, improve, construct and administer all national irrigation systems. It was given the power to investigate all available water resources in the country in order to utilize them for irriga- tion and to collect water rates from the beneficiaries of the national irriga- tion systems. The staff and responsibilities of the Irrigation Division of the Bureau of Public Works were taken over by NIA. Presidential Decree No. 552 of 1974 further widened NIA's scope by giving it "broader powers and authority to undertake concommitant projects such as flood control, drainage, land reclamation, hydraulic power development, domestic water supply, road or highway construction, reforestation and projects to maintain ecological balance, in coordination with the agencies concerned." 3.02 The governing body of the NIA is the Board of Directors, composed of a Chairman and five members. The Minister of Public Works is ex officio Chairman of the Board and the Administrator is Vice-Chairman. The Ministers of Agriculture and of Planning and the General Manager of the National Power Corporation (NPC) are members of the Board. Management of the NIA is the responsibility of the Administrator appointed by the President of the Philippines. The Administration has gone through a series of reorganizations, the latest of which began in 1978, to cope with the large increase in personnel resulting from the enhanced emphasis on expansion of irrigation in the seventies. In the ten years between 1970 and 1979 the total number of employees rose from about 8,700 to about 35,500, a 305% increase. Head office staff increased by 284%, regional personnel by 139% and employees working on foreign-assisted projects rose by 1,023%. The various functions of the Administration have been classified under the following four groupings: (a) Development and Implementation; (b) Engineering and Operations; (c) Finance and Management; and (d) Administration and Personnel. An Assistant Administrator is responsible for each group of activities. The first two groupings are technical and cover all phases of the irrigation project cycle, from identification, through planning, preparation, design and construction to operations and maintenance. The second two groupings include all the support functions. 3.03 The main purpose of the 1978 reorganization was to correct the dis- parities in conditions of service, quality of personnel and perceived career prospects which had developed between regular NIA staff and personnel attached to the numerous ad hoc offices set up to implement foreign-assisted projects. At the time of the reorganization the special projects staff was slightly larger than the regular staff; there were in effect two parallel staffs, with a substantial amount of wasteful duplication. The reorganization was also motivated by the perceived need to begin to shift emphasis from construction to operations and maintenance to meet the changing requirements of the eighties. Such a shift implied the devolution of greater responsibility to  - 18 - the Regional Offices which coincided with the avowed governmenL policy of decentralization. While the reorganization has many positive aspects, it has resulted in a more complicated and diffused chain of command, as Regional Directors and Project Managers are now responsible to three or four Assistant Administrators for various aspects of their work, whereas in the past they usually dealt with only one. 3.04 The Assistant Administrator for Development and Implementation is responsible for the identification, preparation and construction phases of projects, in theory through the Regional Directors, but more frequently directly through special project offices. Notable exceptions to this rule are the two National Irrigation Systems Improvement Projects (NISIP I and II), which have always been implemented by the Regional Directors under the overall supervision of the Assistant Administrator for Engineering and Operations. A principal aim of these projects was to strengthen the capa- city of the Regional Offices and to bring their effectiveness in line with that of the various special offices created to build foreign-assisted projects. 3.05 The Project Development Department (PDD), under the Assistant Administrator for Development and Implementation, is responsible for identi- fying potential projects, bringing them forward in the pipeline and preparing them up to full feasibility level. The PDD started life in the early sixties as the NIA counterpart group to the United States Bureau of Reclamation team of specialists engaged under USAID to prepare the feasibility report for the Magat River Multipurpose Project. Over time the group grew to its present size and status and has been responsible, either on its own or with consultant assistance, for identification and preparation of most of the projects built or under construction since 1970. With the gradual exhaustion of more easily developed projects, the emphasis in the PDD's work has changed to longer-term subsector planning, together with support to consultant teams working on feasibility studies of large monolithic projects with relatively long gestation periods. Because of its range of experience, level of competence and familiarity with the requirements of both multilateral and bilateral donors, the PDD frequently provides leadership or trained personnel for the preparation phase of multi-agency projects, even when the irrigation component may be small. 3.06 Since the beginning of the major expansion of irrigation in the early seventies, NIA has shown excellent ability to handle the construction of large works. The agency has assigned the best of its staff, large experienced contractors have been employed and highly competent consultants have been used to assist with design, bid evaluation, contract administra- tion and construction supervision. The agency's record of performance with the construction or rehabilitation of smaller, more dispersed works has been much more uneven. The level of NIA technical and supervisory staff has been variable, contract administration has been much slower, the quality of contractor has been generally poorer and the long distances from the Manila/ Central Luzon construction market of most of the smaller projects have  - 19 - discouraged the better contractors from competing. Performance has tended to deteriorate over time as the agency's resources have come under strain from the increasing work load. Design and construction management capabil- ities have become limiting factors, with the deficiencies being accentuated in the last few years by the brain drain of skilled engineering staff to the Middle East and the private sector The agency's contract management proce- dures, which were designed to handle a small volume of work at a deliberate pace, have been overwhelmed by the quantity of work and the demands for adherence to tight timetables. Increasing the personnel in proportion to the volume of work has made the system even less responsive to control. The management of NIA is very conscious of the problem and is in the process of redesigning the procedure to make it more responsive to requirements and subject to constant step by step monitoring. 3.07 When the construction phase of a national irrigation system has been completed, responsibility for operation and maintenance (O&M) of the system passes to the appropriate Regional Director who is answerable to the Assistant Administrator for Engineering and Operations- Historicaly O&M on the national systems has suffered from chronic and severe shortages of funds and the near absence of any efficient mechanical equipment. As a result, main- tenance has been inadequate to keep up with normal wear and tear, except where emergency grants have been made available to deal with severe typhoon damage, in which case maintenance has actually been reconstruction. The general lack of funds and equipment to properly maintain the systems, together with the lack of physical facilities and discipline to ensure adequate equitable and timely distribution of water to farmers, has led to a lack of credibility of the agency with its clients which has been reflected in the traditionally low rates of collection of water charges. Inability to collect has led to a vicious circle of decreased funds for maintenance less maintenance, greater farmer frustration and lower payments of water charges. 3.08 The magnitude of the problem is suggested by Table 3.1. Aside from the low and quite variable collection rates shown, an important fee-related effect is suggested by the sharp drop in collections from 1975 to 1976. This was largely a result of the large increase in fees from P 60 per double- cropped ha to 2 cavans of paddy per ha in the wet season and 3 cavans per ha in the dry season, worth P 240 to double-cropping farmers in 1976. Farmers who were used to paying a nominal fee resisted the change since no improvement in services accompanied it most Upper Pambanga River Project farmers never experienced the fee subsidy and their payments were better partly as a result. National irrigation system fees averaged P 152 per cropped hectare in 1980, not enough to cover that year's O&M expenditure of P 164/ha of irrigated land. Bank-assisted projects are authorized to charge 7.9 cavans! double-cropped ha within five years of project completion. The collection rate of fees on communal irrigation systems is believed to be substantially lower than that on national systems because collections for many systems have been handled by a separate agency with little interest in aggressive collection efforts systems are more dispersed and farmers less accessible, farmers have often felt that systems designed without their advice did not well meet their needs and were not truly their responsibility, and  - 20 - Table 3.1: COLLECTION OF WATER CHARGES ON NATIONAL IRRIGATION SYSTEMS /a Amount Collected as Percent of Fees Due Upper Pampanga Cotabato Irrigation All national River Project Project systems (78,400 ha) (8,900 ha) (current account) 1974 n.a. 71% 36 1975 n.a. 30% 41 1976 23% 8% 21 1977 34% n.a. 26 1978 28% n.a. 34 1979 29% lb n.a. 36 /a Presents figures only for years for which data were available. /b Represents collection for dry season crop only. maintenance problems have been even worse than those on national systems. It should be noted, however, that all communal irrigation fees can be credited to capital recovery, since irrigators bear O&M costs directly, while at current collection rates national irrigation system fees are entirely consumed by O&M costs, allowing no capital recovery at all. A recent reversion of most collection responsibilities to NIA and a new commitment by NIA to developing strong Irrigation Associations with close farmer involvement in system design is improving the collection record (the current account collection rate reached 42% in 1981) and O&M efforts. 3.09 The rehabilitation, upgrading, and expansion of the national sys- tems which has been taking place since the early seventies will produce the physical prerequisites for reliable, timely and equitable delivery of water to all users, providing that the O&M staff have the knowledge, trust of the farmers and equipment and funds to operate and maintain the relatively sophisticated systems now coming into production. The Bank-assisted Upper Pampanga System (UPRIIS) and the ADB-assisted Angat-Magat Project are the only two modern systems now in operation, and NIA is in the process of learning how to operate and maintain them efficiently. The Bank-assisted Communal Irrigation Development Project is designed to improve the conditions which underlie poor operation, maintenance and fee collections on communal systems. 3.10 Both the UPRIIS and Angat-Magat systems are designed for rota- tional irrigation, based on a 50 ha rotation unit, consisting of 5 subunits  - 21 - of 10 ha. The key man in the NIA operational staff dealing with the farmers is the Water Management Technician (WMT). These are mostly agricultural graduates whose main functions are to improve water management practices, to educate the farmers to make better use of water, and to gradually prepare the farmers to assume more active roles in the operation and maintenance of the facilities at the field level. The standard staffing pattern is one WMT to 10 rotational units, or 500 ha, although a denser coverage is provided where holdings are very small or where the technical competence of the farming community is low. In 1975 NIA established with Bank assistance (Loan 1080-PH) a program for training WMTs for the group of Central Luzon systems centered on the UPRIIS. The one year WMT training, which emphasizes learning by doing, teaches the trainees water management, crop production and farmer organization. The training program has been greatly extended since its inception and there are now four regional training centers in Northern Luzon, Central Luzon, Panay and Mindanao. To date over 400 WMTs have graduated from the training, and it is planned to process a further 600 candidates by 1985 so that all foreign-assisted projects, which will account by then for 80% of the national systems, will be fully staffed with WMTs. In systems where trained WMTs are operating, field efficiencies are gradually approaching the target levels of 58% in the wet season and 70% in the dry season, crop yields of about 4 ton/ha of paddy are being reached before the estimated full development date and cropping intensities are climbing to design levels. Surveys in UPRIIS have shown that trained WMTs were not being called upon to use the full range of skills they had been taught because their supervisors did not understand the scope of the training and because of their own deficiencies did not know how to use it. The majority of the supervisory staff are civil engineers, most of whom have little operational experience, have only recently been transferred from construc- tion and have a limited comprehension of the crop production functions of irrigation systems. The need to extend the training upwards through the operations hierarchy has become apparent, and NIA is planning to initiate such training starting with the zone engineers, i.e., the level immediately above the WMTs. 3.11 In order to ensure efficient utilization of the expensive capital investments being made in irrigation, NIA has been the prime mover in devel- oping five-year agricultural development plans for newly developed or rehab- ilitated systems. The plans come into operation in the year construction is completed and are reviewed and revised annuallly. An Input-Output Monitor- ing Program (IOMP) was established with Bank assistance (Loan 1227-PH) as an adjunct to the agricultural development plan for the UPRIIS. Provision for IOMP has since been included as a regular feature on all subsequent Bank- assisted projects. The main objectives of the monitoring program are: (a) to serve as an early warning system that will monitor the flow and buildup of input supplies into the project area so that any shortfalls are detected with enough lead time to allow for remedial action before a crisis situation develops;  - 22 - (b) to measure the quantity and quality of input use; and (c) to monitor production and yield levels and others benefit flows. The program has been successfully implemented in the UPRIIS and adjoining Aurora-Penaranda Irrigation Project and is currently in its sixth year of operation. While the IOMP has worked very well at identifying weak spots in the input supply and delivery system, it has not always proved possible to remedy the deficiencies as speedily as would be desirable as responsibility for inputs is fragmented between many agencies of the Government and private enterprise. When problems have been identified which are outside of NIA's jurisdiction, a process of negotiation and persuasion has had to replace immediate executive action. An interesting result of the introduction of the WMT training and the IOMP has been to create within the operations staff a constituency with a keen interest in the crop production functions of irrigation rather than in the traditional civil engineering aspects. 3.12 As the ambitious construction program of the seventies begins to run down and the newly built or rehabilitated systems come on stream, NIA is being faced with the need for a major change in emphasis from construction to operations and maintenance. In achieving this objective, the agency is confronted with a number of problems, amongst which are: (a) determining the level of O&M required to optimize and safeguard the irrigation capital investment which is compatible with the recurrent funds likely to be available; (b) attracting and retaining high quality staff in the professionally less glamorous O&M activities; (c) rationalizing and improving the whole process of water charges assessment and collection; and (d) lessening of the agency's dependence on public funds by diver- sification and improvement of its sources of income. NIA, with Bank assistance (Loan 1809-PH), is engaging consultants to carry out a comprehensive study of the levels of O&M required to meet the agency's needs now that the new systems are coming into production. The Irrigators Assistance Department (IAD) of NIA carried out an elaborate study of the factors affecting the collection of water charges under the provision of Loan 1367-PH. The study showed that higher collections of water charges would depend on the following preconditions: (a) improved credibility of the agency with its clients through greatly increased reliability and consistency of its services;  - 23 - (b) a credible package of rewards and punishments to back up the collection system; and (c) a practical, simple and efficient billing and collection system. The agency is currently running a pilot program in the Angat River Irriga- tion System designed to meet the requirements identified by the study. The program would be extended to the neighboring UPRIIS and other completed projects in the event of its success. Since 1975, water charges in the Philippines have been denominated in terms of paddy which has provided a degree of indexation against inflation and has freed NIA from the difficult task of frequent recourses to the President in order to raise the level of water rates. In view of the political unpopularity of increasing water rates and the traditional difficulties of collection, NIA management has been considering alternative sources of revenue. The NIA charter allows the agency considerable latitude to engage in potentially profitable activities connected to its main field of work. Over the the last few years the agency has developed excellent specialized engineering capabilities such as drill- ing, grouting and tunnelling and these are now being hired out for profit in the Philippines. Similarly, NIA has recently set up a subsidiary consultant firm to market its project preparation know-how both in the Philippines and in neighboring countries. In the Bank-assisted Watershed Management Project (Loan 1890-PH) NIA is working on an agroforestry venture with the residents of the Pantabangan watershed which could prove to be very profitable for both participants. Other Institutions Involved in Irrigation 3.13 In addition to NIA, several institutions are involved in irriga- tion development. They include: The National Water Resources Council (NWRC), the Farm Systems Development Corporation (FSDC) and the Bureau of Flood Control and Drainage of the Ministry of Public Works. 3.14 The National Water Resources Council, created in 1974, has as its members the Minister of Public Works and Highways who is Chairman, the Ministers of Natural Resources, Energy, Agriculture and NEDA, as well as the Administrator of NIA, the President of the National Power Corporation and the General Manager of the Metropolitan Water Works and Sewerage System. It is responsible for formulating the rules and regulations for the use and management of water resources and for coordinating all activities related to water development in the Philippines. Specifically it is responsible for registering and issuing water permits for the water sources throughout the country. The Council is also responsible for coordinating the different programs and projects of the various agencies involved in water resource development, including not only irrigation, but also domestic and industrial uses. With UNDP aid, the NWRC is currently gathering and publishing basic information on river basins for planning purposes. These publications can provide a good starting point for a more programmed approach to the development of specific basins.  - 24 - 3.15 The Farm Systems Development Corporation, created in 1975, has as one of its major functions the organization of farmers into Irrigation Associations in areas where NIA has constructed a communal system or where pumps have been sold on credit to the farmers. Up t- 1980, FSDC had organ- ized 1,400 Irrigation Associations (IAs) covering a total area of 190,000 ha and with a membership of about 70,000 farmers. Since 1976 FSDC has also been responsible for collecting, on NIA's behalf, 90% of the costs of con- structing communal systems. In 1981, however, NIA assumed responsibility for all activities related to foreign-assisted communal systems. While FSDC initially concentrated on the organizational function, it is diversifying rapidly and becoming involved in a wide variety of programs including agri- cultural extension activities, cottage industry development, storage and processing, low-cost energy production and rural credit. Lately it has become the rural arm for the Ministry of Human Settlements' BLISS program (Bagong Lipunan Sites and Services). This wide diversification of activi- ties militates against a more effective role in its main function of organ- izing and supporting the lAs. It also accounts for the thin spread of FSDC Institutional Officers compared to the technical staff provided by NIA. Problems developed in the arrangements for division of responsibilities between the two institutions. FSDC became involved in construction activi- ties and NIA developed its own organizational capability to parallel its construction responsibilities. These events were triggered, among other reasons, by the mounting number of communal systems for which construction has been completed but that have not been turned over to the farmers either because there is no functioning organization to assume responsibility or because of disagreements on the system-s design, caused by lack of farmer involvement at the initial stages of construction./l The role of the two agencies in the preparation, construction and operation of communal irrigation systems was recently reviewed and redefined. By 1985, NIA will assume all responsibilities for systems of over 100 ha, and FSDC will take over all responsibilities for most systems under 100 ha. 3.16 The Bureau of Flood Control and Drainage in the Ministry of Public Works and Highways is in charge of all flood control projects in the coun- try. Except for river training activities which have some effect on agri- cultural lands, the bulk of its activities are concentrated in urban areas, In the last three years the Metro-Manila Flood Control and Drainage Project has absorbed between 40-60% of the Bureau's budget. Recently, however, the Bureau was given the chair of the Committee for the Small Water Impounding Management (SWIM) Program to coordinate the activities of the different agencies involved in the program, which include NIA, FSDC, the National Electrification Administration, the Ministry of Agriculture (Bureau of Soils), the Ministry of Natural Resources (Bureau of Forest Development), /1 Of the 350 communal projects completed in the last four years, only slightly more than 50% have been turned over to the users; the rest are still under dispute, mostly due to organizational problems.  - 25 - and the Ministry of Human Settlements. The program is aimed at constructing small earthen dams up to 20 m high built at the head of a watershed which are used for flood control as well as for irrigation, fish production, energy, etc. By 1981, 10 such projects were completed- Their irrigation and productive impact is small since an average of only 100 ha are being irrigated and between 50 and 80 farmers are benefitting per project Unit costs have tended to be high and the economics of this form of development remain doubtful. 4 IRRIGATION DEVELOPMENT: PRESENT SITUATION AND POTENTIALS Background 4.01 There is a long tradition of irrigation in the Philippines. Even before the arrival of the Spaniards, the original inhabitants had developed expertise in water control, as witnessed by the 25,000 ha of mountain slopes converted into rice terraces in Northern Luzon.- In Spanish times, about 30,000 ha of church lands around Manila were brought under irrigation. Government involvement in irrigation began early in the 1900s, but it was not until 1923 that the first irrigation project the Talavera River Irrigation System in Nueva Ecija, was completed. Construction of new systems continued until the Second World War, albeit at a slow pace, and by 1940 ten irrigation systems covering about 80,000 ha had been completed. After the war, the Government stepped up its construction activities so that by 1960 new systems covering about 160,000 ha, had been completed. 4.02 In the mid-1960s the Government took the first steps in an irriga- tion program aimed at making the Philippines self-sufficient in rice. The NIA was formed in 1964 and a USAID-financed study of the development poten- tial of major river basins was undertaken by the United States Bureau of Reclamation (USBR). Numerous projects identified in the USBR study have been or are being built by NIA with financing from the Bank and ADB. NIA's program has included new projects to irrigate rainfed rice land and the upgrading of existing projects. Many of the projects taken over by NIA in the mid-1960s had technical deficiencies which made them very difficult to operate and maintain and this created many problems in delivering a timely and reliable supply of water to the farmers' fields- Upgrading included the provision of control structures, service roads, and construc- tion of tertiary canals and drains. By 1968. when the Bank first became involved in Philippine irrigation) the area covered by national schemes was about 370,000 ha, and communal and other smaller projects covered about 400,000 ha. Table 4.1 shows the present situation in the five main geographic areas, and the three types of systems are described below. 4.03 National Systems are built, operated and maintained by NIA. They range in size from about 100 ha to nearly 80,000 ha. Existing systems have an irrigable area of about 470,000 ha, of which about 360,000 ha have been,  - 26 - or are being, upgraded under Bank and ADB-financed projects. Total cropped area is around 640,000 ha, giving an overall cropping intensity of 137%./l Completed projects include the Bank-financed Upper Pampanga Project (78,400 ha); the ADB-financed Angat-Magat Project (70,000 ha), both in Luzon, and ADB-s Cotabato Project (8,000 ha) in Mindanao. Projects under implemen- tation will add a further 330,000 ha to the national systems. These new projects and those being upgraded will include design standards developed in the early Bank and ADB projects. The canals include all necessary struc- tures to ensure control of water levels over a wide range of flow. Particu- lar attention is given to tertiary canals and farm ditches which are designed to permit rotational irrigation. The service unit is about 50 ha, but each 10 ha subunit has a turnout from which farm ditches carry water to each farm. Service roads, to facilitate access for operation and mainte- nance, are provided along main canals (5 m wide on each bank) and along laterals (3 m wide on one bank). Important features of the national systems are NIA's Water Management Technicians (WMTs) who assist the farmers in setting up irrigation schedules and advise them on water management and cultural practices to fit the schedule. /1 See page 34 for definitions of cropping intensity and their importance in estimating irrigated area of rice.  - 27 - Table 4.1: AREAS SERVED BY IRRIGATION SYSTEMS, 1980 ('000 ha) Existing Total National systems pump & area to Under Existing other Total be ser- Exist- construc- communal irrigation area ved in Regions/b ing /b tion Total systems systems served future/c Northern 80/d 110 190 165 40 285 395 Luzon (40o) Central 200/e 20 220 115 45 360 380 Luzon (35) Southern 55 75 130 80 40 175 250 Luzon (50) Visayas 70 25 95 60 35 165 190 (45) Mindanao 65/f 100 165 130 30 225 325 (35) Philippines 470 330 800 550 190 1,210 1,540 (205) /a Northern Luzon includes Region I & II except Pangasinan Province; Central Luzon includes Region III and Pangasinan Province; Southern Luzon includes Region IV & V; Visayas includes Region VI, VII & VIII; and Mindanao includes Region IX, X, XI, and XII. /b Figures in parenthesis are areas under rehabilitation. T7 Excluding communal systems under construction. /d 40,000 ha upgraded under the completed ADB Angat-Magat project. 7e 110,000 ha upgraded under completed Bank & ADB projects. /f 8,000 ha upgraded under completed ADB Cotabato project. 4.04 Communal Systems are owned and operated by the farmers themselves. Presently, there are nearly 5,700 such systems covering nearly 600,000 ha./l Northern Luzon has the largest proportion with more than 2,700 systems covering 165,000 ha, followed by Mindanao with 130,000 ha, and Central Luzon with 115,000 ha. The average cropping intensity is estimated at around 130%. Communal systems vary in size from a few hectares to as large as 4,000 ha, but over 70% covering 24% of the area are smaller than 100 ha. More than 70% of the communal systems were built before 1965 with some /1 Not all this area can be considered operational, as it is based on poorly-updated provincial records. RIDs project listings yield a total of 470,000 ha.  - 28 - dating back to Spanish times. There are wide differences in performance between communal systems, depending on natural conditions (topography, hydrology, soils) and the standards of construction and maintenance. If river flows are dependable and the river can be easily diverted to flat tracts of paddy land, a communal project can provide fairmers with an ade- quate and reliable supply of irrigation water. Where natural conditions are favorable, as in many of the older and well-established schemes, any defi- ciencies in construction and maintenance can usually be remedied at fairly low cost. Improvements often include replacing temporary dams with perman- ent diversion weirs, and general upgrading of canals including structures and service roads. Since most of the best sites have already been devel- oped, new developments will generally encounter less favorable topography and less dependable sources of water. There is, however, scope for rehabil- itating, upgrading and extending some of the older systems where defects in the original design and construction make them difficult to operate and maintain. The Communal Irrigation Development Project provides an opportunity to determine the extent of this problem and the potential for new projects. 4.05 Government assistance to communal systems dates back to the early l900s; however, since NIA-s creation in 1964, a stepped-up program has been in effect. About 2,000 systems covering about 340,000 ha have received help from NIA in design and construction. When the systems are completed, opera- tion and maintenance is passed on to an Irrigation Association set up with help from NIA or the Farm Systems Development Corporation (FSDC). 4.06 Pump Irrigation and Other Private Systems. These schemes are constructed and supervised by the Irrigation Service Unit, formerly under the Bureau of Public Works and now under the administrative supervision of NIA. It is estimated that about 150,000 ha are presently irrigated by these schemes, but no reliable inventory exists for these projects. Nor is there a clear indication of the areas truly served by these systems or the poten- tial for future development. Most pump systems take water from reservoirs and rivers; groundwater pump systems are far less numerous, but are to be expanded in number. Present Situation and Potential by Regions 4.07 The present situation and future potential for irrigation are described below for each of the five main geographic units in the Philippines. 4.08 Northern Luzon includes Ilocos and the Cagayan Valley. Ilocos is mostly mountainous with short rivers and narrow coastal plains, and its irrigation potential is rather small. Fifteen existing national systems cover about 15,000 ha. All but one of these systems are now being upgraded and extended under the Bank-financed National Irrigation Systems Improvement Project (NISIP) Stage I. The only major project identified in Ilocos is the Palsiguan Multipurpose Project. It would consist of a 150 m high rockfill dam across the Palsiguan River to store about 380 MCM, a power plant with an  - 29 - installed capacity of 60 MW and an irrigation system of about 22,000 ha of which 14,000 ha would be newly irrigated lands. NIA has also identified five smaller projects in Ilocos to irrigate a total area of about 10,000 ha of currently rainfed lands. 4.09 Most of the Cagayan Valley's potential is already being developed under the ongoing Magat River Multipurpose Project and the Chico River Pro- ject which, when completed, will irrigate 102,000 ha and 20,000 ha, respectively. Further development of the Chico River would require a reservoir upstream of the existing scheme; this would add about 15,000 ha of new irrigation together with some hydroelectric power. Another storage project proposed is the Matuno reservoir on the Matuno River in Nueva Vizcaya. This would provide year-round irrigation to about 17,000 ha of presently irrigated land. NIA has also identified twenty small projects covering a total area of about 90,000 ha, of which 70,000 ha would be newly irrigated lands. 4.10 Central Luzon, often referred to. as the "rice bowl" of the Philippines because of its rich soils and large expanse of flat lands, is drained by the Agno and the Pampanga Rivers. Although some large schemes have been carried out, Central Luzon still has a considerable irrigation potential. Of a total potential irrigable area of 820,000 ha, about 360,000 ha are presently irrigated, 160,000 ha of which are either communal or pump schemes. When the ongoing irrigation projects are completed, close to half the overall potential or around 400,000 ha will be irrigated. The Central Luzon Study, carried out by NIA with the help of consultants financed by the Bank (Loan/Credit 984/472-PH) has identified 34 irrigation projects covering 440,000 ha of which 330,000 ha would be newly irrigated lands. The plan is based on the full utilization of the surface and groundwater resources in Central Luzon as well as bringing in water from the neighboring Cagayan River Basin where there is a surplus of water. Three major projects are included in NIA's plans. They are Balog-Balog (32,000 ha) on the Tarlac River, Lower Agno (70,000 ha), and the Casecnan Trans Basin Diversion Scheme (50,000 ha). A feasibility study has been completed for Balog-Balog. The project involves the construction of a 110 m high dam on the Bulsa River to store 625 MCM of water and provide year-round irrigation to 26,000 ha in the Tarlac Irrigation System area and another 6,000 ha of wet season irrigation. Some 18,000 ha of the total of 32,000 would be upgraded and 14,000 ha would be newly irrigated. The project would also have an installed capacity of about 40 MW and generate annually about 100 GWh. It is currently scheduled to begin construction in 1984 and is expected to be completed by 1988. The rice supply projections made in Chapter 5 assume that the feasibility study estimates of Balog-Balog rice production will be achieved. The Lower Agno Irrigation Project would involve the upgrading of 27,000 ha of presently irrigated land and the extension of irrigation to 43,000 ha of rainfed paddy land. This project would require construction by the National Power Corporation of the 190 m high rock-fill San Roque dam across the Agno River. This dam would store about 1,300 MCM and provide water for an overall cropping intensity of 190% on the 70,000 ha service area. A hydroelectric plant at the damsite would have an installed capacity of 390 MW and generate 900 GWh annually. Project feasibility studies for  - 30 - the power and irrigation component were completed in November, 1981, but the project has been deferred due primarily to the high cost of its power canponent. Nevertheless for illustrative purposes the feasibility study estimates of Lower Agno's incremental rice production are included in the Chapter 5 projections. A third major project involves importing water from the Casecnan River Basin (part of the Cagayan Basin) into Central Luzon through the construction of a storage reservoir on the Casecnan River and a 22 km tunnel from the reservoir to the existing Pantabangan Reservoir. This would allow the installation of a 100 MW power plant at the outlet of the Transbasin tunnel as well as increase the capacity of the existing power plant at Pantabangan by an additional 100 MW, generating a total of 700 GWh annually. The water stored would provide year-round irrigation to an existing 25,000 ha plus an additional 25,000 ha of wet season irrigation. A reconnaissance report was prepared in 1977 and feasibility studies began in 1981. In addition to these three major projects, NIA has identified three other storage projects to serve about 50,000 ha of newly irrigated lands. 4.11 Southern Luzon encompasses Southern Tagalog and Bicol. The Southern Tagalog region is composed of the southwest part of Luzon island up to Quezon Province and the islands of Mindoro and Palawan. The portion in Luzon is highly developed and includes the capital city of Manila and one of the largest lakes in Southeast Asia, the 40 km wide Laguna de Bay used for irrigation, fishing, water supply, power and recreation. This region has 17 irrigation systems covering 41,000 ha including the Laguna and Cavite Friar lands, dating back to Spanish times. Five systems, covering almost 5,000 ha are being rehabilitated and extended under the Bank's ongoing NISIP II Project, while seven systems covering about 10,000 ha are included in ADB's Laguna De Bay Development Project Stage I. An additional 5,000 ha will be irrigated under this project, and under the ADB-financed second stage another 14,000 ha will be upgraded. There is little potential for further irrigation development other than rehabilitation of the rest of the existing systems. 4.12 Mindoro Island, separated from southern Luzon by only a 20 km wide channel, is relatively undeveloped. The island is divided by a mountain range into two provinces, Mindoro Oriental and Mindoro Occidental. Rivers are short and swift with unstable courses. Presently, NIA has two irriga- tion systems on the island, covering about 10,000 ha which will be upgraded under the Bank-financed Mindoro Rural Development Project (Loan 1102-PER). The recent Bank-financed Medium Scale Irrigation Project (Loan 1809-PH) provides for constructing new national systems on about 28,000 ha of which 22,000 ha would be in Mindoro Oriental and almost 6,000 ha in Mindoro Occidental. About 300 km southwest of Mindoro is the relatively undeveloped island of Palawan. The island measures 425 km long but only 20-40 km wide and is also divided by a rugged mountain backbone. The largest pockets of level to near-level lands are in the northern end of the island and along a narrow coastal strip south of Puerto Princesa. It is in this area where the Medium Scale Irrigation Project will provide irrigation facilities on 10,000 ha of mostly rainfed lands. Other than these ongoing projects, there is little additional potential in these two islands.  - 31 - 4.13 The Bicol region consists of three parts: the Bicol Peninsula on the island of Luzon, the Catanduanes island on the east and the Masbate island on the southwest. The Bicol River is a major feature of the penin- sula. Its basin is about 130 km long and has a drainage area of 312,000 ha. About 50,000 ha are currently irrigated, including 12 national systems covering 12,000 ha. The Bicol River occupies the flat central plain and flows in a general northwesterly direction from Lake Bato to the Bicol estuary which drains into San Miguel Bay. 4.14 The Bicol River Basin Development program, conducted in 1976, identified seven project areas with a total irrigable area of 70,000 ha including the existing schemes. USAID is currently financing one project, the Libmanan/Cabusao Project covering 4,000 ha, and the Naga-Calabanga and Rinconada Integrated Area Development Projects (17,300 ha) have been included in ADB-s Bicol River Basin Development Project appraised in 1979. Of the other projects, only the Baliwag-San Vicente Integrated Area Development Project appears attractive. It would improve irrigation on 3,400 ha of currently irrigated lands and extend irrigation and drainage facilities to an additional 8,500 ha. 4.15 The Visayas, one of the most backward and economically depressed areas of the country, includes three administrative regions scattered over six major islands: Panay, Negros, Cebu, Bohol, Leyte and Samar. The largest portion of flat land in the Visayas, extending over about 150,000 ha, is located in the central part of Panay where ten existing national systems covering about 40,000 ha of irrigated land are located. Six of these systems are presently being upgraded and extended under the Bank-financed NISIP II and Jalaur I projects. A second stage of Jalaur is being prepared. This involves construction of a 140 m high dam which would store about 370 MCM of water and would provide year-round irrigation to about 36,000 ha, of which 11,000 ha would be newly irrigated lands. The feasibility study of Jalaur II made estimates of incremental rice output which are included in the Chapter 5 projections of rice supply. NIA has identified ten additional irrigation projects covering some 22,000 ha for possible future development. 4.16 The Ilog-Hilabangan River Basin is located in the western part of Negros Island and with its 120,000 ha of cultivated land is the major agri- cultural area in the island. However, this is mostly a sugarcane area where little irrigation development has occurred. Only four national irrigation systems, two on the west side and two on the east side, are operating in the island. They cover 13,000 ha, and two are being upgraded and extended under NISIP II. As noted in para. 2.16, NIA has identified nine projects covering about 56,000 ha, most of them of the rurr-of-river type and mostly located on the west side of the island. They are likely to involve large sugar-growing areas. 4.17 Leyte is the only other island of the Visayas with some irrigation potential. It is also divided into a west and east side by a rugged moun- tainous backbone. Along the northeast coast there is another north-to-south  - 32 - 30 km long mountain range. Between these two mountain ranges and in the northern portion of the island is the plain of Leyte (about 120,000 ha). Presently there are 11 national irrigation systems in operation covering about 10,000 ha. Seven of them, covering an area of 3,000 ha, are being upgraded under the Bank-financed NISIP I and three new systems covering 3,500 ha are being constructed under the same project. There is some fur- ther irrigation potential in North Leyte along the Sangputan River and in the plains facing Carigara Bay, and NIA has identified a 4,500 ha project in this area. 4.18 In the islands of Bohol, Cebu and Samar, only communal systems are in operation. The Bank-financed Rural Infrastructure Project includes 24 such systems in Panay, covering about 6,400 ha, and 10 in Bohol covering 1,600 ha. Another Bank project, the Samar Rural Development Project, pro- vides for construction of new facilities on about 2,000 ha in the island. The Catubig Valley in the north of Samar has some potential for development, but drainage rather than irrigation is needed because annual rainfall is about 4,000 mm. Consultants financed by the Bank are working on a development plan for this basin. 4.19 Mindanao, the second largest island in the Philippines, encom- passes four administrative regions. Apart from Central Luzon, it is the only major part of the country where there is still substantial potential for future development. The major impediment, however, for a smooth development of these lands is the security problem present in much of the island. Two large valleys, a plateau, and several coastal plains contain most of the irrigable area in the island. The Cotabato Valley covering 500,000 ha in the southwest is the largest valley in Mindanao. However, about 70,000 ha are marsh land including the Liguasan marsh along the Buluan River and the Libungan marsh along the main course of the Cotabato River. Along the eastern part of the island run the contiguous Agusan and Davao Valleys of 100,000 ha and 250,000 ha, respectively, while the Bukidnon plateau in the north-central part of the island covers about 500,000 ha. Along the east coast, on the seaside of the eastern mountain range and all along the northern shore there are many small coastal plains of varying width, while small scattered coastal plains as well as intermountain basins are present on the Zamboanga Peninsula. 4.20 Presently, 17 national irrigation systems covering about 50,000 ha are operating in the area. Under the Bank-financed NISIP II project, 10 systems, covering 26,000 ha, are being upgraded and extended. A major thrust of ADB's program of assistance to NIA is in Mindanao with one com- pleted and eight ongoing projects covering 121,000 ha. These projects have begun to realize the island's irrigation potential in the major areas des- cribed above. NIA has identified 27 projects to irrigate some 270,000 ha. About eight projects covering some 100,000 ha have already been pre-iden- tified for ADB financing, while NIA is preparing, for possible Bank financing, a feasibility study for a second mediuim-scale project covering some 60,000 ha throughout Mindanao.  - 33 - 5. A PROGRAM FOR THE FUTURE Government Objectives 5.01 The Government-s principal development aims are clearly stated, including the attainment of high, sustained economic growth with self-suffi- iency in food; improvements in the balance of payments; and reduction of income disparities specifically through increased development of lagging regions, especially rural areas, and improvement of the living standards of the poor. Achieving these objectives is, however, constrained by resource limitations, and in some cases objectives are not wholly compatible. Irriga- tion investments have an important role to play in both growth and distri- butional aims, but in determining what these investments should be, the impact of different approaches to irrigation development has to be considered, the potential contribution of alternative approaches to achieving the same goals has to be weighed, and an assessment is needed of the trade-offs and comple- mentarities between each thrust. For example, a decision to concentrate funds on large-scale irrigation projects on efficiency grounds (lower investment costs per incremental ton of crops produced) could have negative effects on income distribution, and perhaps indirectly on health and domestic peace. Similarly there may be tradeoffs between new irrigation investment and direct or indirect approaches to increasing productivity in either existing irriga- tion or rainfed areas (such as investment in agricultural extension and research, land tenure reform, or crop diversification), while a population policy which slows the increase in demand may give a much higher return than an expansion or intensification of irrigation. Alternatives may be judged by many criteria including the internal rate of return, foreign exchange require- ment, recurrent costs, time to onset of benefits, impact on the supply-demand balance, and ability to reach target groups in the population. The discussion which follows first considers the implications of the Government's current plans for irrigation development, in particular their impact on rice production, examines some alternatives to new investments in surface irri- gation, and suggests some new directions. Rice Production and Consumption Projections 5.02 The need for future irrigation investments rests on reasonable estimates of the need for increased production of irrigable crops. The most important of these is rice, occupying 96% of irrigated area, and rice self-sufficiency is a primary objective. 5.03 For this review, the future growth of domestic rice consumption has been estimated on the basis of the expected population growth, national income growth and the income elasticity of demand for rice. The specific assumptions used in estimating future demand for rice are as follows: (a) Population growth rate is projected to decline from 2.62% p.a. in the 1980-85 period to 1.89% p.a. from 1995 to 2000. This is from  - 34 - Bank projections based on the 1980 census and recent fertility trends. It is accepted by the Philippine authorities; (b) Average national income growth to 2000 is assumed to be 5.5% per annum throughout: this compares with recent rates of 8.3% for the 1975-78 period and 3.6% for the 1978-81 period. With population growth declining, the per capita income growth rate per annum is thus assumed to rise from an average of 2.8% from 1980 to 1985 to over 3.5% from 1995 to 2000; and (c) The income elasticity of demand for rice has been taken as 0.1, the MOA figure. This figure takes into account the rising demand for rice products. Thus, the annual per capita rice consumption growth rate is expected to rise from 0.28% to 0.35% over the two decades. 5.04 The probable growth of rice production has been calculated on the basis of expected changes in the physical area and yields of irrigated, rainfed lowland and upland palay./l A small improvement in the conversion rate of palay into rice available for human consumption has also been assumed. These assumptions are summarized below and elaborated in Annex 2, Tables 5-7. Projections of future rice production have been derived as follows: /1 There has been some controversy over the areas of land irrigated and cropped. In a draft version of this report the figures for the "irrigated area" were those used by NIA in its reports. The Ministry of Agriculture, which also generates data on the irrigated area, suggested that these figures were too high. The difference between the two versions is, however, reconcilable. The NIA term refers to the service area of irrigation schemes, comprising the total area developed including not only irrigable land, but also land taken up by roads, canals and ancillary buildings. Also, in some schemes which were designed without adequate studies of water availability, it includes areas of land which do not receive irrigation water even in the wet season. The NIA concept of "cropping intensity" is the ratio of the cropped area (the area of cumulative plantings of crops over the year) to the service area, cur- rently averaging in the Philippines about 137%. The Ministry of Agri- culture defines the irrigated area as the crop area actually receiving irrigation water in the wet season. The ratio of this area to the cropped area is obviously much higher than NIA's cropping intensity -- on the order of 175% -- but the same expression has been used by the two agencies. With the agreement of MOA and NIA the term "cropping intensity" is here applied to the NIA definition, and, for want of a better term, "cropping ratio" is used in this report to convey the MOA meaning. The difference between cropping intensity and cropping ratio has been larger (over 25%) than it would otherwise have been because assumptions on water availability, on the basis of which the systems have been designed, have often been optimistic. This problem is already being tackled by NIA and, therefore, the difference should be less in future schemes.  - 35 - (a) MOA figures for area cropped in the wet season have been adopted for the existing irrigated areas (this represents a decrease of about 25% from the NIA area figures used previously); (b) for projected additional areas, the areas of new irrigation schemes noted by NIA in its September 1982 Corporate Plan to be under implementation (i.e. on which design or design and construction expenditures have been incurred) have been taken into account; for these the proportion of service area actually cropped has been assumed to rise because, in the outer years, more time will be available to effect design improvements. Three major multipurpose projects (Lower Agno, Balog-Balog and Jalaur II) are assumed to add nearly 127,000 ha to total irrigated cropping (i.e., combined wet and dry season areas), of which 108,000 ha would be in rice;/l (c) the dry season irrigated cropping has been taken as a nearly constant 76% of the wet season irrigated area. This is based on the assumption that relative prices of rice and crops which could sub- stitute for it will remain constant, and that diversification into other crops will come mainly from a higher overall cropping ratio rather than through a lower rice cropping ratio; (d) wet season rainfed lowland area has been assumed to fall by the amount wet season irrigated area rises; (e) for upland area, MOA Policy Analysis Staff projections have been adopted, and the 1990 cropping assumed to hold for the year 2000, indicating a small net reduction in upland rice cropping. (f) yield levels and growth have been differentiated according to the season (wet and dry) for each of 3 areas of production (irrigated, lowland rainfed, upland rainfed), based on MOA survey data and projections. Irrigated yields represent weighted averages of national, communal and pump system yields. In each case the rate of increase to 2000 is markedly lower than that which prevailed in the 1970s (see Figure 1), reflecting the view that growth in that decade was exceptional, caused by the rapid acceptance of HYVs over 73% of the rice area; and (g) the disappearance rate in seed, feed and waste has been held constant at the historical level (10.8%) but milling recovery rates have been marginally raised to reflect expected improvements in milling plant. 5.05 Table 5.1 summarizes the production and consumption projections which result. The table demonstrates that, on the basis of these data, just by completing the irrigation systems already under construction or for which funds have already been earmarked (plus the three multi- purpose projects noted above), the Philippines could increase its recent annual rice surplus over domestic demand to nearly 600,000 tons of milled rice in 1985, nearly 800,000 tons in 1990, and /1 These are project feasibility study estimates. Rainfed rice cropped area would be reduced by over 61,000 ha. Annex 2, Table 6 contains further details.  - 36 - Table 5.1: SUMMARY OF RICE SIJPPLY AND DEMAND PROJECTIONS Crop year 1980 1985 1990 2000 Production ('000 mt) Irrigated 4,313 6,392 7,740 9,223 Lowland rainfed 3,020 2,568 2,562 2,768 Upland 503 312 458 549 Total Palay Production 7,836 9,272 10,760 12,539 Total Rice Production 4,543 5,418 6,335 7,494 Consumption Population (millions) 48.41 55.20 62.23 76.21 Per capita consumption (kg) 86.5 87.8 89.4 94.0 Total Consumption ('000 mt) 4,187 4,845 5,563 7,165 Surplus ('000 mt) 356 573 772 329 (% of consumption) 8.5 11.8 13.9 4.6 Source: Annex 2, Table 3. still be running a surplus of over 300,000 tons in 2000./i The surpluses would peak between 1990 and 1995, and would continue to about 2010. Unless export or new domestic outlets are established, the cumulative milled rice surplus from 1980 to 2000 would exceed 11 million tons. /1 It is to be expected that there would be fluctuations around the pro- duction trend lines caused by the vagaries of weather. However, regression analysis of fluctuations around the production trend over nearly 20 years indicates that such fluctuations would cause, on average over the 1980-2000 period, a rice deficit in only about one year in ten. This risk rises to about one deficit in six years over the 1995-2000 period. Clearly the normal yearly surpluses would easily allow for the maintenance of stocks to cover these small risks of deficit.  - 37 - 5.06 This analysis is, for simplicity, a static one. In the absence of outlets for this excess of supply over domestic demand, the existence of the stocks this surplus would create would put downward pressure on prices of rice to farmers, with negative effects on rice farm incomes and production. The analysis also abstracts from the effects of changes in demand for competing staples, principally corn and wheat, as prices of those products vary, and from changes in relative prices for rice and inputs at the farmgate. Implicitly these are assumed to hold constant. 5.07 The results are highly sensitive to two sets of assumptions made in the projections: the cropping ratio and yield growth of irrigated rice. The balance is fairly insensitive to other assumptions: to changes in rainfed yield growth, deferral of the three large multipurpose projects (Lower Agno, Balog-Balog, Jalaur II), lower milling recovery ratio, and changes in per capita income growth, national income growth, and income elasticity of demand for rice. The sensitivity of the rice supply-demand balance to four of these variables in 1990 and in 2000 is illustrated in Table 5.2, and tests of all these variables are shown in Annex 2, Table 4. 5.08 The analyses show that if the irrigated rice cropping ratio in 2000 were 167% rather than its projected level of 177%, a deficit of 210,000 tons would result. A shift of 84,000 ha (8%) out of dry season irrigated rice in the year 2000 would eliminate the surplus and produce exact self-sufficiency. If the average growth of yields of irrigated rice were to be 20% slower than expected (reaching only 3.4 tons/ha and 3.7 tons/ha wet and dry season, respectively) a moderate annual deficit of 200,000 tons could be expected by 2000. On the other hand, if irrigated yield growth were 20% faster than projected, reaching an average of 3.9 tons/ha in the wet season and 4.1 tons/ ha in the dry season, even in year 2000 (by when consumption would be catching up with supply) the annual surplus would be over 850,000 tons. None of the tests produces a projected rice deficit in 1990, when the base case surplus is near its peak. 5.09 Considering that NIA estimates that average national system yields have already reached 3.7 tons/ha in 1981, and given the considerable effort that is now being initiated through improved extension, fertilizer usage, higher yielding varieties, systems rehabilitation and water management, the projected overall yields for the entire irrigation system for 2000 appear reasonable. Also, the cropping ratios adopted in the projections are close to the average which was obtained over the past decade and there is no persuasive argument to suggest major change in the next 20 years. Consequently, it appears that the projected balance is a reasonable one. However, both cropping ratio and yields depend primarily on the price of palay relative to that of other crops and of inputs; and thus the balance will also be dependent on these price ratios. It is clear that price ratios must constantly be monitored and care taken with public interventions which may reduce the relative profitability of rice. There will also be a need to examine the requirement for further expansion of irrigated area as surplus tendencies shrink during the 1990s.  - 38 - Table 5.2: SENSITIVITY ANALYSIS OF RICE SUPPLY-DEMAND PROJECTIONS Effect on surplus Resultant net surplus Test Case ('000 tons rice)/a ( 000 tons rice)/a 1990 2000 1990 2000 1. Base case - - 772 329 2. Irrigated rice yield: annual growth rate 20% higher or +291 +528 481 to 1,063 -199 to 857 lower 3. Irrigated rice cropping ratio ten percentage points lower -466 -541 306 -212 (from fall in dry season area) 4. Lower Agno, Balog-Balog and -/b -166 772 lb 163 Jalaur II not operating by 2000 5. Income elasticity of demand for -36 -115 736 214 rice is 0.12 (vs. expected 0.10) /a All figures apply to milled rice available for human consumption. Supply is calculated by subtracting 10.8% of the palay harvest for seed, feed and waste, and applying a 66% and 67% milling recovery ratio to the remainder for 1990 and 2000, respectively. /b These projects are assumed to have a production impact only after 1990 in the base case.  - 39 - Supporting Policies 5.10 The expected values of consumption and production presented above do assume that certain government policies and actions continue to operate to reduce the rate of growth of tota