Jordan - Northeast Ghor Irrigation and Rural Development Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 3991 PROJECT PERFORMANCE AUDIT REPORT JORDAN NORTH EAST GROR IRRIGATION AND RURAL DEVELOPMENT PROJECT (CREDIT 498-JO) June 21, 1982 FILE COPY 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. ABBREVIATIONS AND ACRONYMS ACC - Agricultural Credit Corporation AMC - Agricultural Marketing Corporation DCA - Development Credit Agreement EGMC - East Ghor Main Canal ERR - Economic Rate of Return FAO - Food and Agriculture Organization IDA - International Development Association JVA - Jordan Valley Authority JVC - Jordan Valley Commission JVFA - Jordan Valley Farmers' Association KfW - Kreditanstalt fur Wiederaufbau - the German foreign aid agency LCA - Liaison and Advisory Committee NRA - Natural Resources Authority ODA - Overseas Development Administration - the British Government's foreign aid agency OPEC - Organization of Petroleum Exporting Countries PCR - Project Completion Report PPAM - Project Performance Audit Memorandum RPM - Reinforced Plastic Mortar pipes SAR - Staff Appraisal Report USAID - United States Agency for International Development FOR OFFICIAL USE ONLY PROJECT PERFORMANCE AUDIT REPORT JORDAN NORTH EAST GHOR IRRIGATION AND RURAL DEVELOPMENT PROJECT (CREDIT 498-JO) TABLE OF CONTENTS Page No. Preface ......................................................... i Basic Data Sheet .............. o............................. . .i i Highlights ................................ ....... . iv PROJECT PERFORMANCE AUDIT MEMORANDUM I. SUMMARY ............................................. 1 II. MAIN ISSUES ........................... 5 A. Project Design ....................... ........ 5 B. Project Scope and Costs ................................ 6 C. The Irrigation System ............................ 6 D. Marketing and Market Development ................ o. 10 E. The Project's Economic Rate of Return .. 11 F. Water Charges; Cost Recovery ..................... 13 Annex 1 - Comments by Jordan Valley Authority (Telex) .. 14 Annex 2 - Comments by National Planning Commission (Telex) ..... 17 Annex 3 - Comments by Jordan Valley Authority (Letter) . 19 Annex 4 - Comments by Organization of Petroleum Exporting Countries ....................................... 26 PROJECT COMPLETION REPORT I. Summary/ *............................................ II. Background .................................. ......... 28 III. Project Formulation ................................ 29 IV. Implementation ................. .. ..................... 30 V. Agricultural Development .............................. 38 VI. Institutional Performance .......................... 42 VII. Economic and Social Impact ............................ .....48 VIII. Bank and Consultant Performance ............................. 53 IX. Recommendations and Lessons Learned .................. 55 Annexes I-VIII 57 Map 1/ Included in the PPAM. 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. os 1 PROJECT PERFORMANCE AUDIT REPORT JORDAN NORTH EAST GROR IRRIGATION AND RURAL DEVELOPMENT PROJECT (CREDIT 498-JO) PREFACE This is a performance audit of the North-East Ghor Irrigation and Rural Development Project in the Hashemite Kingdom of Jordan, for which Credit 498-JO was approved on July 2, 1974 in the amount of US$7.5 million equiva- lent. The final disbursement in respect of this loan was made in August 1980. The audit report consists of an audit memorandum prepared by the Operations Evaluation Department (OED) and the Project Completion Report (PCR) dated June 26, 1981. The PCR was prepared by the Europe, Middle East and North Africa Regional Office. The audit memorandum is based on a review of the Appraisal Report (339a-JO) dated May 10, 1974, the President's Report (P1403-A) of May 24, 1974 and the Development Credit Agreement (DCA) No. 498-JO dated July 25, 1974. The draft Staff Appraisal Report (SAR) for a proposed follow-on project has also been consulted for pertinent information. Internal Bank memoranda as contained in project files have been consulted and Bank staff associated with the project were interviewed. An OED mission visited Jordan in November/December 1981. The mission had discussions with officials of the project executing agency, the Jordan Valley Authority (JVA), senior research, extension and marketing officers of the Ministry of Agriculture, the dean of the Faculty of Agricul- ture at Amman University and the Jordan Valley Farmers' Association (JVFA). Several project area farmers were visited. The information obtained during the mission was used to test the validity of the conclusions of the PCR. A draft report was sent to the Government of Jordan and to the Organization of Petroleum Exporting Countries (OPEC) who have co-financed the project. The comments received from the Government are included in this report as Annexes 1, 2 and 3. OPEC's comments are in Annex 4. The audit finds that the PCR covers adequately the project's salient features and generally agrees with its conclusions. In addition to summariz- ing the objectives and achievements of the projects, the PPAM reviews irriga- tion development in one of the areas included in this project (the "Conversion Areas"). It concludes that because farmers are changing over rather slowly to using facilities provided by the project, and the necessarily speculative nature of forecasting not only the rate, but also the method whereby farmers will use the new facilities, it would be premature to estimate an economic rate of return to this component. Since however the conversion has major implications for development strategies in the whole of the Jordan Valley, - 11 - progress should be closely monitored not only by JVA but also by potential financing agencies. The audit also expands on the need for a market research and development component. The valuable assistance provided by the Government of the Hashemite Kingdom of Jordan, the Jordan Valley Authority and the farmers and officials met during the mission is gratefully acknowledged. - iii - PROJECT PERFORMANCE AUDIT REPORT JORDAN NORTH EAST GHOR IRRIGATION AND RURAL DEVELOPMENT PROJECT (CREDIT 498-JO) BASIC DATA SHEET KEY PROJECT DATA Appraisal Actual or Actual as % of Estimate Estimated Actual Appraisal Estimate Project Costs (US$ million) 17.4 40.3 232./ Credit Amount (US$ million) 7.5 7.5 Date Board Approval ----------------------------07/02/1974----------------------- Date Effectiveness 10/28/74 01/08/75 182 Date Physical Components Completed 03/31/77 03/79/b -- Proportion then completed (%) -- -- 60 Closing Date 06/30/79 06/30/80 36 Economic Rate of Return (%) 29 15.LS. 51 Financial Rate of Return (%) NA NA NA Institutional Performance Slightly below expectations Agronomic Performance benefits are slow to accure Number of Direct Beneficiaries (year 12) 25,000 32,300 129 CUMULATIVE DISBURSEMENTS FY75 FY76 FY77 FY78 FY79 FY80 FY81 Appraisal estimate (US$ million) 1.4 4.6 6.9 7.4 7.5 Actual (US$ million) 0.1 1.0 4.1 5.5 6.6 7.5 Actual as % of estimate 2 14 55 73 -- -- Date of final disbursement ------------------- August 1980 ------------------- Principal repaid to (mo./day/yr.) (US$ million) August 31, 1981: NIL MISSION DATA Date No. of Mandays Specializations Performance Types of Mission (mo./Yr.) Persons in Field Represented/d Rtn/e Trend-/ Problems/A Identification 09/72 2 10 ad -- -- -- Preparation 03/73 5 113 abedd -- -- -- Appraisal 09/73 4 100 abcd -- -- -- Subtotal 11 223 Supervision 1 03/75 2 10 b d 3 3 F M Supervision 2 08/75 3 15 a c d 2 1 F M T Supervision 3 02/76 3 13 a c d 2 2 F M T Supervision 4 10/76 3 21 a c d 2 1 F M Supervision 5 04/77 3 15 a c d 2 3 F M Supervision 6 11/77 3 20 a c d 2 1 M F Supervision 7 03/76 1 8 b 2 2 M Supervision 8 11/78 1 3 d 2 3 M Supervision 9 03/79 1 5 d 2 2 M Supervision 10/h 04/80 2 7 a b 2 1 M Subtotal 117 TOTAL 340 OTHER PROJECT DATA Borrower: Hashemite Kingdom of Jordan Executing Agency: Jordan Valley Authority Fiscal Year: January 1 - December 31 Name of Currency (abbreviation) JD (Jordanian Dinar) Currency Exchange Rate: Appraisal Year Average US$ 1.00 = 0.32 Intervening Years Average US$ 1.00 = 0.32 Completion Year Average US$ 1.00 = 0.30 Follow-on Project: Name: Jordan Valley Irrigation II Project Loan/Credit Number Loan/Credit Amount (US$ million) ) Appraised but not processed Date Board Approval /a Taking cost of all components from project description. Cost of irrigation component for which IDA disbursed, rose from US$6 M to US$13.8 M equivalent. /b Excluding on-farm development. 7 The PCR estimates ERR to be 13%; the audit, 15%. /d a = agriculturist; b = agricultural economist; c = economist; d = irrigation engineer. /e 1 = problem-free or minor problems; 2 = moderate problems; and 3 = major problems. f I = improving; 2 = stationary; and 3 = deteriorating. /g F = financial; M = managerial; T = technical; P = political; and 0 = other. /h Preparation for PCR.  - iv - PROJECT PERFORMANCE AUDIT REPORT JORDAN NORTH EAST GHOR IRRIGATION AND RURAL DEVELOPMENT PROJECT (CREDIT 498-JO) HIGHLIGHTS The project was designed to assist with the Stage I rehabilitation and betterment of the east bank of the Jordan river. Because of its mild climate, the area is a natural venue for out-of-season vegetable production, which has been making rapid strides since the early 1960s. But limited water supplies have always restricted the extent of development. To overcome this constraint the irrigation component of this project was designed to tap waters of three small tributaries of the Jordan river. The pressurized pipe cum sprinkler system was expected to irrigate some 1700 ha, of which previously only 720 ha were intermittently irrigated. A further 1000 ha, previously surface irrigated from a canal, would also be converted to the same system. Other components of the project were credit for on-farm development, the provision of facilities for produce grading and packing, assisting agricul- tural research and extension and providing housing, schools, public buildings and roads. The project cost nearly 2-1/2 times as much as estimated at apprai- sal which made it necessary to involve other donor agencies besides the Bank Group: USAID, OPEC, ODA, KfW and Dutch technical assistance finally partici- pated. Much of the cost increase was due to rapid inflation after 1973-74. However, all the planned works were completed. On the economic side, the results are as yet not final. The bene- fits on land not previously irrigated are likely to be good: the economic rate of return for this component is now estimated to be 16%. However, on previously irrigated land project benefits are not yet readily apparent. Their full realization will be contingent on further developments over the next few years. The potential is undoubtedly great but its exploitation depends on many technical, social and commercial factors. The overall ERR has been recalculated at 13% compared with an appraisal estimate of 29%. Other points of interest are: - the universal adoption of sprinkler irrigation for vegetable produc- tion requires much discretion because many species are susceptible to leaf diseases (PPAM para. 13 and PCR para. 5.09); - conversion to an expensive buried piped water distribution system to replace a functioning surface system will only pay if rapidly accepted by farmers and by an equally quick change of cropping patterns to better exploit the flexibility and water savings in- herent in the system (PPAM paras. 13-17 and PCR paras. 5.01-5.13); - v - - there is an overwhelming need to support large-scale production of perishable commodities (vegetables in particular) with a well- organized marketing system, including not only provisions for grading and packing but also market research and development (PPAM paras. 18-20). - 1 - PROJECT PERFORMANCE AUDIT MEMORANDUM JORDAN NORTH EAST GHOR IRRIGATION AND RURAL DEVELOPMENT PROJECT (CREDIT 498-JO) I. SUMMARYI! 1. The project area is in the Jordan Valley, below sea level where temperatures correspond to those in areas several degrees of latitude nearer the Equator. This makes it a natural venue for exotic and out-of-season fruit and vegetable production. The Valley's production provides some 70% of Jordan's vegetable requirements; surplus from the Valley is sold to neighbor- ing countries and, increasingly, to the Gulf states. The change-over from rainfed to irrigated farming started in the early 1960s when the first phase of the East Ghor Main Canal (EGMC) was commissioned: this canal runs along the entire eastern bank of the Jordan river. It receives its water supply from the Yarmouk river, the main tributary of the Jordan, and from some smaller side wadis (seasonal streams) and rivers. Development was rapid until 1967 but the war, and civil disturbances that lasted until 1971, caused the farmers to leave the area; also, irrigation and other structures sustained considerable damage. In 1971 the Government formulated a two-stage rehabili- tation and development program. Financing of the Stage II development still awaits outside creditors, who, in their turn, await the resolution of inter- national water rights and other riparian issues. The Stage I development was divided into four projects. USAID supported two; KfW, the German foreign aid agency, and IDA have each financed one. 2. The IDA project was designed to feed a buried pipe distribution system from waters of three side wadis, pressurized by gravity.2/ These side wadis have previously irrigated about 720 ha of land; when in spate, the surplus water escaped into the Jordan river. This system was to provide irrigation to 1760 ha above the EGMC (including the 720 ha it previously supplied); the area is out of the EGMC's command. The area is referred to as the "New Lands". Water use efficiencies would be increased by converting from the traditional zig-zag type surface irrigation to more efficient sprinkler irrigation on about 1000 ha, the so-called "Conversion Areas". In both areas the pressurized pipes would serve hydrants placed at the highest point of the farms; the plan was to connect them to a portable sprinkler system, which would be purchased by the farmer with credit funds partly provided by the project. Part of the land where surface irrigation would continue (about 4940 1/ Adapted from the PCR. 2/ One of the three side wadis, the Wadi Arab, later became the venue for a project, supported by Japan, to construct a storage dam. Water for the IDA-supported project would still be supplied, but under different arrangements and structures (see para. 3 (a) (i) below). - 2 - ha) would be levelled to increase water use efficiencies; tile drainage would be provided where necessary to control waterlogging and salinity. Other components would provide physical facilities for produce grading, packing and storage; assist agricultural extension and research; provide housing, school and public health facilities, domestic water supply and community centers; and farm roads, to encourage the inhabitants' return to the valley. 3. It soon became clear that total project costs would exceed the estimated US$17.3 million equivalent by a considerable margin. Completion of all the components was seen to cost in excess of US$40 million; the modified irrigation component alone was about US$12.4 million against an original estimate of US$5.5 million. Therefore, other donors have been invited to participate; some of the components assisted by IDA were also slightly modi- fied. Eventually the following developed: (a) the IDA funds, augmented by US$1.6 m equivalent supplied by the OPEC Special Fund and administered by IDA, were used for: (i) the construction of two diversion weirs on side wadis;l/ (ii) the primary and secondary trunk pipelines for the irrigation system; (iii) engineering consultancy services; (iv) equipment and material for operation and maintenance and for the agricultural extension and research services; and (v) seasonal short- and medium-term agricultural credit, much reduced from original estimates. (b) the USAID incorporated into its projects the sprinklers required in the IDA project area as well as in the section undertaken by them; (c) the KfW took on the housing, school building and farm road com- ponent; (d) the United Kingdom's ODA also joined in the development and financed one of the markets and grading facilities. 4. Modifications during execution included the abandoning of the land levelling component, which, on further investigations, was no longer con- sidered necessary since in most cases the zig-zag system would be converted to 1/ The third would have been the Wadi Arab (see footnote to para 2). - 3 - furrow irrigation and not level basins or borders. The tile drainage compo- nent was much delayed as the need for it turned out to be less urgent than expected, and was in fact started only after the project under review was comple ed.1/ Further augmentation of water supplies from one of the side wadis2' was made part of a dam project, supported by Japan. 5. The project is yet to achieve all its objectives. Irrigation of formerly rainfed lands is undoubtedly a considerable benefit. However, its realization is delayed by the slow rate of reallocating the newly irrigable land amongst the many claimants.-3 Further, the replacement of surface irrigation by a pressurized system, and therefore water savings, has taken place only on some 15% of the conversion lands (May 1981 data). Many of the farmers who had converted by then did so in citrus orchards and have installed a fixed sprinkler system. However, the audit was advised by JVA engineers in November 1981 that an increasing number of the larger, more prosperous farmers are beginning to utilize the system, either through portable sprinklers or, by installing pressure relief valves, through drip pipes. The emphasis on sprinklers may have been the wrong technology for most of the crops grown at present (see para 13 below). The agricultural credit provided by the project was duly taken up: repayment rates are an acceptable 80-85% and are further improving. The agricultural research and extension services have not been providing the backup to farmers as expected; however, commercial firms selling agricultural chemicals and seeds are advising their customers on the best use of their products. It must be accepted that the farmers would set a high standard for any advisory service that could meaningfully assist them: they are already practising high-technology, sophisticated production methods. 6. In the PCR, economic rates of return have been separately calcu- lated for the new lands and the conversion lands. They conclude that for the former the ERR is a satisfactory 16%. However, in case of the conversion lands, the PCR, in the audit's view, may have underestimated the benefits of the project. The PCR estimates the ERR on this component to be 5%; the audit estimates it to be around 10%. In addition, the project has set the stage for considerable further agro-technological improvements that would be more difficult without having converted the open system to a closed one. The issue is further discussed in paras 21 to 25. 7. A notable feature of the project is the building up of the executing agency, the Jordan Valley Authority (JVA) (until 1977 known as the Jordan Valley Commission), which is now charged with the formulation, execution and subsequent operation of development programs in the entire Jordan valley. During project implementation, and despite the considerable administrative 1/ Progress has recently been quite significant (see Annex 3, section 6). 2/ The Wadi Arab (see footnote to para 2). 3/ On the reasons for delay, see Annex 3, section 5. burden placed on it, the authority has risen to its task as an engineering and administrative organization and has, in recent years, performed well. However, it was not able to persuade the agricultural extension and research services to develop technical packages which would further improve the farmers' already high production standards. It may well be argued that it would have been better to try and improve the Ministry of Agriculture's management of its extension and adaptive research services as part of the project.1/ Also, especially in matters pertaining to crop agronomy and production techniques, it is the audit's opinion that Bank supervision missions could, with advan- tage, have striven harder to involve the Faculty of Agriculture at the Uni- versity of Amman. Although a comparatively recent institution, the Faculty makes an excellent impression, demonstrates a keen awareness of farmers' problems and could have made a significant contribution towards developing agronomic packages that would benefit the farmers in the valley.2/ 8. One problem that has arisen during the early operation stage relates to the material used for pressure pipes: reinforced plastic mortar (RPM) instead of a more conventional material. This relatively new material was introduced as a potential cost saver. During the two-year "running-in" period while the contractors were still operating the system, failures were frequent, aggravated by "water hammer" due to intermittent operation, causing interrup- tion of supplies and making costly repairs necessary. At present, however, in the IDA-supported project area, only one failure has been reported in the whole of 1981, although the adjoining system at Zarqa, operating at a higher pressure, is said to average one failure every 22 days. The JVA believes that by now all pipes that were inadvertently damaged during construction have been replaced. They are encouraged by the more recent results, the reduction of 1/ The JVA point out (Annex 1) that "the project was approved.. .in 1974 at the time of the Jordan Valley Commission.... The role of the Commission was to end as each project in the Valley development was completed. All projects were to be handed over to the relevant Government Agency for operation and maintenance.... Equipment, supplies and materials [for agricultural extension and research] were purchased...and handed over to the Ministry of Agriculture. ..in 1975." The JVA's involvement in agri- cultural extension and research therefore dates from 1977 only, when it was established under Law No. 18 (PCR para. 6.02). It therefore remains an open, unanswerable question whether JVA would still have been given the task of operating the agricultural extension and adaptive research service if the Ministry of Agriculture's operation had been technically supported under the project, as the audit suggests. 2/ JVA (Annex 1) comment on this issue as follows: "...There had been attempts by the JVA, the Ministry of Agriculture and the Faculty of Agriculture to coordinate the roles of each in agricultural research and a task committee was formed to define roles and relationships. These attempts were triggered by a project sponsored by USAID. However the momentum of this cooperation was hampered by frequent changes in per- sonnel and lack of enthusiasm, and the project was ended in 1982 without much results." - 5 - the "water hammer" through more continuous operation and no longer consider the RPM pipes a problem in the IDA project area.-/ II. MAIN ISSUES A. Project Design 9. Partly with the benefit of hindsight, a legitimate question to raise is whether the high capital expense of installing a closed irrigation system in the conversion lands (as noted, already served at the time of appraisal by quite a good surface distribution system) was justified (PCR paras 5.12 and 7.09). At the time of project appraisal, it was estimated that without the project, yields would increase by about 2% per annum, whereas the installation of a closed system would considerably accelerate this increase. This somewhat pessimistic without-project assumption must be viewed in the light of con- temporary events: after the 1967 war and the disturbances in 1970 and 1971, the valley was deserted, much of its infrastructure damaged; Bank staff who visited the area at the time informed the audit that they had every reason to be cautious in estimating the rate of unaided recovery and advancement. In the event, yields everywhere increased far more rapidly than expected; the low rate of utilization to date of the closed system in the conversion lands gives no definite indication that, with it, yields would have increased even more. 10. On theoretical grounds, the advantages of a closed distribution system are reduction or almost complete elimination of conveyance losses, a very sizeable reduction of operational losses between the main canal and the farm boundary, and also the greater ease of operating a drip or sprinkler field irrigation system from a farm hydrant rather than from a reservoir constructed on the farm and using pumps to feed the sprinklers. The sprinkler or drip system gives greater flexibility to grow expensive, high-technology crops - the growing of which, however, depends partly on a well-organized marketing system. Sections C and D of the audit further discuss the issue. The issue of determining which is, or would have been, the right project design, finally comes down to the rate and method of adoption assumed for the full exploitation of the closed system. This is necessarily speculative; experience to date is anything but encouraging. The JVA, in discussions with the audit mission, stated that they have evidence of a breakthrough in farmer attitude and acceptance and further stated that they believe that during the last few months the rate of conversion to sprinkler and drip irrigation is much faster in the conversion area than outside it. Obviously, events over the next few years will have to bear out this impression. 1/ The audit understands that RPM pipes, of the specifications used in this and the Zarqa project, are no longer manufactured. Also, that based on the U.S. precedents, JVA's optimism may prove premature and that fail- ures will continue due to deterioration of the pipes caused by inter- laminar separation of the glass fibre in a pipe subjected to excessive deformation. -6- B. Project Scope and Costs 11. The project was appraised in September 1973. Most projects appraised at that time have encountered considerable cost overruns; this one was no exception. However, the audit wishes to draw attention to the way the Bank chose to cope with expected cost overruns. 12. When it became apparent that the costs of the project would exceed estimates and that the Government was unable to meet the increased expendi- tures, the Bank encouraged d'scussions with other external aid agencies to obtain additional financing.V Eventually USAID, ODA of the United Kingdom and KfW of West Germany took over some of the components and the OPEC Special Action Fund also lent US$1.6 m (para 3). However, the Bank did not endeavor to maintain formal control of the project by inviting the other donors to co-finance rather than to take over some project components in their entirety, i.e. parallel finance; it only tried it - and failed - with USAID. The exception was the OPEC Special Fund which agreed to IDA being the execut- ing agency; a formal agreement was duly concluded. However, by then, IDA decided to supervise closely only the components listed in para 3 (a) above and reviewed progress on the other components less often and in less detail. The credit agreement was not amended to reflect this and therefore it may be concluded that IDA was in technical breach of the DCA. The JVA states, and the audit has no reason to disagree, that execution of the project, as appraised, did not suffer as a result of IDA's partial withdrawal. However, similar violations in other countries could affect project implementation and would also be a partial abrogation of the Bank's role as a development agency upon which borrower countries rely for impartial technical expertise as much as - possibly in some cases even more than - a source of concessional finance. C. The Irrigation System 13. On-farm water distribution. The project has aimed to convert sur- face irrigation in the "conversion" lands and provide a supply in the "new" lands specifically designed to take sprinklers. This would make it easier for farmers to reduce on-farm water losses compared to surface irrigation and would, if adopted on a large scale over the valley in the future, enable expanding the cropped area and/or cropping intensities. However, there is no indication that consideration was given to the well-known fact that sprinkler 1/ On this subject JVA comment as follows (Annex 1): "...The Government was never unwilling or unable to meet the increased expenditures and was committed to meet them either from its own sources or other sources.. .The costs of domestic water supplies and of the farm roads were met by the Government of Jordan alone. We were not aware that the Bank encouraged discussions with other lenders. Such discussions were initiated by the Government of Jordan and only the approval of the Bank was sought...." -7- irrigation encourages leaf diseases, especially in tomatoes and cucurbits,1/2/ currently the two most commonly grown annual vegetable crops in the valley. Therefore farmers' reluctance to invest in sprinklers is readily understand- able. On the other hand, the increasing use of plastic tunnels and, to a lesser extent, plastic greenhouses favor the use of a closed on-farm irriga- tion system. In areas not served by a closed distribution system farmers construct ponds fed by the farm watercourses from which they operate either sprinklers (when crops responding well to this system of irrigation predomi- nate) or install drip pipes. 14. The main water reticulation system. The issue whether the open channels now conveying water from the EGMC to farms in the "conversion" area should be replaced by buried pipes - or, in the context of this project - whether the water of the side wadis, tapped by this project and piped to the "conversion" lands should have been discharged into EGMC to augment its supplies and taken to farms in the existing open network, suitably improved, was discussed in Section A. Apart from the issues raised there, principally hinging on the time frame assumed for acceptance by the majority of the farmers, of a closed on-farm irrigation system (para 10), consideration must be given to the all-important aspect of water saving; water is by far the major limiting factor to the development of the Jordan valley. 15. Determining water losses in open conveyance systems of the kind used in the Jordan Valley: secondary and tertiary canals lined with concrete slabs and unlined channels between the farmers' outlet and the farm, is usually controversial. In principle, and if the system is well constructed and well maintained, losses due to seepage are moderate in the lined channels: about 30 1/m2/24 h. While substantially higher per unit area in unlined field channels, these are relatively short (in the Jordan Valley, typically about 100 m) and therefore, in sum, are not of major importance. The imponderables are the operating losses due to inattention, operator error, physical damage (accidental or deliberate) to channels and structures, or water escapes simply caused by the fact that farmers are not able to use the water when their turn comes, for temporary but unavoidable reasons. These operating losses are highly variable: measurements made in one year may be completely misleading in another. Therefore, at best, an informed estimate has to be made. Based on experience elsewhere, bearing in mind the staffing and efficiency of the JVA organization and making allowances for the relatively short (2.5-3 km average) distance of farms to the EGMC, the audit estimates the loss to be 1/ Generic term for plants in the marrow family (squash, melons, marrow, cucumbers, etc.) 2/ JVA do not agree with this contention and state that "...if certain irrigation and cultural practices are followed" no yield reduction would result. Therefore "research should be directed towards development of such packages." (Annex 3, section 1). - 8 - around 10-12%.1/ In contrast, system losses in closed pipes are almost nil and operation is easier: inadvertent physical damage is less likely and unless a farmer turns on his hydrant, water cannot escape from the system. While in the past there has been extensive willful damage and vandalism (PCR para 4.14), JVA expects this to stop once the system is more universally accepted and used. Therefore, the audit's estimated water losses between the line of the EGMC and the "average" farm in the conversion lands would be about 2%. Consequently savings due to converting the water delivery system may be around 8% to 10%.1/ 3/ 16. Other considerations in evaluating the pros and cons of the two systems are: (a) Power saving. The topography allows the system to be pressurized by gravity; if farmers have to pump from farm reservoirs fed by an open system, pressure can only be built up by using motors consuming petroleum or electricity. 1/ In view of the importance of the issue, it is to be regretted that there is no recorded information about the amount of water leaving the EGMC, which could be compared to the quantities paid for by the farmers, i.e. the water actually arriving at the farm outlet. Such information, over several years, would serve to validate - or contradict - the audit's assumptions on conveyance and operating losses. These assumptions are not accepted by regional staff; however, in the absence of data from JVA they could do no more than replace it with their own assumption (see however footnotes 2/ and 3/). 2/ Regional staff state that at present (1981) the system in the valley is so well maintained and operated, that total operational and conveyance losses may be below this figure. JVA provided no actual measurements either to the project completion or the audit mission (see however footnote 3/). 3/ JVA responded to the foregoing as follows: "After using the surface methods of irrigation for more than 20 years in the East Ghor Canal area the on-farm efficiency in the project area ranges between 33% and 52% with an average of about 43%. Weakness of extension during this period did contribute to this low efficiency, but the most important single factor was the land topography. Most of the irrigated area (about 80%) has slopes of more than 1%. In much of the area slopes are complex, efficient surface methods can be designed and managed on slopes of about 0.7%. Land smoothing and levelling to varying degrees is essential for surface irrigation on 80% of the area. Institutional measures to raise irrigation efficiencies are not sufficient alone. With land levelling on-farm efficiencies can be raised to 65%, with sprinkler irrigation efficiencies can be raised to 75% and with drip irrigation efficiencies of 85-90% can be reached. The JVA is still convinced that the decision to convert surface methods into either sprinkler or drip methods is the right decision." The reasons for JVA's conviction are detailed in Annex 3, section 2. -9- (b) Logistic problems. Farmer-operated individual pumps would either require a costly and vulnerable electricity distribution system, or a maintenance and repair organization for internal combustion engines which, in JVA's opinion, may tax the ability and capacity of local spares and repair organizations. The alternative, carrying sufficient spare pumps, is not only expensive but also logistically difficult in a valley over 100 km long and where all the settlements are along the single north-south axis. However, in the 18 km exten- sion area of the EGMC farmers operate about 100 such individually owned pumps. Regional staff who have visited this area state that the farmers have not reported any undue difficulties in keeping their systems operating. (c) Saving of operating costs. JVA maintain that the demand for a closed field irrigation system - sprinklers or drip, depending on the proposed crops - is by now increasing rapidly (see para 10 above) and the demand for surface irrigation will diminish con- siderably in a few years. They further maintain that when demand for surface irrigation diminishes, economies of scale will make a closed, pressurized trunk distribution system the logical choice.1/ The Region state that the break-even point would appear to be after about 80% of the farmers utilize the closed system. Extrapolations from past achievements (paras. 5 and 10 above) do not support JVA's more recent optimism on conversion rates; whether the change of farmer attitude is /ndeed significant will have to be seen over the next few years.V (d) The PCR states that, as all main canals and most of the distrib utaries were already lined, it would have cost relatively little further to modernize the system (PCR para 7.09 and para 9). Un- doubtedly this is correct; however, it makes little allowance for possible physical damage, blockages and spills (accidental or delib- erate) to which open canals and structures are exposed; to this list JVA add the likely damage caused by grazing animals. In addi- tion, all structures require careful surveillance to ensure their uninterrupted operation (para 15). All such problems can be elimi- nated by a closed system, albeit at considerable capital cost. Admittedly a closed system is also subject to damage, and this one, using reinforced plastic mortar (RPM) pipes, has been particularly vulnerable in the past (see para 8). Repairs, in extreme cases, can take quite long. 17. As noted in para 13 above, opting for a trunk distribution system primarily suited for sprinklers may have been a misjudgement. However, since the use of pressure relief valves enables the use of drip pipes, the remain- ing, and probably temporary problem is the large stock of unsold sprinklers. 1/ See Annex 3, section 3. 2/ See Annex 3, section 4. - 10 - Final determination whether, in concept, conversion was the right decision, has to be postponed for several years. The economics of the concept obviously hinge to a major extent on the rate of adoption and the exploitation of the production flexibility of the system (para 10). This issue is further examined in Section E. D. Marketing and Market Development 18. The project, as defined in the SAR, has sidestepped problems asso- ciated with marketing by confining itself to providing support only, instead of facing the major issue of building up a market forecasting, intelligence and development organization. There is no doubt that the packing and grading facilities, eventually built with Dutch and British aid, were needed and are good, but are, in themselves, insufficient. 19. The valley serves domestic as well as export markets, the latter being at present mainly in nearby Arab countries. At present, only a minor portion of the production reaches Europe. Competition in the export markets is intense and reaching them depends not solely on commercial considerations. Therefore, from some authority: be it the Ministry of Agriculture, JVA or Jordan Valley Farmers' Association, farmers expect a market research and forecasting service on which they could base their decisions not only regard- ing the dispatch of current production but also planning the kind of crops to grow. The Agricultural Marketing Organization (AMO), a semi-autonomous body attached to the Ministry of Agriculture, considers a wider spread of exports, principally into Europe, to be desirable, which would imply also a change in the crops planted. AMO advised the audit mission that an FAO-supported project has organized such a service, which, however, came to an end with the closing of the FAO project. 20. Disposing of the fruit and vegetables produced in the project area at present, and, even more, the ones that farmers can produce more easily with access to a more controllable and technically superior irrigation system (para 10), is probably the most challenging task a marketer can face. Pro- ducers for such a market are mostly sophisticated farmers with highly devel- oped technologies and a flexible outlook: there is no traditional attachment to any crop or any one farming system, unlike in communities that produce principally staple foods and fibres. This flexibility is amply demonstrated in the recent agricultural history of the valley; paradoxically, it is not the farmers who at present resist switching from cabbages, tomatoes and squashes to strawberries, green beans and peas and even flowers, but without a much better organized, and more forward-looking, marketing structure they realize this is simply not possible on a meaningful scale. In market planning it is important to realize that growers of perishable commodities rarely expect to sell all of their production at a standard price; produce is marketed on quality and timeliness. Therefore, penetrating a market is not contingent on whether that market is saturated or not, but on the expectation of being able - 11 - to enter it with a product of higher quality than that of potential competi- tors!l or at a better time (before or after season); moreover, the edge of quality and timeliness is of importance on a week-by-week basis. Therefore, the audit believes that, notwithstanding the fact that the Bank is involved in similar projects in other countries, and that these countries are also actual or potential competitors in markets where Jordan competes, or should compete, introducing a market research and development component would have been a vital, and in all respects justifiable component of the project; it should certainly be included if and when a follow-on project is finalized. The Stage II project documents, at present unprocessed awaiting the resolution of outstanding issues, (para 1) have addressed the marketing issue in detail. E. The Project's Economic Rate of Return 21. In reference to the project's ERR discussed in paras 7.09-7.13 of the PCR, the audit accepts the PCR's estimate of 16% for the new lands (para 6 and PCR para 7.10).2/ However, the audit believes that benefits accruing to the conversion lands may have been underestimated. In material terms, this contention hinges on three issues: (a) the extent of water savings of the closed system compared to likely developments without the project (para 15); (b) the extent to which farmers can, more easily, change to more efficient farming systems and introduce more remunerative crops by utilizing project works, and, most importantly (c) on the accuracy of fore- casting the time when the closed distribution system will be fully utilized in the "conversion" areas. 22. Concerning water savings at the field level, the general assumption in pertinent literature is that on-farm efficiency increases by about 10%. Numerate, substantiated information from Jordan - admittedly from one farmer only, whose main, but by no means sole, crop is citrus - records an average saving of 37% on six of his farms, rangint from 12% to 55%. JVA engineers stated that other farmers using sprinklers have taken about 22% less water since conversion; those using drip, as much as 35% less. Based on this infor- mation, the audit is satisfied that a 10% on-farm reduction in water use saving with a closed system, as compared to the open system it has replaced is, if anything, a conservative estimate. 23. Field losses with the open system can only be reduced by careful land shaping and even more carefully planned field layout: the "zig-zag" 1/ The region believes that the potential competitors' edge on quality is likely to take some time to overcome. 2/ If the allocation, and, therefore, bringing into production of these lands is further delayed, the ERR would also be affected. 3/ The informants did not specify whether the sprinklers were fixed or portable, see para. 24. - 12 - system of irrigation which is the traditional response to the fairly steep slopes, is wasteful. Much improvement has already taken place in response to the introduction of tractor cultivation, but further land shaping, which would be necessary, is limited in scope in some parts by shallow soils, steep topography and potential damage caused by exposing the subsoil. 24. In estimating the ERR on the "conversion" lands, and in view of the fact that "conversion" so far has taken place on only 15% of the area (para 5), the PCR necessarily had to make its own estimates on its nature and timing. These were as follows: (a) the change-over period would be four years (para 7.08); (b) water savings would be 10% (para 7.11); (c) when changing, farmers would convert as follows (Annex VI): to drip 15%; to portable sprinklers 40%; to fixed sprinklers 45%. Based on all the assumptions listed, with the exception of estimating overall water savings at about 20% instead of 10%, the audit estimates the ERR on the conversion lands to be not less than 10%. If, encouraged by a good marketing and market development system, farmers would feel more confident in changing their cropping patterns in a manner better to exploit the much improved and flexible water delivery system, the ERR would be higher. 25. The re-estimated ERR clearly refers to the line of development followed: converting the open system, as it was, to a closed system. The audit, for all the operational reservations enunciated in Section C, does not contend that the possible modernization of the open system would not have yielded an ERR higher than 10%; it merely wishes to emphasize that based on data, given to the audit by JVA on actual water savings in the "conversion" lands of this project, tend to support a 10% ERR rather than the 5% estim- ated by the PCR. However, it is by no means certain that water savings of the order given by JVA, and clearly referring to progressive, go-ahead farmers who have already converted, can be achieved by all the farmers. 26, In concluding the discussion on the issue of the ERR, it is neces- sary to note a more general point of major importance. Whether converting to a closed irrigation system in the Jordan Valley - for which the conversion lands in the project under review is best regarded as a pilot operation - is the best way to develop the area, will depend materially on the following factors, not necessarily listed in order of importance: (a) the rate of conversion, (b) the actual water savings per cropped hectare, (c) whether yields are higher in conversion areas than in surface-irrigated areas and (d) whether new higher-value and more profitable crops are introduced more easily and rapidly into the conversion areas than into surface-irrigated areas. Therefore, it is clearly in the best interest of JVA as well as the Bank and other potential donors to continue to monitor carefully developments in the conversion lands for the next several years before committing heavy investments. - 13 - F. Water Charges; Cost Recovery 27. Cost recovery through water charges is a perennial problem in most Bank projects!' and this one is no exception. While a water charge is in fact being collected - and the recovery rate is good - collected fees do not cover the O&M costs of the project. Yet, the water charges actually collected in other countries where charges are based on volume used and not area irri- gated (Mexico, Morocco and the Western States of the U.S.) are roughly similar to what is being charged in the Jordan Valley. Nevertheless the audit is satisfied that economies in water use on the scale that would permit sizeable increases in cropping intensities will not occur until charging full costs will compel farmers to rationalize water use. Little progress is evident towards this objective and Bank efforts to resolve this issue have, to date, yielded no results. 1/ See Water Management in Bank-Supported Irrigation Project Systems: An Analysis of Past Experience (OED Report No. 3421) dated April 16, 1981.  -14- ANNEX 1 Page 1 JUhiAi VALLEY kUTihOR ITY TELEX iiO. 04(0 DATEL iAY 10,19b2 ATTENTIOi: i. SHIV S. KAPUR, DIRECTOR OPERTlúo EVALUATljio DEPARTiNENT TLE v.ORLÄ DÄNK RE: PROECT PCE FWRiAsÅCE AUtIT REPORT Nv'THIAST UHu(R lanúATIWio ANýD RURAL DEVELOPVIENT PROJECT (ChELT 4c-JO) JEAkda.KArUR: wE mAVE REVItWEL THE AUDIT DRAFT REPORT Oi THE AjOVE PRCJECT Alm ARE FLEASE TO SENLD YOu uUR COiENTS WHICH ARE RELEVANT TO THE ROLE OF THE JOADAN VALLEY AUTHURITY, AND HOPE THAT THEY BE TAKEN INTE CUILR'ATIO IN THS FINAL REPORT. WE WILL LE PLEASED TO AiwEr AinY quESTIOi5 THAT YOU NIAY HAVE. PARA 7. rAGE 5 RE: ESEAiCH AND EXTENSION Ti- PRýJECT wAs APPayPE AND THE LOArl SIGNEL IN 1974 AT THE TIME OF THE JORSAN VALLEY CWHISSION WHICH WAS TRANSFORmED TO AN AUTHORITY IN 1i77. TiE rRLE uF THE GUiMiSSibäON WAS TO END AS EACH PROJECT IN THE VALLEY DEVELOPi¼¿_T A COiPLETED. ALL PROJECTS WERE TO BE hANDEi OVER TU THL RELEVANT G i NT AGENCY FOR OPERATION AND i-iAltT i' T4E CDIs,SCL TE uF AGRICULTURAL RESEARCHIAND EXTENSION tv TnE Phû C AS Nû EXCEPTlwN. EcUleENT , SUPPLIES AND iATERIALS ERE PuRunASco uNijER ihE PRuJECT AN wERE riALDEL OVER TO THE 1l'dSTRY tF AGR ICLT&iuE. AG6ICuLTURAL RESEARCH AND LXTENSION WAS NEVER ,iEAIT Tu rt THE RESPW1S10LITY OF THE JORDAN VALLEY COMAISSION dc ROLE WAS TO WCiK -uT, Iis GüOPERATION WITH THE MINISTRY OF AG61CULTuKE, TnE REsulkciEiEvTS FOR EQU PENT AND SUPPLIES, TO PUChAbE TriEii A TO HAoD ThE SAME TO THE dihlSTRY OF AGRICULTURE, wHICn THE CON4ISSION LID AND COilPLETED IN 1975. - 15 - ANNEX 1 Page 2 WITH REGARD TO THE INVOLVEMENT OF THE FACULTY OF AGR ICU'LTURE IN THE RESEARCH C PONENT, THE EANK SUFERVISION FISSIONS LROUGHT IT UP IN THELa2 FRKUENT CONåULTATIjNS. AT THE TIME, AND SINCE 1977 Th-ERE HAD 3 BEEN ATTEMPTS BY THE JVA, THE HiNISTRY OF AGRICULTuRhE ANýD THE FACULTY OF AGRICULTURE TO COO'LINATE THE ROLES OF EACH IN RICULTURAL RESEARCH AND A TAzK COMMITTEE WAS FORMEL TO DEF INE ROLES AND RELETlUNSH IPS. THESE ATTErPiTS WERE TRIGGEREDl t>Y A PrKQuECT SPOQREL BY UbAIjD. HOWEVER THE OMENTUM OF THbIS COUPERATIUN VAS hAiPERED -Y FREQuENT CHANGES Iii PERSONNEL AND LACK OF ENITHUSIASi, AND THE PROJECT wAS ENDED IN 1964 WITHCUT MUCH RESULTS. PROJECT SCOPE AND COST PARAGRAFH i, PAGE b THE COST OVERRUM wAS PRIMiARILY DUE TO THE COST INCREASES OF MATER IALS AND SERV ICES ALL OVER THE WORLD BETWEEN 1973 AND 1977. THE PROJECT WNAS APPRAISED IN 1973 AND CONSTRUCTION COK1ENCED IN 1976. THE 1973 COSTI ESTIHATE vAS ASCUT DOLLARS 16. MILLION AND IN 1976 WAS AbCUT ÅiLLARS 30 MILLIGN. THE GOVERNM1ENT WAS NEVER UNWILLING NûR UNABLE TO MEET THE INCREASED EXPENDITuRES ANiA vAZ OMITTED TO MiEET THEi EITHER FROM ITS jVd SUURGES OR OTHER bOURCES. THE JVA PRESENTEJ To uSA ID THE V ILLAGE LEVELOPPNENT tJ OJECT AND INLUDED IN IT PARTS OF THE RURAL DEVELOPMENT POENT OF ThlE IDA PROJET. THE A ID V I LLAGE DEVELOPNENT PROJECT wAS VALLEY WIDE AND LATER EXPANDED Tu THE SOUTHERiN GHORS AND WAD 1 ARAoA, AND viAS NOT L IMITED TO THE IDA PROJECTAREA. THE COSTD OF DO,iESTIC WATER SUPPLY AND OF THE FARm ROADS WERE iiET LY THE GOVERNuENT OF JORLAN ALONE. wE WERE NOT AWARE THAT THE BANK ENCOURAGED DLISCUSSIûNS WITH OTnER LENDE.RS. SUCH L ISCUSSIONS WERE INITIATED BY THE GOVERNNENT OF JORDAN AND ONLY THE APPROVAL OF THE SANK WAS SOUGHT TO HAVE USAli INVOLVED IN THE FINANCING OF THE SCHOOLS AND HEALTH CENTERS IN IDA PROJECT AREA AS PART OF USAIL VALLEY WIIE VILLAGE DEVELOPMENT PROJECT. - 16 - ANNEX 1 Page 3 PARAGRAPH 6.03 PAGE 19 THE REPORT MENTIONS THAT JVA WAS DEFICIENT IN ITS TECHNICAL PREPARA- TION OF THE PROJECT... ETC.. THE DELAY IN PROJECT PkEPARATION WAS LIMITED TO THE IRRI6ATION CONPONENT. THE DESIGN OF IRRIGATION wORKS ON WADI ARAL DEFENDED ON WHETHER OR NOT A FUTURE DAM ON WADI ARAB WILL BE BUILT. THIS AGAIN DEPENDED ON WHETHER OR NOT THE MAQARIN LAM ON THE YARMOUK RIVER wILL .E. BUILT. IN 1974THERE WAS TALK BETWEEN JVA AND THE BANK TO FINANCE THE FEASIBILITY STUDIES OF MAQARIN DAM. ANTICIPATING FAVOURABLE CONCLUSION OF THESE TASKS THE JVA DECIDED ON THE IRRIGATION WORKS ON WADI ARAB LY EXCLUDING THE POSIBILITY OF THE CONSTRUCTION OF A LAiN ON WALI ARAb. HOWEVER, IN LA LATE QOUR THE BANK INFORMED JVA THAT IT WILL NOT FINANCE THE FEAsI- bILITY STUDIES OF NAQARIN DAN AND A CHANGE IN THE DESIGN WORKS OF THE PROJECT mAD TO BE MADE. THE REPORT FURTHER MENTIONS THAT JVA wAS NEVER ABLE TO ASSUME MUCH OF A ROLE IN AGRICULTURAL RESEARCH. WE BELIEVE THAT THE wRITERS OF ThE 0 REPORT ARE UNDER THE IMPRESSION THAT THE JVA IS ENTRUSTED WITH THE RESPONSIBILITY OF AGRICULTURAL RESEARCH WHICH HAS NEVER BEEN THE CASE. TnE ROLE FUF THE JVA ( THE JORDAN VALLEY COMMISSION THEN) IN AGRICULTURAL RESEARCH W A SUPPORTING FUNCTION TO THE iNISTRY OF AGRICULTURE. WITH OUR BEST REGARDS PRESIDENT JORDAN VALLEY AUTHORITY R 248423 WORLDBANK 21692 JVC JOMMwMi  - 17 - ANNEX 2 Page 1 248423 WORLDBANK (i/ _0 21319 NPC JO 16 MAY, 1982 ATTN: MiR. SHIV S. KARUP 'DIRECTOR OPERATIONS EVALUATION EPARTMENT SUB: PROJECT PERFORMANCE AUDIT REPORT ON JORDAN NORTH EAST GHOR IRRIGATION AND RURAL DEVELOPMENT PROJECT (CREDIT 498-JO) REFERENCE YOUR LETTER OF 10 MARCH 1982 REQUESTING OUR COMMENTS ON THE DRAFT AUDIT REPORT. AS YOU HAVE SENT COPIES OF THE DRAFT REPORT DIRECTLY TO ALL THE CONCERNED AGENCIES, WE TRUST THAT THEY WILL PROVIDE THEIR COMMENTS DIRECTLY DUE TO THEIR GREATER DETAILED INVOLVEMENT IN THE PROJECT. NPC COMMENTS ARE AS FOLLOWS: AAA PROJECT PERFORMANCE AUDIT MEMORANDUM. (1) PAGE 1 OF SUMMARY, FOURTH SENTENCE SHOULD READ QUOTE THE CHANGEOVER FROM RAINFED TO IRRIGATED FARMING STARTED IN THE EARLY 1960'S WHEN THE FIRST PHASE OF THE EAST GHOR MAIN CANAL (EGMC) WAS COMMISSIONED. UNQUOTE. (II) PAGE 5, PARA 7, SUGGEST THAT FOURTH SENTENCE SHOULD READ QUOTE IT MAY WELL BE ARGUED THAT IT WOULD HAVE BEEN BETTER TO TRY AND IMPROVE MMMM ETC. UNQUOTE. ALSO, THE FIFTH AND SIXTH SENTENCES CONCERNING THE FACULTY OF AGRICULTURE ARE CONSIDERED CONJECTURE AND SHOULD BE DELETED. (III) PAGE 8, PARA 12, THE FIRST SENTENCE SHOULD READ IN PART QUOTE ... AND THAT THE GOVERNMENT WAS UNABLE TO MEET THE INCREASED EXPENDITURES, ..ETC. UNQUOTE. - 18 - ANNEX 2 Page 2 (IV) PAGE 14, PARA 19. THE FOURTH SENTENCE SHOULD READ IN PART QUOTE THE AGRICULTURAL MARKETING ORGANIZATION (AMO), A SEMI-AUTONOMOUS BODY ATTACHED TO THE MINISTRY OF AGRICULTURE ... ETC. UNQUOTE. ALSO, IN SECOND LINE OF THIS PARA 19, PLEASE CORRECT ''PERSIAN GULF'' TO READ ''ARABIAN GULF''. BBB PROJECT COMPLETION REPORT: -- PAGE 15, PARA 5.04, THE FIGURE OF JD 45 MILLION IS IN ERROR. WE HAVE ASKED JVA TO PROVIDE YOU WITH THE CORRECT FIGURE. -- PAGE 19, PARA 6.03, THE FOURTH SENTENCE IS SUBJECT TO QUESTION UNLESS FURTHER ELABORATED. THE DEFICIENCY SHOULD BE MORE SPECIFICALLY STATED. JVA WILL DOUBTLESS BE COMMENTING ON THIS. BEST PERSONAL REGARDS, H. ODEH PRESIDENT, NPC. 248423 WORLDBANK 21319 NPC JO - 19 - ANNEX 3 Page 1 THE HASHEMITE KINGDOM OF JORDAN Jordan Valley Authority 41472 - 41473 Tel. 142228 -42386 t V - t%tvy Amman - P.O. Box 2769 Telex 21692 JVC JO JVC JO Y Cable : JOVAKO No. JVA/ C. - - . Date /3 /5/1982 - Mr. Shiv S. Kapur, Director Operations Evaluation Department, World Bank, 1818 H Street, S.W., Washington, D.C. 20433, U.S.A. Subject: North East Ghor Irrg. and Rural Development Project - Project Performance Audit Report. Dear Mr. Kapur, In reference to your letter of March 10, 1982 and further to our earlier telex on the subject, attached are some further comments on the audit report and the project completion report of the North East Ghor Project. Taking this opportunity to thank you and the Bank for the fruitful cooperation in our development programs. Sincerely yours, President JVA. Encls: Comments on the audit report. FN/GK. - 20 - ANYEX 3 age 2 JVA Comments on the World Bank Project Performance Audit Repot, March 10, 1982 (PPAR) and the attached Project Completion Re c-t (PCR) 1. AdaptabilME of Sprinkler Irrigation to Tomatoes, Cucurbits and Eggplants (PPAR Para: 5. 13, PCR Para: 5Z') Fears of diverse effects of sprinkler irrigat:n on certain vegetable crops like tomatoes, cucurbits and ggplants are theoritical and not substantiated by any resear-ch. In fact such worries were publicizedin California a long time ago when sprinkler irrigation was developed. There are theoritical counter arguments to these fears (Reference No. 1 Page 538-540.) In practice the JVA knows of Tomatoes being produced under sprinkler Irrigation in California, in Florida and in Utah, USA. In Israel it is reported that "the yield of canning tomatoes in the Negev region was 68.0 tons/ha with tricle irrigation compared with 56.0 tons/ha with sprinling" Ref. No. 2 Page 384). This proves that the above fears are not founded on actual research or practice especially if certain irrigation and cultural practices packages are followed. Research should be directed towards development of such packages. 2. Conversion to Sprinkler or Drip Irrigation After using the surface methods of irrigation for more than 20 years in the East Ghor Canal area the on-farm efficiency in the project area ranges between 33% and 524 with an average of about 431. Weakness of extension during t-:Lif period did contribute to this low efficiency, but the mcs. important single factor was the land topography. Most of the .rrigated area (about 80%) has slopes of more than 1%. In muc. of the area slopes are complex, Efficient surface methods can be designed ../2 - 21 - ANNEX 3 Page 3 and managed on slopes of about 0.7%. Land smoothing and levelling to varying degrees is essential for surface irri- gation on 80% of the area. Institutional measures to raise irrigation efficiencies are not sufficient alone. With land levelling on-farm efficiencies can be raised to 65%, with sprinkler irrigation efficiencies can be raised to 75% and with drip irrigation efficiencies of 85-90% can be reached. The JVA is still convinced that the decision to convert surface methods into either sprinkler or drip methods is the right decision: The following factors went into evaluation of this decision: a) On-farm efficiencies as discussed before. b) Costs of and rate of land 3evelling required. c) Impossibility of land levelling in shallow soils of class 3 land (about 10% of the area) d) Loss of a cropping season during land levelling. e) Loss of productivity for some years after land levelling. f) Labor requirements for irrigation. In studies conducted in California (PG & E Company) and in Jordan (JVA) irrigation labor requirements for surface irrigation were estimated to be about 5 times the requirements for sprinkler irrigation. With the increasing cost of labor this factor becomes a major one. g) Loss of land under surface irrigation for farm ditches, tail ditches and dykes for borders (about 0.25 ha per 3-ha farm) - 22 - ANNEX 3 Page 4 h) Cost of lined farm ditches and gates for a good surface irrigation system. i) Night irrigation: Labor is almost impossible to find for night irrigation due to scarcity of labor in the Jordan Valley. It is easier to set sprinkler or drip irrigation during the night and leave the system unattended, than to try surface irrigation during the night. For security reasons parts of the irrigation area are restricted during the night and only unattended systems can be used at such times. Night irrigation is a must as the irrigation system is sized for continuous operation. j) Yields under the alternative irrigation systems. Because of more uniformity of wetting, better germination and higher plant population yields under sprinkler or drip are higher than yields under surface irrigation. k) Extension requirements. It is easier to train farmers to use sprinkler or drip systems efficiently. Many of the decision variables in surface irrigation are incorporated into the system design of sprinkler or drip systems. 3. Collective Versus Individual Conversion The JVA is still convinced that the collective conversion alternative through presurized pipe networks is more attractive from the economic stand point as well as from the practical stand point. The JVA and the World Bank studied the economics of the alternatives and both studies showed that the collective - 23 - ANNEX 3 Page 5 alternative was more economical than the individual alternative if all the land is converted. When the Bank assumed 50% of the Farmers to stick to surface methods the individual conversion was more alternative. The present trend of the farmers is towards more drip areas and there is no reason to assume that all through the life of the system 50% of the farmers will continue to use the surface methods. The collective system is even more attractive in the gravity areas where pressure is generated without energy cost which had risen since the Bank's study. 4. Slow Conversion Rates (PPAR Par. 16 (c); PCR Par. 4.02) It is admitted that the rate of conversion to sprinkler by farmers was slow at the beginning, but it is gaining momentum at present. Conversion is left to the farmezs. Free choice and it is known that Farmers change slowly. This is accentuated by the absence of any extension work along this line. From recent increasing rates of conversion to sprinkler or drip it is expected that the rate will be satisfactory in the future. 5. Time lag of land reallocation to farmers (PPAR Par. 5; PCR Par. 4.17, 4-18) Boundaries of new Farm Units, although set at design and layout stage, are only finalized after construction and as-built drawings are produced. Land reallocation can only be made after these final boundaries are known. A lesson learned for future development is to schedule construction and produce as built drawings lateral by lateral so that land reallocation can be started during construction instead of waiting until the whole project is completed. ++/The number of farms which purchased sprinkler equipment is now 220 out of 315, in addition 5 farm have now drip irrigation. - 24 - ANNEX 3 Page 6 6. Tile Drainage JVA is carrying out by its own staff, investigation design and construction of tile drains in the project area. The areas that have now tile drains is 420 ha out of 1080 ha needin tile drains in the project area. This is based on recent surve carried out by JVA. According to JVA plans, tile drains will be constructed for about 200 ha in 1982, while tile drains for rest of the area will be executed during the period 1983-1985. 7. PCR Par. 8.05 last line. "Weakness of the Agriculture Ministry of Jordan." This is a wrong generalization. The research and extension program was weak but that can not be generalized to include all other functions of the Ministry of Agriculture. 8. PCR Para. 5.04 Loss in tomatoes because of the yellow leaf curl virous is estimated at J.D. 3 million and not J.D. 45 million. - 25 - ANNEX 3 Page 7 References 1. Arnon, I 1972. Crop production in dry regions Vol. I. Leonard Hill, London. 650P. 2. Yaron, B., E. Danfors, and Y. Vaadia. (ed.) 1973. Arid Zone Irrigation. Springer-Verlag Berlin. 434P.  - 26 - ANNEX 4 The Opec Fund for International Development April 5, 1982 Mr. Shiv S. Kapur Director, Operations Evaluation Department, The World Bank 1818 H Street, N.W. Washington D.C. 20433 USA Reference: Project Performance Audit Report on Jordan North East Ghor Irrigation and Rural Development Project (OPEC Fund Loan 52 P; IDA Credit 498-JO) Dear Mr. Kapur, Thank you for your letter of March 10, 1982, and the attached report. Our staff has just completed the review of the above report, as well as the Project Completion Report prepared by EMENA Region, and has found both reports adequate in their coverage of the various aspects of the project. We therefore have no further comments other than to commend you for the efforts made in such an important task. With best regards, Sincerely yours, Dr. J Pabloerez Castillo AssisAnt Director-General Operations Management A. : 3 I 6-0 P. Box 995 -1011Nicenna-Aitstria: Cable: ()pLLffind:Te1cx: 1,31734 Ruid AJ1-34831 Fund A.Tl.: 31,55 36{)  - 27 - JORDAN NORTH EAST GHOR IRRIGATION AND RURAL DEVELOPMENT PROJECT (CREDIT 498-JO) PROJECT COMPLETION REPORT Europe, Middle East and North Africa Regional Office  - 28 - II. BACKGROUND 2.01 Only 12% (60,000 ha) of cultivable lands (out of a 500,000 ha total) in Jordan are potentially irrigable. About 60% of these lands are in the Jordan Valley. The entire Valley is below sea level from about -200 m at Lake Tiberias to about -400 m at the Dead Sea. Its topography and temperate climate in winter is highly favorable for early season vegetable production. The gradual transformation of the Valley from subsistence farming to commercial agriculture started in the 1950s with the construction of the EGMC which is supplied by the international Yarmouk River. The EGMC runs all along the East Bank of the Jordan Valley. 2.02 Prior to the 1967 war, development of irrigation in the Jordan Valley had resulted in a threefold increase in agriculture's contribution to Jordan's GDP over a 10 year period. During the 1967 war considerable damage was sustained in the Valley. This was followed by the civil disturbances of 1970. Normal farming operations were restricted and as a result, a large part of the Valley's population left. In 1971 the population started returning to their holdings but were severely handicapped by the poor state of civil infrastructure and irrigation works, the shortage of housing, and the lack of social services for their families. 2.03 To accelerate the recovery of economic activity in the Valley and to further develop its productive potential, the Jordanian Government in 1971 prepared a three-year Rehabilitation and Development Plan (1973-1975) for the Jordan Valley. It approached various governments and international agencies, including the Bank, to request assistance in financing the Plan /1. 2.04 The Bank reviewed the Plan and sent its first identification mission to Jordan at the end of 1972. At that time other governments and financing agencies had already committed funds or expressed interest in financing parts of the Plan. The Kuwait Fund had agreed to finance construction of the King Talal Dam on the Zarqa River and USAID declared its intention to finance irrigation development downstream of this dam. The German Government also expressed interest in providing financial assistance for development in the south of the Valley. The Bank was asked to consider financing the remaining works to be undertaken in the northern portion of the Valley, north of Wadi Yabis, for which funds had not yet been committed by other donors. /1 This Development Plan for the Jordan Valley was to be implemented in two stages: Stage I would utilize the unregulated flows of the Yarmouk and of some side Wadis and the regulated flows of the Zarqa River to irrigate 18,000 ha; Stage I projects included the King Talal Dam on the Zarqa River and the four following irrigation projects: (i) North East Ghor financed by IDA; (ii) Zarqa Triangle financed by USAID; (iii) the 18 km extension financed by USAID; and (iv) Kafrein-Hisban financed by KfW. Stage II would develop the Valley's remaining land and water resources by building the Maqarin Dam on the Yarmouk River and modernizing and extending the irrigation system; implementation of the Stage II project is still pending resolution of riparian issues. - 29 - III. PROJECT FORMULATION 3.01 The project was identified by the Bank and prepared by the JVA with assistance from FAO/IBRD Cooperative Program. JVA prepared technical feasibility studies for the engineering works. The project area proposed for Bank consideration stretches from Wadi Yabis north to the the Yarmouk river east of the Jordan River (see Map IBRD 15734-PCR). The original project, conceived by JVA and the Bank, included the rehabilitation of surface irrigation facilities on about 5,000 ha under command of the EGMC and the installation of a pipe network for gravity-fed sprinkler irrigation on approximately 1,700 largely unirrigated ha above this canal, where surface irrigation was considered impractical because of the steep slopes and shallow soils. The preliminary design prepared by JVA included three independent irrigation schemes served from the unused Wadi Ziglab storage reservoir and from river run-off from Wadi Arab. 3.02 During preparation, the irrigation component was substantially modified. JVA entered into negotiations with the consulting firm NEDECO-Dar Al Handasah to prepare the detailed design of the irrigation works. The draft agreement provided for the design of the sprinkler system for an additional 1,000 ha lying below the EGMC which were already surface irrigated ("Conversion Lands"), as well as a study to select the most appropriate area to be converted to sprinkler irrigation on the basis of cost and land grading limitations. This study on conversion lands was later considered unnecessary by JVA and excluded from the final agreement between JVA and their consultants with the concurrence of the Bank in April 1973, as JVA had decided to eventually convert all area surface-irrigated by EGMC into sprinkler irrigated. JVA also requested the appraisal mission to consider the inclusion of a dam on Wadi Arab as a part of the project works. 3.03 The appraisal mission agreed to this revised irrigation component, but did not agree to consider the dam on Wadi Arab because its flow had very little annual variations. Its contributions to the final design were: (a) replacing the concept of a pumped system envisaged in the original proposal with a gravity-fed pipe system; and (b) connecting the three individual distribution systems of the original proposal through a main pipe line to optimize water use from river run-off of Wadis Arab and Jurum and from the reservoir on Wadi Ziglab. This integrated system also allowed flexibility to make use of the supplies developed by the construction of the dam on Wadi Arab. 3.04 The project approved by the Bank on July 25, 1974 (Credit 498-JOR) covered 7,700 ha of which about 86% was already surface-irrigated. A pressure pipe system for gravity-fed spinkler irrigation was to be installed on 1,000 ha of land being irrigated by surface from the EGMC (conversion lands) and on 1,760 ha of land above the canal, of which only 720 ha were under intermittent surface irrigation from side wadis and the rest unirrigated (the entire 1,760 ha are referred to as "new lands"). To improve production on the gravity-irrigated area under command of the EGMC and not to be converted to sprinkler (4,940 ha), the project included land leveling on about 3,000 ha and the drainage of about 520 ha. The project also included rehabilitation of some cross-drainage works on EGMC, and smaller components in agricultural credit, marketing, research and extension, and social infrastructure. The project was to benefit a population of about 25,000. As appraised, the project included the following: - 30 - (a) Irrigation Works (i) diversion weirs on Wadis Arab and Jurum; (ii) buried carrier pipes and distribution networks to serve by sprinkler irrigation a total of 2,760 ha; (iii) farm roads (30 kms); (iv) land leveling of about 3,000 ha and tile drainage of about 520 ha; and (v) minor works on the existing East Ghor Canal. (b) Rural Development (i) treated water supply system for about 10 villages; (ii) 3 new health centers and extension of an existing center; (iii) about 150 schoolrooms; (iv) a vocational training center; (v) improvement of about 60 km of roads; and (vi) 10 community development centers; (c) Two assembly markets at Wadi Yabis and North Shuneh; (d) Equipment and materials for maintenance of irrigation works and roads; and for agricultural research and extension; (e) Agricultural credit to help finance loans for agricultural production and on-farm development, including portable sprinkler equipment; (f) Consulting services to assist project staff with the preparation of detailed design and supervision of construction. 3.05 Credit 498-JOR was declared effective on January 8, 1975, about 2 months after the proposed aate and after conditions in section 7.01 of the Credit Agreement had been met, i.e. enactment of a law establishing the Jordan Valley Farmer's Association and the election of its Executive Board. The Borrower submitted all documents on time. However, since the Parliament was not in session, the Law establishing the Jordan Valley's Farmers' Association (JVFA) was, under Jordan's Constitution, only temporary; after several exchanges of correspondence between the Borrower and the Bank's Legal Department, the Bank was persuaded to accept the temporary law as fulfilling the Credit Agreement condition. IV. IMPLEMENTATION Targets Achieved and Changes in Project Scope 4.01 Considerable changes in project scope were introduced during project implementation, principally as a result of cost overruns in the irrigation works which necessitated the elimination of most of the non-irrigation related components. As a consequence, the project ultimately consisted of: - 31 - (i) construction of two diversion weirs, primary and secondary pipe lines and a distribution network serving 2,808 ha (2,760 ha at appraisal); (ii) engineering consultant services; (iii) equipment and materials for operation and maintenance and for agricultural research and extension; and (iv) credit for incremental seasonal agricultural production and medium-term on-farm development loans. The planned improvements on the non-converted lands, minor rehabilitation works, assembly markets and rural development infrastructure components were all dropped from the project, and the credit component was substantially reduced. 4.02 Irrigation Works. The project succeeded in meeting the implementation targets for the installation of the pressure pipe system. The system can serve a net area of approximately 2,808 ha, including 1,030 ha of converted lands below the EGMC and 1,778 ha of new lands above the EGMC. However, at the time of the completion mission, only 12% of the conversion lands was being irrigated from the pipe system, while the remainder was still being surface-irrigated from the EGMC. None of the new lands was irrigated from the system pending completion of land distribution. The limited use of the pipe system in the conversion lands may be attributed to: (i) frequent breakdowns of the pipe system (see para 4.16); and (ii) reluctance of the farmers to adopt sprinkler which they feared would damage the quality of their highest-value vegetable crops (see para. 5.09). 4.03 The only major design change in the irrigation component following agreement on the engineering contract was in the plans for the Wadi Arab diversion weir. The first supervision mission of March 1975 was advised that JVA was considering constructing a dam on the Wadi Arab with Japanese financing, which would require a change in the design of the diversion weir. At issue was whether the diversion weir proposed for the Bank project would be submerged by the new reservoir and, if so, whether the weir should be relocated or a temporary structure should be built. This question dragged on for more than 3 years until completion of the tender documents for construction of the dam. JVA ultimately decided to temporarily pump from the EGMC into the North East Ghor main line and to divert Wadi Arab water into the EGMC until completion of the dam. The temporary pumping station was completed in April 1981, two years after completion of the pipe system. 4.04 Non-Converted Lands. At the time of the first supervision mission in March 1975, two months following loan effectiveness, JVA requested that the proposed improvements in che non-converted lands--land-leveling and tile drainage--be dropped from the project. Since JVA had, by that time, decided to convert to sprinkler irrigation throughout the Valley, the land leveling component was no longer considered necessary. The Bank agreed that it would be difficult to level those project areas which have "relatively steep slopes, shallow soils and perennial orchards and that final decision whether such - 32 - areas should be sprinkler irrigatea .... should be delayed." At the same time, JVA requested that the tile drainage works be excluded from the major construction contract in view of its preference to undertake these works under force account with technical assistance from USAID. The Bank also agreed to delay these works provided adequate provisions were made to monitor the situation so that tile drainage could be installed should waterlogging occur. The Bank, however, did not follow up on these aspects, since the non-converted areas were to be included in the Jordan Valley Stage II Project, which was then under preparation. Since then, tile drains have been installed only on about 130 ha through JVA force account. Additional drainage works are needed as evidenced by waterlogging observed in the project area during the completion mission. 4.05 Assembly Markets. Originally the project provided for construction of two assembly markets to be adjoined to two existing grading and packing stations on Wadi Yabis and North Shuneh. The total cost for establishing two modest markets covering I ha each with a shed of 800 m2 was estimated at about JD 51,000. These markets were dropped from the project and funds reallocated to the irrigation works. Since then, JVA decided to undertake a more ambitious program to improve the marketing of Valley produce including three new grading, packing and marketing centers which would also serve as wholesale markets. The first one, financed through the Government of the Netherlands, at Al Arda, is now completed (JD 1.8 million). The second, at Wadi Yabis, financed by the UK Overseas Development Authority is now under construction at a total cost of JD 1.5 million. A third one is planned for the southern part oi the Valley. 4.06 Rural Development. While the original project design included financing for a variety of social infrastructure works, these funds were ultimately reallocated to help finance cost overruns in the irrigation works component. Nevertheless, the Jordanian Government succeeded in carrying out a major investment program to improve various social services in order to attract families and government employees to the Valley, much of which has been with USAID assistance: -- 208 schoolrooms have been constructed in the project area, compared to the 150 originally planned under the project. Construction began in late 1976 and was completed in 1978, ana facilities were promptly equipped and staffed, permitting a rapid increase in enrollment; -- 4 health centers, to have been financed under the Bank project, have also been constructed in the project area. Two of the centers provide comprehensive health care services, each including 20 beds, an X-ray unit, dental clinic, surgical facilities, blooa bank, and a maternity/child care unit. The other two are smaller, with less comprehensive facilities. Although there have been no staffing problems with the two smaller centers, the Ministry of Health has reportedly been reluctant to provide full staffing for the larger centers, where some of the services are underutilized. It would appear that these centers have been over-aesigned for the needs of the Valley population. - 33 - -- Domestic water supply, contemplated in the Bank project for 10 villages, has been provided to 11 villages, fully financed by the Jordanian Government. Service is planned for five more villages. The system is supplied by spring water, instead of the irrigation pipeline as originally planned. Following an outbreak of cholera in 1977-78, the planned use of community taps was abandoned, and house connections were provided, with users paying for meter installation ana water fees. -- With Jordanian Government financing, JVA has proceeded to contract for the construction of new farm roads and improvement of existing roads throught the project area. Under the project-financed irrigation construction contract, earthworks were completed for roads in the new lands. 4.07 Credit: The project originally included a US$760,000 credit component to meet needs for both medium-term credit (on-farm sprinkler equipmenc, on-farm development, farm machinery), and incremental seasonal credit. When USAID agreed to provide funding for on-farm sprinkler equipment under a separate, Valley-wide project, this credit allocation was reduced to US$400,000 to cover remaining medium term requirements and incremental seasonal credit. This allocation was fully disbursed by the Agricultural Credit Corporation (ACC) in three and a half years, from date of loan effectiveness in January 1975 to May 31, 1978. Over 90% was disbursed for medium-term investment, 60% of which was for tree planting and land improvement (see Annex VIII). The balance was disbursed for incremental seasonal credit. This credit component was administered entirely by ACC; the Jordan Valley Farmers Association (JVFA) did not assume any responsibilities for credit distribution or collection, as was envisaged at appraisal, since JVFA's collection function was to have been tied to its marketing operations, which were never implemented. 4.08 In summary, while the scope of the IDA project was substantially reduced in the course of implementation, most of the original targets for improvements in a number of sectors in the project area have been met. (See Annex I). This overall success can be credited to the readiness of other donors to provide additional financing and to the Jordanian Government's evident commitment to Jordan Valley development. The other donors' participation in the development of the Northeast Ghor has had the advantage of simplifying management of the investment program by permitting consolidation of procurement and construction contracts for equipment and works planned on a Valley-wide scale (e.g., schools, health centers, provision ot on-farm sprinkler equipment). However, it appears that the ready availability of external financing has encouraged, in a few cases, construction of facilities (e.g., health and marketing centers) which are not being fully utilized at the present time. Construction and Procurement 4.09 The irrigation works were completed during the first quarter of 1979, about 2 years behind appraisal schedule and only 6 months behind the - 34 - contractual completion date. A completion certificate was later issued by JVA establishing the beginning of the maintenance period on June 16, 1979. 4.10 The main cause for the delay on completion date of works stems from the delay in finalizing the detailed design and bidding documents. Appraisal of the project was based on a very preliminary design prepared by JVA staff. The appraisal report dated hay 10, 1974, indicated that tender documents would be ready by about September 1974. However, by that time, JVA had only just entered into a final agreement with the engineering consultant. Bidding documents were not ready until July 1975. The process of preparing bids by contractors, awarding and approving the contract took one more year, which is unusually long for these types of works. The order to start construction was issued on August 9, 1976, about 17 months behind schedule. 4.11 Procurement for equipment and civil works financed by the Bank was carried out in accordance with the Bank procedures for ICB. All civil works construction, including the two diversion weirs, the pressure pipe distribution network, rehabiliation of the EGMC and road alignments in the new lands were grouped under a single contract. Of 21 prequalified bidders, five contractors from four different countries submitted tenders. The consultants' evaluation of the bids resulted in a recommendation to accept the lowest responsive bid from the Korean Cho Suk Construction Company. Its JD 4.19 million bia was 36% above the original estimate of JD 3.07 million. Equipment for operation ana maintenance and for research and extension was mostly supplied through local agents of foreign manufacturers. 4.12 The overall quality of design and construction was good, with two exceptions. First, the pressure regulators, originally called for in the consultants' design plans could not accommodate zero flow and had to be replaced to avoid overpressurizing the farm mains. This problem was quickly identified and resolved by the contractor. 4.13 A potentially more serious defect may be the choice of reinforced plastic mortar (RPM) pipes proposed by the contractor for part of the distribution network. While the bidding documents called for asbestos cement pipes for the pressure distribution network, the contractor had the option to submit bids for other materials. Cho Suk proposed using RPM for pipes with a diameter over 600 mm, asbestos cement for pipes between 500 and 125 mm and polyvinyl chloride (PVC) for pipes of 100 mm. These two latter types are well known and of proven material. However, plastic pipe was a recent technology, and its field experience was limited. No mention is made in the consultant's bid evaluation report on the performance of these pipes. JVA, with the consulting engineer's assistance, sought outside technical assessment of RPM for similar works financed through USAID at the time of awarding contracts, and was satisfied that RPM, through testing, appeared to be an adequate material. It was therefore selected on the basis of its lower cost. However, JVA reports 14 cases of breaks in the RPM pipe in the project area since completion of construction, and in the last year, 60 breaks in the RPM pipes in the USAID financed Zarqa Triangle Project. While these breaks may have - 35 - been caused by water hammer arising from poor operation, they more likely reflect defects in the RPM material. /1 4.14 In addition to these problems, a large share of meters have been seriously vandalized because they were not protected in concrete boxes. Many of these were replaced by the contractor during the contractual one-year period of maintenance. Since the system was handed over to JVA's Operation and Maintenance division, vandalism has not stopped. Cost of additional repair is estimated at JD 200,000. JVA expects that damages will drastically be reduced after distribution of lands since responsibility for maintaining and repairing turnout equipment will be assigned to the farmers themselves. Redistribution of Lands 4.15 Land reform laws passed since 1959 are associated with the development of irrigation in the Jordan Valley. These laws consolidate land ownership and redistribute farm units in order to obtain a pattern of farms compatible with a geometric irrigation system. The objective of the reform was to establish upper and lower limits on the size of land holdings and to redistribute farming units from large to small farmers. These laws were applied to the lands under command of the EGMC during its construction between 1958 and 1966. Application of the laws include the following steps: land surveys of existing fragmented holdings; preparation of the as-built drawings of the irrigation network defining the boundaries of the new farm units; land consolidation and redistribution; transfer of the new holdings to farmers; and a balancing of owners' accounts compensating them for expropriation of land ana charging fees for provision of irrigated services. 4.16 The distribution of new lands in the North East Ghor Project was seriously delayed because of JVA's survey procedures and the almost simultaneous completion of the four Stage I irrigation projects (totalling about 9,400 ha). The number of JVA survey teams was limited. Their role was to prepare the as-built drawings of the irrigation network to be superimposed onto the original holding boundary maps. These drawings for the new lands were not received by the JVA Land Reform Office until June 1980. Distribution to farmers started in December 1980. Completion of the program is scheduled for July 1981. 4.17 At the time of the completion mission, only 85 farm units out of 451 (18%) had been distributed in the new lands. Each unit is 3.8 ha which can be held by one or more owners. If owners are assumed to hold equal shares of each unit, the average holding size among the lands already distributed is roughly 2.4 ha. Over 60% of those receiving land own less than 2 ha, and another 30% own between 2 and 4 ha. Almost all of those receiving land previously owned land in the same area. Roughly 18'. have increased the size of their holdings, although most of these are still below the desired holding size (1 farm unit, or 3.8 ha); a smaller fraction have smaller holdings than /1 A serious cause of breaks discussed in an ASCE Conference could be the gradual corrosion of glass fiber or interlaminate separation arising from excessive deflection of the pipes during installation. - 36 - previously, even though many of these are still considerably in excess of the desired holding size (i.e., over 10 ha). Thus it appears that the land distribution program as being carried out in the new lands of the project is succeeding in its redistribution objective, but that the upper and lower limits on size of holding are not being uniformly applied. 4.18 As the result of the delay in land distribution, two full production seasons have been lost, which has lowered the economic rate of return on the new lands irrigation investments by 5%. This delay could be eliminated by a change in JVA survey procedures. According to current practice, only the carrier pipes, main and secondary pipe lines, are surveyed at the time of preparation of the detailed design for the pressure network. The alignment of the farm main pipes, which establishes the farm unit boundaries, are determined during construction. Thus, delineation of new farm units for redistribution cannot begin until the pipe system has been built. In order to accelerate the land distribution process, boundaries for the new farm blocks could be finalized during preliminary design of the network. Land redistribution could then proceed independently of construction. This is a procedure which has been succesfully used in all Moroccan irrigation projects since 1960, permitting distribution of land parcels by the time construction is completed. Project Costs and Financing 4.19 At appraisal the total cost of the project was estimated at US$17.4 million. The actual cost of the project components financed by the Association is estimated at US$16.3 million instead of US$12.2 million. In view of the substantial changes in the scope of the project, the only meaningful comparison between estimated and actual project costs to be made is for the pressure system irrigation works and consultant services. The Table below compares original appraisal estimates with the approximate actual project costs. Comparative Costs (us$ ,000) Appraisal Actual Cost Component Estimate (IDA-Financed Project Only) Irrigation Works 5,458 12,382 Tile Drainage 305 - Land Levelling 1,046 Main Canal Rehabilitation Works 218 - Equipment and Materials 784 650 Markets 218 - Credit 3,483 1,040 Rural Development Infrastructure 3,429 - Consultants Services 566 1,424 Administration 1,922 889 Total 17,429 16,385 - 37 - 4.20 The total actual cost of irrigation works under the project (excluding spare parts) amounted to US$11.9 million equivalent, or 125% above the appraisal estimate (which included roughly 45% physical and price contingencies) for the items actually financed. The cost of consulting services (US$1.4 million) is about 150% above the appraisal estimate. When compared to cost overruns on similar projects in the region over the same time period, these cost overruns are not substantial. Cost per ha of the irrigation system, including engineering and administration, is about US$5,150. This is over twice the appraisal estimate and about 60% above the cost of comparable works (including costs of on-farm sprinkler equipment) in Morocco, for example. The higher cost of the North East Ghor Project is due to the unfavorable shape of the project area requiring a 35 km long mainline, the lack of large diameter pipe factory in Jordan and the higher labor cost. 4.21 The SAR provided for the financing of 82% of the foreign exchange cost through an IDA credit of US$7.5 million and the balance of the foreign exchange cost and all local costs from Government resources (92%) and farmers (8%). As a result of the cost overrun on irrigation works, the Borrower made arrangements with other donors for the financing of the rural development component, the marketing centers, and the on-farm sprinkler equipment. The OPEC Fund provided US$1.65 million to finance part of the gap in foreign expenditure requirements for the irrigation civil work contract. These OPEC funds were administered by the Bank. 4.22 Reflecting the successive changes in the project scope, there were two substantive amendments to Schedule 1 of the Credit Agreement. Amendment No. 1 of December 1976 reallocated funds on the basis that USAID funds were available for the rural development component and tb the portable farm equipment of the project. As a consequence, US$1,690 million were transferred to Category I (civil works) and the allocation for agricultural credit in Category IV was reduced from US$760,000 to US$400,000. The percentage of disbursement was also changed at the request of JVA from 50% of total expenditures to 50% of foreign expenditures to accomodate cost overruns under Category I. Amendment No. 2 of November 1977 increased the funds for civil works to US$5,650,000 and for consultant services to US$1,000,000 and raised the percentage of disbursements under Category I to 75% of foreign expenditures. Final disbursement as compared to appraisal allocations are shown in the following Table. A comparison of the actual disbursement schedule and that projected at appraisal are shown in Annex II. - 38 - Allocation of the Proceeds of the Loan Appraisal Actual Actual Disbursements Allocation Disbursements as % of appraisal Category 07/25/74 Estimates -------- US$ million ------ % I. Civil Works 3,600,000 5,613,645 156 II. Equipment & Materials 760,000 444,056 58 III. Consultants' Services 470,000 1,042,298 221 IV. Agriculture Credit 760,000 400,000 52 V. Unallocated 1,910,000 - - Total 7,500,000 7,500,000 100 4.23 The original closing date of June 30, 1979 was first extended to December 31, 1979. The Bank then agreed to a further six months to extension to June 30, 1980 to permit full disbursement of the proceeds of the Credit already committed or paid by the Borrower. V. AGRICULTURAL DEVELOPMENT Production Patterns 5.01 Agricultural production in the Jordan Valley has developed rapidly over the last five years. Average yields have increased for virtually all crops, and cropping intensity for the Valley as a whole has increased from 106% to 158% between 1975 and 1979. Increased productivity is a result of various factors: (a) strong demand in regional export markets and favorable prices have led to a marked shift from cereal crops to fruit and vegetable production; (b) rapid adoption of new cultivation techniques, including use of drip irrigation and cultivation of vegetables under plastic; and (c) more extensive use of purchased inputs--fertilizers, herbicides, fungicides and pesticides. These developments are most noticeable in the middle and southern areas of the Valley, where its warmer climate permits earlier harvesting and farmers can capture peak prices for off-season vegetables. Returns on investments in purchased inputs, new irrigation and production technologies are consequently higher in these areas of the Valley. 5.02 In the north, adoption of new techniques has been somewhat slower but much progress has been made nevertheless. Farmers in the project area have shared in the Valley's response to market opportunities for off-season vegetable production. In the conversion lands, over one-fourth of the cultivated land area has shifted from wheat since 1975 to vegetables and citrus in roughly equal proportion, with a slight increase in cropping intensity from 1.36 to 1.44. Despite two drought years, average yields for major crops have shown substantial improvement; average yields for 1974-76 and 1977-79 have increased 58% for tomatoes, 55% for cucumbers, 47% for squash, - 39 - and somewhat less for cabbage, cauliflower, and eggplant. These increases are largely attributable to the farmers' use of improved seed varieties, better fertilizer application, and a sharp decline in labor intensive zig-zag irrigation in preference to straight furrow irrigation which has freed up farm labor for more productive land preparation and cultivation tasks. It also ensures a more even distribution of irrigation water. 5.03 Comparing these developments with those projected at appraisal, the most striking differences are: (i) the notable absence of maize in the cropping pattern, which was anticipated to represent over 10% of the net cropped area; and (ii) the substantially higher share of area under citrus, nearly 2.5 times that estimated. It is also worth noting that yields achieved in 1978 and 1979 for the most important vegetable crops (tomatoes, eggplant, cabbage, and cauliflower) and for bananas equal or exceed yields expected in the SAR at full development in 1983 (see Annex III). These developments have undoubtedly been encouraged by market forces. Prices for all vegetables, except beans and pepper, have at least doubled in current terms since appraisal, and more than tripled for citrus and bananas (see Annex III). This underscores the extent to which farmers in the Valley are able, with limited public sector extension services, to shift to new crops and improved farming practices in response to market opportunities. 5.04 There remains, however, much room for further yield improvement. The most serious constraint to agricultural production is plant disease, not only in the north, but in the whole Valley. Yellow leaf curl virus, which affects early tomato plants (principally in the south) has resulted in a loss of JD 4-5 million in tomato production this past cropping season./l If this disease could be controlled, tomato yields could nearly triple under ideal conditions. White spot disease is beginning to seriously hurt cucumber and squash production, and the problem of nematodes is growing throughout the Valley with the intensification of vegetable production. While improved plant protection through use of disease-resistant varieties and production of disease-free seedlings in private nurseries is practised, considerably more needs to be done in the areas of research, seedling management, improved use of insecticides, and crop rotation. 5.05 A second serious constraint on further agricultural development is the high cost of farm labor. Farm wages currently range from JD 0.5 per hour (roughly US$1.50) for Jordanian laborers to JD 0.4 plus subsistence (housing and food) for foreign workers, mainly Egyptian. This is 2.5 to 3 times the appraisal estimate (JD 1.0 per day for hired labor). The abundant employment opportunities for semi-skilled labor available in the service sector (and in Saudi Arabia and the whole Gulf region) have raised wages throughout the economy and farm labor costs have risen accordingly. Even at these wages, it is difficult to recruit agricultural workers from within Jordan and farmers have increasingly relied on unskilled foreign laborers, now estimated at 3,500, or 10-15% of the Valley's agricultural work force. /1 According to preliminary estimates from Ministry of Agriculture. The JVA reports the correct figure is JD 3 million. See Annex III to the PPAR. - 40 - 5.06 Total labor supply in the Valley appears to be considerably below the requirements of intensive agriculture. Based on preliminary figures from the 1978 census (which covers only the fall cropping season), there were an estimated 21,500 agricultural workers employed, including unpaid, mainly family workers. There are thus less than 1.3 workers for each ha of irrigated farm-land, which, assuming 300 days of labor per worker per year, is roughly two-thirds of the labor employed in open field vegetable cultivation under similar conditions in Tunisia, for example The most noticeable effects of the relatively high cost of farm labor in the Valley with its consequent low labor inputs in production are poor weed control, inadequate land preparation, and poor crop husbandry in general. Further improvement in yields will require increased mechanization of certain tasks (land preparation, spraying of herbicides and insecticides, and harvesting). Scarce labor would then be available for such tasks which cannot be performed by mechanical means (weeding and seedling preparation and management), as well as for those associated with the new technologies of drip irrigation and cultivation under plastic. The high cost of farm labor should also favor a progressive shift to labor-saving irrigation techniques such as sprinkler (see para. 5.11). 5.07 Major shifts in cropping patterns in the project area are unlikely, because of a continuing strong demand in regional export markets for off-season fruit and vegetables. The present cropping pattern, including that projected for the new lands, is likely to stabilize at the present distribution of fruit and vegetables crops since further increases in citrus production will be limited by availability of irrigation water. (JVA has suspended granting new licenses for new citrus plantings.) There may be some shift to grape production, judging from the enthusiastic farmer response to a Ministry of Agriculture seedling distribution program in 1980. Grapes, which require considerably less water than citrus, would be a suitable substitute where there are restrictions on new citrus plantings. Rate of Adoption of Alternative Irrigation Techniques 5.08 At the time of the completion mission, only a small fraction of the project area farmers were utilizing the pressure pipe system provided by the project. Operation of the system had not yet begun in the new lands due to delays in land distibution, and only 38 of 297 farp units in the converted lands were irrigated from the pipe system in 1980,L all for sprinkler irrigating citrus. On-farm portable sprinkler equipment purchased from JVA with ACC credit under USAID's Sprinkler Equipment Project was being used on 25 of these units. The remainder had been equipped with permanent (buried) sprinkler sets which landholders had imported directly from Cyprus. 5.09 The evident reluctance of the farmers, not only in the project area but throughout the Valley /2 to adopt sprinkler appears to stem from fears /1 JVA advises (see PPAR Annex 3, para. 4) that, as of May 1982, 220 out of 315 farm units had purchased sprinkler sets. If these are in fact being used, this shows substantial progress in the rate of adoption. /2 None of the farmers in the 18-km extension project financed by USAID to the south of the project area have elected to irrigate by sprinkler from the pipe system; it is estimated that at least half of the farm units have shifted to drip systems after starting with short-furrow irrigation during the first year. - 41 - that sprinkler irrigation would adversely affect the yields and quality of their vegetable crops. While research findings are far from conclusive, there is some evidence that sprinkler can damage tomatoes, cucumber, squash and melons and increase plant disease and weed growth. (See Annex IV). It is not surprising that, with the exception of citrus growers, farmers in the conversion lands of the project area have chosen to continue to irrigate by surface from the canal. 5.10 At the time the completion mission was in Jordan, JVA decided to cut off distribution of water to the conversion lands through the open canal system, obliging the farmers to irrigate from the pipe system starting from iid-March 1981. It is likely that strict enforcement of this precipitous decision cannot be achieved as it will provoke an adverse reaction from farmers who are not prepared to shift suddenly from one method to another. A certain flexibility during the transition will be needed to permit them to obtain portable equipment from JVA or other on-farm equipment such as solid or permanent sprinkler equipment, a hose system for drip or gated pipes, and to make necessary credit arrangements. 5.11 Predicting future rates of adoption of sprinkler irrigation in the project area is difficult. Since the rate of discharge from the pipe system is very lowL!, labor requirements for surface irrigation are very high, at least double those for drip and sprinkler. Due to labor shortages, farmers will effectively have to choose between drip and sprinkler. Sprinkler is most likely to be adopted for citrus, which represents over 40% of the conversion lands and a projected 30% of the new lands. Given the high capital cost and relatively high labor requirements of drip, portable sprinkler systems may also be adopted for some vegetable crops, especially on the new lands where steeper slopes would discourage extensive use of costly drip irrigation equipment. Drip systems will, however, be needed for tomatoes, squash, cucumbers, and melons, which do not thrive under sprinkler. 5.12 In retrospect, the decision to install a new collective pressure pipe system for sprinkler irrigation in the conversion lands as the least cost alternative for improving water efficiency and production can be questioned. First, the existing canal distribution system is in good shape and reasonably well operated. Overall system efficiency appears to have improved in recent years simply through adjustments in JVA's operational procedures and better irrigation practices by farmers themselves. With the goal of water saving and reducing operating costs in mind, an alternative approach utilizing the existing investment in conveyance facilities would have offered substantial investment cost savings. This approach would have been to continue using the existing surface canal distribution system while building certain features (small reservoirs) and installing more adapted water control equipment to improve /1 4 1/second, which could be increased by some 60-80% when water is delivered directly at the farm hydrants. - 42 - water management flexibility in water distribution. Furthermore, farmers desiring to use sprinkler or drip irrigation for specific crops and when convinced of the values of these methods could install individual pumps. An economic comparison of full collective and individual conversion alternatives indicated that the individual conversion alternative enjoys a saving of initial public works investment of about JD 950 per ha (US$3,180) and an annual operating cost savings of JD 12 (US$40) per ha at a minimum. Another reason for concern is the design of the pressure pipe system for optimal use of sprinkler which therefore limits its adaptability to a surface irrigation method. In view of the relative lack of data on the suitablity of sprinkler for the range of crops to be grown in the project area, the selection of a relatively inflexible design carries a certain risk. 5.13 This issue is a major one facing JVA as it considers the conversion of another 12,000 ha of lands now being surface-irrigated from the canal network. These considerations were addressed during the appraisal of the Jordan Valley Stage II Project and alternative methods of progressive conversion from surface to pressure irrigation were proposed by the appraisal team. In view of the widespread adoption of drip irrigation in the southern portions of the Valley, and the evident problem of sprinkler irrigation for some crops, the team proposed that the concept of a generalized sprinkler system be dropped in favor of a more flexible irrigation system in which the decision whether to use surface, drip, or sprinkler would be left to the farmers. VI. INSTITUTIONAL PERFORMANCE Project Management - Jordan Valley Authority 6.01 The Jordan Valley Commission (JVC), in charge of the economic and social development of the entire Jordan Valley, was established under a 1973 law when the project was appraised. The appraisal report proposed that JVC would receive project funds and act as the management agency in exercising financing control over project funds, but responsibility for construction and operation of project works would be assigned to the respective technical ministries: (i) the Natural Resources Authority (NRA) for construction and O&M of the irrigation system; (ii) the Ministry of Agriculture for research and extension; (iii) the Agriculture Marketing Organization for the management of marketing centers; and (iv) ACC for the credit component. These provisions with the exception of (iv) were not fulfilled as a result of changes in project scope and institutional jurisdictions. 6.02 Under Law No. 18 of 1977, one year after the start of the project, JVC became officially known as the Jordan Valley Authority (JVA) and entrusted with the total development program for the Valley, including design and construction of all of agricultural and service infrastructure. It also assumed certain responsibilities for research and marketing - 43 - activities. JVA's expanded staff needs were met by incorporating the NRA irrigation staff, recruiting staff for the Social Development Directorate, and seconding Ministry of Agriculture staff to assist in research and planning. JVA, however, is obliged by law to hand over the public services infrastructure to the relevant national organizations two years after completion of construction and is prepared to do so. Because of the increase of JVA responsibilities, the Liaison and Advisory Committee (LCA), contemplated at appraisal to coordinate project activities among executive agencies, was no longer necessary and was never convened. 6.03 JVA's performance in managing the project was mixed. It was efficient and thorough in carrying out its responsibilities in contract supervision and was able to quickly build up a capable engineering staff. It was also highly successful in making financing arrangements with other donors to accommodate the successive changes in the scope of this project. However, it was deficient in its technical preparation of the project, /1 which delayed commencement of works, and did not maintain detailed separate financial records on the different aspects of the project. It was also never fully able to assume much of a role in agricultural research, a responsibility which, in retrospect, it was perhaps unrealistic to have entrusted to JVA. Operation, Maintenance, and Water Management 6.04 The JVA's Directorate of Irrigation is responsible for the O&M of the surface and sprinkler system. The 0&M Division is based at Deir Alla in the Jordan Valley and has a staff of about 1,150 people. As shown in the following table comparing JVA's staffing and O&M costs with other irrigation projects, these costs are high. /2 Ha per O&M Farm Turnouts O&M Costs Per Employee Per Ditchrider 1000 m3 Water Delivered (us$ ) Jordan Valley 18 13 30.18 Tunisia (Lower Medjerda) 66 12 22.40 Algeria (Bas Cheliff) 43 10 24.20 California (4 gravity supply irrigation 660 120 4.10 districts) Morocco (Doukkala) 350 60 n.a. Nevertheless, given the operational requirements of the system and the flexibility it affords farmers, JVA's O&M Division has achieved a reasonable standard of operational efficiency. One of the reasons given by /1 The JVA has observed that its preparation of irrigation works in the Wadi Arab was delayed because their design depended on a decision pending at the time as to whether the Stage II project would be undertaken. See Annex I to the PPAR. /2 Source: Staff Appraisal Report of the Algeria Bas Cheliff I Irrigation Project. -44- JVA for its high staffing needs is the time required to adjust and check several times a day the settings of the Constant Head Orifice (CHO) gates in order to deliver a constant flow into the farms/l. Overall efficiency of the EGMC and lateral canals, as measured, varies between 75 and 80%. 6.05 In addition, JVA provides a flexible water supply service to accommodate farmers' irrigation needs. The basic water scheduling system is a weekly rotation system with fixed frequency and variable flow and period of delivery. The farmers may request up to a maximum of 25 1/s during 6, 12, 18, or 24 hours. During the seedling period, water is delivered twice a week. Time of delivery is also changed every month for reasons of equity between farmers to avoid a situation where any one farmer has always to irrigate during night time or the warmest hours. During water shortage, water demand is readjusted to the prorata share of area planted per farm unit with priority given to licensed permanent tree crops. 6.06 JVA expects a substantial increase in O&M efficiency in the future. It has recently taken measures to improve efficiency by providing transportation to the staff, improving farm access roads along the distribution system, and using wireless communications. More important savings will come from shifting from surface to pipe system. After completion of pipe system serving 9,400 ha (over the last two years), the irrigable surface increased by 63% from about 13,300 ha to 21,700 ha; the O&M staff has, however, only increased by 15%. The number of farm units a ditch tender can supervise with the pipe system is reported to have increased threefold compared to the surface system. Although the present systems are far from being fully used, JVA does not intend to increase O&M staff. For the more distant future, JVA is planning to improve efficiency of the main system by providing automatic operation of the main canal. 6.07 Considerable data are available in the O&M Division for evaluating the efficiency of on-farm water management, but these have not been analyzed in recent years. The last detailed assessment of actual on-farm efficiency was made by the NEDECO consulting firm in 1966 which resulted in an overall figure of 45%, and an estimate of water consumption of 15,830 m3/ha. At appraisal, the common method of surface irrigation was the zig-zag method which was reported to be very inefficient. It is evicent that new cultivation and irrigation practices have dramatically improved water efficiency. The zig-zag method has now been largely abandoned for furrow irrigation. The 3-year drought between 1976 and 1979, and the extension of irrigated areas resulting in lowest flow deliveries for many years have also contributed to improving farmer efficiency and made them aware of need for water savings. The table below gives a rough indication of the increase in water use efficiency. Between 1974 and 1980, water use per hectare increased by only 15% over the period, while overall intensity /1 The Neyrpic type distributors largely used in North African countries are precalibrated and do not require any adjustments if upstream water level remains within certain limits. Operation staff can be substantially reduced by use of these simple devices. - 45 - increased by nearly 50% and the share of cultivated area planted under citrus (which requires 5 to 6 times more water than vegetables crops) increased by 75%. During the same period, the annual value of water sold to farmers has incresed by 33%, while the cropped area has increased by 71%. Area Overall Water Sold Area Water Under Cropping Calendar Year To Farmers Irrigated Use Fruits Intensity (million m3) (ha) (m3/ha) % % 1974 75.7 12,000 6,300 12 /a 1.06 /a 1978 94.0 12,400 7,600 15 1.18 1980 101.0 13,800 7,300 21 /b 1.58 /b /a 1975 data. /b 1979 data. Cost Recovery 6.08 A water charge of 3 fils per m3 is charged to the farmers for water consumed on-farm. For the surface system, JVA's O&M field staff reports daily on the amount of water distributed based on the flow rate fixed at the farm gate which is verified and adjusted by field checks two to three times per irrigation. These figures are then checked against the cropping pattern for each development unit. For farmers using the pipe system, volumetric meters are used; where these are inoperative, JVA bases its consumption estimates entirely on the cropping pattern, applying a formula which is reported to be 85-90% accurate in calculating actual water consumption. 6.09 Farmers are billed monthly, with payment due in 45 days. Water is shut off if payment is not received. JVA reports that its collection rate is close to 100%. It attributes this success to an efficient collection system, the effective sanction of cutting off water, and to the fact that water charges represent such a small proportion of total production costs (4% as compared to 15% in the Morocco Sebou Project) /1. 6.10 At the current rate of 3 fils per m3, a 100% collection rate covers roughly 35% of estimated O&M costs. In 1980, JVA's O&M budget was approximately JD 910,000; total water charges billed in the Valley (including penalty fines for stealing water and charges for special /1 There is, however, a consistent pattern of arrears over part of the year, as farmers often delay payment of water charges after October when the rainy season begins, and do not pay outstanding charges until March, when they need to begin irrigation. - 46 - repairs) were JD 327,000. When fixed at 3 fils in 1974, this rate covered nearly all of JVA's O&M budget. But while that budget has increased fourfold since that time, water charges have remained fixed. JVA fully recognizes the need to adjust to current water rate to cover at least 100% of O&M costs and in principle some proportion of investment costs. Its Planning Analysis section is now preparing a study of farmers' capacity to pay higher water charges to serve as a basis for a decision on an upward adjustment. 6.11 Since the Credit Agreement (Section 3.06) calls for JVA to collect water charges of no less than 6 fils/m3 in the first year of irrigation in the sprinkler-irrigated area, JVA will need to double these charges by the end of the 1981/82 cropping season, during which the pipe system will be in full operation in the project area. A further increase is likely to be necessary to comply with the Credit Agreement condition of full O&M cost recovery. In addition, the Credit Agreement provides for the Borrower to fully recover investment costs with 5% interest over 40 years. Given the substantially higher capital costs of the system than were anticipated at appraisal, JVA is actively considering additional recovery methods, including a betterment levy of JD 170/ha to help cover costs of the farm main pipes /1. Agricultural Support Services 6.12 Credit. ACC has shown an impressive capacity to keep pace with the strong growth in credit demand arising from the rapid intensification of agriculture in the Valley. Its total lending level (for both seasonal and investment credit) has jumped from JD 2.25 million in 1978 to JD 4.0 million in 1979 and JD 4.6 million in 1980. While much of this increase reflects recent inflation rates, some is attributable to the increase in on-farm investments. Medium-term credit as a share of total credit extended in the northern Valley increased from 69% in 1974 to over 90% in 1980, and the average size of medium-term loans increased from JD 1,362 to JD 7,368 in that same period. ACC reports that its overall collection rate ranges between 80-85% for the Valley, and somewhat higher for seasonal credit. ACC foresees no difficulty in meeting continuing strong demand for investment credit in the future with continuing capital contributions from the GOJ budget and prospects of a third loan from the KFW. 6.13 ACC's current credit policy requires, in most cases, land as collateral. This effectively excludes sharecroppers and tenant farmers from access to ACC credit. ACC recognizes the need to expand services to these groups and is reportedly looking to JVFA to perform this role. To this end, it has recently undertaken to make a loan to JVFA at 4 1/2-5% to be used to make in-kind seasonal credit available to JVFA clients at rates comparable to ACC terms. This loan would be repayable in 18 months, although it is likely that ACC would continue to provide JVFA with additional resources on a revolving fund basis. /1 In addition to the JD 200/ha levy which is charged for "irrigation rights" on new lands being brought under irrigation. - 47 - 6.14 Research and Extension. Performance in research and extension has fallen far short of the needs for the Valley. There are currently 30 extension agents working in the Valley, including 26 from the Ministry of Agriculture and 4 from JVFA. On average, each agent is responsible for 700 irrigated ha, an exceedingly low ratio for intensive vegetable cultivation. (In a region in Tunisia with comparable production patterns, the ratio is 1 agent for 50 ha.) It was reported that in the northern part of the Valley, Ministry of Agriculture agents rarely were able to make field visits more than 2-3 days each week, typically made only at farmers' requests. It is not surprising that the great majority of farmers in the Valley continue to rely almost entirely on technical advice received from commercial input suppliers. 6.15 In the area of research, the Ministry of Agriculture maintains its main station at Deir Alla, with a substation at Wadi Yabis serving the Project area. The Ministry's existing programs focus on variety testings, plant protection, and seedling distribution. The only research being done in water management is under the direction of JVA, which maintains a small research plot at the Ministry's Deir Alla station, and a separate station at Wadi Yabis. Under an AID-financed technical assistance project, JVA conducted a research program in water management technology which covered a wide range of technical practices. Of particular interest for the project were comparative tests of furrow, sprinkler, and drip irrigation with regard to water-use efficiency and yields and quality of major vegetable crops. Considerable emphasis was given to on-farm demonstration of alternative practices. Unfortunately, this program was terminated in early 1981, one year earlier than planned, amid much controversy over the validity of the trials on comparative on-farm irrigation systems, which showed yields for vegetables under sprinkler were consistently lower than under drip or furrow irrigation. Since there was no effective control of the amount of water applied to the plants in these tests, these findings are not considered meaningful. While other tests produced interesting and more valid results, little has been done to follow up this work in any systematic fashion. Agents in the Valley note that few of the participating farmers have in fact adopted any of the new practices which were successfully tested on their own farms. 6.16 A major issue which appears to have hampered the progress in research and extension in the Valley has been the ambiguity surrounding the institutional arrangements for research and extension activities. Since 1977, when the law expanding JVA's authority was issued, responsibilities for these activities were shared between JVA and the Ministry of Agriculture. While JVA's research function were limited to on-farm water management problems, this arrangement has not proved practical, in part because of JVA's own staffing limitations, and in part because of a failure to develop effective means to coordinate the research being done by the respective agencies. A proposal for reorganizing national agricultural research is being discussed with the Government within the framework of the Bank's agriculture sector work /1 . /1 Rainfed Agricultural Subsector Memorandum, IBRD, April 1981 - 48 - Jordan Valley Farmers' Association 6.17 The Jordan Valley Farmers' Association (JVFA), established in 1974 as contemplated in the SAR to provide selected services to Valley farmers, now operates out of six offices, including its headquarters office in Deir Alla, and plans to open three more branches in 1981. While its staff was increased from 30 to 60 last year, it is still far below the levels needed to provide the range and coverage of services originally planned. Its current membership stands at 4,000 farmers, or two-thirds of its target clientele of 6,000. 6.18 JVFA's principal service to farmers has been in input supply, where bulk purchases, in some cases directly from overseas (US) suppliers, has permitted JVFA to sell well under prices offered by local commercial suppliers. In fertilizers, for example, which account for roughly 50% of JVFA sales, JVFA prices typically range from 10-18% below commercial prices. JVFA input sales in 1980 totalled over JD 450,000, more than doubling 1979 sales of JD 200,000. (Since sales in both years were from existing stock, this increase reflects sales volume and not price increases.) 6.19 JVFA also provides seasonal credit to members under an AID project designed to broaden credit access to smaller farmers. The number of seasonal loans extended to members increased from 350 in 1978, to 432 in 1979, and to 650 in 1980. Even with this growth, JVFA's credit services are reaching only 16% of its total membership. While the opening of new branch offices and the injection of new lending capital from ACC (see para. 6.13) should encourage greater coverage, JVFA's credit program may be constrained by staffing limitations in capacity to process loans and supervise collection. The current collection rate is estimated at 77%. 6.20 JVFA has yet to assume any of the production marketing functions which were originally conceived as central to JVFA's role in the Valley due to delays in the planning and financing of the three new marketing centers in El Aarda, Wadi Yabis, and South Shuneh. The cause of much of this delay has been the extended planning period required for the design of these relatively sophisticated facilities. Once completed, these facilities will be operated by JVA for two years. It is not clear at this time whether the centers will then be turned over to JVFA, as originally planned, or to a joint public-private sector marketing board. JVA reports that the decision will depend to a large extent on an assessment of JVFA's capacity to take on the complex administrative tasks of operating the facilities. VII. ECONOMIC AND SOCIAL IMPACT General 7.01 Looking at the impact of the Stage I development program in the Valley, of which this project was a part, it is evident that the general level of development activity and standards of living have risen significantly. Non-agricultural employment increased by 90% between 1973 and 1979, mostly as - 49 - a direct result of public sector infrastructure investment. The percentage of paid labor in agriculture grew from 46% of total agricultural labor force in 1975 to 57% in 1978; the declining reliance on family labor has permitted family members to seek more remunerative employment elsewhere or to pursue higher levels of education. The rate of on-farm investment has grown markedly with the rapid diffusion of new irrigation and cultivation technologies, particularly drip irrigation and plastic houses and tunnels. Increased traffic and transport services point to growth in general commercial activity. 7.02 These measures of economic progress are paralleled by indicators of improved levels of social welfare. School enrollment increased by 61% between 1973 and 1979, and it is estimated to have increased by another 20-25% during the 1979/80 academic year. These increases are almost twice the rate of population growth over the period. Particularly striking are the 119% increase in secondary school enrollment and the relatively higher growth in enrollment of female students, evidence of upward mobility among second-generation residents. Literacy rates also showed significant gains, and the availability and quality of health services have improved considerably, including the provision of domestic water supply throughout much of the Valley. 7.03 These achievements in the provision of social services are all the more impressive considering the rate of population growth in the Valley. Between 1973 and 1979, the Valley's population grew from 64,000 to 87,000, an increase of 35%, well above the natural growth rates for the country. This fast population growth has been both the consequence of and necessary to the rapid growth in the Valley; agricultural production following the expansion of irrigated area under Stage I, and the substantial improvements in educational, health, and water supply facilities have undoubtedly helped to attract newcomers, not only in agricultural activities but also in related support services, both public and private. 7.04 Because the northern part of the Valley has benefitted from a longer history of irrigated agriculture, it enjoys somewhat higher living standards than the middle and southern parts of the Valley. Enrollment levels are higher, especially among females, illiteracy rates are lower, and there is less reliance on family labor. The relatively advantaged position of the north may, however, be eroding as the adoption of new technologies in the middle and southern areas has permitted higher cropping intensities and outstripped productivity gains in the north. This pattern can be expected to continue unless the north can begin to share in the productivity increases associated with improved technologies. Given the north's climatic disadvantage in competing with early-season production, the more expensive technologies (i.e., greenhouses and irrigation) are not likely to be as widely adopted in the north. The smaller farm size in the north may also inhibit investment in labor-saving equipment. Future research may need to consider the special needs of the northern farmers for lower-cost improvements in technical practices, and extension will need to be considerably strengthened to accelerate the rate of adoption of improved practices. - 50 - 7.05 An especially interesting aspect of development in the Valley is its impact on the structure of farm management and the agricultural labor market. With the upward mobility of the second generation of landowners and increasing labor requirements of intensified vegetable production, there has been a sharp increase in paid agricultural labor in the Valley, much of which is foreign labor. Together with the increased role of high technology in agriculture, management has become a more complex and critical feature of Valley farming. There has been, thus, a fast-evolving role for "professional" farm managers which is being filled by service sector employees from Amman and by recent immigrants typically from the West Bank who have brought with them extensive experience in irrigated agriculture. It is this trend that partly accounts for the increase in paid labor hired on a relatively longer-term basis in the Valley, and may be associated with the evolution of a more modern and efficient agricultural production system. 7.06 Parallel to this development is a change in the nature of sharecropping and tenant/landlord relationships in the Valley. Sharecropping, traditionally an important feature of Valley agriculture, has assumed new significance with the introduction of the new high-cost technologies since it permits sharing of risks and costs. Over 90% of the land where drip irrigation and plastic covers are being used in the Valley is being farmed under sharecropping arrangements. Given the shortage of skilled agricultural labor and the premium on experienced farm managers, sharecroppers may enjoy a higher degree of bargaining power than is typically assumed and participate actively in most production decisions. Sharecropping thus appears to be an efficient adaptation to the transition to modernized agriculture. This development may imply the need for more flexibility in institutional credit policy, which at present restricts availability of credit primarily to landowners who can pledge land as collateral. The Project 7.07 The economic impact of the project has been evaluated by re-calculating the rate of return to reflect changes in both project scope and assumptions concerning future production. Since the project as implemented did not include significant components for research and extension, credit, or supporting social and economic infrastructure, which were ultimately financed by other donors and the Borrower, only those benefits derived from the irrigation network itself were considered. The projection of benefits reflects a considerable degree of speculation since, at the time of the mission, only a small fraction of the farmers in the Project area were irrigating from the pipe system. 7.08 In projecting benefits, it was assumed that there would be no marked change in the market demand for Valley produce; therefore, the cropping pattern in both the conversion lands and the new lands would reflect the existing distribution of fruit and vegetable production. In the new lands, however, the cropping patterns assumes a somewhat lower cropping intensity, owing to rougher terrain, and a lower percentage of the area in citrus, reflecting limitations on irrigation water supply. (See Annex V for cropping patterns.) Substantial increases in yields were assumed in both the with and - 51 - without project cases reflecting relatively low base yields and the high production potential of the area. (See Annex V.) The projected yields at full development are substantially higher than those estimated at appraisal, reflecting the fact that current yields in the Valley are better than expected in the without project situation. The phasing of projected benefits assumed maximum yields are reached over 12 years, maximum cropping intensity on the new lands is reached over 5 years, and the shift to alternative on-farm irrigation systems (sprinkler and drip) is made over a 4-year period, depending on the crop and system selected (See Annex VI). The assumed shift to drip and sprinkler by all farmers is predicated on JVA's recent decision to cut off water from the canal system and the low discharge rate from the pipe system which will effectively preclude irrigating by surface methods (see para. 5.11). Actual investment costs of the irrigation system network were used in the analysis, including costs of the construction contract, consultancy services, and project administration. Replacement costs were included at 5% of initial investment for the first two years following construction, and 2% thereafter. Operation and maintenance costs were projected at JD 60 per ha. All costs and benefits were valued at 1979 prices. (See Annex VII for farmgate prices of agricultural production). 7.09 On the basis of these assumptions, the overall ERR is estimated to be 13%. This is a substantial reduction from the 29% return estimated at appraisal, reflecting the large cost overruns on the irrigation works, the delay in production benefits on the new lands, and the change in project scope by which the relatively low-cost high-yielding investments in land leveling and tile drainage on the non-converted lands were dropped from the project. Another important factor accounting for the lower ERR is the revised assumption concerning yields on the conversion lands in the without project case. While the SAR assumed annual yield increases of only 2% without the project, actual experience has shown substantially higher yield increases on the conversion lands due to factors which cannot be attributed to the project. Therefore, the revised ERR calculation has assumed that future yield increases on currently irrigated land would have occurred even if the investment in the pipe system had not been made, substantially reducing the incremental benefits of the project. Given this assumption, the nature of the project benefits on the conversion lands differ significantly from those on the new lands; a separate analysis was done for each. 7.10 New Lands. The economic IRR for the new lands (1,760 ha) was estimated at 16% showing the benefits of a shift from wheat to off-season vegetables and fruit on 730 ha, more intensive cultivation of vegetables on 530 ha now irrigated from the side wadis, and new cultivation of 500 ha. Bringing these lands under permanent irrigation permits an increase in overall cropping intensity from 68% /1 at present to 125%, and a tripling of the area in fruit production (largely citrus and grapes). /1 This figure is based on 1980 production data; it is unusually high given abundant rainfall in the past low cropping seasons which has permitted additional cultivation of rainfed wheat and a larger area irrigable from the side Wadis. - 52 - 7.11 Conversion Lands. As explained above (para. 7.09), the rate of return for the conversion lands (1,000 ha) assumed no change in cropping pattern or in yields as a result of converting from the surface system to the pipe system. Incremental benefits attributed to the project and considered in the analysis were: a savings in operation and maintenance costs from JD 90 to JD 60 per ha, a savings in irrigation labor requirements for sprinkler and drip systems compared to the existing surface system, and a 10% reduction in irrigation water consumption (combined off- and on-farm savings), valued as the net value of production on an incremental 10% of irrigable land. Based on these assumptions, the estimated economic rate of return for the conversion lands is 5%. Sensitivity tests showed the IRR was not sensitive to increased labor costs or to lower operation and maintenance costs, but was moderately sensitive to benefits from water savings. However, these benefits would have to nearly double to 20% water savings to obtain a 10% return. This would imply a combined irrigation system efficiency of 85%, which, while theoretically possible, is highly improbable. FAO reports that sprinkler systems in use in the US exhibit 60% overall water efficiency under hot and dry climate conditions. Researchers in the Jordan Valley have reported on-farm water losses alone of 10-15% due to wind and an additional 10% due to evaporation. 7.12 Impact on Farm Incomes. Since only a small fraction of the farmers were irrigating from the pipe system at the time of the completion mission, it is too early to make any meaningful projections of the impact of the project on farm family incomes. Much will depend on the selection of on-farm irrigation methods, the level of water fees to be charged, and the on-farm efficiency of water use, for which no information is yet available. Based on the assumptions made in the economic analysis, however, the net value of production on the new lands could be expected to increase five times for an average farm, and considerably more for farmers cultivating only rainfed wheat, showing the production benefits of increased cropping intensity and a shift to fruit and vegetable production permitted by year-round water availability. 7.13 The impact of the project on family incomes on the conversion lands will be limited to savings from reduced irrigation labor and water requirements, which will be insignificant in most cases. Fruit growers will benefit disproportionately; given the high water consumption requirements of fruits, the value of water and labor savings is correspondingly higher. Citrus farmers could realize total savings of up to JD 138/ha assuming a shift to permanent sprinkler and water charges at 6 fils/m3. This saving represents approximately 15% of total production costs. The savings for vegetable farmers would be considerably less, ranging from JD 4/ha for squash under drip (or less than 1% of total production costs) to JD 34/ha for eggplant under portable sprinkler. These savings on conversion lands are not adequate to cover the amortization costs of acquiring necessary on-farm irrigation - 53 - equipment /1. If water charges are increased to 12 fils/m3, the value of these savings would just cover amortization costs of sprinkler equipment. VIII. BANK AND CONSULTANT PERFORMANCE 8.01 The Bank's performance in this project must be viewed in the context of the favorable atmosphere which prevailed in Jordan after 1971 for economic development after years of uncertainty. This atmosphere encouraged the Bank to respond to the expectations of the Jordanian Government, as well as those of other donors, to take an active role in the Rehabilitation and Development Plan for the Valley. At a donor conference held in November 1972, the Crown Prince expressed the hope that the Bank would find the Jordan Valley Project (Stage I) ready for appraisal and would take the lead for the appraisal. Because other donors had already made commitments to other projects in the Plan, IDA financing was limited to the northern sector of the Valley. 8.02 The eagerness of the donors and the Government to act swiftly in an area of high potential and priority for Jordan led to a hurried appraisal of the North East Ghor, which was one of the least prepared in the Plan proposed by Government. According to an FAO/CP Cooperative Programme mission which visited Jordan in December 1972, the designs prepared by the Jordanian Government were not at feasibility standard. But the Bank staff did not consider this a major problem because the study prepared by the same consultant for the two USAID projects in the Valley was well advanced and could provide reliable estimates for the IDA-financed project. The project was promptly appraised in September 1973 without further preparation and without an updated comprehensive preparation report by FAO/CP reflecting design changes requested by JVA at time of appraisal. 8.03 Problems during project implementation reflected its lack of sufficient preparation resulting in: (i) the Borrower's request for substantive changes in project scope only two months after the loan was effective; (ii) an 18-month delay in starting physical implementation to prepare tender documents; and (iii) unneccessary postponment of land distribution procedures after completion of irrigation works. Had more time been allowed for a thorough project appraisal, cost overruns might have been reduced and the lag in project benefits on new lands could have been minimized. 8.04 The most serious question is why the Bank and JVA, together with all other parties involved in the project, did not question the agronomic suitability of sprinkler irrigation for the Valley and, consequently, supported the adoption of sprinkler for the conversion lands at appraisal. Only the water management and economic advantages of sprinkler were considered during preparation and design stages. While the steep slopes and shallow /1 Assuming 80% credit financing at 6% over 10 years, which are the terms for portable sprinkler equipment procurement under the USAID project. - 54 - soils of the project lands above the EGMC may have justified adoption of sprinkler on the new lands, regardless of its agronomic disadvantages, a more careful assessment of sprinkler may have discouraged its adoption on lands already being surface-irrigated from the EGMC. It is striking that even the irrigation experts recruited by JVA to review the design criteria and preliminary layout prepared by their consultants for all on-going Stage I projects did not raise any questions on the use of sprinkler irrigation on the crops proposed, particularly since many technical publications available at the time had discussed the disease and weed control problems associated with sprinkler. 8.05 Several factors may help to explain the failure to consider these aspects for the North East Ghor as well as for the three other Stage I projects. One is the lack of coordination among the responsible agencies for preparation of the project. The FAO/IBRD Cooperative Programme was in charge to assess the project's overall economic and financial justification as well as its agricultural and organizational aspects while JVA and their consultants were in charge of preparing the technical aspects of the irrigation and non-agricultural components. Second, the special advisor hired by JVA to review the detailed design was asked to examine only the effect of sprinkler irrigation on bananas, citrus and tall-stemmed crops and the wind effect. Third, the development of cultivation under plasticulture (greenhouses and small tunnels) which excluded the use of conventional sprinkler systems was not foreseen at appraisal. Fourth, the cropping pattern projected in the SAR included a significant share (nearly 15%) of cereal crops for which sprinkler is ideally suited. With the development of new export markets for off-season fruit and vegetables, cereal crops have virtually disappeared in the current cropping patterns. Some areas in the south are now nearly 100% planted with crops not well adapted to sprinkler. Part of the explanation may also lie in the strong engineering and construction-oriented organization of JVA during implementation stage and the weakness of the Agriculture Ministry of Jordan./l 8.06 In the later stages of the supervision of the North East Ghor Project and the preparation of the Jordan Valley Stage II Project, the Bank, however, has been instrumental in focusing attention of the Borrower on the potential problems of sprinkler and, more broadly, on the need to address the non-engineering aspects of irrigated agriculture through a more comprehensive research and extension program. The dialogue between the Bank and the Borrower was effectively continued in the context of the preparation of the Jordan Valley Stage II Project. 8.07 The consultant was diligent in performing engineering services along the design criteria set up by the Borrower. The responsibility of the consultant in the selection of the RPM pipes is unclear. The lack of technical data in the bid evaluation report does not indicate whether it had adequately investigated this relatively new technology. Nor is it clear whether either JVA or the consultant were aware of the special requirements for installing RPM pipes and took appropriate precautions during supervision of construction. The Bank, too, perhaps carries some responsibility in this matter, since, in its review of the proposed contract award, despite the new technology proposed, it did not ascertain of its feasibility and reliability. /1 Specifically its researach and extension program. (See Annex III to the PPAR). - 55 - In restrospect, the Bank should have requested a substantiated technical evaluation of the RPM pipes which JVA selected only on price comparison. This is crucial when a relatively unproven technology is proposed. IX. RECOMMENDATIONS AND LESSONS LEARNED 9.01 Since the project-financed irrigation works were not yet in service at the time of completion mission, it is premature to draw final conclusions about the project. Projections regarding agriculture production and rate of adoption of alternative irrigation methods are highly speculative. An impact evaluation of the project should be scheduled in about two or three years and certainly before the Bank considers financing of any follow-up project for the installation of pressurized pipe irrigation system on currently surface-irrigated lands. 9.02 The Borrower will need to increase water charges to 6 fils/m3 by the end of the 1981/82 cropping season (para 6.11). The Bank should, as soon as possible, review with JVA its proposals for adjustments in water charges and betterment levies now being studied. It should aim to seek JVA's compliance with the principle of full operation and maintenance cost recovery as stipulated in the Credit Agreement, while recognizing that a restructuring of water charges may require some time to implement. 9.03 Inadequate agricultural research and extension is presently the major constraint to the agricultural production in the Valley. The Bank, together with other donors should now aim to support an applied research program project in the Valley and to establish a dynamic and competent extension service. 9.04 This project illustrates several consequences of premature and incomplete appraisal. In particular, it demonstrates the drawbacks of appraising a major civil works project when preparation is carried out only to the feasibility stage. It is likely that present Bank requirements (OMS 2.28) for preparation of final engineering designs prior to project approval, would have resulted in a more rapid implementation of the project and a less substantial cost overrun. It also demonstrates the importance of giving careful attention to institutional and procedural aspects of key implementation actions. Because the procedures and practices for land distribution were not examined in detail, an opportunity was lost to make simple reforms which could have avoided the two-year gap between construction and production on the new lands. 9.05 More importantly, the experience of this project proves the difficulties of selecting an appropriate on-farm irrigation technology and the need to examine non-engineering aspects of such a selection. While it is premature to conclude that sprinkler irrigation was an inappropriate selection in the Northeast Ghor, two years of experience with the pressure distribution system in the south of the Valley and limited results of research on water - 56 - management point to the need for flexibility in selection of on-farm irrigation methods and system design, especially in the Jordan Valley where the farmers are quick to adopt innovative techniques and new crops in response to market opportunies. When it became evident by mid-1977 that Valley farmers were electing to irrigate by drip systems, the Bank suggested that JVA study the possibility of adapting the design for the irrigation outlets to permit low pressure irrigation as an alternative to sprinkler. In retrospect, it is unfortunate that implementation of four sprinkler projects covering about 9,500 ha started simultaneously without the benefits of a small scale pilot project demonstrating the value and the problems associated with this irrigation method A priority project to be developed in the Jordan Valley should consist of a research and extension project, encompassing water management, irrigation methods, weed control, and plant protection.  - 57 - ANNEX I KINGDOM OF JORDAN NORTH EAST GHOR IRRIGATION AND RURAL DEVELOPMENT PROJECT PROJECT COMPLETION REPORT Implementation Results SAR Actual or Expected (including those not financed under the IDA Project) I. IRRIGATION WORKS Area Served by Pipe System (net) - New Lands (ha) 1,760 1,778 - Conversion Lands (ha) 1,000 1,030 Farm Units Using Sprinkler (%) 100 12 Improvements in Surface-Irrigated Area - Land Levelling (ha) 3,000 - - Tile drainage (ha) 520 130 Ia II. FARM ROADS - New (km) 30 (gravel) 79 Ia (seal-coat) - Improved (km) 60 (gravel) 90 a (seal-coat) III. RURAL DEVELOPMENT INFRASTRUCTURE - Marketing Facilities 2 assembly markets 1 wholesale market /a - Water Supply Systems 10 11 /a - Health Centers 3 new, 1 improved 4 new /a - Schoolrooms 150 208 /a - Vocational Training Center 1 1 /a (in planning stage) - Community Centers 10 8 /a (in planning stage) IV. PROJECT BENEFICIARIES Project Area Population 25,000 32,300 Farms Served 1,800 holdings 748 farm units /b /a Components not actually financed under IDA Project /b Actual number of farm householders being served is not yet available pending completion of land distribution on new lands. (0593E) - 58 - ANNEX II KINGDOM OF JORDAN NORTH EAST GHOR IRRIGATION AND RURAL DEVLOPMENT PROJECT (Credit 498-JO) PROJECT COMPLETION REPORT Comparison of Schedule of Disbursement IDA Fiscal Year At Appraisal Actual and Quarter (uS$ '000) 1974/1975 Sept. 30, 1974 40 - Dec. 31, 1974 200 - March 1975 480 - June 1975 1420 - 1975/1976 Sept. 30, 1975 2220 - Dec. 31, 1975 3440 - March 1976 3970 100 June 1976 4550 100 1976/1977 Sept. 30, 1976 5390 300 Dec. 31, 1976 5800 400 March 1977 6390 800 June 1977 6850 1000 1977/1978 Sept. 30, 1977 7050 2600 Dec. 31, 1977 7350 3600 March 1978 7420 3900 June 1978 7470 4100 1978/1979 Sept. 31, 1978 7500 4310 Dec. 1978 7500 4960 March 1979 7500 5540 June 1979 7500 5540 1979/1980 Sept. 31, 1979 7500 5770 Dec. 1979 7500 5890 March 1980 7500 6550 June 1980 7500 6590 1980/1981 Sept. 30, 1980 7550 ANNEX III - 59 - KINGDOM OF JORDAN NORTHEAST GHOR IRRIGATION AND RURAL DEVELOPEMENT PROJECT PROJECT COMPLETION REPORT Comparative Production Estimates Cropped Area Yields (% of Total) (t/ha) SAR /a Actual (1980) lb At Appraisal Actual (1977779 Current Projected Average Highest Field Crops Wheat 4.3 4.1 2.5 3.5 1.2 2.1 Maize 10.4 - 4.0 8.0 - - Subtotal 14.7 4.1 Vegetables Tomatoes 19.8 5.4 17 25 19.4 24.6 Eggplant 15 23 19.1 26.0 Pepper 20.7 22.2 15 23 19.1 26.0 Cucumbers 10 20 14.3 18.0 Squash 9.3 6.6 10 20 14.4 15.6 Cauliflower/Cabbage 5.7 8.3 12 23 17.3 23.8 Onion 2.2 12 23 13.1 15.4 Potatoes 5.3 2.0 12 17 6.7 7.1 Beans 4.1 1.6 12 15 6.0 6.2 Other - 8.1 Subtotal 69.0 56.4 Fruits Citrus 12.5 35.6 17 28 14.5 17.6 Bananas 3.8 3.6 15 22 14.1 30.0 Other - 0.3 Subtotal 16.3 39.5 /a At full production in 1983 /b In conversion lands only. (0596E) - 60 - ANNEX IV KINGDOM OF JORDAN NORTH EAST GHOR IRRIGATION AND RURAL DEVELOPMENT PROJECT (Credit 498-JO) PROJECT COMPLETION REPORT Suitable Methods of Irrigation for Different Crops Tomato Drip system or surface. Tomatoes respond well to a constant water supply, drip reduces water stress, gives a higher yield, improves quality of fruit. Sprinkle irrigation is not a suit- able method because it increases foliar diseases, (early and late blight), reduces fruit setting, washes off insecticides and fungicides, will produce cracked fruit. Eggplant Surface is the preferred method. Drip is used if an early pro- duction using mulch is required. Sprinkle will cause disease problems, will dirty the fruit, will create difficult conditions at harvesting time when the soil is wet. Pepper When plastic mulching is applied, it is better to irrigate with drip. Drip will give the same advantages as for tomatoes. Sur- face can be a suitable method. Sprinkle will increase foliar diseases and dirty the fruit. Cucumber under tunnel and in greenhouses Drip will be the preferred method. Sprinkle will increase foliar diseases especially downy mildew and make the fruit dirty. Potato Drip is too expensive. Surface is the preferred method because it is inexpensive as the plants have to be hilled up making the furrows, so the same furrows can be used for irrigation. Sprinkle can also be used. Squash Surface is the most common system because less expensive. Drip will be used in existing system already installed for another crop. Sprinkle will increase foliar disease and require more pest control. -61 - ANNEX IV Page 2 Watermelon Surface will be the preferred method because less expensive. Drip can be specifically used for early production. Sprinkle will increase diseases, wet the fruit and decrease the quality. Cabbage/Cauliflower They are indifferent to the different methods. Surface is the cheapest method. -Where drip or sprinkle are installed for other crops, they can be used. Sprinkle has the disadvantage of making the heads rot. Broadbean It is a low return crop. It is advisable to irrigate them with the cheapest method. They can be grown under sprinkle and drip. Greenbean Surface is the preferred method. Sprinkle will increase diseases and dirty te pods. Drip will not give the same advantages as for tomatoes. Onion Surface will be the preferred method. Sprinkle will also be a suitable method. Citrus Sprinkle is the preferred method because it is water and labor saving. Surface and drip can also be used. Banana Drip is the preferred method because it avoids water stress, gives fruit of better quality. Grape Drip and sprinkle are the preferred methods because this crop is generally grown on slopes. - 62 - ANNEX V KINGDOM OF JORDAN NORTH EAST GHOR IRRIGATION AND RURAL DEVELOPMENT PROJECT (Credit 498-JO) PROJECT COMPLETION REPORT Agricultural Production Projections Used in Economic Analysis Cropping Patterns Yields (t/ha) Conversion Lands New Lands (1_760 ha) Current Projected /c Crop (1,000 ha) Current Projected /b Area (ha) % /a Area(ha) % /a Area(ha) % /a Citrus 435 43.5 190 10.8 530 30.1 15 35 Bananas 45 4.5 - - 10 0.6 22 30 Grapes - - - - 90 5.1 15 Other 5 0.5 30 1.7 Sub-total - Fruits 485 48.5 220 12.5 630 35.8 Eggplant 265 26.5 120 6.8 490 27.8 19 35 Cabbqge/CauLifLower 105 10.5 60 3.4 200 11.4 16 29 Squash 65 6.5 50 2.8 230 13.1 10 24 Tomatoes 65 6.5 35 2.0 215 12.2 17 43 Potatoes 25 2.5 15 0.9 70 4.0 12 29 Beans 20 2.0 10 0.6 45 2.6 6 12 Cucumbers 15 1.5 10 0.6 45 2.6 14 28 Melons 5 0.5 - - 70 4.0 17 25 Other 125 12.5 10 0.6 45 2.6 10 20 Sub-total - Veg 690 69.0 310 17.7 l4 80.3 Wheat (I) 500 50.0 285 16.2 - - 1.2 1.6 (R) - - 445 25.3 - - 0.8 1.5 Uncultivated - - /d /d Total Cropped Area 1,675 167.5 1.20 71.7 24 16.1 Cropping Intensity 1.44 .67 1.25 (excl. perennials) /a % of total cultivable area /b To be reached over 5 years 7c Full development over 12 years. d 500 ha of new lands are presently uncultivated - 63 - ANNEX VI KINGDOM OF JORDAN NORTH EAST GHOR IRRIGATION AND RURAL DEVELOPMENT PROJECT (Credit 498-JO) PROJECT COMPLETION REPORT Assumptions Used in Economic Analysis Note on Shift to Alternative on-farm Irrigation Methods 1. Since only a fraction of the farmers in the project area were irrigating from the pressured pipe system at the time of the completion mission, the projections in the economic analysis had to incorporate assumptions concerning the selection of alternative on-farm irrigation methods after the pipe system is under full operation and farmers in the conversion lands are no longer able to irrigate from the canal network. The assumptions made were based largely on agronomic and engineering data collected by the Bank's 1979 appraisal mission for the Stage II Jordan Valley Development project. On-farm irrigation investment costs and estimated labor savings considered in the IRR reflect a simplification of the expected trends, which are summarized below: (a) All citrus orchards will be irrigated by sprinkler, which is known to be well suited to citrus. Farmers will select the permanent (buried) system for mature orchards, where it is difficult to move the pipes under a portable set-up. Portable sets will be used for new plantings given their considerably lower costs. (b) Drip systems will be used on those vegetables which are susceptible to damage under sprinkler irrigation, namely tomatoes, squash, cucumbers, and melons. While drip systems are expensive (see para. 2 below), they are well adapted to additional relatively low-cost farming improvements--such as mulching, raised-bedding, and application of soluble fertilizer and other chemical treatments--which can substantially increase yields and reduce labor required for weeding and chemical spraying. The assumed rate of adoption of drip irrigation for these crops in the analysis is considerably more rapid than might otherwise be expected simply because, with the exceedingly low water discharge available from the pipe system in the project area, surface irrigation will not be practical and the farmers will effectively have no choice but to shift to drip system. This system is now widely adopted in the southern part of the valley. (c) All other vegetables--principally eggplant, cabbage, cauliflower, beans, and potatoes--are assumed to be irrigated under portable sprinkler systems, which is the least costly alternative to surface. - 64 - ANNEX VI Page 2 2. Capital costs of these alternative on-farm irrigation investments are based on estimates provided by JVA's Engineering Division and expressed in 1979 prices. The permanent (buried) sprinkler system is costed at 1,580 JD/ha, with an assumed life of 20-25 years. The portable sprinkler system (2 lines for a farm unit of 3.8 ha) is costed at 340 JD/ha, with a life of 8 years. The drip system, including a filter, is costed at 1,230 JD/ha; the filter is assumed to last 10 years, while the pipes and nozzles are assumed to be replaced every fourth year. Savings in irrigation labor requirements when shifting from unimproved surface with a 20 liter/second discharge to these alternative systems generally fall within the range of 50-80%. The greatest savings--with a nearly 100% reduction in irrigation labor requirements is for citrus under permanent sprinkler. For squash and cucumber, however, assumed to be under drip labor requirements, increase somewhat (see attached table). (0588E) KINGDOM OF JORDAN NORTH EAST GHOR IRRIGATION AND RURAL DEVELOPMENT PROJECT PROJECT COMPLETION REPORT Reduction in Irrigation Labor Requirements /1 Crop Labor Requirements (hrs/ha) Labor Savings Unimproved Surface at 20 1/s Portable Sprinkler Permanent Sprinkler Drip Hrs/ha Reduction Citrus 272 55 5 217/267 80/99 Bananas 400 96 304 76 Grapes 208 69 139 67 Tomatoes 96 53 43 44 Cn Squash 40 46 (6) Cucumbers 40 46 (6) Melons 144 60 84 58 Other 128 58 70 54 Eggplant 80 19 61 76 Cabbage/Cauliflower 40 12 28 70 Beans 32 10 22 70 Potatoes 80 19 61 76 Other 80 18 62 78 /1 Based on data collected by Bank appraisal team for proposed Stage II Jordan Valley Development project. (0589E) M t1 - 66 - ANNEX VII KINGDOM OF JORDAN NORTH EAST GHOR IRRIGATION AND RURAL DEVELOPMENT PROJECT PROJECT COMPLETION REPORT Farm-Gate Prices used in Economic Analysis /1 (JD/ton) SAR PCR Wheat 30 60 Tomatoes 30 73 Greenbeans 82 179 Eggplant 24 51 Potatoes 34 75 Pepper 40 64 Cauliflower 25 64 Squash 28 68 Cucumbers 49 156 Beans 48 62 Onions 26 52 Oranges 30 91 Bananas 62 192 Grapes - 109 Melons 71 /1 Weighted average of prices for summer and winter production seasons, based on average wholesale market prices over 1971-73 for the SAR and 1978-79 for the PCR. (0596E) KINGDOM OF JORDAN NORTH EAST GHOR IRRIGATION AND RURAL DEVELOPMENT PROJECT (Credit 498-JO) PROJECT COMPLETION REPORT ACC - Medium-Term and Seasonal Credit (in JD) 1975 1976 Jan-June '77 July 77-May 78 Total Medium-Term Loans Amount % Amount % Amount % Amount % Amount % Tree Planting 27,273 33.5 28,880 26.2 3,760 13.0 16,140 19.4 76,053 25.1 Land Improvement 30,665 37.7 46,345 42.1 9,400 32.6 19,925 23.9 106,335 35.0 Farm Machinery 15,100 18.6 18,600 16.9 14,500 50.2 14,875 17.9 63,075 20.8 Plastic Tunnels 2,300 2.8 3,500 3.2 - - 24,800 29.8 30,600 10.1 Farm Buildings 6,020 7.4 6,700 6.1 1,200 4.2 2,620 3.1 16,540 5.5 Pipes & Canal-Lining - - - - - - 4,875 5.9 4,875 1.6 01 -4 Poultry Farms - - 6,000 5.5 - - - - 6,000 2.0 I Total 81,358 100.0 110,025 100.0 28,860 100.0 83,235 100.0 303,478 100.1 Average Loan Size /l 1,432 2,124 2,575 /a 2,412 /b 1975 1976 1977 1978 1979 1980 Seasonal # Average # Average # Average # Average # Average # Average Loans /I Amount Loans Loan Amount Loans Loan Amount Loans Loan Amount Loans Loan Amoy Loans Loan Amoun Loans Loan 50,554 103 491 36,510 88 415 32,865 112 293 56,530 113 500 54,225 92 589 45,780 60 763 jl For all loans extended from Wadi Yabis branch office, which includes Krisech, outside the projeek area. /a For 1977 jb For 1978  _____BRD 15734 (PCR) kORDAN s Y R A NORTH EAST GHOR IRRIGATION & RURAL DEVELOPMENT PROJECT New Lands Not Converred and Srage ] Lands Carer ano Main oelmres - Prorect Arec Bounda ry -- -e-r -- - E-st Gh.r Cano - R ) Darnr and, Resera Wel] -ORTHMo-M Yarmo. k Ded Sea Ro-d --Orner Roods - C(aoor1 Me'ers SoNreh - nre-naoal B odare ,n l . MBrsh.y NEW LANDS - r-gated 720 ha Rarfed 1,40 h CONVERSICN ýANDS 0 ha NOT CONVERTED LANDS 4940 ha 770 h, 0 m M 25Kiy 32°31, T J R K E Y 30K- sesR y~~~ y R__ _l__ _A 35 R EEe J U RD A N.. 3 3 -, -EGyP .Y 51 WAD DRAB1 ARAB 5 lAA IRE*1. '-C y .