Turkey - Participatory Privatization of Irrigation Management and Investment Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No: 31391 IMPLEMENTATION COMPLETION REPORT (SCL-42350) ON A LOAN IN THE AMOUNT OF US$20 MILLION TO TURKEY FOR A PARTICIPATORY PRIVATIZATION OF IRRIGATION MANAGEMENT AND INVESTMENT PROJECT February 10, 2005 Environmentally and Socially Sustainable Development Unit Europe and Central Asia Region This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EQUIVALENTS (Exchange Rate Effective September 30, 2004) Currency Unit = USD TL 1,000,000 = US$ 0.833 US$ 1 = TL 1,200,000 FISCAL YEAR January 1 December 31 ABBREVIATIONS AND ACRONYMS ACC Assistant Consultant Coordinator CC Consultant Coordinator DSI Devlet Su Isleri (State Hydraulic Works) GDP Gross Domestic Product GOT Government of Turkey GDRS General Directorate of Rural Services Ha Hectares ICB International Competitive Bidding IIT Irrigation Investment Transfer (including management of irrigation investment) IMT Irrigation Management Transfer to users IRR Internal Rate of Return IC Irrigation Cooperative (GDRS) IS International Shopping LTCs Long-term Consultants M Million NCB National Competitive Bidding NS National Shopping OCC Opportunity Cost of Capital OECD Organization for Economic Cooperation and Development O&M Operation and Maintenance PIM Participatory Irrigation Management PPII Participatory Privatization of Irrigation Investment (including management of Investment) PPIMIP Participatory Privatization of Irrigation Management and Investment Project PRP Pilot Rehabilitation Project RC Regional Coordinator SCF Standard Conversion Factor TA Technical Assistance TOR Terms of Reference WUA Water Users' Association (DSI) WUO Water Users' Organization (includes WUAs, ICs and village authorities) Vice President: Shigeo Katsu Country Director Andrew N. Vorkink Sector Manager Joseph Goldberg Task Team Leader/Task Manager: Giuseppe Fantozzi  TURKEY PARTICIPATORY PRIVATIZATION OF IRRIGATION MANAGEMENT AND INVESTMENT PROJECT CONTENTS Page No. 1. Project Data 1 2. Principal Performance Ratings 1 3. Assessment of Development Objective and Design, and of Quality at Entry 2 4. Achievement of Objective and Outputs 7 5. Major Factors Affecting Implementation and Outcome 20 6. Sustainability 21 7. Bank and Borrower Performance 21 8. Lessons Learned 23 9. Partner Comments 24 10. Additional Information 24 Annex 1. Key Performance Indicators/Log Frame Matrix 26 Annex 2. Project Costs and Financing 28 Annex 3. Economic Costs and Benefits 29 Annex 4. Bank Inputs 34 Annex 5. Ratings for Achievement of Objectives/Outputs of Components 36 Annex 6. Ratings of Bank and Borrower Performance 37 Annex 7. List of Supporting Documents 38  Project ID: P009072 Project Name: TURKEY PRIVATIZATION OF IRRIGATION PROJECT Team Leader: Giuseppe Fantozzi TL Unit: ECSSD ICR Type: Core ICR Report Date: February 23, 2005 1. Project Data Name: TURKEY PRIVATIZATION OF IRRIGATION L/C/TF Number: SCL-42350 PROJECT Country/Department: TURKEY Region: Europe and Central Asia Region Sector/subsector: Irrigation and drainage (72%); Central government administration (28%) Theme: Other environment and natural resources management (P); Rural services and infrastructure (P); Rural policies and institutions (P); Participation and civic engagement (S) KEY DATES Original Revised/Actual PCD: 09/20/1995 Effective: 01/20/1998 04/24/1998 Appraisal: 02/18/1997 MTR: 12/01/2000 12/01/2001 Approval: 10/14/1997 Closing: 12/31/2002 06/30/2004 Borrower/Implementing Agency: REPUBLIC OF TURKEY/DSI Other Partners: STAFF Current At Appraisal Vice President: Shigeo Katsu Johannes Linn Country Director: Andrew N. Vorkink Ajay Chhibber Sector Director: Laura Tuck Kevin Cleaver Team Leader at ICR: Giuseppe Fantozzi Joma Mohamadi ICR Primary Author: Elliott Hurwitz (consultant) 2. Principal Performance Ratings (HS=Highly Satisfactory, S=Satisfactory, U=Unsatisfactory, HL=Highly Likely, L=Likely, UN=Unlikely, HUN=Highly Unlikely, HU=Highly Unsatisfactory, H=High, SU=Substantial, M=Modest, N=Negligible) Outcome: S Sustainability: L Institutional Development Impact: SU Bank Performance: S Borrower Performance: S QAG (if available) ICR Quality at Entry: U Project at Risk at Any Time: No  Introduction Turkey's agricultural sector is of considerable importance in terms of its share of the nation's GDP, about 14%, as well as employment, about 40%, and over one-third of agricultural GDP is dependent on irrigation. Bank analytic work in the early 1990s (see list of studies in Section 10) determined that key problems in the irrigation sector were unsatisfactory operation and maintenance (O&M) services due to inadequate government funding and inefficient investment. Regarding O&M, the main problems were inadequate funding and low recovery of funds from users, which had resulted in major deterioration of irrigation schemes. Regarding irrigation investment, the main difficulties were negligible cost recovery, poor planning of new schemes, and extensive delays in project completion. The principal agencies responsible for irrigation development and management are: the State Hydraulic Works (DSI), which develops and manages very large irrigation systems; the General Directorate of Rural Services (GDRS), responsible for minor irrigation works and on-farm development of many irrigation works completed upstream by DSI; and a variety of small organizations commonly referred in Turkey as Water Users Organizations (WUOs). These WUOs manage irrigation schemes of different sizes. Typically Irrigation Cooperatives (ICs) under the tutelage of GDRS have a command area of 600/700 ha, while Water User Associations (WUAs) under DSI have about 3,000 ha), report to different agencies, and operate under three different laws: law no. 442 regulating village authorities (that manage less than 1% of irrigated land), law no. 1580 regulating municipalities and WUAs that manage 54% of irrigated land, and law no. 1163 regulating ICs that manage 24% of irrigated land. The remaining 22% of irrigated land (approximately 1 million ha) is represented by schemes (usually very small, about 5 ha) constructed and managed by the farmers themselves without the support of any WUO. In part under the Drainage and On-Farm Development Project, the Bank encouraged the government in the early 1990s to transfer O&M responsibility to WUOs, an approach that had been found highly effective in Mexico and other countries. The Participatory Privatization of Irrigation Management and Investment Project (PPIMIP) was set within the context of an extensive program of reform in agriculture aimed at achieving higher productivity and efficiency and decreasing government intervention. 3. Assessment of Development Objective and Design, and of Quality at Entry 3.1 Original Objective: The objectives of the Participatory Privatization of Irrigation Management and Investment Project (PPIMIP) were to: 1. strengthen irrigation-related institutions by supporting: (a) DSI and GDRS in providing guidance and technical support to Water Users Organizations-WUOs (a definition that encompass Water Users Associations-WUAs and Irrigation Cooperatives-ICs); and (b) WUOs in fulfilling their responsibility for irrigation management and investment (including the purchase of O&M equipment and improving irrigation systems) in an efficient and sustainable manner; 2. relieve the public sector of its traditional responsibility for funding and subsidizing the costs of irrigation O&M; 3. initiate a process of reducing public sector responsibility for funding and managing irrigation investment; and 4. promote efficient and sustainable utilization of irrigation systems which would contribute to improved agricultural productivity. 3.2 Revised Objective: The objectives were not revised. - 2 -  3.3 Original Components: Component A: Operation and Maintenance Equipment Component A is the core of the project. Provision of grants through DSI to WUAs to finance purchase of irrigation and O&M equipment. Grant rates (share of total cost provided by the project) were expected to reach 28% in average and range from 15% to 45% by a complex schedule involving region, type of equipment and type of irrigation system (gravity or pump). This was the largest component of the project for which originally US$12.37 million or 62% of the loan amount were allocated. It was a detailed and well designed component. Component B: Institutional Strengthening l Provision of advisory services, training and equipment to DSI and WUAs to facilitate implementation of Component A and to prepare irrigation improvement schemes; l Provision of advisory services, training and equipment to GDRS and to an IC (Uluborlu Irrigation Cooperative) to facilitate implementation of Component C and prepare irrigation improvement schemes. This component was not sufficiently detailed in the original project documents. The original amount allocated for this component was US$5.22 million or 26% of loan amount. This was clearly oversized if we assume that the component was mainly aiming at providing TA and equipment to facilitate implementation of the remaining US$15 million project. It was also not clearly targeting specific identified larger needs of the implementing agencies or farmers organizations, since no WUOs or public agencies training needs assessment (or similar evaluation) was ever carried out during preparation. Component C: Pilot Drip Irrigation Scheme Provision of grants to a specific irrigation cooperative, Uluborlu IC, to finance the purchase and installation of drip irrigation systems. As provided in the original component C, irrigation of 200 ha and purchasing on-farm drip equipment was foreseen. The grant rate was set at 40% for civil works and 16% for goods. On the average, 28% was to be provided from project funds and 72% from the WUO budget. For this component, a sum of US$2.02 million or 10% of loan amount was originally allocated. The component was well designed but very strictly built around the peculiar needs of the foreseen investment in Uluborlu IC. 3.4 Revised Components: The components were revised and adapted quite often during project life and the loan agreement was amended twice (as shown below in Table 1) to allow the major changes to be effective. Component A was indeed well designed and started working effectively from the beginning of the project. Nevertheless, the first project amendment in June 2001 adopted uniform grant rates at the top limit of the original range, 45%, for all regions and all types of equipment. Grant rates were made uniform in order to provide better clarity and equitability, and therefore also enhance project and implementing agency credibility, among WUOs in the cost sharing arrangements. The decision of setting the uniform grant rate at the top limit of the range was decided in response to the added financial burden imposed on all WUOs as a result of the 2001 financial crisis in the country that weakened substantially the WUOs financial possibilities. As a final result the grant rate for O&M equipment provided under this component was about 38% on average. In a worldwide irrigation context, where farmers typically pay a small proportion of O&M costs, the fact that WUOs here were willing to pay more than half of the cost of new O&M - 3 -  equipment is really rather exceptional. Even more exceptional results could be obtained in the future by decreasing again the grant rate now that the effects of the financial crisis seem to be finished. As a matter of fact during calendar year 2004 a huge amount of WUA grant applications were addressed to the project that remained unsatisfied. These would have required an additional amount of US$8 million at the current grant rate of 45% to all be accepted. Component A was, also in June 2001, expanded from covering only the purchase of irrigation and O&M equipment (which became Sub-component A1) to include financing for rehabilitation and minor improvements to irrigation schemes in what was called the Pilot Rehabilitation Program (PRP or Sub-component A2). These rehabilitation and minor improvement interventions were funded on a 50% basis from the loan, and the remaining 50% from the participating WUOs. This was done on the basis of the findings of a survey undertaken in 2000 in seven regions representative of various types of irrigation systems in the context of the project to assess the need for rehabilitation and the extent of the interest of WUOs in investing to upgrade the operational efficiency of the irrigation systems serving them. The intention of the Government was for this pilot program to pave the way for a larger project which would have supported a national effort to assist WUOs in the rehabilitation and upgrading of the irrigation and drainage infrastructure on a country-wide basis. This pilot has proven to be successful, and even in this case the high amount of farmers contribution makes of this experience a quite unique model in this field for many other countries, and not only developing ones. As a result of the high level of requests from the farmers and of the two combined amendments of June 2001 and November 2002 Component A1 actually disbursed US$14.01 million or 70% of the loan amount and component A2 about US$1.92 million or about 10% of loan amount. Component B was quite quick initially in disbursing the allocated amount for provision of equipment to DSI and GDRS but found serious obstacles at the moment of defining which training activities should be financed and targeted to whom. Component B was proposed for a revision in 2001 trying to strengthen the link between proposed training activities to the needs, and not only technical needs, of the WUOs. A first agreement was reached with DSI about the urgent need of carrying out a WUOs training needs assessment but successively, this agreement was lost mainly due to: (i) the approaching of the end of the project and time constraints in implementing under the project any solution to what the needs assessment could have identified; (ii) the necessity of organizing an inter-departmental group within DSI for carrying on the training needs assessment (PPIMIP was implemented by the DSI O&M department); (iii) the pressing request for additional funds to be reallocated to the successful component A; and (iv) a general distrust in DSI about effectiveness of large amounts to be spent on TA for WUOs. Therefore, all through the project life this component was the subject of lengthy discussions between the Bank and the implementing agencies and as a final result only US$3.28 million were disbursed or 16% of loan proceeds, mainly for equipment for the two implementing agencies and for salaries of the PCU consultants (CC and ACC). This report considers that the main problem here was poor design and agreement/ownership at entry, which even intensive effort during supervision could not compensate for. For Component C, the originally-identified IC for drip irrigation (Uluborlu) did not follow through. It appears that the farmers there expected some financing from the local administration. This is indeed nothing surprising given the fact that ICs have always, in the past, obtained substantial rehabilitation and construction investments through a mix of financing by GDRS and local administration at zero cost for them. Consequent to Uluborlu declining to participate, the LA was first amended to introduce a set of eligibility criteria for participation enlarged not only to ICs but also to WUAs, to include sprinkler irrigation and to increase the grant rate to 40% . It was then substantially downsized in November 2002 when the large investment of an eligible WUA was included in the PRP (WUAs report directly to DSI and - 4 -  DSI was in charge of implementing component A2). Eventually, only 3 ICs participated in the sprinkler irrigation part of the project for a very modest amount of about US$50,000. It should be noted that while the financed IBRD portion of the project was US$19.7 million, the overall final project cost was US$45.1 million, including the contributions of the government agencies involved (about US$1.1 million), and especially, of the beneficiaries, US$24.3 million. 3.5 Quality at Entry: Quality at Entry is considered unsatisfactory, although slightly and although it only marginally affected achievement of overall project objectives. The project was consistent with the 1997 CAS, which supported "privatization" of irrigation. It was also consistent with the Bank and the government's overall approach to the agricultural sector, which sought to reduce government involvement in the sector. The goal of devolving responsibility for management and financing of irrigation O&M was adopted prior to project inception and was supported by substantial analytical work (see section 10 for list of studies). The objectives were quite realistic and were clearly stated in the project documents. The basic project plan was generally sound and focused around the core of the project, component A. However, there were three aspects of the project design that were unrealistic. First, making establishment of a "Unified Legal Framework" for all WUOs a condition of effectiveness was not realistic considering the history of such endeavors involving the Parliament (in the event, the condition was waived and such a law has not yet been passed). Second, the Bank was not realistic when it set the grant rates for Component C and even more when it designed the whole component around the needs of such ICs, to the point to name it explicitly in all project documents and to refer to it as the sole beneficiary of the pilot. Moreover, Uluborlu cooperative, accustomed to GDRS services at little or no cost, was not prepared to pay 60-80% of the funding of these schemes and was indeed actively looking for having local administrations to pay for the remainder of the costs not covered by the Loan or GDRS. GDRS should have alerted the Bank that this plan was impractical. Finally, contents of component B were not specified enough and, more important, could not find its rationale on the basis of a needs assessment or similar activity. Those shortcomings did not affect achievement of project objectives since the main project activities (equipment and pilot rehabilitation) aimed at supporting WUOs in implementing their basic function (that is implementation of O&M activities) produced positive results per se independently from the policy setting and the legal environment around them. Also, in implementing the project a large number of WUOs across Turkey were forced to learn by doing and to upgrade themselves, so that they are now obviously more capable and self- confident than they were before the project. Of course these effects would have been further enhanced by a unified legal framework and by a partial restructuring of DSI in its role towards WUAs, but this would have meant introducing the seed of an institutional reform that was not explicitly among the agreed main aims and objectives of the project, which was probably impossible. The principal government agency involved in the project was DSI, which implemented activities covering US$17.8 million or 89% of Loan amount, and its support for and involvement in the project was strong, although not all departments of the agency were as positive and cooperative as its top officials. GDRS had a much smaller role, US$2.2 million of loan amount or 11%, and was much less enthusiastic. Technical and economic aspects of the project were satisfactory, although with some shortcomings. The project focused on the purchase of O&M equipment--the largest component--with smaller amounts of funding going to institution building and the pilot drip irrigation component. The O&M equipment purchase component was appropriately detailed and proved successful. However, the SAR lacked a TOR - 5 -  for a training needs assessment for the institution building component, and while the TOR was initiated during the project, the activity was never undertaken. Also, the SAR did not specify consultant qualifications or job descriptions, nor specify whether international consultants or national experts were to be used, which later became an issue between the Bank and the Government. Procurement support and financial management were adequately arranged prior to project inception. The risk factors identified in the SAR lacked realism and relevance. For example, reversal of the transfer of irrigation schemes, potential lack of WUOs commitment, and difficulty in finding consultants with relevant experience do not seem to be significant risks while, as experienced during project life, lack of financial capabilities of the WUOs or control of local administrations over the WUAs general assembly were indeed critical. Especially with regard to finding consultants, the existence of successful programs in Mexico and Israel indicated adequate availability of expertise. As a final point, Monitoring and Evaluation was strongly overlooked. The PPIMIP had in fact little in the way of meaningful monitoring indicators, establishment of initial benchmarks, or systematic data collection. The project was designed to promote participatory involvement of WUOs in irrigation investment, and it would have been much stronger in this regard if the M&E had been better developed. The project design can therefore be faulted in this regard with consequences that spilled over the design phase. In fact later it was quite difficult during the implementation phase to convince DSI O&M department that accurate M&E needed to be carried out since those activities had not been agreed from the beginning and DSI claimed that there were no resources allocated under the project for this purpose. - 6 -  Table 1: Project Components and Funding Allocation: Original and after Amendments Stage/ Major Grant Rate IBRD Changes Introduced Component/Activity (share of total cost Funding provided by the project) (US$m) Original Component A-Operations and Maintenance Average grant rates 28%, 12.37 Equipment: from 15%-45% by region, (1) Provision of grants through DSI to WUAs to type of equipment and finance purchase of irrigation and O&M equipment. type of irrigation system. " Component B-Institutional Strengthening " 5.61 " Component C-Pilot Drip/Sprinkler Irrigation " 2.02 Scheme Amendment 1: 6/01 Component A-Operations and Maintenance Grant rates increased Component --increase grant rates Equipment and Pilot Rehabilitation Program: from an average of 28% A1--10.42 --introduce component to maximum of 45% of A2 to finance rehab (A1) Provision of grants through DSI to WUOs to O&M equipment Component and improvement purchased by WUOs. A2--1.95 --expand eligible areas finance purchase of irrigation operation and maintenance equipment under component. Grant rate standardized for types of equipment, (A2) Provision of grants through DSI to WUOs to irrigation and regions. finance irrigation rehabilitation and minor improvement works. Component B-No change Same as Component A 5.61 --Uluborlu WUO Component C-Pilot Drip/Sprinkler Irrigation Same as Component A 2.02 removed, other WUOs Scheme: Introduced grants to eligible ICs for made eligible drip/sprinkler irrigation systems (vs. only drip --sprinkler systems systems earlier). Amendment also noted removal of added to drip Uluborlu IC (which elected not to participate) Amendment 2: 11/02 Component A--No change in activities funded No change Component --eliminated A1--14.09 geographic constraints Component on participants A2--1.92 --extended closing date Component B--No change in activities funded No change 3.28 Component C--No change in activities funded No change 0.04 4. Achievement of Objective and Outputs 4.1 Outcome/achievement of objective: Introduction Outcome is assessed as overall satisfactory, as the project strengthened WUOs by providing support for O&M equipment and for rehabilitation, which significantly enhanced WUOs credibility in the minds of the farmers and, in turn, allowed collection of revenues that enhanced financial sustainability, reduced water losses and improved development of technical and managerial capacities. The rehabilitated irrigation schemes resulted in a greater quantity of water available (due to lower losses) and also permitted a more reliable supply (due to fewer breakdowns).This enhanced credibility and service validated the bottoms-up, participatory approach of the WUOs and contributed to the achievement of objects 1(b), 2, and 3.The project contributed to a paradigm shift, already underway prior to project inception, which changed the mind-set of farmers--in the past accustomed to receiving irrigation services at little or no cost--(1) to pay for the water they used; and (2) to become more pro-active and supportive of WUOs in their efforts to take - 7 -  greater responsibility for managing and paying for the O&M of irrigation schemes. It should be noted that the increased tendency for farmers to pay for the water they used comprised a significant shift.Prior to 1993 (when the paradigm shift began), farmers normally paid only for rehabilitation or new investment (and the payment scheme used resulted in farmers paying considerably less than full cost) and did not pay for O&M costs. While still far away from paying for full O&M cost of water (for example bulk water is still provided by DSI for free to its WUAs) WUOs have increased notably the recovery of the O&M cost supported by them at secondary and tertiary level. Objective (1): Strengthen irrigation-related institutions by supporting: (a) DSI and GDRS in providing guidance and technical support to WUOs; and (b) WUOs in fulfilling their responsibility for irrigation management and investment (including the purchase of O&M equipment and improving irrigation systems) in an efficient and sustainable manner. Achievement in the area of Objective 1 was overall satisfactory, with significant benefits achieved under objective 1(b) substantially outweighing deficiencies in objective 1(a). Objective 1(a) Strengthen irrigation-related institutions by supporting DSI and GDRS in providing guidance and technical support to WUOs. Although some benefits were achieved in this area, achievement of sub-objective 1(a) is assessed as unsatisfactory.On the positive side, long and short-term consultants supported DSI, GDRS and WUOs in all aspects of project activities.Study tours and other training benefited these organizations, DSI and the WUAs are now more capable in selecting appropriate equipment for maintenance of irrigation infrastructure after the experience of purchasing equipment under the project grants.And DSI local managers and technical staff gained experience in selection and implementation of rehabilitation projects based on WUAs' demand and commitment to share investment cost. The results and lessons learned were disseminated by implementation in 6 different DSI regions as well as through national WUAs Heads meetings in Adana (in November, 2002) and Antalya (in May, 2004). However, achievement of objective 1(a) was unsatisfactory largely because insufficient training was provided to WUOs. As they were rapidly acquiring new responsibilities and financial obligations, they were not provided with a commensurate augmentation of their expertise.It was intended that DSI would provide the training needed, and while the agency did provide some training, it was too rooted in the older, "top-down" paradigm, and too theoretical, to be effective.The training the WUAs needed was in e.g., management, irrigation technology, and water pricing.While the PCU developed a TOR for a "training needs assessment" for the WUAs as well as DSI itself, these were never implemented, in part due to insufficient agency support as well as confusion surrounding the extension of the project in late 2002 (see sec. 5.1). (It should be noted that WUAs largely made up for the lack of formal training by "learning-by-doing."See discussion of achievement under Objective 3, below). An additional factor in assessing Objective 1(a) was that the government did not pass a WUO law during the period of project implementation, despite this having been a condition of effectiveness (later waived). Objective 1(b): Strengthen irrigation-related institutions by supporting WUOs in fulfilling their responsibility for irrigation management and investment (including the purchase of O&M equipment and improving irrigation systems) in an efficient and sustainable manner. Achievement of this sub-objective was satisfactory.The provision of O&M equipment and rehabilitation of - 8 -  irrigation schemes was very successful in meeting this objective because it provided a tangible demonstration of the capabilities of the newly-empowered WUAs. Farmers were more willing to participate and pay their water bills because they were assured of more adequate and more reliable service, and--based on anecdotal data-- increases in efficiency and productivity (for example in Adana region, after the rehabilitation works in 5 WUAs cropping patterns are clearly shifting from cotton to oranges with an increase of financial net margins of about 150%).The PPIMIP helped to build confidence between WUOs and farmers.Evidence of achievement in this area includes: l The much higher portion of rehabilitation costs paid by WUAs in Turkey compared to other developing countries l Greater pro-activity in undertaking O&M by WUAs that manage their own work compared to those managed by DSI l Higher collection rates at WUAs participating in the project than the general population WUAs paid half of rehabilitation costs: It should be noted that WUAs paid 50% of the cost of rehabilitation and minor improvement works provided under the project.This is an exceptionally high ratio by the standards of other developing countries (except Mexico).In no other developing or middle income countries do water users pay as much as 50% of such costs, and Turkey is ahead of some industrialized countries as well (e.g., France, where payments are typically 25-30%, and Germany, where user payments for drainage infrastructure average around 40%). In a country where irrigation services have typically been provided free by a highly centralized agency, the autonomy and self-reliance now achieved by Turkish WUAs are actually quite stunning. More pro-active O&M:Figures 1 and 2 show maintenance activity undertaken in 2003 in areas managed by DSI (Figure 1) and in areas managed by WUAs (figure 2) as a percentage of estimated needed amount of O&M to be implemented at the beginning of the same year (in both cases, for DSI and for the WUAs the estimation has been made by DSI). As can be seen, the WUA-managed areas were more active in undertaking O&M:in 5 of the 6 categories of activity, WUA-managed areas were more active, in many cases by a large margin.These data seem to validate the point that participatory irrigation schemes generate better maintenance than those provided by the central government. Figure 1: Maintenance Works in Irrigation Areas Managed by DSI, 2003 (%) 77 77 80 52 44 60 32 40 20 20 0 lines Canals Roads roads lining Canalet Irrig. replacement Drainage Service Upgrade Concrete - 9 -  Source: DSI O&M Evaluation Department Figure 2: Maintenance Works in Irrigation Areas Managed by WUOs (%) 97 89 100 81 80 62 55 60 41 40 20 0 Irrig.ationdrainage upgrade concrete Replacement lines Roads roads liningcanalet Source: DSI O&M Department Higher Collection Rates--Collection rates were higher at WUOs that participated in the project and rehabilitated existing irrigation schemes. The tangible benefits provided to WUOs, especially rehabilitation of existing irrigation schemes, increased farmer satisfaction by providing better and more reliable service, which resulted in higher collection rates.This is illustrated in Figure 1 below.The data show that the 1997-2002 collection rate for all irrigation systems transferred from DSI to WUAs varied from 50% to 67%, with no clear trend; the rate for 2002 was 60%.While time series data for the 280 WUAs participating in the O&M sub-component were not available, 2003 data for all 13 participating WUAs under the PRP showed an average collection rate of 82%.This higher collection rate was validated by interviews with WUAs and government officials, who unanimously attributed the higher collection rate to the changed mind-set of farmers and their greater willingness to support O&M for which they had greater responsibility and ownership. - 10 -  Figure 3: Collection Rate: 1997-2002 (nationwide) vs. 2003 (13 WUOs in project) 90 80 %,et 70 60 ra 50 40 llection 30 co 20 10 0 1997 1998 1999 2000 2001 2002 2003 Source: 1997-2002, DSI O&M Department; the "collection rate" is the percent of billings that are received in a given year. NB. Only collections for the relevant year are included--not collections for previous years. 2003 data provided by individual associations to PCU. Sharing of Equipment: Interviews with 5 WUAs in the Adana region, held during ICR preparation, indicated that WUAs are generally willing to share O&M equipment that is not in current use with nearby WUOs. Several WUAs also reported that use of Bank procurement procedures saved considerable money in the purchase of O&M equipment, compared to the procedures that had been used earlier. These WUAs stated that they would continue to use such procedures for purchases outside of the PPIMIP. Objective (2): Relieve the public sector of its traditional responsibility for funding and subsidizing the costs of irrigation O&M Achievement in this area was satisfactory, with: l A major and continuing decline in O&M spending by the Government l A large decline in the gross area provided with O&M services by the Government l A more modest reduction in the size of DSI's O&M staff and facilities. While the PPIMIP cannot claim credit for all of the decline in O&M spending, by demonstrating the efficacy of WUOs and contributing to the paradigm shift in the mindset of farmers, it was a facilitating factor in these trends. Decline in O&M Spending: Table 2 below shows that DSI O&M expenditures have fallen dramatically over the last 14 years. While they fell 62% from 1990 to 1997, they declined a further 63% during the period of the PPIMIP. Table 2: DSI Operation and Maintenance Expenditures, by Year - 11 -  Year Billion 2003 TL 1990 80,384.3 1995 65,607.3 1996 38,335.8 1997 30,373.7 1998 23,185.5 1999 29,370.3 2000 26,581.9 2001 28,107.1 2002 25,736.6 2003 8,613.8 Source: DSI So to the extent that PPIMIP contributed to WUAs taking greater responsibility for the financial support of maintenance and equipment purchase, this contributed to a reduction in the fiscal burden of these functions. Trend in Size of Area Maintained by DSI: The most important factor in the large decline in DSI O&M expenditures was the decline in the size of the areas it serviced. Table 3, below, shows the trend in the size of the area currently provided with O&M by DSI. As can be seen, the total irrigation area developed by DSI rose by 47% from 1990 to 2004. However, the area provided with O&M services by DSI fell dramatically, starting before project inception. During the period of the PPIMIP, the area provided with O&M services by DSI fell by 59%. By 2004, DSI support for O&M had become negligible, covering around 4% of all irrigable areas (command areas) developed by DSI, with the balance supported predominantly by WUAs, and to a small extent by villages or municipalities. Table 3: Irrigation Ratios in Areas Developed and Operated by DSI Year Total irrigation area Currently provided with O&M by DSI developed by DSI (ha) Command Area Irrigated Area Irrigation Ratio (%) (cumulative, all (gross-ha) (net-ha) management entities) 1990 1,626,170 1,251,251 857,499 69 1995 1,897,850 543,650 304,524 56 1998 2,154,918 248,486 97,645 39 1999 2,202,562 247,699 102,792 41 2000 2,251,625 214,910 86,398 40 2001 2,296,350 198,718 75,980 38 2002 2,340,197 200,014 71,146 36 2003 2,353,360 102,265 35,124 34 2004 2,393,862 102,104 32,488 NA Source: DSI O&M Department A further trend revealed in Table 5 is the steady decline in the irrigation ratio of the land for which DSI continues to provide O&M services. In effect, DSI has transferred O&M responsibility for the more attractive plots, while retaining responsibility for the less attractive areas. - 12 -  Downsizing of DSI: DSI is moving to downsize its O&M staff and facilities, and while its O&M operations are shrinking, the decline has not been commensurate with the rapid fall in the size of the areas maintained or in expenditures. DSI's ultimate goal is to devolve all responsibility for O&M--and the associated capital equipment--to the WUAs. Since fewer facilities will be needed to manage O&M, DSI is also moving to divest facilities it no longer needs, and reduce its staffing. In the Adana region, for example, DSI has closed 13 local offices in the last 8 years, of which some have been sold and the remainder are in the process of being sold. The DSI staff in the Adana region has been modestly reduced, mainly by attrition, as shown in Table 4, below. Table 4: DSI Staffing in Adana Region, 1994-2003 1994 2003 Total DSI regional staff 3,500 2,700 DSI staff working on O&M 1,180 922 Source: DSI Adana regional office The decline in DSI staff nationally has been proportionately larger: from 1993 to 2003, the number of DSI staff working nation-wide on O&M declined from around 10,000 to 6,000. And total DSI national staffing has declined from around 35,000 in 2001 to 31,000 in early 2004. The decline in DSI involvement in O&M is part of a broader trend of reduced government involvement in the agricultural sector, including reduced price support, etc. Figure 4, below, shows this trend; budgetary support for agriculture has declined, in real terms, from about 3.5 trillion TL in 1999 to around 1.5 trillion TL in 2003. Figure 4 - Total Agricultural Expenditure, Real 1999 Trillion TL 4,000 3,500 3,000 2,500 2,000 1,500 1,000 500 0 1999 2000 2001 2002 2003 Source: World Bank, Review of Impact of Reform of Agricultural Subsidization, 2004. Note: "Agricultural" includes current and investment expenditures on rural infrastructure (roads, drinking water, and electrification), forestry and environment, as well as all "traditional" agricultural sector expenditures and transfers. - 13 -  Objective (3): Initiate a process of reducing public sector responsibility for funding and managing irrigation investment Achievement in this area was satisfactory. As described in earlier sections, PPIMIP provided a highly-effective demonstration of how WUOs could manage, and partially fund, irrigation investment and rehabilitation. As noted, WUAs contributed more than 50% of these funds, an exceptionally high proportion among developing countries. And WUOs demonstrated by on-the-job-training that they could develop the requisite management, technical, and planning skills to implement these projects successfully. (The increased efficiency produced by PPIMIP is discussed under Objective 4). PPIMIP investment to rehabilitate irrigation schemes is summarized in Table 5. Thirteen WUAs received 16 grants averaging US$239,000, which they matched. (In one instance, 7 WUAs collaborated to obtain project funds that were used to rehabilitate a trunk canal that served all seven.) The rehabilitation included rebuilding damaged sections of canals, building canalets, replacement of faulty valves and other hardware, and replacement of earthen canal beds with concrete. Some of the benefits of the pilot investment program have been discussed earlier; the rehabilitation program was regarded by all persons interviewed by the ICR mission--WUAs staff, government officials, and farmers--as a highly cost- effective way to raise efficiency and productivity in the irrigation sector. Government satisfaction with PPIMIP--In part due to the fiscal savings resulting from devolution of responsibility to WUOs, the government has expressed considerable satisfaction with the project: l The government has expressed interest in pursuing a follow-on project of a similar type l The government believes that the PPIMIP effectively demonstrates the benefits of public-private partnerships, and put forth the project as an example of "best practice" in this area at the Development Summit in Johannesburg in 2000. The government is pursuing the replication of the concept in the area of licensed warehousing by the Agricultural Sales Cooperatives Unions (potentially with Bank support). - 14 -  Table 7. Statement of Expenditures by WUOs and Grants to Contractors under PRP to date No. of Item Name of WUO Total paid to Region (B