Philippines - First and Second National Irrigation Systems Improvement Projects

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 10669 PROJECT PERFORMANCE AUDIT REPORT PHILIPPINES NATIONAL IRRIGATION SYSTEMS IMPROVEMENT PROJECTS I & II (LOANS 1414-PH & 1526-PH) MAY 22, 1992 Operations Evaluation Department This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. CURRENCY EQUIVALENTD Name of Currency: Peso (P) NISIP I NISIP II Appraisal Estimate US$1.00 - P7.44 US$1 - P7.40 Intervening Years Average US$1.00 - 10.88 US$1 - 12.33 Completion Year Average US$1.00 - 20.50 US$1 - 20.70 ABBREVIATIONS DOH = Department of Health ERR - Economic Rate of Return IBRD = International Bank for Reconstruction and Development IIMI M International Irrigation Management Institute IRR = Internal Rate of Return IRRI = International Rice Research Institute IOSP - Irrigation Operations Support Project NIA = National Irrigation Administration NISIP I - National Irrigation Systems Improvement Project I NISIP II - National Irrigation Systems Improvement Project II O&M w Operation and Maintenance OED = Operations Evaluation Department PCR " Project Completion Report PPAR = Project Performance Audit Report SAR = Staff Appraisal Report FISCAL YEAR OF BORROWER January 1 - December 31 THE WORLD BANK FOR OFFICIAL USE ONLY Washington, D.C. 20433 U.S.A. Office of Director*General Opeations Evaluation May 22, 1992 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Performance Audit Report - PHILIPPINES National Irrigation Systems Improvement Projects I & II (Loans 1414-PH & 1526-PH) Attached, for your information, is a copy of a report entitled "Project Performance Audit Report on Philippines - National Irrigation Systems Improvement Projects I & II (Loans 1414-PH & 1526-PH)". prepared by the Operations Evaluation Department. Attachment 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 PROJECT PERFORMANCE AUDIT REPORT PHILIPPINES NATIONAL IRRIGATION SYSTEMS OMPROVEMENT PROJECTS I t II (LOANS 1414-PH & 1526-PH) TABLE 9 CONTENTS Paite No. Preface . . . . . . . . . . . ....... . . . . . . . . . . Basic Data Sheet . . . . . . . . . . . . . . . . . . . . . . . . . . . Evaluation Summary . . . . . . . . .. . . . . . . . . . . . . . . . vii I. Introduction . . . . ........... . . . I II. Project Objectives .............. . . . . . . . III. Implementation ....... . . . ............... 3 IV. Impact........ . . . . . .... ...... . . . 5 V. Sustainability............... . . . . . . 12 VI. Findings and Issues . ............... . . . . . . . 12 Annet - Summary of Observations from III Report . .. . . . 15 Figures 1. Figure 1: NISIP I - Areas, Yields and Production . . . . . 17 2. Figure 2: NISIP II - Areas, Yields and Production . . . . . . . 18 Tables: 1. Table L: NISIP I - Production Summary . . . . . . . . . . . . . 19 2. Table 2: NISIP II - Production Summary . . . . . . . . . . . . . . 20 3. Table 3: NISIP I - Irrigated Areas, Yields and Production . . . . 21 4. Table 4: NISIP II - Irrigated Areas, Yields and Production . . . . 22 5. Table 5: Rice Price Projections . . . . . . . . . . . . . . . . . 23 6. Table 6: NISIP I - Audit Economic Analysis: Most Likely Outcome . 24 7. Table 7: NISIP I - Audit Rconomic Analysis: Appraisal Rice Prices 25 8. Table 8: NISIP I Audit "oonomic Analysis: Declining Irrigated Areas without Project . . . . . . . . * . . 26 9. Table 9: NISIP II - Audit Economic Analysis: Most Likely Outcome . 27 10. Table 10: NISIP II - Audit Economic Analysis: Appraisal Rice Prices 28 11. Table 11: NISIP II - Audit Ec omic Analysis: Declining Irrigated Areas without Project . . . . . . . . . 29 12. Table 12: Contribution of Revenue Sources to Total Revenue . . . . 30 Attachment: Comments from the Borrower . . . . . . . . . . . . . . 31 MAP IBRD 12507R2 IBRD 23307 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. PROJECT TERFORMANCE AUDIT REPORT PHILIPPINES NATIONAL IRRIGATION SYSTEMS IMPROVEMENT PROJECTS I & II (LOANS 1414-PH & 1526-PH) PREFACE This is the Project Performance Audit Report (PPAR) on the first and second National Irrigation Systems Improvement Projects. The first project was financed by Loan 1414-PH for the equivalent of US$50 million, approved May 5, 1977 and signed May 13, 1977. The original Closing Date, December 31, 1982, was extended to December 31, 1985 to permit completion of project works. The final disbursement was made on July 1, 1986, and the undisbursed balance of US$11,786,932.76 was cancelled on July 3, 1986. The second project was financed by Loan 1526-PH for the equivalent of US$65 million, approved February 28, 1978 and signed March 15, 1978. The original Closing Date, 31 December 1984, was extended to December 31, 1986 to permit completion of project works. The final disbursement was made on July 13, 1987, and the undisbursed balance of US$16.28 million was cancelled at that time. The PPAR is based on the Project Completion Reports prepared by the Borrower, with Overviews added by the Bank's Asia Regional Office,!/ on the President's Reports and the legal documents, on the transcripts of the Executive Directors' meetings at which the projects were considered, on study of projects' files, and on discussions with Bank staff. An OED mission visited the Philippines in June 1991 and visited four of the 50 sub-projects. The mission discussed the effectiveness of the projects with Philippine Government officials from National Irrigation Administration (NIA), National Economic Development Authority, Ministry of Finance, and Philippine Council for Agricultural, Forestry & Natural Resources Research & Development, as well as private citizens and representatives of international organizations. The PCRs provide basic data on the project experiences. The PPAR supplements these data with more recent data on land use and yields and corrects erroneous land use and yield data in the SARs and PCRs from NIA records. The PPAR reviews the way the irrigation systems are being operated and tries to !1 The NISIP 1 completion report submitted to the Board (Report f8390 of February 28, 1990) contains an abridged version of the Borrover's PCR. The NISIP 2 completion report submitted to the Board (Report #7820 of June 7, 1989) contains only the Preface, Evaluation Summary and Overview. In both cases, the Borrower's E%x is available from files. explain why the regulating structures constructed under these projects are sel'om used and why production increases as a result of these projects have been so modest. Following standard OED procedures, copies of the draft PPAR were sent to the Borrower and its executing agencies for comment. The PPAR was revised in the light of co.mnents from the Department of Agriculture - Policy and Planning Division. These comments are also reproduced as an Attachment. PROJECT PERFORMANCE AUDIT REPORT PHILIPPINES NATIONAL IRRIGATION SYSTEMS IMPROVEMENT PROJECT I (LOAN 1414-PH) BASIC DALa SHEET KEY PROJECT DATA Appraisal Actual or Actual as % of Expectation Current Estimate Appraisal Estimate Project Costs (US$ million) 107.2 76.0 71.0 Loan/amount (US$million) 50.0 38.211 76.0 Date Board approval 0/03/77 Date effectiveness 08/09/77 Date physical components completed (2) 12/31/81 06/30/86 200.0 Proportion then completed (2) 45.0 90.0 Closing date 12/31/82 12/31/85 154.0 Economic rate of return (2) 20 Negative 75.0 Institutional performance Satisfactory Agronomic performance Satisfactory Number of direct beneficiaries (year 1987) 240,000 N/A CUMULATIVE DISBURSEMENTS FY78 FY79 FY80 FY81 FY82 FY83 FY84 FY85 FY86 Appraisal estimate (US$ million) 2.4 12.1 25.3 38.5 46.5 50.0 Actual (US$million) - 7.9 10.3 15.2 20.1 22.5 29.1 34.7 38.2 Actual as % of estimate - 65.3 40.7 39.5 43.2 45.0 58.2 69.4 76.4 Date of final disbursement 7/2/86 Principal repaid (to 12i31/eo) 20.36 (US$ million) The undiabursed balance of US$11,786,932.76 vas cancelled effective July 3, 1986. - iv - KISSION DATA k' Date No. of Vandays Speialtsaton Performsaeo tEteeton uplyr) Persons u£pd rueeu.YrtnW8 Idenstltit Preparetta 04/74 S 20 E(2),A,8e(2) - Appraisal 10 & IllIN 6 150 A.Zn,Eo(3),H * * Subtetal 170 * . SupeTviston 1 0177 a A.t Supervislo 2 07177 4 40 A,3(3) 1 Supesites1o 3 03/78 2S E(4). 1 1 Superviton 4 09178 2 8 A. a Superytsto 5 11/78 1 5 1 1. Superviato 4 11/76 2 10 En(2) 2 2 V Sup*wvision 7 02/179 4 0 8(4) 2 2 t Supervt to4 06/79 2 Zn(2) 2 2 t Superv*aion 9 10/79 2 1 A,9 1 1 supertatm 10 03/80 2 10 A's 2 2 Mr Superviteas 11 05180 4 20 Zn(4) 2 1 V Suprviteo 1 10180 1 S A 3 1 t Supervitso 13 03/81 2 is Zn(2) 2 2 t Superviston 14 02/62 1 8 Eo 2 1 MT Supervison 1 10162 3 15 Zn(3) 2 1 M Supervitsio 14 10183 1 7 En 2 1 Mr Supervision 17 07184 1 7 En 3 2 WR Supervieton 1o 03/65 1 8 En 2 iS Supersviton 19 0648 2 10 En(S) 2 * Subtotal M 24 Total 413 OTHER PROJECT DATA Borrower: Republic of the Philippines Executing Agency: National Irrigation Administration (NIA) Fiscal year of Borrover: January 1 - December 31 Name of currency (abbreviation): Peso (P) Appraisal year average US$1.00 - P7.44 Intervening years average US$1.00 * P10.88 Completion year average US$1.00 a P20.50 Follow-on project: Name: National Irrigation Systems Improvement Project II Loan numbers 1526-PH Loan amount (US$million): 65.0 Date Board approval: 02/28/78 k' Total Staff-weeks booked to the project ares Identification/Preparation - 15.2; Appraisal - 71.2; and Supervision - 125.2. A - agriculturalist; Ec - economist; En * engineer; H - public health specialist (Schistosomissis control). I - problem-free or minor problemi 2 - moderate problems; and 3 - major problems. I - improvingi 2 n stationary; 3 - deteriorating. l F - finaLcial; K = managerial; T a technical; and 0 - other. -v - PROJECT PERFORMANCE AUDIT REPORT PHILIPPINES NATIONAL IRRIGATION SYSTEMS IMPROVEMENT PROJECT II (LOAN 1S26-PH) BASIC DATA SHEET EY ?ROJECT DATA Appraisal Actual or Actual as 2 of Expectation Current Estimate Appraisal Estimate ProjecL Costs (US$ miilion) 140.0 80.7 58 Loan amount (US$ million) 65.0 48.7 11 75 Date Board approval * 02/28/78 Date effectiveness - 06/20/78 - Date physical components completed 12/31/83 06/30/87 160 Proportion then completed (2) 62 100 Closing Date 12/31/84 12/31/86 129 Economic rate of return (2) 23 2 160 Institutional performance Satisfactory Agronomic perforance Satisfactory Number of direct beneficiaries (Year 1988) 316,000 N/A CUMULATIVE DISBURSEMENTS Y FY80 FY61 FY82 FY83 FY84 FY85 FY86 -FY8 FY88 Appraisal estimate (US$ million) 0.3 10.0 23.0 38.0 51.0 62.0 65.0 65.0 65.0 65.0 Actual (US$ million) 0.3 11.7 15.5 17.4 21.6 33.1 38.7 44.3 46.4 48.7 Actual as 2 of estimate 100 117 67 46 42 53 60 71 75 Date of final disbursement 7/13/87 Principal repaid to 06/30/88 5.46 (US$ million) The undiebursed balance of US$16,261,974.69 wa cancelled effective July 14, 1907. - Vi - MISSION DATA Date No. of Man-Jays Specializatiole Performan?e .ype of Mission (molyr) Persone in field represented r' ig Trend l problems / Identification NA Preparation NA Appraisal 07 & 03!77 150 En,Ec,A(2),H - - - Subto:al 5 150 Supervision 1 07/78 3 15 En(2),A 1 2 - SupArvision 2 02/79 1 10 En 1 2 T,M Supervision 3 10/79 3 24 En(2),A 1 1 - Supervision 4 03/80 2 10 A,H - - - Supervision 5 0/80 3 20 En(3) 1 1 - Supervision 6 10/80 1 5 A - - - Supervision 7 03/81 2 14 En(2) 2 2 M,F Supervision 8 02/82 1 9 En 2 1 H,T Supervision 9 10/82 3 15 En(3) 2 2 M,T Supervision 10 10/83 1 7 En 2 2 H,T Supervision 11 07/84 1 8 En 2 1 M,T Supervision 12 03/85 1 8 En 2 1 H,0 Supervision 13 04/86 2 10 En(2) 2 - - Supervision 14 04/87 2 14 En,Ec 2 - . Subtotal 26 169 Total 31 319 OTHER PROJECT DATA Borrower: Republic of the Philippines Executing Agency: National Irrigation Administration (NIA) Fiscal year of Borrowers January 1 - December 31 lame of currency (abbreviation): Peso (P) Appraisal year average US$1.00 - P 7.40 Intervening years average US$1.00 - P 12.33 Completion year average US$1.00 - 20.70 Follow-on project: NONE Total Staff-weeks booked to the project are: Identification/Preparation * 14.2; Appraisal = 58.5; and Supervision 100.7. L A ! agriculturalist; Ec - economist; En - engineer; H = public health specialist (Schistosomiasis control). 1 - problem-free or minor problem; 2 - moderate problems; and 3 - major problems. 1 * improving; 2 - stationary; 3 - deteriorating. F - financial; M - managerial; T - technical; and 0 - other. - vii - PROJECT PERFORMANCE AUDIT REPORT PHILIPPINES NATIONAL IRRIGATION SYSTEMS IMPROVEMENT PROJECTS I & II (LOANS 1414-PH & 1526-PH) EVALUATION SUMMARY Introduction 1978 through 1983, was supported by a US$65 million loan approved in 1978. 1. A 4eficit in the production of rice, the main staple crop, was a per- Imvlementation sistent problem for the Philippines in the 1970s. These projects were the 3. Due to the occurrence of most eighth and ninth in a series of Bank- conceivable uatural and unnatural de- assisted irrigation projects designed lays (para. 3.1), implementation of to help the Philippines achieve rice both projects took about twice as long self-sufficiency. While the earlier as originally expected. The majority projects had been for individual large of major contractors did not fully projects, the National Irrigation Sys- complete their contracts. While peace tems Improvement Projects (NISIP) I and order problems, unexpectedly and II were focussed on the many re- severe weather conditions, frequent latively small existing systems judged design changes and unusually high suitable for upgrading. price escalation were some of the causes, the main reason was the vul- Objectives nerability of these contractors to the prevailing difficult financial situa- 2. The NISIP projects were designed tion and the lack of readily available to improve the performance of 50 irri- credit during the project period. The gation systems in several regions of smaller contractors proved less vul- the Philippines. They mainly com- nerable and, together with NIA's own prised re%abilitation, upgrading and torce account teams, managed to com- extension of existing system and plete most of the work left by the construction of new irrigation infra- larger contractors, as well as an strncture; strengthening of the O&M arrpy of smaller works. capabilities of the National Irri- gation Administration (NIA); and pro- 4. With the exception of these grams to reduce the prevalence of implementation delays, the physical schistosomiasis for some existing aspects of implementation of both irrigated areas on Leyte and Mindanao, projects were rated satisfactory in two of the worst endemic areas in the the PCR's of both the Borrower and the country. NISIP I, slated to cost Bank. US$107 million and to be implemented from 1977 through 1981, was supported 5. As a result of slow start-ups, by a US$50 million loan approved in disbursements did not commence until 1977. NISIP II, slated to cost US$140 the second project year of either pro- million and to be implemented from ject, and then contiaed at lower than - l,iii - projected levels throughout. The were written (in 1988 for NISIP I and works for both projects were completed 1987 for NISIP 11), but they have not in mid-1986, 4 112 years later than done so. projected for NISIP I and 1 1/2 years later for NISIP II. The cost of NISIP 9. The principal disappointment was I was US$76 million, 292 less than the very modest expansion in areas projected. US$11.8 million (24%) of irrigated in the dry season. In re- the loan was cancelled. The cost of trospect, it is difficult to under- NISIP II was US$81 million, 42% less stand how these dry-season expansions than projected. US$11.8 million (25%) were supposed to occur. Most pro- of the loan was cancelled. Jects' &.,stems are run-off-river, gravity systems with no capacity to 6. Cost savings vis-a-vis appraisal store water from one season to the estimates stem partly from reductions next. Nothing in NISIP I or Il to some project components (para. 3.8) changed that. The large number of and partl- from falling local costs -'n cross-regulating gates installed by US$ terms when the Peso was repeatedly the project might theoretically have devalued during the long implementa- made it possible to ration and dis- tion period. tribute the same amount of available water better in dry seaison. However, Results in virtually all cases, they are not operated. Canals are allowed to run 7. The agricultural results of the full whenever water is available. two projects were disappointing. Actual paddy production at full devel- 10. Based on actual results and opment was 762 of appraisal estimates, projecting present trends into the a shortfall of 24%. However, incre- future, economic returns to irrigation mental production was only 17% of for growing rice are negative for appraisal estimates, a shortfall of NISIP I and 22 for NISIP II. The dif- 83%. Most of the discrepancy stems ference stems from NISIP I's 35% from a very large underestimation of higher average cost per hectare. production without the project in both There was a 100% over-run on 0&M the SARs and in the preparation re- equipment procured under NISIP I, but ports. Those estimates are at cdds this did not affect the economic with yields and areas cultivated on returns (para. 4.18, note 9). the sub-projects before the NISIP projects ae recorded in NIA's own 11. Buildings, service facilities, records. No justification has been equipment, vehicles and spare parts to suggested for taking a etarting point strengthen NIA's 0&M capability were substantially below what wa& already provided. In the field, system oper- being achieved before the NISIP pro- ations seem to be close to the basic jects. This information is presented minimum which were practiced beforc graphically in the Summary Diagram the project; several thousand farm which follows, and in Tables 1 and 2. turnout gates are seldom operated, as the simple rotation of flows between 8. Some of the discrepancy results secondary canals means they can just from the irrigated areas' having in- be left cpen. Mai-tenance is mostly creased only marginally due to the limited to cutting grAS on canal project, and because yields were banks and around some suzuctures, and essentially static throughout the pro- de-silting. Although 0&M expenditures ject period (paras. 4.14-5). The PCRs are close to projected levels, idi- projected that areas and yields would cations are that more than 802 of start expanding tight after the PCRs these are personnel expenses, and that - ix - the personnel involved spend much of pattern of allocating the dry season their time collecting Irrigation water between them could be worked out Service Fees, rather than in teahnical under circumstances in which less than O&M functions. 3/4 of the farmers could have assured water in any one dry season. 12. The schistosomiasis control components of both projects were 15. The small actual increase in extremely successful, with the preval- dry season irrigated area indicates ence of this disease reduced from that the improved level of water man- 18-302 at appraisal, to 10-15% at agement was simply an unrealistic completion and 2-5% in 1988, making it expectation. Moreover, if the pro- safe to say that schistosomiasis has jected 50% increase in area could ceased to be a major public health have been achieved without any signi- problem in either of the two project ficant increase in the water supply, areas where control programs were there is little evidence to suggest implemented. that the upstream farmers would have willingly accepted the consequences - - Sustainability reduced dry season water deliveries. It appears that neither those who did 13. The irrigation systems of NISIP the feasibility studies for these pro- I and II seem to be working reasonably jects, nor those who appraised them, well. The level of management is low gave due thought to these consider- but it could be near optimal given the ations. prevailing circumstances. Water is a less critical constraint here than it 16. It is not so much the projects is in the arid regions of the world. which have failed as the appraisal Nevertheless, many farmers are getting process. There have been two prin- water and paying for it. Neither cipal areas of failure. The first is irrigated areas nor yields seem to be the presentation of incorrect pre- declining. Although economic analysis project areas and yields. Good mon- shows that the investments were not in itoring data, showing substantially the national interest, the situation higher figures, go back to years well appears to be entirely sustainable. before appraisal. Because project analysis starts from an incorrect Findings base, it counts as benefits production which already existed. The second, 14. The accomplishment of the the over-estimation of achievable project's agricultural production increases in area irrigated, is most objectives depended on two -ritical likely to have resulted from assumptions: a) that without build- unrealistic optimism and un-supported ing any additional storage, improved assumptions regarding the future water control would cause the same behavior of and interactions between amount of water as originally used for farmers and system managers. These dry season irrigation to serve areas incorrect assumptions led analysts to about 50% larger in that season, and; project benefits which have not b) that there would be a sufficient materialized. level of organization and social co- hesion among farmers that an equitable 1. Borrower's Department of Agriculture feels that this statement a ambiguous. See Attachment. See also paras 5.1-2 below. -x- NISIP I 25 300 / Appraisal RC.R. 200 158 / 181 ('000 Tons) 137 AUDIT 121 100 01 1980 1985 1990 796 700 NISIP II PC. 600 516 500 ------- 427 AUDIT / 460 400 389 ('000 Tons) 300 Appraisal 200 100 01 1980 1986 1990 PRODUCTION SUMMARY DIAGRAMS PROJECT PERFORMANCE AUDIT REPORT PHILIPPINES NATIONAL IRRIGATION SYSTEMS IMPROVEMENT PROJECTS I & II (LOANS 1414-PH & 1526-PH) 1. INTRODUCTION 1.1 The Bank's agricultural sector strategy for the Philippines in the early 1970's included supporting the Borrower's objectives of attaining self-sufficiency in cereal production, particularly of rice and maize, and strengthening of sector supporting institutions. Self-sufficiency in cereal production was considered important not only to strengthen the balance of payments, but also to raise incomes for the rural population. 1.2 It was judged that the Philippines would be unable to reach self-sufficiency in rice without substantial investments in irrigation expansion and improvement. A 1974 Sector Surveyl' estimated that of a total of 960,000 ha that could be served by existing systems, only 630,000 were served in the wet season and 254,000 ha in the dry-season. The existing systems were reported to have inadequate structures to provide the level of water control required by modern high yielding rice varieties. Maintenance was minimal due to shortages of staff and funds, and the lack of access roads for maintenance machinery. An aggressive rehabilitation and extension program, planned to reach 50,000 ha per year for 10 years, was accepted as essential to meeting the national objectives. II. PROJECT OBJECTIVES 2.1 NISIP I was designed to: (i) increase production on 49,400 ha of land in Ilocos, in the Cagayan Valley of Luzon, and on Leyte (see map). Twenty-two existing systems were to be rehabilitated and upgraded (27,600 ha) and extended (18,300 ha) by construction of irrigation, drainage and road infrastructure. A further 3,500 ha was to be brought under irrigation by construction of three new schemes. Expected costs of this component, about US$65.7 million. (76% of project costs) including physical contingencies, also included about US$13.9 million worth of equipment for force account construction, O&M of the project facilities and strengthening of regional 0&M capabilities. (ii) reduce the schistosomiasis risk to about 475,000 people on Leyte, one of the worst-affected areas in the Philippines. The climate and coastal plain topography are particularly hospitable to the snail i' IBRD. Agricultural Sector Survey. May 2, 1974 (Report No. 39a-PH). - 2 - vector of this disease, and the low levels of hygiene and constant exposure of the rural population to water, combine to produce an optimum situation fnr high infection levels. This component, expected to represent about 18% of project expenditures, improved drainage, reduced the snail population through mollusciding, and lessened contact with water by providing road and foot bridges. Public hygiene wac improved by environmental and health education, and the parasite reservoir was reduced by diagnostic screening and chemotherapy. (iii) strengthen NIA's operation and maintenance capability in the project area through provision of offices, stores and housing for O&M (4.3% of project cost). (iv) examine the feasibility of land consolidation in an area of high land fragmentation, on a pilot basis (less than 1% of project costs). (v) study alternative on-farm facilities to serve areas characterized by landholdings of under 2 ha and different cropping intensities (less than 1% of project cost). Thus the project was intended to accelerate socioeconomic development by increasing farm incomes, generating additional farm and agriculture-related employment opportunities, improving farm-to-market roads, all contributing to meeting Government's objectives of increased food production and more balanced regional development. 2.2 NISIP II's design and purpose were identical except that its irrigation component was to cover a much larger area. (i) The twenty six systems to be rehabilitated, upgraded, and extended by NISIP II covered a nominal area of 80,900 ha. Twelve systems were in Southern Luzcn (in the provinces of Quezon, Batangas, Camarines Norte, Albay and Sorsogon); 4 were in Western Visayas (in the provinces of Aklan, Antique and Negros Occidental); and 10 in Mindanao (in the provinces of Zamboanga del Sur, Bukidnon, Davao del Sur, North and South Cotobato). Their improvement was estimated to cost US$71.8 million, again 76% of total project costs. (ii) to reduce the schistosomiasis risk to about 140,000 people on four Mindanao systems with infection rates ranging from 22 to 35% (see Map). (iii) three feasibility studies of substantial future irrigation projects. 2.3 The total estimated cost of NISIP I was US$107.2 million. Bank Loan 1414-PH for US$50 million was approved by the Executive Directors on May 3, 1977, and became effective on August 9, 1977. It was to finance the entire foreign exchange component (47% of total project costs), with the Borrower financing the - 3 - balance. It was anticipated that the project would be completed in about 4 1/2 years (by December 1981) and that the Loan would be fully disbursed by December 31, 1982. 2.4 NISIP II was expected to cost US$138 million. Bank Loan 1526-PH for US$65 million to cover the entire foreign exchange fraction of the project was approved by the Executive Directors on February 28, 1978, and became effective on June 12, 1978. It was estimated that implementation would take 5 1/2 years and that dA.sbursements would be complete by December 31, 1984. 2.5 Most of the 48 schemes slated for rehabilitation, up-grading and extension -- which constitute the bulk of the two projects -- share certain characteristics. Almost all are in high rainfall areas (more than 2000 mm/year) and on short rivers with steep upper catchments and flat, but short lower courses next to the sea. Consequently, there are no good sites for storage dams. Therefore, most of the schemes cannot store water; their irrigation is limited to the run of the river. To give a clearer idea, the average scheme had a theoretical command area of about 2,500 ha, of which, at the projects' starting point, about 1,500 ha were irrigated in the wet season and a little less than 1,000 ha were irrigated in the dry season from stream flow. Rice was overwhelmingly the predominant crop as the projects started and was expected to become more so as small areas of other crops, mainly coconut and sugarcane, were converted to rice. System improvements were e;,pected to allow increased in paddy yields and in area devoted to paddy, particularly in the dry season. Project preparation was thorough and included detailed plans of what was to be done on each scheme. 2.6 There was no provision for an agricultural development component in either project. Farmers were already growing paddy and were expected to understand the production possibilities offered by improved irrigation (or by irrigation for those slated to get it for the first time). The SAR of NISIP I devoted 4 paragraphs and a 5-page Annex to discussion of agricultural supporting services, and the Loan Agreement required the Borrower to prepare and implement an Action Program. The SAR of NISIP II devoted just two pages to a discussion of agricultural supporting services (none of them project specific) which were generally available in the provinces. In both cases, it was assumed that required agricultural support services were already in place; no part of the proceeds of either Loan was allocated for this purpose.' III. IMPLEMENTATION 3.1 Both projects started very slowly because of difficulties in organizing implementation through the regional offices whose staffs were not V Borrower's Department of Agriculture feels that the projects should have included training on farming in improved irrigation systems and on the operation and maintenance of those systems. See Attachment. familiar with foreign-financed project procedures. During implementation, the projects experienced every conceivable delay, from typhoons, floods and earthquakes, through slow engineering design, cumbersome local procurement procedures, problems in right of way acquisition and exceptionally poor performance by major contractors, to non-cooperation of farmers and, as the Borrower's PCR calls it, severe disruptions to peace and order by "unidentified lawless elements." 3.2 All work relating to the irrigation part of the projects and the civil works required for the schistosomiasis programs was implemented by NIA. Health related aspects of these programs were implemented by the Department of Health (DOH). 3.3 The Closing Date of NISIP I was formally extended by three years to the end of 1985, but construction was not complete until June 1986. The final disbursement was made on July 2, 1986 and the undisbursed balance of US$11.8 million was canceled. NISIP II was formally extended to the end of 1986, and construction continued until June 1987. Considering the difficulties encountered and the nature of the projects, implementation performance was rated satisfactory in the Bank's PCR Overview. 3.4 Irritation Development. NISIP I rehabilitated and upgraded 21 existing irrigation systems, including extensions to most of them, and constructed three new systems, to serve a nominal area of about 44,000 (89% of the appraisal estimate). The NISIP II PCR and Overview reported that about 76,300 ha had been developed (94% of the appraisal estimate). The peace and order situation prevented the completion of a new diversion structure for ore major scheme. The service areas of others were rnduced due to water shortages, unsuitable topography and reluctance of land owners with tree crops to convert to rice production. With these modifications, the planned facilities of the schemes were pretty much constructed as per the designs in the preparation reports. 3.5 O&MSupport. Support for strengthening NIA's Regional O&M capability was provided by allowing, In addition to the equipment required for the maintenance of the systems rehabilitated and extended under the project, enough further equipment to maintain the non-project systems in each region, amounting to about 36,500 ha under NISIP I and a similar area under NISIP II. O&M infrastructure including buildings for offices, stores and staff quarters, and other facilities such as water and power were also provided in all regions. 3.6 Schistosomiasis Control. As indicated above, these programs were focussed on Leyte and Mindanao. The main engineering task here was to open up existing drainage ways, increase the intensity of drains in cropped areas, and improve access to facilitate maintenance. Under NISIP I about 170 km of main and secondary drains were constructed or rehabilitated, 140 km of embankments and roads and 240 footbridges were built. Some 15,210 toilet slabs, 7935 toilet bowls and 987 handpumps were also provided. Under NISIP II 88.6 km of main and secondary drains, and 70 footbridges were constructed. The public health - 5 - components included snail control, health education, diagnosis and treatment by the DOH. 3.7 Other components. A survey of farmer response to the proposed land consolidation pilot project of NISIP I met strong resistance and this component was accordingly canceled; the same fate met the on-farm facilities study. Under NISIP II an input-output monitoring program designed to provide bench-mark data and monitor the flow and buildup of farm inputs before and during each cropping season to provide warning of impending shortages was implemented. In addition, feasibility studies were prepared by NIA, with assistance from foreign consultants, of three large multipurpose schemes and a trans-basin diversion scheme. 3.8 The final project cost of NISIP I in dollar terms was US$76 million (Pesos 888.5 million), 71% of the appraisal estimate (US$107.2 million or Pesos 804 million). When the Philippine peso was devalued during the implementation period, local costs did not rise proportionately. Other explanations for the reduction in cost include the reduction in service area and the exclusion of some drainage works related to the schistosomiasis control program. 3.9 NISIP II finally cost US$80.7 million (Pesos 996.8 million), 58% of the appraisal estimate (US$140 million or Pesos 1038 million). Explanations for the cost reductions are the same as for NISIP I. 3.10 Despite overall cost savings, there were some areas of substantial increase.Y1 It was estimated at appraisal that the procurement of equipment for construction and O&M under NISIP I would cost less than US$14 million equivalent. Actual expenditures under this category apparently went to about double this (ca. US$31.2 million). These amounts covered the procurement of 134 pieces of heavy earthmoving equipment, 49 dump trucks, 120 station wagons, pickups and jeeps, and a long list of other equipment including a nine-seater executive aircraft. In NISIP II, the appraisal estimate for equipment procurement by the Government was US$10.2 million, with actual expenditures of US$13.6 million.!' 3.11 Due to the many delays in implementation, disbursements were spread over 10 and 9 years for NISIP I and II respectively, instead of 4 1/2 and 5 1/2 years as originally expected. IV. IMPACT 4.1 The schistosomiasis control programs which were estimated to require 18% of total expenditures, ended up taking considerably less (circa 2.5%) because 11 Direct comparison on a line by line basis is rendered difficult by both different formatting and major arithmetic mistakes in the final sumnary of expenditures (Table 8.1 of NISIP I PCR). i1 Part of the explanation of the cost over-runs for procurement on both projects may be that NIA seems to earn a significant fraction of it's income from renting out equipment to private contractors. See Para. 4.16. -6- of both reduced drain construction and improvements in treatment technology. These programs were however remarkably successful, with prevalence rates in Leyte being reduced from 18Z, at appraisal in 1977, to 11.8% in 1984 and under 5% in 1988 in the worst affected areas, and in Mindanao to below 2.5%. 4.2 Road construction, which represented up to 50% of the civil works element of many of the individual schemes (though less than 5% of total project costs), was clearly of benefit to farmers, many of whom had previously had difficulty getting their produce to market, particularly in those places where crops are harvested during the wet season. The canal-side roads also facilitated system operation and maintenance, and certainly cortributed to stabilizing at least one bank of each major canal. 4.3 The irrigation and drainage infrastructure and 0&M strengthening components, slated to be 75% of total project expenditures but actually much more, had less apparent impact. For NISIP I, the area irrigated simply failed to develop as anticipated (See Table 1 and Fig 1). The SAR shows pre-project irrigated rice areas of 26,200 ha in the wet season and 13,100 ha in the dry season.Y These figures were projected to increase to 49,400 and 35,950 respectively; representing increases of 23,200 ha in the wet season and 22,850 ha in the dry season. In fact, the comprehensive monitoring data available!/ show an increase from 1979 to 1990 of less than 1,000 ha in the wet, and 7,500 in the dry seasons. 4.4 Explanations for this shortfall are complex. One is that the actual pre-project irrigated area was much larger than indicated in the SAR. NIA data from 1978 show a dry-season irrigated area of 20,772 ha, 59% greater than the figure shown by the SAR.Yi V The distinction between 'vet' and 'dry' seasons needs to be treated with great caution. With the availability of non-photosensitive varieties, planting and harvesting occur in every month of the year on most systems. As yields are close to 4.0 tons/ha in all seasons, it is simpler, for purposes of production accounting, to focus on the total area cropped annually. From a water supply point of view, however, one must remember that, even in the wetter regions, there are months when the rainfall is well below the level required to sustain crop evapotranspiration. In many, if not the majority, of systems, the area which can be irrigated in these months is constrained by low streamflow. NIA has excellent data for irrigated areas and crop yields covering the period of the projects and more -- a fact which makes possible the analysis which follows. Z At this point, data consistency problems are encountered because the audit mission was given three sets of figures for irrigated area. The PCR presents "actual" areas irrigated for the period 1979-1986, while the tables reprinted from NIA's monitoring system give "actual" and "benefitted " areas for the period 1983-1990. For the period of overlap, the "actual" figures differ, and there is no data on areas "benefitted" prior to 1983. Under these circumstances the only way to estimate benefitted areas for the years 1979-83 was to reduce the actual areas given in the PCR in the same proportion as the difference between actual and benefitted areas in the period 1984-1990. This has no effect on the economic analysis as there were apparently no project benefits before 1983 (irrigated areas declined), but it does obscure the overview of the first years of the project. - 7 - 4.5 Part of the anticipated increase in wet season area did not develop because "unidentified lawless elements" interrupted the construction of a major new diversion structure for the Abulog Scheme in Cagayan Province. No more than 8,200 of the 14,700 ha of this scheme have ever been recorded as irrigated. Other decreases were attributed to reductions in the area of some schemes because of water shortage and some to the choice of farmers with standing treecrops to maintain those crops and not switch to rice cultivation. Net of Abulog, these reductions amounted to about 2,750 ha (PCR Fig. 7.1). All of these reductions total 9,250 ha. 4.6 Close questioning of responsible NIA officials regarding the accuracy of the irrigated area figures yielded the information that the recorded area is that on which there is a reasonable prospect of collecting Irrigation Service Fees. Apparently there are a number of farmers who, reasonably enough, refuse to be counted as receiving irrigation service because they can not be assured of dry-season water. On this basis, it seems possible that the actual wet season area irrigated (occasionally, and only if needed) may be bigger than that reported. 4.7 With respect to the not yet explained shortfall in dry-season irrigation, the truth is almost certainly a combination of two factors: (a) there is actually less water available in the dry season than anticipated at appraisal (entirely possible because, although the hydrological preparation appears to have been done thoroughly and professionally, there were no long-term streamf low records available for many systems); and (b) the available dry-season flow is being managed with less precision than expected (not just a possibility, this is a virtual certainty). 4.8 NISIP II fares somewhat better under examination. The SAR shows pre-project irrigated rice areas of 31,200 ha in the dry season and 46,800 ha in the wet season, for a total of 78,000 ha irrigated annually. These figures were projected to increase to 55,100 and 74,000 respectively, for a total of 129,100 annually, a combined increase of 51,100 cropped ha. In fact the monitoring data available show (when sugarcane is netted out) a change from about 97,000 ha in 1979 to 115,000 ha in 1990, an increase of about 18,000 ha, which is only 35% of the projected increase of 33,100 ha (Table 2 and Fig. 2). Again, as in NISIP I, the SAR understated the actual irrigated area without project, this time by 20%. A small part of the shortfall -- 4,606 ha -- is explained in Section 3.03.1 (p.21) of the Borrower's PCR as being due to reductions in the official service area of some schemes due to: - perceived water shortages; - changes in land use (urbanization); - reluctance of some farmers to change coconut and sugar land to paddy; - some land being at elevations too high for gravity supply; - exclusion, abandonment of some areas because of the deteriorating -8- sad p3ace and order situation in Region XII (North Cotobsto, Mindanao). 4.9 As in NISIP I, the two primary factors explaining the area shortfall vis-a-vis SAR estimates of some 33,000 ha are: first, again, the SAR understatement of pre-project areas and, second, an overestimation of post-project areas. Both thn Borrower's PCR (Section 10.03.1 - p.142 and Table 10.5 p.168) and NIA's monitoring data, on which the PCR was presumably based, indicate that about 102,000 ha were irrigated under pre-project conditions which, when the sugarcane is deducted, leaves about 96,000 ha of rice. That is 18,000 ha mo:e than the 78,000 on which the SAR and the Feasibility Report are based (Table 7.1, p.44). 4.10 Some of the failure to develop additional wet season irrigated area may be due to the lack of incentive for farmers. Local rice prices declined in real terms from 1975 to 1988, falling by 50% over the period. This reflected the world market, although, in surplus years (1977-83), farm-gate prices were generally below, and in deficit years (most years after 1984) generally above world levels.81 4.11 Operation & Maintenance. Despite the major expenditures for strengthening O&M, current operations seem to be at the most rudimentary level, and maintenance is mostly limited to "vegetation control" (cutting grass on canal banks and around structures) and de-silting. Main canal gates seem to be .operated", i.e. closed, on most systems only when the gate keeper sees a big storm coming and anticipates flood flows which need to be kept out of the canal system. Below this level, operations seem to be limited to either gate operation or, more often, simply the insertion of stop-logs at a few critical points as required to arrange rotation of available supplies between laterals serving rotation units of 30-160 ha. Below this level, several thousand gate-equipped farm turnouts are simply left open, so that water flows through them so that each lateral gets its turn, e.g. on "Tuesday of each week", or "Thursdays and Saturdays" as the case may be. 4.'! In those (apparently few) systems where any regular water measurement at all occurs, this takes place at the diversion structure and/or head of the main canal. The relatively complex procedures required to manage the water, including measurement of flows, combined with rainfall measurement and calculations of water required and supplied to various parts of the system, seem to have been accepted as theoretically correct and desirable, but either impossible or too difficult to implement in practice. On those systems where an effort is made to measure water available or supplied, this is done in the river above the diversion structure and/or at the head of the main canal, where staff gauges have been installed to permit observation of water levels. To convert these gauge readings to estimates of flow quantity in un-lined open channel sections, some of which change shape continuously, it is necessary to go through a calibration exercise which involves measurement of channel dimensions and flow 1' IBRD. Philippines - Irrigated Agriculture Sector Review. In draft. - 9 - velocities. Current meters are required for this purpose but it seems that these are not available for use by the managers of most systems, so that, even if those managers want to know accurately how much water is available for distribution at a particular time, this would not be possible. Although there was considerably more spent on equipment for O&M than expected (para 3.10 above), it seems that no one remembered the current meters!Il 4.13 Actual and current expenditures on O&M are close to the levels projacted as being required for both projects. Interestingly, this seems to be happening because the collection of irrigation service fees has been pursued so effectively that finds for 0&M are coming largely from the farmers rather than from NIA. Upon examination it turns out that the majority of "O&M" expenditures are either for personnel or for the transport costs of moving them around the project area, and much of the time of the personnel is spent in the collection of irrigation service fees. So while a substantial fraction of the salary burden has been shifted from NIA to the farmers, which seems good, NIA staff spend much of their time collecting funds largely required to pay their own salariesl Nevertheless, the farmers seem to be getting some service for which they are willing and able to pay. On one of the four schemes visited, an area which had been turned over to the Irrigators Association for management a year earlier had been turned back to NIA when the farmers concluded that they preferred it that way. 4.14 Whatever the explanat tons for the current very basic level of water management, the facts for NISIP I, as shown by the area and yield monitoring data presented in the PCR, are that the wet season area reported in 1990 was only 26,773 ha, as opposed to 26,200 shown for pre-project conditions in the SAR, and the dry-season irrigated area in 1990 was 21,482 ha versus 17,182 ha monitored in 1979 (and 20,772 monitored in 1978). However, with the pre-project area reduced to allow for the differences between the "actual" and "benefitted'.i data series, and with yields over the period increasing slightly (see Fig.1), production went up by about 50,000 tons, from 130,000 to 180,000 tons. This was of course only one-fourth of the nearly 200,000 ton increase projected in the SAR and 31% of the 162,000 tons projected in the PCR. (Fig.1c). 4.15 For NISIP II, as summarized in Table 2 and illustrated in Fig. 2, rice areas in both seasons have increased marginally, while yields have remained largely unchanged, so that here the increase in paddy production has been about 60,000 tons. This figure must be viewed in relation to the SAR projection of a 280,000 ton increase, and PCR projection of 540,000 tons! 4.16 It would be normal to seek for the explanation of production's having failed to increase as expected in the possibility that there had been a severe constraint on inputs (agricultural chemicals/credit) in the years following 11 See Attachment for Department of Agriculture counent which states that the report's discussion of O&H here (para 4.11-12) and at para 3.5 and 3.10 is not clear. .L See footnote 7 above. - 10 - implementation. Examination of these parameters however suggests that, while performance in this respect has been spotty, the overall supply situation has not changed a great deal. The PCR for NISIP I is a bit evasive and incomplete in this respect, but for NISIP II it is stated clearly that "the majority of the farmers were already using high yielding seed varieties and agro-chemicals", and though there were reports of declining input use in some areas, production loans increased from Pesos 62 and 49 million to Pesos 124 and 121 million in the wet and dry seasons respectively. The input supply situation must have been reasonably good to start off with, otherwise pre-project yields would not have been so high, and this must have been largely sustained, o-herwise they would not have increased to the limited extent they did. 4.17 The treatment of with and without project conditions in the Borrower's PCR for NISIP II is perhaps worthy of comment. Apparently the analysts responsible for preparing this PCR picked up the discrepancy between actual pre-project areas and yields and those shown in the SAR. The only way that the project could then be shown to be a success was to further inflate the estimates of areas and yields to be achieved at full development. Thus, where the SAR presents an increase of 282,600 tons, from 234,000 to 5i6,000 tons/yr, the Borrower's PCR presents an increase of the same 280,000, oriy this time it is from 516,400 to 796,420 tons per year (Section 11.01, p.215). The difficulties of accommidating to these "changes" led to a number of inconsistencies in the PCR. For instance, the total dry-season irrigated area for 1979 shown in Table 10.5 is 47,386 ha while the corresponding figure in Table 11.2a is 31,200 ha. Production during the appraisal period is presented as 516,400 tons on p.215, but only 256,570 tons on p.224. The significance of these discrepancies escaped those who prepared the Bank's PCR Overview. 4.18 When economic returns are recalculated to reflect our current knowledge of what has happened, notable the areas actually irrigated, the yields actually obtained, and the lower-than-expected actual and projected future price of rice, NISIP I's net benefit streamL" is negative in all but a few years (in which it is barely positive). It has no rate of return. Economic returns to NISIP II are 2%. NISIP II is less disastrous principally because the average unit costs incurrd under NISIP I were about 35% higher ($1450 vs $880 per ha) than for NISIP II. 4.19 NISIP I had a far greater relative expenditure and cost over-run for O&M equipment. Although the total service area of NISIP II at 60,000 ha was more than double that of NISIP I at 27,000, the reported total outlay for 0&M U1 Econom±t- returns are for the rice irrigation components of both projects. These were expected to consume about 3/4 of projects' costs but, in fost, accounted for much higher portions. Excluded portions are schistosomiasis control, NIA "O&H" equipment not used for NISIP I & II schemes, and studies. The problem of which "O&M" equipment to attribute to the schemes and which to exclude from the calculation was solved in the sconomic analysis by attributing to the cost stream an O&M charge/hectare of irrigated land being operated and maintained based on actual costs translated into economic costs. See Borrower's PCRs for details. Given any reasonable valuation of the economic and/or private benefits of not having schistosomiasis, the schistosomiasis component had a very high rate of return indeed. - 11 - equipment under NISIP I was more than double that reported for NISIP II -- US$31.2 million vs $13.6 million. 4.20 Is the Audit economic analysis reasonable in comparing irrigation results with the without-project assumption based on maintenance of pre-project status quo? After all, studies made in the course of preparation for the current Irrigation Operations Support Project (Loan 2948-PH ) indicated that declines in irrigated area averaging about 1.5% per year had occurred on 13 National Irrigation Systems similar to those included in NISIP I and II. n-i2 Might the irrigated area of the 46 schemes rehabilitated and upgraded under NISIP I & II have declined in area 1.5%/year i-a the absence of the projects? The Audit asked NIA to see whether any such trend could be discerned by looking at similar schemes not part of NISIP I & II. They could not. 4.21 Nevertheless, despite the absence of evidence, the Audit recalculated the economic returns on the assumption that, without the projects, irrigated areas would decline 1.5%/year for 17 years, holding all other assumptions the same, to see what the effect of this conjecture would be. Under this conjecture, NISIP I would yield an economic return of 1% and NISIP II, of 10%. 4.22 It is often suggested that rice irrigation projects and their designers are being unfairly criticized since they would have been successful if only rice were now worth what it was expected to be worth now when they were designed and appraised. To explore this suggestion, the A-adit recalculated the economic returns at appraisal projections for rice prices, holding all other assumptions the same. Under this counterfactual situation, NISIP I would have yielded an economic return of 6% and NISIP II, of 16%. Thus, NISIP I would still have had an unacceptably low economic rate of return, although not disastrously low, while NISIP II, with its lower costs/ha, would have passed muster. 4.23 These calculations are summarized below: NISIP I NISIP II Economic IRR (i) Current best estimate negative 2% (ii) Assuming irrigated area would have declined 1.5%/year without project 1% 10% (iii) Assuming SAR rice price projections 6% 16% The figures underlying these and associated calculations of net present worth are presented in Tables 6 through 11. (No allowance is made for income from .L 10SP Working Paper No. 1. - 12 - equipment rentals to outside contractors which, at about 20% of Irrigation Service Fees, represent a significant fraction of NIA's income. See Table 12.] V. SUSTAINABILITY 5.1 The irrigation systems of NISIP I and II seem to be working reasonably well. The level of management is low but it could be near optimal given the pravailing circumstances. Water is a less critical constraint here than it is in the arid regions of the world. Many farmers are getting water and paying for it. Neither areas irrigated nor yields seem to be declining. This appears to be an entirely sustainable situation. 5.2 NIA's institutional structure has been built up, under not only these projects, but as part of a broader nation,,l program which NIA has implemented with Bank support over the last 20 years. The success and sustainability of this infrastructure which, inter alia, achieves an uncommonly high degree of self financing -- the collection of Irrigation Service Fees (see para 4.11 above and Table 12) were reported in the PCRs to be running at about 58% of collectibles for both projects -- are worthy of comment. VI. FINDINGS AND ISSUES 6.1 For NISIP I, leaving aside the three new schemes covering 3,500 ha, and the extension to the Abulog Scheme which did not materialize, the project recommended in the SAR was to rehabilitate and extend a series of existing schemes serving less than 30,000 ha, of which about 20,000 were already irrigated in the dry season, to a final service area of about 40,000 ha, of which 30,G00 would be irrigated in the dry season. For NISIP II the project recommended in the SAR was to rehabilitate and extend a series of existing schemes servirg about 55,000 ha, of which about 47,500 were already irrigated in the dry season, to a final service area of about 75,000 ha, of which 55,000 would be irrigated in the dry season. 6.2 The accomplishment of these increases depended, for both projects, on two critical assumptions: (a) that without pi,.v'iding significant additional water e.sUrces the flow originally sserd 1n the dry season coult be caused by imr,-."ed water management, to serve about 50% more a-ea in thal. ason, i.d; (b) that there would bo, a sufficient level of sccial cohesion and organization among farmers that an equitable pattern of all-cating dry-season water between them could be worked out (and implemented in a disciplined fashion by project managers) under circumstances in which only 3/4 of the farmers coule have assured water in any one dcy season. - 13 - The small actual increases in dry-season irrigated area show that the improved level of water management was simply an unrealistic expectation.L/ Even if the projected 50% increase in area covered could have been achieved in the absence of any significant increase in the water ,supply, there is little evidr .ce to suggest that the farmers would have willingly accepted the consequences. The reason we attempt to manage water in any situation is because we are dissatisfied with the natural or current distribution in terms of quantity, time or location. If a scheme with an existing pattern of distributing a fixed quantity of dry- season irrigation water is to be "managed" in a different way, there is a high probability that some farmers will benefit and others will suffer from the resulting redistribution of supply. This is unlikely to be acceptable to the losers except in the most unusual social and political circumstances.'1 6.3 Though there is substantial failure in NISIP I & II, it is not the projects that have failed so much as the preparation and appraisal process. There appear to have been two principal areas of failure: (a) the misrepresentation of low pre-project areas and yields, and (b) the overestimation of possible increases in area irrigated. 6.4 It is difficult to understand how pre-project areas and yields could have been so under-estimated. Monitoring data go back to years well before appraisal. Comparison of pre-project irrigated areas and yields between the Feasibility Reports on which the SARs werd based and the SARs themselves show that the SARs basically accepted the figures presented in the Feasibility Reports.L 6.5 The fact that the feasibility reports lowered figures for areas irrigated and yields is indisputable. Motives are impossible to prove, but what the feasibility report team did improved the apparent returns to the proposed investments. Similarly, the fact that the appraisal missions did not pick up 2' This problem has cropped up in other OED audits, notably of run-of-river, gravity-fed systems on short, steep rivers. See, for instance, Report #7956 of June 30, 1989, Performance Audit Report: Indonesia Irrigation Prolects X. XIV. and XV (Loans 1578 & 1811 and Credit 995). Because rainfall was usually adequate on the wet season, the schemes did not much improve paddy output then; because storage reservoirs are prohibitively costly, schemes were not built to carry water over to the dry season and, therefore, could not much Improve paddy output then either. See also Report #7949 of June 1989, Performance Audit Report: Philippines - Rainfed Agricultural Development (Iloilo) Project (Loan 1815), which argues that an IRRI-inspired rainfed paddy development project was as productive, and at much lower unit costs, as an adjacent irrigation project with the same physical characteristics as most of those under consideration here. The Department of Agriculture would have liked to have seen further examination of how these wrong water-management assumptions came to be made. See Attachment. -W Sanyu Consultants. NISIP Package I Feasibility Report. Quezon City, September 1976; and NISIP Package II Feasibility Report. Quezon City, June 1977. - 14 - these discrepancies is indisputable. Their oversight helped the projects to pass muster in the Bank.-L 6.6 The over-estimation of achievable increases in service areas appears to have been primarily the result of over-optimistic and un-realistic assumptions with respect to the potential for improving water management, as well as of the future behavior of farmers and system managers, and of interactions between them. Some sense of the institutional environment in which the Feasibility Reports were prepared and appraised is given in a recent perceptive IIM Report,- of which the relevant observations are summarized in Annex. -L The Borrower's Department of Agriculture wishes that the audit had gone further in finding the explanation for the use of faulty data. See Attachment. !Zl Murray-Rust, D Hammond and W. Bart Snellen. Performance Assessment Diagnosis. Final Report. Colombo: IIMI, International Institute for Land Reclamation & Improvement, and International Institute for Hydraulic & Environmental Engineering, <1991>. A report prepared for the International Commission on Irrigation and Drainage. - 15 - ANNEX SUMMARY OF OBSERVATIONS FROM IIMI REPORT "PERFORMANCE ASSESSMENT DIAGNOSIS" Total investments in Third World Irrigation now exceed US$15 billion equivalent; most of this has been provided by external funding at lower than commercial interest rates. * While these funds came largely from public sources, their expenditure has benefitted huge numbers of private enterprises - notably contractors and consultants, enabling many of these entities to reach their profit objectives. * Investment in major irrigation facilities is also attractive to financing agencies to the extent it provides the opportunity to disburse large amounts of funds over relatively short time periods, helping to meet their lending targets which are to them an important measure of success. * The national irrigation agencies also welcome new investment in facilities because it brings in new funds and increases their own sphere of influence. * The key that opens the door to all these "benefits" is a favorable outcome to the feasibility study (and appraisal process). * Because the instructions to conduct a feasibility, the study itself, and the interpretation of it's results are all done within a relatively small group of interested parties, there is every reason to expect that the estimates of agricultural benefits will be optimistic, to generate the high apparent ERRs required to cause funds to flow., For the same reason there may be a tendency to understate costs or to reduce them by specifying less costly designs (or to understate pre-project conditions). * In a business environment nothing is gained by setting higher standards than can realistically be achieved; setting unrealistic targets in the planning stage simply leads to wrong investment decisions, which can undermine the company's long term profitability and eventually perhaps even it's survival. * In the case of investments for irrigation development however, there seem to be organizations able to get away with unrealistic planning assumptions. This is because by the time construction is completed most of these organizations' own objectives are already achieved, and their own involvement with the system has come to an end. ~. 1~ FIGURF 1 NISIP l . Fig. 1. - AREAS, Y1ELDS AND PRODUCTION 50 49400 44400 25 26200 27900 a. IRRIGATED ET EAS N P.C.R. RICE AREA 0 AU D I T ('000 Ha.) 13100 DRY\SEA O Appraisal 25 ~ 1500 50 35950 -- 38100_- - 1980 1985 1990 5.0 3.9 4.0 2.5 2 3.75 b.YELD WET SASO N PC.R. (Tons/Ha.) 0.0 A U D I T 25 Appraisal 2.5 6-& 3.751~ 5.- 3. - -4.0 - A 1980 1985 1990 200 191 ---- 178 - 150 -- 100 84 ..101 50 5 WET S*EASON PC.R. 0 A UDIlT SPRODUCTION 50 -- 4-Y S EA N Appraisal ('000 t)FRY ksI\ 100 , 79 150 134 4 - 1980 1985 1990 - 18 - FIGIURE 2 NISIP 1. Fig. 2 - AREAS, YIELDS AND PRODUCTION 75 74 72.5 50 5 - 47---- 22 25 /*I a. IRRIGATED WET SEASON PC.R. RICE AREA 0 A U D I T ('000 Ha.) 2DRY SEASON Appraisai 25.-' 31 50[ 45 55\ 50 75 6.3 1980 1985 1990 5.0 - -. 5.5 - . 4. '- --4.0- 2.5 2.5 --- b. YIELD WET SEASON RC.R. (Tons/Ha.) 0.0 A U D I T 2.5 2.4.0N Apprasal 5.0 - --4-- I - -.-- 6.0 - 1980 1985 1990 400 399 300 296 200 20 2 40 100- - c. PRODUCTION WET C-AS2 ON PC.R. (-000 t) 0_ A U D l T 100 73 -RY SE AO Appralsal 200 180 220 205 300 400 398 _ _ 1980 1985 1990 NISIP I - PRODUCTION SUMMARY TABLE APPRAISAL P.C.R. AUDIT PRODUCTION Season Area Yield Production Area Yield Production Area Yield Production (Ha) (T/Ha) ('000 T) (Ha) (T/Ha) ('000 T) (Ha) (T/Ha) ('000 T) Wet: 49400 3.87 191.10 44409 4.00 177.64 27000 3.75 101.25 AT FUU DEVELOPMENT Dry: 35950 3.73 134.10 38057 3.76 143.09 21000 3.75 78.75 Total 325.20 320.73 180.00 FUTURE WITHOUT Rainfed 23200 1.69 39.27 18239 1.69 30.77 900 2.10 1.89 PROJECT Wet: 26200 2.54 66.46 26200 3.22 84.22 26100 3.75 97.68 Dry: 13100 2.60 34.08 13100 2.60 43.81 14500 3.75 54.38 Total 139.81 156.80 154.14 Increment 185.30 161.93 25.86 PRE Rainfed 23200 1.48 34.3 18239 1.48 26.99 900 1.80 1.62 PROJECT Wet: 26200 2.21 58.00 26200 2.21 57.90 26100 3.25 84.83 Dry: 13100 2.18 28.50 13100 2.16 28.56 14500 3.00 43.50 Total 120.80 113.45 129.95 Increment 204.40 207.28 50.06 NISIP II - PRODUCTION SUMMARY TABLE APPRAISAL P.C.R. AUDIT PAUOUCTION Season Area Yield Production Area Yield Production Area Yield Production (Ha) (T/Ha) ('000 T) (Ha) (T/Ha) ('000 T) (Ha) (T/Ha) ('000 T) Wet: 74000 4.00 296.00 72511 5.50 398.81 60000 4.00 240.00 AT FULL DEVELOPMENT Dry: 55100 4.00 220.40 66270 6.00 397.60 55000 4.00 220.00 Total 516.40 796.41 460.00 FUTURE WITHOUT Ralnfed 27200 1.49 40.6 10249 1.41 14.42 8000 . 2.50 20.00 PRK3JECT Wet: 46800 2.75 128.50 60142 3.82 229.71 52000 4.00 208.00 Dry: 31200 2.76 88.00 46673 3.93 183.64 45000 4.00 180.00 Total 255.10 427.77 408.00 Increment 261.30 368.64 52.00 PRE Rainfed 27200 1.39 37.9 27200 1.49 40.57 8000 1.80 14.40 PFJECT Wet: 46800 2.52 117.80 46800 2.78 129.93 52000 4.00 208.00 Dry: 31200 2.50 70.10 31200 2.76 86.07 45000 4.00 180.00 Total 233.80 256.57 402.40 Increment 282.60 539.84 57.60 NISIP I Irrigated Areas, Yields and Production WESEASO [RYSEASON YEAR Area Avg. Yield Production Area Avg. Yield Production Area Avg. Yield Production (Ha) (T/Ha) (T) (Ha) (T/Ha) (T) (Ha) (T/Ha) (T) APPRAISAL 1979 26200 2.20 57640 14400 2.20 31680 40600 2.20 89320 1980 33900 2.50 84750 21600 2.50 54000 55500 2.50 138750 1981 41700 2.80 116760 29000 2.80 81200 70700 2.80 197960 1982 49400 3.10 153140 35950 3.10 111445 85350 3.10 264585 1983 49400 3.40 167960 35950 3.40 122230 85350 3.40 290190 1984 49400 3.70 182780 35950 3.70 133015 85350 3.70 315795 1985 49400 3.90 192660 35950 3.90 140205 85350 3.90 332835 1986 49400 3.90 192660 35950 3.90 140205 85350 3.90 332865 1987 49400 3.90 192660 35950 3.90 140205 85350 3.90 332865 1988 49400 3.90 192660 35950 3.90 140205 85350 3.90 332865 1989 49400 3.90 192660 35950 3.90 140205 85350 3.90 332865 1990 * 49400 3.90 192860 35950 3.90 140205 85350 3.90 332865 AUDIT 1979 26099 3.27 65344 17181 3.03 52058 43280 3.17 137402 1980 26168 3.37 88186 16856 3.19 53771 43024 3.30 141957 1981 24596 3.60 88546 16003 3.39 67301 41499 3.51 145847 1982 26363 3.44 90689 18352 3.58 65700 44715 3.50 156389 1983 23757 3.59 85288 14411 4.48 64561 38168 3.93 149849 1984 26066 3.77 98269 15771 3.03 47786 41837 3.49 146055 1985 25616 3.42 87607 15278 3.26 49806 40894 3.36 137413 1986 25841 37 87084 20003 3.12 62409 45844 3.26 149494 1987 23215 3.37 78235 18674 3.66 68347 41889 3.50 146581 1988 26045 3.37 87772 16739 3.60 60260 42784 3.46 148032 1989 26826 3.44 92263 20135 3.77 75909 46961 3.58 168172 1990 27927 3.59 100207 21482 3.71 79698 49409 3.64 179905 1. Actuals for years 1979-85 from PCR Table 10.4 for years 1986-90 from NIA Monitoring Data NISIP II. Irrigated Areas, Yields and Production WET SEASCN DRY SEASON YEAR Area Avg. Yield Production Area Avg. Yield Production Area Avg. Yield Production (Ha) (T/Ha) (T) (Ha) (T/Ha) (T) (Ha) (T/Ha) (T) APPRAISAL 1979 .46800 2.52 117930 31200 2.50 78000 78000 2.51 195936 1980 52000 2.75 143000 35000 2.75 96250 87000 2.75 239250 1981 57500 3.00 172500 40000 3.00 120000 97500 3.00 292500 1982 63000 3.25 204750 45000 3.25 146250 108000 3.25 351000 1983 68500 3.50 239750 50000 3.50 175000 118500 3.50 414750 1984 74000 3.75 277500 55000 3.75 206250 129000 3.75 483750 1985 74000 4.00 296000 55100 4.00 220400 129100 4.00 516400 1986 74000 4.00 296000 55100 4.00 220400 129100 4.00 516400 1967 74000 4.00 296000 55100 4.00 220400 129100 4.00 516400 1988 74000 4.00 296000 55100 4.00 220400 129100 4.00 516400 s 1989 74000 4.00 296000 55100 4.00 220400 129100 4.00 516400 1990 74000 4.00 296000 55100 4.00 220400 129100 4.00 516400 AUDIT 1979 52393 3.88 203285 45017 4.11 185019 97410 3.99 388304 1980 49991 4.18 208962 40781 4.08 165569 90772 4.13 374531 1981 50363 4.21 212028 42457 3.98 168980 92820 4.10 381008 1982 48870 3.82 186683 42172 4.07 171642 91042 3.94 358325 1983 54522 4.18 227902 42162 3.91 164853 96684 4.06 392755 1984 54914 3.97 218009 48777 4.03 196571 103691 4.00 414580 1985 54506 3.92 213664 53266 3.95 210401 107772 3.93 424065 1986 60424 3.67 221756 55060 3.60 198216 115484 3.64 419972 1987 59304 3.97 235437 50816 3.95 200723 110120 3.96 436160 1988 58546 3.98 233013 51324 3.98 203243 109870 3.97 436256 1989 59864 3.98 238338 56937 3.96 225471 116821 3.97 463809 CD 1990 58122 4.05 235394 50514 4.07 205592 108636 4.06 440986 /“〕〕！〕、·〕〕〕〕〕〕〕〕〕〕〕〕〕〕〕〕〕〕〕〕〕〕〕〕〕〕〕〕〕〕 【’&&&&&&“華“&&&&&&&“二’&&&&&&! 聖莖奮莖聖呈呈莖呈邑聖邑呈邑邑呈至薑藝邑至至呈莖呈賽覓賽覓認彎 諺實 d •． 心 . ,O■．-- ～■’騙口間口 妒唱， e日，麗。。饑。一留，•觀”矓．t州。。－一•”網戲”- 織屆，。。。。。一•”，•戲”矓綱州。。。一”開•觀”- 一’，閱•。。•一”,－一州州州。。－-.&“觀”- ．儲露個口儲個網•劍開開，個鰓電個個參開唱．館個n纏闢口口劇勵關唱個『馴口奮口•口總 ，健口個魄磚口個啼馴間個個開開電偶奮參開奮個祠騰個口開勵騙馴勵馴開。奮開劍唱個言個開 ，鰓辟馴目仰個個勵網開膩，個開電個參開電O啊開nd開開口個瞭開開唱『口口口電啊曾個口 ,”痲。。購一開，”戲個，，一州一。。－一•『■•戲•。．．•• ，露偶個口仰寫口際開開個n口個瞽電個奮．開奮0網口h口開開口劇．開開個h口臼奮個奮．口口口 ．個偶個開仰個口繪•口口開奮開個電個奮珍開奮綢個牌奮•開口口劇．開開口『口口個奮個唱．召口口 ，露口個目仰寫個冷馴開口，個開奮亡r．磚個•祠臼啊劇開口個個•口口吧啊開闢偶，個個．•口口 ,‘•‘網”一麗，，觀”〝州州織馴一“開。••戲”甲 ，露購個目仰啊個坤開開口口口寫馴電償．開奮綢奮留個d闢口口口勵個開個『口開個奮綢唱申•口口 ．寫購個口婦口個珍開開口州俘徊開電個奮珍開奮朋電口個纏啊口口口瞭劍開個『口開鬍電個奮口開口 ，寫。。併一綢開臘，•“奮•，,&，•響，•，織。’一•開•實馴口個奮個，響•• ．健囉•口仰留個．綢口馴n．個U奮個唱參歸電個實寫口奮d鰓細審口．綢開個州網留奮個奮個口口 ，貿膩個口口視個勵個開啊，啊開電個唱廖開奮口寥審口奮劇開抑電個勵個開口奮口個奮開唱•自口口 ，買露個口露個個勵憫開綢唱潤劍瀾個審開奮中神霹奮朋臘騙口個陣個開個實個陶個奮驪勵個目 ，鰓鰓個寫卹個個險個開馴「•膠U電個奮珍開嚐個胞個唱口總騙個個勵細鼴口h闢盤電憫勵口目 ，霄露個口卹細口廖劍開膩，口口尼個唱，開悶O胞口n劍颼個個口勵個開口n網留電個n勵口瀾口 ，紹偶個臘留•口廖個開觀奮闢個奮口n中開甩個跑自n劇開他個口勵劍開口個個留奮個勵開唱 計鰓口個口卹口口廖劍開膩，口召奮個奮審綱悶•視口奮口越個留劇口開吧個h開他電開言開視 •偶偶他目開個開勵儲口個n•口鰓奮開書開奮唱館個唱個閱口寫中斗開唱也常開口偶電個奮口口口 。必·痲。。。一爍•，“戲”矓州州。。。矚•『•”。。－ •，”。•辭－一•。••縱”矓一州。。。一。『••縱”- 必．。•。。。•．•，闢“，個，。“，•，。。。“．•”&&,&，細• N．。。。。•一•，闢一”矓州州。馴。償。，開戲”- •，一個個匯•個劉瞭個口開，口鰓電麗奮珍開唱個調膠個•開論個網口口目“個奮個電個奮．個口口 &t&，。。•織一戲”一州戲併。。－一觀”豐 面痲必。。。一個，闢觀••一矓一。。。一。，，州個，& ,j健口口目仰審個勵口口口，細爛電個響開奮綢奮留口馴煙口口個口口口羈個開鰓，個曾倘臘騙 合露牌口口•奮個勵網口個h閑留電個『瞥闢悶綱奮個唱權開口口馴審個口。賽開留唱個奮瞭口口口 ．牌偶自團仰口個啼個鰓付，口個奮個，開電•n個實I馴嗡口口個勵憫開寫觀網寫唱個奮珍個劍細 ,&&‘州一編•，闢戲•，m州中鐵•一“&．『••戲“，豐 ．騙必么。一閱•，”戲”戲州州。。。一州“。””• •口館引濾仰口口口“口個奮闢鰓奮個審開電飼奮口，個奮屆口口口d口開口奮寫臼奮開奮p個口l &&&&，織馴••戲”州州州償州“騰•”••綱”甲 必。。。。嚇一•，•一”戲一州。。•一•，&－一戲－t一闢 口•一•綢寫O口口•煙圈口騙奮，•電唱審開奮開奮仰奮劇口臼留個勵個口口零個．電劍奮申開電個奮綢奮馴開讓 玲。•‘戲•麗．們••觀齡”州•‘州”••間”，觀”“州叮鰓 必痲一，騰“一總劇”．•戲”•載啊。。－一實••。”•戲戲－ 馴分開個認，口口目口劊勵d開口•勺口•開電個闕參個，開奮開、個開個•口．目覦．閱口閱電劍奮申個奮綢唱開甩個咸 個，粈編開．個．卹奮，口啊目個，個0奮仰•．．奮留電開翅網酌•細個開綢目個，個0電口『．，電鯽奮，•奮馴開個 中”州•購馴嗚•．『■閑觀“朧網•憫•”•“••婉驪州州齡州 ,．一“一。‘”•，•戲””州州•朧，。•，•載鸛焜州巒卹 必，。•。。。一•••。”－一觀“。•－.，•。－一州州－ 寫•…口騙個個個歸個••『．口鰓奮馴闕勵個電卹細儲個個瞌•口個口個•騙奮個個電個奮飼奮口奮自露奮個開日 開計，個口．開盤•奮邊口開口劊n口個盧粈驪，個奮口奮電n開區卹口口口開．總開個日啊個闖申偶電儲唱個乍口“口 •。一“••“• 卜•綢口戶要鉀•騙個喲鉀•唱口同添扣•纏口戶簪徊•口•畸鉀•口口畸卜豳口口畸嗡－d個j咸開口必鉀屹戶唱網劇勵論細口必唱網口頂唱網綢隨勵仲．綢口戶驕卜•纏口同參妒網口口畸仲．唱開讓喝馴目必唱叫吵煙網劇勵，細目j徊口）唱網個勵方 ..”徊潤。口網。月••個口“，昌”，啊，•唱，’間d“寧矗•開悶，&，口，．間••門區••，，飾細細•同區••，■’，昌，,，〕一’門一一一－_－一一－ 乞讓口閱目口調口O誠細個口騙目開馴悶憫酒••簪．0－■甲州勿婦－••．•唱甲呵州卜－－••響•。－■妒，物畸．－•■•唱甲開悶卜•－ 口開一倡以闔必糅國必作劍劇開．鬥－一一一一一一～一·‘一－ 口開認 圃悶〕叩田O露t觔留1 .1奮）N 寫認勵取用州O歸閱儲以細勵（n閑 寫J鑰以 叩斗Pl A觔馴憎以力潤O地C州闖縫．y劉嗎 A了A馴叮議豳議L劉［C跑們田CES, 怏田 一審粈關馴麒磚劇馮一－．•必‘“•一賽口•闖膠劉膩騙口寫－一一．一開膩 個顱州h細自參．•鬨－闢州州目•－啊個磚．•－J&d鑰騙．--－心向闕必→－也．纏“•口卹劉j閱”•訓劇口跚。j開••．閑悶•‘& &”奮““魚自“邊閑奮細開魚向“h細q開“彎滷購口網必“曲•C州h•，•“‘網．糒細闖中向“•網•習細開么響“戲細口‘••馴U目•網由闖魚•勸‘論•，‘•劉口•響膠個國抑個開口網必個 神總細珍“頃畸作畸吃神開細要個細讓f開網論細奮鰍•卹《•口h網物細口飾．喲喝州口•開參嗡奮勵細丰口b中戶開韋作開州個細華作口n唱，戶•叫唱州嗡h•畸唱目口飾•畸唱目口•••韋《州口夢•吋唱州口訪．吋權劇口計•叫唱目開••畸 州開煙劍口‘劉口“寧纏口 ．口歸開口開口讓跚切閱開，矚閑嗡個口•口目“n個亂•馴口呂坤楓潤U個，抑』唱劍口潤口口個奮潤訪口個．纏口 州口屆潤開口讓細煙口亂開口唱U口細開•調必口個啊蘿亂潤U口二開楓綢屆萬屆么開個口魚開日臼個口總口．遛口 編開寫勾屆口開必個目為口口個．口開州口日口口購為口馴口O口個潤騙禺胞纏口J開口馴口他個個口01細．纏月 奮馴勵口瀾開甲為開細矓開為狗口話總調口朋網口n馴”屆屆U唱離個個月呂勻二個』口開口屆自•口劍開勵－一••。 屆個口開奮魚禺繁勵抑觔為細I胞鰓寫仰開潤屆勾潤口欄奮開勵個禺論購為騙口為U論”」網，自煙•幼。口個t．碼自 •個個．觔個綢亂歸唱」勵為0．個綢奮，.1口開•．個開馴口U唱論口個」勵為•．磁總習口唱口個闢蠶口開口認口個U4& 口綢口認月．月吃鑰徊為么調論臘徊審自口蠹劇•屆開口間為潤．月唱勵磁綢鳥認鳥他婦個徊購自個頃月”劍口一。寫觸必口 t劇口亂O呂卹開口唱開亂開唱個闖騙個劇口朋婦口祕，開昌口亂綢欄口口綢騙馴t口J目個劇際口個闢當O劍才曆口瀾紹纏 頗口唱n月』口戲煙個J購魚綢二倡口寫劇總開召自口，開毛個亂閱•口個．儲為開，韶口奮劇口口個閑各口憫口勵開－ 日馴口幼•呂個魚馴藝瀾讓趙露個口個劇口叫口榆口馴騰馴．為口鳥開楓開各韶．纏個奮劇口朋口月口•口幼論4 唱劍．馴口當總讓口倡閱騙綢個口口個劇洶口馴闢月開口馴閱電月乞編勵開必潤為黝」總口方自口口劇開日倡•開口歸認絕 •開．•論•，•－一矚”網山謝州。購謝鏽．抽．珊州。．繙“湯 頗間唱論為潤馴亂幼個劇口U口口口朋論亂口觔•讓劍磁神鳥開騙‘屆勻個目口馴j二倡」中開口細馴騰 口劇唱個綢魚瀾鳥亂”勻劍口口劉口月口開龍亂口觔口鳥劍纏他開為羈屆個U勵個馴審屆酒論個口自自口自劉觔－ 口劇口弓月呂仰口口』寫常偶口．矚瀾觔開n馴口觔口勵劍煙月為為．‘萬奮劇口口口狗蠶留電馴喝口話話卹 州闖纏勻名勿口口UJ審口口口劉口個騙U勵個口訕自勵綢口肆為潤症山閱口，劇口寫劉瞌口倡他馴唱騙騙認仰 頗閑個抑為”口楓朋常口口屆潤口馴跑朋認馴勵觔勵為綢細唱開亂飼屆“月闖口口劉名蠶劍月魚口個細話－ 州略•掬謂以。”•州•總“”繡榭編•潤聯l•月”‘鑲」惱論．喝”一• 月目陰倡論為”口開」，劇口叫口口口開n亂口呂麗寫綢徑開鳥為論。細奮劇口劇口屆口倡馳劇口口幼認儲 口個口唱論鳥為個個Uj奮劍口口口口目開n馴口開•論潤煙萬為為口必開勵奮自目勵觔論屆個口念口口騙認一常必 自個陰磚禺鳥開日亂幼審劇口倡馴口寫鑰開龍亂口觔口為網楓開鳥認屆綢開鱸奮劇目馴目呂二唱勵馴口口翅認一｀＇• 調•••細以”&．••編開甩嬝細馴•開騙““•”.“•騙“潤拙•寫 調偶纏黝鳥個口楓朋，劇口壇劇口劇口開仰馴口開中為綢楓為鳥寫讓曰潤頸劍目口口壇口個日馴開他論話－ 劍•弓論開口口總朋寫劇口”口口口開馳U目開•么闢蠱論為胞個．潤奮引口口劉口日•口魚口口認認－ 馴口日個潤為瀾日口』．常劇纏昌團口口口“界馴口觔煙•闢他鴉鳥館勵自目騙留劇目口認論口臼口你目口細認• 劉開•嗡鳥抽亂認屆馴Uj，網口屆馴口口口朋跑．口開0鳥N個月開•日個U口奮劇曄屆言ql劍自．•日目細－ 網口倡潤為屆開U寫奮劍濾日馴勵日口．•認馴口開•屆綢煙萬為細口．潤夠口唱屆壇，弱盒0口口朋蠶騰 開開纏為為為口籐朋奮劇鱸自馴勵織觔物鳥口馴口亂貿煙開為為日“月勻劇目口細口口紹唱擊鱸勵論騙－ 開仰纏論為nl口纏寫奮劍口．屆口口觀“界亂口訕0寫口曦磁勻亂話口•月奮劇申鳥口跚蠶劍寫念口1認認• 認編弓潤為口月亂。奮劇口緝口口口．．龍個口訕•讓闕煙肩讓仰屆疾叫開個劇口口潤瀾個韶口取口視幼認－ 觔•他萬鳥為口亂。零劇口鑲口細口開鉤U口開個騙綱個勵為n．煙潤黝口緩口口倡日獸細口幼齧• 魚口0倡幼鳥”口臘抑，自唱騙開I.U話開口幼0呂N徊勵亂界闢開口訕卹口劉口口鄴論口口認祕． 讓個亂總開騙為開開，個劍口．開讓口個開跑開口開口讓綢磁開馮館個必月開歸個劉口口目歸開t認馴－ 劉口挪鄴潤鳥開l馴』寫個網總個調口口馳，I龍馴口勾•個攔口萬為為口•月奮劇團口劉凶唱弱寫你口月認騙－ 讓口j編結呂個口楓朋，劇目網勵奮騙開物跑月馴0為N煙開為話口綢L口細卹口劉」個偶奮馴唱勵騙認• 細．糅屆鳥仰細開』馴零自鱸U口日矓鵝騙他屆觔自方綢煙芻為神I囉瀾勻劍口國開細口個屆．口口幼認• 劉口口攤寫為瀾肩臘朋開口叫口屆矚朋n馴自開自寫攔煙萬鳥話屆偶瀾鉑口口劉讓口口二馴開邑細認曾 寫開口徑騙魚個口個』讓個劍口賺開屆口開跑自•馴個亂闕徑開為話論個闕勵奮劇口口劉目個弱月顱口自認驢偉 細口，磁狗‘添鴉馴口』口，劇jl調口口個開鉤口月訕0,N倡潤魚為自憫開隨開嗡寫論論口劍鑰口口胡細－ 合口口倡黝魚為開U口奮劇唱螺跑自口“n馴口U口呂闢磁萬為n 1．月鉑口口騙涌論倡鈴開口抑話－ 口口．唱細為粕，口』．奮引目，啊、個“界纏牌馴口口綢個洶為露總已月奮個口糾開讓日鵝勵馴口口潤留－ 抽劉d撾鳥一口無”個劇口縲口個驪卹認編總訕•屆綢月綢魚寫l口月奮劍申口目讓口個論口寫認認臼 調目樣開鳥寫口個J露個劇隨口馴口口個朋龍開自U自鳥闕煙萬為跑奮•潤擘劇口月自」總懿壇中開口劉認－ 觀論權論認”口口』j零劇磚仰口口口“跑馴口開煙呂闕磁論開，胞‘月柚口口細」口自口中開口朋認惜 閱．•潤為”編開，細．矚口喝”跑“•訕••“．萬亂”.‘涌”.“。州論•．&&& 馴網唱潤鳥個騙亂絲個綢口網開口個開仰開口開•昌個磁潤為n個個闖開奮個口屆寫論自儲口參開口抑騙－ 網••“•”r喝鬨•網•I.&”••細•“&”鯽•．為”•”•嗚••唱‘-.j -&-．中j 屆闖 . j . 旬I 中I ”細l ‘屆”&& 一露，，總系細馴口綢開煙霄個〕鰓唱臼認開憎審個常審個，網總•，個勵劇開戲，•“電開鸞個總 一麗邸•露卹露煙開劇開g」•斗奮儲，口個牌個口實唱個唱馴開．，口勵綢間個實個個以唱個個合口目 一總一膠口口留口開馴開個「口個鰓奮口，C開，I口tt口，d開開口劇勵個開戲奮，口奮開奮中開曉 •總．•口口露胞口劍開留口•馳電作開憫浮個•作個，口目•露個勵劇間口『口奮他奮個〕自個口 一侃開寧口仰伊個矓開開個劇徐個馴奮個開個牌個口露個，劍咸•露個勵馴間群唱開囑奮憫閑．唱開口 一總一，，鯽萬個颼劇開個『口口U奮綢，啄開俘寫口神口奮個颼開口口勵個口唱個r口l開廈個奮．•開閑 一霄－一，開口口臘綢開個h口口U奮口，露開奮，口實日臼，自颼開個馴勵個開囉『口留鰓奮個蒲．乙口唱 一露．口口仰露個鱸•口口個劉目個開奮個，口目濘細•胞個奮劇口•，個勵口口口勿『口奮鰓唱闢曾，個總 ，實開，寫卹寫個矓劇間個劉爍個豳奮臼閱口鰓口神個，口麗•，劇勵劍開自『總寫綴唱驪奮．．開口 ，露．，口仰審個開劇開自實口開豳奮口貿馴牌奮唱胞個，口目騙寫個勵馴間驪『口IU奮個奮．馴口 一覓，，寫卸寫個口開開個劇口口仰奮臼攤個階口口神個奮d闢個個馴勵開口「奮，U奮個唱．他口口 一健一，露露寫個口劇開劇n讓個U電騙開偶，，仰神個奮』覦口口個．自自口卹『口個臼奮個奮申個奮購 一露．,”露口口向開自n讓口馴悶個奮勰口憎個口實鰓口奮自他露口個勵口個口驪「齣“奮開奮．奮口矓 一麗留j口卹總口口口口唱個雙個盤奮個開個1個口祠個奮口觀口屆網．劍閱。奮個U奮個奮申•開豳 一開，9認”寫個口劍開留口口盤奮個閱個浮個•館個，口闢，審馴勵開離個實口奮豔奮驪雙個個 一偶．個鰓口審口口劍霄劇n口審馴電個奮自口浮奮•祠個，網總寫審個．個開劇州口個臼唱開奮開個 一偶留奮認”審個目綢開•『偶，鰓電個奮唱個坤奮•rt臼，纏開屆個馴勵自口認嘗『口奮盤電曉奮由口總 一開留•口仰寫個他劍開個『口開U電啊奮電個牌個•胞個，4目屆個馴參網開個實開言U奮驪唱．開口 一膩藝個開仰．個膩憫開自『口個U奮個奮C偶審認口n個，d鰓騙個口勵劇開個『開l鰓電鰓奮申開總 昤露騰個•仰口個口偶目•細開鰓鸞口奮唱馴悶個O『奮開，口開留個顱．開朧中『口細閱電個當細口 •颼一屆颼開讓劇．名口開口閃口露醒奮開奮矓名n露O網個奮偶口，•．，名口口叩『口個鰓鸞麗唱．個馴目 -．偶屆口臨仰初個•開目個州口個馳電個唱口個騰口唱蓄電屆奮個囉．1個口閱開蔔『口細U電個贊•編 1．露。•網一•••，，蜿劉開•”州”。。。•••『•■州“,’• 。””一•鋼一個一‘．胛興墾竺州豐竺竺墾豐雙竺 長一，“。。－一，買•觀•留一州州償，，。一•『馴觀”做 ．真•口目露視個啼韶目留奮奮寫奮倡瀾個悶寫個界個響口口，，馴留口口口權實口，劍電憫唱．自開認 l•。t，。。。•視•，．州“實一”•”。。”•••r■•h，憫，.．• ，個應劍開仰個口．口團總仰，，開奮闢觀開奮個啊視，•麗磚I卹寫口偶個『口細開電個唱．馴口寫 一麗，口口開屆口，個個實當，鰓奮口矯轉奮•跑•，勰口論露齣．個勰個『口奮開電勰，b頗留個 購個”個開露露馴分個開魚「口個鰓奮網奮．開響煙實個豔奮馴鱸奮怔劇瞭口開rr口個U電開瀾申么開奮 飩偶一個開婦驪奮．•口閑口『口口鰓奮鰓簪口『寫個實唱•奮網總認電口勵個口啊館奮開開電膩奮．個開開 。個一嗡口仰•個．自開口r口，偶奮口『．個，•個奮on個閱■，綢口開啊個實口寫“電個勺合他開馴 仰麗一口總卹審口口口口團開留，偶電開唱申d口勻『啄口奮個曆口審口口日口開飲個審開電劍闖．奮開個 開鰓奮個審口個劇．，寫鱸個訪目開神電•零口奮勻視飼奮•開口祖口口•劉口個n開個鯽電仲閱．個闔個 館視祝個口”寫．啊煙開開間唱口．留電個馴勵煙奮個寫口個個．個個個•口口騙•細騙電開奮申口電個電仰勺個闖審 屆■驪•飼口口騙馴勵口馴寫個戲勵唱鈴電鰓，儲奮個奮•n網口O奮中勵低口開個n勵個牌電馴奮開唱個奮煙n個閱I 個計飼•個開口口口口個觀個門際騙儲電•『言•個憫奮騙，個開辟．唱口個口口唱綱·個電劍’奮．唱個奮留，．開口 開才飼個唱“，寫O寫個勵口口個rD，開電馴雙口常審個個個奮騙口開鰓個唱口開日奮內•個認電個奮珍戲奮個電口奮馴闖個 口勵．，暱辟t個開口奮劇勵口開憲•貿，憫閱•賽口，劍唱個電開寫．，個勵口開口。謝n開電韶響開奮劍唱開電個開細 •計口口口•劇．•個挪口個個個審，0唱驪會•電留奮閱，馴酌•11個開目開奮勵，0奮驢奮•電口n開蒲劇闔審 顱計綢口口•個唱曰奮中鵝名個勵r「口蒲個電開實馴馮個甩d作個際個，．唱已個繪“留總劍電償實馴審劍〕盧神口開審 口計•一“口，抑細魚口個口懿廈開韶鸞個奮瞭唱個n。憤．際颼奮口■個屆•開個韶奮留神開奮騙，儲奮唱瞬 留計．個奮•個唱•露個口口口口州勵口口電口，口神口個口奮馴唱•留劍口目口辟留•留電個『奮，神口個口電口閱口 奮，．一•口悶口個劇目個•口留奮．闢留個唱，開電償個劍，口纏•奮綢口個．個，.，鰓彎儲奮珍開奮口個紹惜個口， 煙計t“籐•口偶開留d口口團，個「口個屆n開閑勵個鸞．，個口唱麗昌•個d臼口．常颼l名視劇〕》口唱口唱口j個jl .．。“•鰓•自“, 卜•唱開戶，徊•唱口戶屆扣．閱口闕徊．唱口同奮卜•綢口同奮神．唱口畸徊•唱口同言扣．騙開J徊口”唱悶叩煙悶個必個鱸U權悶吵q網個闐卜口唱口戶奮扣•騙口，奮卹•唱個同韋唱開閑開J細口必口悶吵巾q“必煙州必個畸吵權叫個必 口d闖即．”甲州•甲革•皺，••用，由，馴，究，•唱．，•J開甲禹•飼，•學』•“馳唱調悶甚•••妒卸物網悶禺’&“甲禹，闢••O•瞭’閱“d•，為••網••開閑，細場網悶名．&，卹向物網門名．”甲 ,，•膩”口•“，。一－•網•，•個。•．．•。－-，叫卸，-－劍闢•哺同同網十•－•目，•。－→，•向綺•－一馴，甲．，•• 口開一斗d戲鬨開攔閱目園勵一一一■•一一－－■一■名潤面鬨開祖j鳥口，••••”■•口口•■ 織開昌 」刀日代總dlno玆U鼠馴他必VC七卹0劍匯10閱NnO日O 勺雪rw闢V〕顫闖0闢以方汨吃nV ldls州 H認IP題 AUD叮石CO閔劍國心州袖U潔S防 馴心STI刀U江．yo圍化冷細田、 奮j婦 ••．•••，••••••••．一馴膩騙馴跚■••••••■•．•一一騙廳口闐露網膩舉開讓“-－一～一馴口實 •閑纏目細h卜•－一必喊州目卜－煙．必••－纏目目目••→•－司州擊自卜•■勾．唱••煙開司口闔口“個團口目口口口唱開口屆馴悶口膩寫 ．口•奮h闖齣觔“d網•奮細闔煙肉‘h細偽••審目•取．黝目．•《目網‘奮j•自•奮h開．向‘‘•奮h闔．向“h細‘•．奮．．戲“州口•嬝‘•‘,d“口口口奮口。闖寫。．間口網•啊 總馴口畸個細寫『開曰珍州口彎O細論口馴個方e開勵•囀唱•口，•囀細口，馴哺唱叫自計•喲界開彎磁馴畸f歸開方作”彎口頗囀戶總開方e開鰍••參神團計•喲細口，．吋口“口留•哺祝目口計•畸口‘口計••勵編叫口么•磚徑目•口•哺 馴馴閑口跑口讓閑 州，露月口月認為口飼眉為卸馴唱O口開口寫開劉口祖口鼠口U口騙口佩•離••．。盧開口認開口口細月•瀾•纏闢 口網口離自開馴口開口涌口劍寫調口綢口•馴開劉口口萬口•口開口認口n口個讓月•‘•口細馴口韋口屆月•唱騙．纏露 口煩口魚口馴口口劉開馴他口系中細陣馮綢幼屆口馴審馴口細他目口綢膩潤網亂口勰•馴匯婦個•言馴口魚口口開煙•口•．4勾 口馴口開」個亂口認闕亂總為潤馴開潤口跚屆拿開潤申叫口．l騙開口論潤開·開魚潤馴他潤口屆個取開討•口瀾唱．口結．纏口 口馴口亂綢閱購眸個煩戚涌纏劍潤唱口d口常口個口寫口日引口口口亂開細總•么開唱開馴甩自細口習騙口萬細口必口．寫口•嗡• ．開口觔煙鳥開由口I觀開唱個．唱口細目口榆開屆甲屬劉口纏口開矓U目侃開也口•••寫開口個口馴口幼口闢口瀾論嗡口 口馴口亂潤U口屆取口．牌月呂口口屆d口細開口口細開個日口開口已騙開唱口勵口，鼠個為月，■，勾啊口瀾讓劇口個抑•劉口他細開 口馴目鳥口U口蠱婦自劍L口呂闢相勵口寫開馴口屆口徊口個口已開閱唱口勵口飼開為綢劉。•魚開口糾口，馴口口口朋口口潤“．口 屆馴口籐仰為臼亂個開劇騙昌闢。越乏勺寫口讓口劍口二二劇口名飼魚開月劇論開凋陽亂開馴闢朋寫口論口屆口劉口個口．目倡鈴牘 頗•亂潤鳥•讓臼“．臼勵”‘略仰勾•”•鬍坤••“鳥“抽陣”〞為”•．••。口口”，一“需 口劇•個』•當口唱細“」』．魚口個江開口口口參口嗡開〔開口口個魚開呂山臼鯽月勵為開，••蠶編口徊唱馴口口．口矓闢馮自劍 & 日馴陰開勵為口視劇L口唱個自•個自，開•口闖目徊口仰讓皂儲亂總騙個飼』曦也啊，••．瀾口編勵團”馴口個口韶認弱 日綢口開勾歸也鑰劇日個屆口口口馴口纏口l個口日儲觔口口鉤開仰口毛個，•．•目斤口闖口細口仰口U口闢口口 開唱開劇開毛編劇織闢開個口調口纏口•卹已婦魚』•斗匹口仰禹口毛口，■．.．潤口口潤口馴口口口魷屆認• 口啊口開叫哺名開劇勵口韋口口口訕口纏口個口呂口觔口口亂論，黑婦毛開，•滷自馴細日口唱劉口個自開話魚 口網闢朋馴口嗡口．他口自劍開屆劉口引•飼讓唱儲U口口亂口飼月嗡口劇隨個劉口口馴鱸瀾口奮嗡口話． 唱劇目U個．•編•‘••．自飼口口劉闢劇勵個闢倡綢亂認”』口仰」口編口•．.．劉幼朋口劉口奮口一鵝遺 口綢唱．5日鈴嗡口劇論口細潤口口劉口個開口口蠱綢魚認屆訕口，點＇必歸馴口開調馴口朋論劉婦個口他認‘ 口網口，‘•．•細』•』‘.．•潤口叫口個個勵口騙屆綢為幼．••．仰』口毛口，•．.．蠶他U潤劉口奮開開話• 口開開‘,.．邊編‘&..．劉口纏口闖口劇口目屆日總認認，••．仰月毛口，•．•屆口口開寫瀾口個總留話• 江飼口“馴•遺綢劇亂啊，口昌臘•劇歸個•呂口言月騙•佩開毛總，■．•，。個劉胞馴口口口”認•， ，間•”‘•‘•劇．&“•寫．一•‘•么論．‘•．臘•‘”馴•“,.‘•寫‘,&‘一N 劍唱朋屆口磁開劇區購，開口劉瞭自歸個口口編為瀾口蘊口卸』自纏口，•」．．口口劉口瀾口口口仰認．｀』 認唱屆騙總嗡昌劇口月勵劉勵口屆闖勵個勵個甩唱觔奮」屆劇口口月勵亂開屆朧自自讓口唱屆悶，馴勵，留訕話• 細口tt自•自毛開．•盧自劉口奮屆劉閱引歸個歸臨倘亂啊論勵飼』•必涌，..．馴口個開口寫讓，口•認－. 審間唱朋馴開必卹．•．．認開口劉口個口個口蠱口鳥儲，．」•飼」•必開，•••留認”屆劉勵個口一認藝 觔口“唱•自唱購劇他．．寫口口自劉自中開調口已口言綢魚購口』口唱綢，•．•細開朋口劉口奮口O留為 馴開‘整．．唱開．•．自日口口口馴勵頗口調口屆綢魚口，•」•個亂總必綢，•••開認It一劇口個論•認潤 劉口，“勵•嗡口•‘•．．萬口哺口徊個口調口日儲讓綢，．••口開必綢，■••劉口口朋開寫口奮騙•鳥觀 調開U‘開唱開中口膚1勵口口歸幼個口奮開唱綢亂口寫鳥口飼」自毛開劉口屆自開唱開儲馴口個劉•認闢 開•開點歸唱目個唱個審口口已馴口個開調口自儲寫月臘•，臨•遺口馴口綢細闢U一細開奮目•留開 劉口•“.．•細d口』••．•劇開口劉勵頗口個口口口為細個亂口飼』口邊綢，•．．認開朋開調口馴•一昌討 亡．口口朋劇口細口劇臘一劉細口細口嗡口調口日綢讓勵當他口，東儲毛開開巳開畫口口朋騙劉口個瀾•認開 細■自“為•．購．••讓細•，開一，•日閱為個細月麗目‘•馴••“目朋•一煙．&- 魚目呂U已開纏口口個唱購劉口口口劉口引磚憫開已開呂綢臘口仰月口嗡口馴略儲由口口朋開細口個目•跚閱 劉口心朋鳥目遞臼個跑口自口口口細口個口憫口日總名口馳口綢肩唱口調勵口屆口口朋口月口奮••認討 細口口顫勵口唱口自離口目口口口細勵、）憫口自口A鸛寫他口界口唱個劉口口自開開朋•細口奮目•話麗 奮口口開鳥開編口劇曦飾馴開歸目個勵憫口蠱儲讓月認編開』自磁偶馴臨啊馴開朋口馴開勻屆•認討 ，口，朋卹開煙綢榭勵開劍開口念口劇口個開名購亂口鑰口魄目唱臼月口月自日口唱朋個劉口個論•留潤 劉口自U口總嗡個劇勵口開開唱馴口引口調口也開魚口離口目跚唱儲屆臘口個口唱朋臼馴口勻口•留－ 自口，開名鑰編•個他口讓口唱月劉口自口調口二總論屆騙口飼口唱•馴勵口寫田口朋開劉口勻論•認潤 開論朋馴開弓綢劇口口口讓自•磚口嗡口馴勵口月呂潤觔開綢』口4口劉口」自勵口口口馴勵調口，口•留潤 認唱朋勾口嗡開自瞌口口涌口口口觔口嗡開調口已口名綢認啊仰』嗡毛口劉滷月口自口唱開編口口個目一認討 闢認開馴開必口劇勵編馴開口劉口嗡口煩開已個認屆魚•開』口J口馴巳鉤盒口口開儲劉口奮口•話－ 劉開口口馴馴開嗡口自口月口瀾目唱口訕口個口調開已開離月認開仰』自遺總馴目目劍開開騙劉口個口•認開 寫口•馴匐目嗡細自口」口讓口唱自念口嗡”調勵口總論潤論儲口」口遺』口劉口••個口口開•劉口個勵一認細 劉口口“勵•唱啊．•」•細開口馴口引購調口口綢論綢論仰卹』口必口馴他個自口口開口劉口個口一認馴 l曰 l卹 ____Id ■•一編I戶 I也 屆嗡 HlsIPU A壯馴口以刀NO取“CA潤州LlrSIS A蓄州舛馱觔SAL以C日馴必絀臼 奮目口 一開口口馴膩鰓口露－一－■－&－一騙間口閱國曆州闢騙口讓“---－一一綱口常 •••鳥由d‘■一••由自闕j→－徊．邊”•－•纏‘目‘‘→－開•也向卹‘－•．．必••。•遲劍圍目馴”•口．．口。‘開••劍口膩寫 細•奮細“魚向“．網．奮細目心哺網h細輪•，‘開口•細口•勵．網‘唱d•．••奮細開勵甲．細開奮細開偽句目州•煙開•奮撾“納“‘目•豳‘細，j•開開•騰個”寫個•黝個•• 作•細鄙唱騙磚牌闕）作開細參口劉h鄙妒•個韋f騙手．卹但口辭■嚇唱•口鰍•畸細口計•吋作開細抑中畸『口開〕州口細申細馴畸『•口方e開口“畸唱•口••畸磁目口訪•嚇唱．口州•卹口‘口『－方個•口訪•卹唱．口飾•叫仲口，－勵 臼開唱口口口網口 開口個口U口口黝開編口呂•屆口．個口審魚口•馴口寫口•誠口U口口為口露總亂卹劉•」•劍口翻開劉口劉跑O•萬自斕闢 頗口個馴論閱口“總儲目個抑調唱啊個口認開論寫口口口卹開亂口騙•他口個口調曆騙口口細口口細口細口口．日開，纏細 口綢口偶口開口職開網亂口徊口馴曦個口細寫調口馴開屆口寫織•開口盒臘潤為口魷•調開口口開，口魚。細口目•，1,．磚勾 口馴口個叫磚開口騙潤開』口魚潤馴口潤開口細勵馴•團口纏朋』0 IJ口論綢口』購魚網劉•“仰口馴n細讓團口••自歸．纏口 口馴口離•U唱認個，么瀾毛細當唱口物仰瀾開寫口劉口，二馳亂潤曆他口申匹論必個，•＇•”n個開認細劉口個．瀾開•纏• 口口開開OU哺俱01楓寫權口個系口奮劉唱開口劉口魚口口他開U唱口馴口開總他綢，■••勻劉口露口細目劇細寫細寫個細纏口 口劇口日儲鳥開鴉簡口顱L口離婦個自口憫口勵。個瀾開口他總開團口細口卹臘口魚潤寫口」口奮個口•口調口認目言細口開朋． 口開口伯開訕口他啊朋馴開鳥劍個勵個零開口二口口讓開口他口U唱口訕口仰」口呂闢劍．•，劇口寫口馴開細n劉萬個話“也口 網口．一魚闢亂口開唱總為潤。個歸口，劇口細口個寫劉口讓闕口認騙口也開綢月U唱細開屆n口留口劉口細口屆口口•騙口 調口U口日觔為目開細歸么仰個編開憚個口口口口勵抽口唱口亂口日劇口網月魚總劉•．•勻個口細口自網口劉口嗡劇口騙劇口 口網•開口亂悶口月朋劇開騙口個矗寫奮劇團，口個目開口口口為個日0口網口屆仰•••．奮劇口寫口口開口韋口，劇口．劇口 口口口鳥開禺口口臘膚唱口自曉潤劇目口目口口口口口唱口他口寫必口網月唱開，•••奮劇口認口1目魚口日細勵認目口 開甩U馴目唱編個日臼，劇口口口個唱他口目口綢觔•日山開網』口也儲劉•．•個劇目馴口口劉勵開口劉口馳讓口 頗唱開為綢唱臼自勵綢奮劇口口口閱唱口細口唱臼觔•口自開開』勵唱開，•．．奮劇目細口開口馴口胞日認屆口 自開口U唱臼唱•細口個奮劇口馴調名口劉口寫唱口訕口日劉開網綢必開•■．．個劇口開劉口口馴口劉口話馴目 頗開口點磚名臼個勵口憫口拿口口劇開華開已口觔卹口劉口仰』自也口••．．奮口口口唱劉勵開口闖．鵝劇口 口目口U編儲唱購個跑口留劇口馴調個口診闢唱開寫論口訕口仰目毛儲，•盧．奮個口馳口馴目開口讓萬話馴勵 口開唱觔名購唱臼自口臼奮劇口馴開個口口闢名總為勻幼』口田』口唱綢，•••個個口開口劉口馴口屆唱話馴口 口口開Ut開唱購劇勵儲，劇口馴闕劇口，開亂總屆」口論開飼月口跑開，•．．個個目開口屆口馴口劉論認口口 口•”訕•‘•劇隨購儲口闢”一•”編••為”“••月‘■馴．•仰•認•一細．。•。• ：綴：二：::::：蠶蠶：二：:::::：鐺 個唱“馴口唱騙．勵月口馴目馴開劇口日磚煙他綢亂寫••』口開』口也開，•」•，劇口開口調口乞口，認奮認個開 •■ut•‘&.．•，••“闢一•”．••‘瀾•馴．•』．‘潤－，劇•潤”。馴•。。一I ••，“卹•‘•．••，••馴“。•”•。亂••‘•”•編馴．•，．•••。“•觔遼。－ 馴口口朋為目必開劇離口個劇口讓目哺個開細磚口他綢鳥綢屆馴口闕綢嗡開馴巳鑰，．口屆口劉勵細口屆嗡細劊口 細口閱寫口權個劇巳婦憫口馴闢自口，闢唱口亂口論臼田盧•毛開劉■」．寥劇口黝口月口開開馴口認讓口 口開肚訕口禹口劇巳臼個劇口細闢必口細陣口口口魚口日．自自開』口唱綢劉•」•個劇口寫馴歸馴口馴韶調目寫劇口 開口朋勾口唱口個跑綢個個口馴開劇口個叩q唱口勵月認綢鉀』口也購開巳購奮劇口貿唱馴口劍口寫口認馴口 細開朋名開媲口個臨口憫目，口口自口細陣編已開鳥勵鑰儲仰d口遺口馴巳購個劇口為藝馴口魚開口馴口，．寫勵 個•朋昌輪也口個日儲，劇口口口個劇口細磚口名騙讓鉤論騙鯽口唱購劉口月自奮劇口瀾鄴認自觔口馴總細劉口 口開OU為口嗡綢劇勵口，個口細調劇口調”唱唱口魚開認口網』口唱總劉瀘口．甲劇口開唱口開開口馴開舊細論 馴口自觔勾開纏口個巳個奮劇口馴開嗡口劉開細已編認口魚臼口』口唱磚開甩開個劇口騙口調口劍口馴口祕個自 個口屆朋亂開細』口個離口奮個口抑口目個口細口織唱綢鳥口論臼飼』口也儲個臘啊個個口開個劉口細口屆開細口口 口口寫I•屆口毛口劇目臼憫口口口闢個口口n煙唱綢亂騙離儲飼』口毛開開巳編個劇口開祕劉口開口潤口話細口 劉日口I．屆口纏臼個他開憫目魚口口個自劉口口唱卹乞月認口鯽』口曦綢月跑購個劇口開唱潤口開唱開自認亡屆 口口•“勾口瓖口．口口口飼口口口闖個口口”煙已開寫月細口O開』口唱口盒■••仰個觔口口萬劍藝劉開舊寫口 口口口開鳥開也開劇他啊奮劇口，口口州j口磚I口購屆」口鴉馳鯽月遺個馴巳購奮劇口屆矚劉勵細目馴口觔屆口 劉口，”劇口嗡開個離口奮劇口馴闢口口口潤口個讓屆開編開」口編目劇口勵個劇口屆．馴目劍目馴口認劉口 馴圈朋鳥開嗡口個口開奮口口細目閱自口寫n唱‘念皂開為儲開綢必綢馴口勵奮個口寫開馴口馴開調口認劉口 口口，朋屆口唱口自口口，個口，口口劇目•閱日口勵月個鳥輪田月編口馴勵口奮自口綢個馴開劍目調口話細口 開口I．為口毛口．口」口馴口•口闢個口個n咽唱口馳細觔口口』口唱口劍•••個劇口認個馴開言劇目讓口認口口 口口開馴開權口個矓口個個口細口目自口劉磚纏唱綢魚瀾騙口調月口遺口馴口臼奮馴口抽開馴口馴口開口認口口 •細開奮口唱口劇勵口奮劇口馴開個口馴馴由口亂口跑鱸”」口嗡個劇口開個劇口調個開自闕口目口認個日 口口•開劉開細月劇口口奮個口綱開網•留細唱口鳥口觔總飼綢q綢馴口口個劍口細鄴開•馴開開口鸛劉口 馴口口綢馴開唱口個隨總個個口馴闢口口劉開口日儲鳥瀾涌目開』口編悶馴開月，劇口開開馴開細目口馴口認劉口 個口“鳥開必口劇勵口憫目馴開嗡口細口口唱口鳥月論口•飼牌編儲馴朧口奮劍口開口馴口馴開開自認馴口 口口•自』口．- －一－IH ·l, I『 I州 I中 .1卜‘ &，匹0 細騙口Pll AU叫叮日X馴劍國cA袖U.y潤s -NINol劇“[G〞路。～馴n開價”一” 劇嗡． 一憫目開啊勵開口‘－•－一－■一．一一’ •。•。”．亡：一rr~,----.---－網口••，‘••．•闕闢•．．•－••．.. －■→州‘&‘必．•－－•細自“‘•－&.‘』．’一－二‘一～&_～更豐 .．•，&“一．&－一‘一一一。一以＿~_~_＿必二不于二二－一“．••個韶開開口劉．,“•一－ 編亂二石蘇名記·之二騙j乞，竺二竺竺吧二h甲禿竺細＿亡，州‘.--.．一－－一必二一必必’ 一－一－-．一一，、．,.，一，.,.－…痲面州兩痲i痲騙’騙而騙－痲訕森編不鳥病認•高么』亂 編閑一 一口寫一 一一‘。．一一一＿_＿一口開 二澀仕仕豐憤豐竺，,&-.-.．一不－.-.－· 二他亡＝＝豐仕豐豐豐竺墾門‘，一‘痲痲不石二；二，二 騙屆二器豐仕亡仕豐竺押州編編熊－．購•－一必騙；;：二 二，=:=＝仕仕豐豐亡竺竺些‘&．頭必必騙不萬；:：二 二二二仕亡乏仕計豐竺竺巒胛，.－二騙必騙面不斗；:：二 二二：森豐亡豐仕仕豐竺雙變”&“騙“汗不不必；露： 編二江訌需寫二：豐仕豐仕豐徑吋吧狎－－一，．必瀾必騙；言 二二二二【器亡上亡亡仕豐＿甲雙，&&&．一面必必不；荔兀 二二二二】囂亡豐豐仕亡豐壁雙，”。。．-．必必必召；二： 二二亡亡仕仕吃豐竺竺變巒‘k.t．二j面i面汗汗必二•：二 二亡亡開豐他豐竺竺豐巒州t..-．騙涌騙面汗；不：: 二j‘仳亡仕仕仕竺＿豐門‘&&．痲騙而不石不；二 江豐仕仕豐豐巒日竺豐吧整甲‘-－二，一必必日：; 二黔亡豐仕竺豐竺墾州”&．痲騙面不不；:: ．二豐仕竺豐豐竺竺豐巒“,--．一馴一必必；二二 ’二·仳亡二仕仕亡豐墾，,&．一“石汗；；不 二豐亡：仕仕仕仕竺亡墾巒豐門．．一。。痲不；; 二兀亡亡仕亡仕！豐豐豐墾豐門‘,&，。必痲騙石； 不訌亡亡亡亡仕仕竺巒塑州門雙．。－－一。必不；;I 二】他亡亡亡二豐竺豐雙巒件“,-－一必必必；;＿•＿ 二豐亡仕亡豐竺竺巒豐鰓編”-.-－一。必日；二挫 二業亡二仕仕豐竺竺豐”&．痲騙瀾不汗斗；二， 二仳亡仕豐仕三仕巒巒變甲編“顯．-”一。i必；:. ：仳：仕豐！三豐竺豐州”&,.‘面必不；；取‘ 二二亡亡豐豐竺豐變豐中“&&“編－一。。必面；; 二仕亡二仕仕豐豐竺彎”&.‘騙騙不不；二森 二豐亡仕仕亡亡竺竺巒嬰‘&”二馴。。j日；: 二上亡亡豐豐竺竺巒甲”,&.．一面必日不；:: 二豐仕亡豐仕仕竺竺巒雙，－二必騙騙騙不；;:: 二二亡二仕竺竺竺豐雙豐，，一面騙日不；;: 二囂二亡二豐仕仕亡巒豐綺豐“,”一。一屆日石； 二仳二他豐豐三竺竺巒‘編“-.．一。。騙日：: 二仕亡二仕竺竺巒墾押豐“．一騙騙日不；;: 二囂二亡亡亡仕豐仕豐豐莖編“朋．．一。黝面必；二 粩二：二二仕竺亡豐吃整，&．一讓騙必不；;: 二：亡二仕亡仕竺豐豐豐雙整•”.．一。。騙日；: 二豐亡亡仕仕竺竺竺墾雙墾”&,－一騙必不；二： 二豐豐吃豐豐竺竺巒巒州編關“二州。。必日二： 二黑亡亡豐仕竺竺豐豐雙豐“.“必不不；二： 二豐仕．仕豐仕豐竺竺巒豐編“&-．一。必騙必；: 二仳’亡豐仕竺竺竺豐竺墾”，一騙騙不不；:: 二認亡亡豐乏豐仕竺三豐雙巒巒門二，一。必必；; 二二亡豐仕竺豐竺豐變變開，－一爾痲痲必；; －一”一•償一”&,.-..-．一面必；; 口口一魷• ―藝 ·10 ∥“ 二―H e National irrigation AuthdritY -Contribution of revenue sources to total revenue, In percent 1976 1977 .1978 1979 1980 1981 1982 1983 1984 1985 1986 (percent) Revenues Irrigation fees collected 25.2 29.9 24.1 19.46 3 ' 3.9 20.1 17.9 2Z.2 22-1.1 30.2 45.3 Compunal amortization U.S. U.S. n.a.i U.S. n.a. U.S. U.S. t.9 1.6 1.6 2.1 Pump amortization U.S. U.S. -U.S. U.S. 4.4 2.8 1.6 1.6 1.5 1.1 1.4 Equip.ment rental 14.1 14.9 8.3 5.1 9.0 6.3 1.8 1.5 6.5 8.2 9.2 Interest earnings U.S. U.S. 1.1 1.2 k9.8 20.9 20.1 37.2 41.2 46.6 16.2 hanagement fee U.S. U.S. U.S. U.S. U.S. U.S. 32.1 24.8 15.7 7.4 14.1 Other non-operating Income U.S. U.S. 24.9 54.2 10.0 42.3 20.5 4.9 3.0 4.9 6.6 Total direct revenue 39.3 44.8 CM6 80.3 17.1 9ZA tOO.0 100.0 98.4 100.0 91.5 Gov't ON subsidy VIA 49.8 55.2 40.4 1947 22.9 7.6 0 0 0 0 0 Calamity fund to VIA 10.8 0 0 0: 0 0 0 0 1.6 0 2.5 Total subsidies 60.7 55.2 40.4 19:7 22.9 1.6 0 0 1.6 0 2.5 note: Income from pump and communal 4mortitation and mgt fees. In saw year,. were )Loped Into other non-operattng lncome '- - 31 - ATTACHMENT Page I of 3 COMMENTS FROM THE BORROWER (RETYPED VERSION OF THE FAX RECEIVED ON APRIL 7, 1992 FOR READABILITY) Republic of the Philippines DEPARTMENT OF AGRICULTURE Office of the Secretary Elliptical Road, Diliman, Quezon City 03 April 1992 Mr. Graham Donaldson Chief, Agriculture and Human Development Division Operations Evaluation Department International Bank for Reconstruction and Development 1818 H Street, N. W. Washington, D. C. 20433 U. S. A. Dear Mr. Donaldson: Please find attached our comments on the draft Project Performance Audit Report (PPAR) on the National Irrigation Systems Improvement Projects I and II (NISIP I & II). We hope that you will find these comments useful in the preparation of the final report. We look forward to receiving a copy of the final report. Very truly yours, (signed) V. BRUCE J. TOLENTINO Undersecretary Policy and Planning - 32 - ATTACHMENT Page 2 of 3 Comments on the Draft Project Performance Audit Report on the National Irrigation Systems Improvement Projects I & II (NISIP I & II) 1. General Comments The draft Project Performance Audit Report (PPAR) on the National Irrigation Systems Improvement Projects I & II (NISIP I & II) gives a complete overview of the two projects. It includes, among others, discussions on the project objectives, components, implementation, project impact, and sustainability. The topics discussed provides the reader an understanding of the two projects. The PPAR revealed several issues/findings which the project Staff Appraisal Report (SAR) and Project Completion Report (PCR) failed to show for one reason or another. However, it was not able to examine these findings deeper other than mention them in the report. The report should have examined these issues further, the findings of which might be helpful in the packaging and implementation of future projects on the part of borrowers, and in the screening of feasibility studies on the part of the bank. The PPAR should include, among other topics, a discussion on how these apparent defects in the SAR preparation could be corrected, inasmuch as one of the objectives of the audit report is for the bank to learn whatever lessons from previous experiences. Furthermore, the inclusion in the report of a separate section on recommendations is suggested, where these lessons learned from previous projects implemented are discussed in detail. II. Specific Comments 0 The reasons or causes for the use of incorrect data in the Staff Appraisal Report (SAR) should be further examined. The incorrect data set had a lot to do with project approval and apparent failure of NISIP I & II. The audit report maintains that it was not so much as a failure in both projects but more of a failure in project preparation and appraisal. -33 - ATTACHMENT Page 3 of 3 The audit report mentions the availability of "good monitoring data" long before project appraisal. Why did the project appraisal staff choose to use incorrect data despite the availability of another data set? Further explanations are required here. o Another weakness in project preparation was the validity of the assumptions on water management. The assumption was found to be wrong by the audit report. That these assumption passed muster by the bank should have been examined further by the audit team. o The projects apparently did not include any component on extension support to farmers. This component should have included the dissemination of information on the advantages of improved irrigation and the provision of farmer training courses on the operation and maintenance of irrigation systems, including topics on improved water management. It was, however, assumed that: (1) farmers are already aware of the benefits that can be derived from improved irrigation; and (2) that the agency mandated to provide extension services have or will provide extension services that are necessary to go with the infrastructure. Again, this assumption was not challenged by the bank such that no funds were allocated for extension services. At the very least, the project should have included measures for inter-agency provision of services. o On the operations and maintenance (O&M) component of the project, it was noted that actual expenditures on the procurement of equipment were more than the appraisal estimation. In fact, the O&M component of the projects was the only component in which actual expenditures were more than planned. Although this was the case in O&M, some very basic equipment were left out in the procurement process such as current meters which are required In the measurement of how much water is available for distribution at any time. That these actually happened are not clearly discussed in the report. O The section on project sustainability is ambiguous. The audit report must state clearly whether or not the irrigation systems established and rehabilitated under the two projects can be sustained or not, considering that most if not all are dependent on rainfall and run-of-river. IBRD 12507R2 116 12T 12C PH I LlPPINES AIRPORTS NATIONAL IRRIGATION SYSTEMS m -- RO/DS IMPROVEMENT PROJECTS SYSTEMS LOCATIONS RAJLROADS NAmONAL A NISIP I CAPITAL - REGION & NISIP 11 BOUNDARIES Ptv Lanag A INTERNAMONAL Din PREVIOUS IBRD PROJECT AREAS BOUNDARIES A CA R kIORIVER iTION PROJECT Sn lkgon N do b GAT RIVER AUT JRPOSE PROJECT KILOMETERS 0 100 200 300 á | | | | |16- I I III6 TARLAC IRRIGATIONSYST95i i- UPP6R PAMPANGA RIVER PROJECT MILEs 0 50 100 150 200 IMPROVEMENT PROJEC1I locI RORA PENARANDA IRRIGATION PROJECT MAN PHIL IPPINE SEA - pNDORtR an Iv \x DEVELOPM JECT SCUTH CHINA S s SEA N~o- viny JAUR RR ON PROJECTVI PrimC~aa 11 { CHINA l1 c22 Ix f*d " X1 kf PHILIPPINES kAA N1 BELINEI ~ e ff.- d. n l pT or I N 0 A naet ro rcxagn rIo,f r,no, b onJe 12OB12E4199 <=CTOBER 1991 IBRD 2330 121-00' 121' 30' PH ILIPPIN ES LUZON kNATIONAL IRRIGATION SYSTEMS IMPROVEMENT PROJECTS .00. SUB-PROJECTS VISITED BY AUDIT MISSION Main Canals Diversion Dams (Existing) N Proposed Diversion Dams Roads Roilways MIDANAO Rivers Rehobilitoted Area n Extension Area Antipolo BATAAN R l Z A L -1 3 Cavite30' Ci ty o Binongonan• Imflus SiniloonC- Rosorio Poete -n Nric nra Rosa Buena .Santa Cr Vista LA GUNA DE Po sonian 8 A Y C A l T E lombo DMauCAA Lo* PAtLC0,-, ...I A G U N RIVER aucbandy IRRIGATION .ICRyRyway SYSTEM To Nasugbu.Tnun .SnPbo,. R100o>TAo)A Taya s o00' Pogbilao - AGNAS RIVER BAnAusl Tioong YIRR A 3 Or BinubusL iir, lpa Lucena C ndeariaDUMACAA Cuencc RIVER IRRIGATION Toae SYSTEM Rosario HANAGDONG (2,500 ha) A A N .RIVER IRRIGATION At sho'{0hG) 0 0 3 for conmin of the ~ readr an " änpdot fi 10 2.0 3,0 Iof I M uo KILOMETERS my ~f mte kg9s0tu "u of 121*00' 12r30' OCTOBER 1991