Turkey - Irrigation Rehabilitation and Completion Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 3515 PROJECT PERFORMANCE AUDIT REPORT TURKEY IRRIGATION REHABILITATION AND COMPLETION PROJECT (CREDIT 281-TU) June 23, 1981 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. FOR OFFICIAL USE ONLY PROJECT PERFORMANCE AUDIT REPORT TURKEY IRRIGATION REHABILITATION AND COMPLETION PROJECT (CREDIT 281-TU) TABLE OF CONTENTS Page No. Preface .........................******************************* Basic Data Sheet ......................... *. ******.***************** .ii Highlights ....................*********............................ iv PROJECT PERFORMANCE AUDIT MEMORANDUM I. SUMMARY ........................ 1 II. THE ISSUES ...................... 4 A. Cost Recovery ..... .......... 4 B. The Limited Area Expansion *9#*e*oo**oo*ee**uee ..... 6 C. Trade-off Between Power Generation and Irrigated Agricultural Development 0.......... .... 7 D. Information on Reimbursements .....o.000000000 ...... 9 III. WATER MANAGEMENT IN THE TOKAT PROJECT AREA ............ .. 10 Attachment Some Minor Disagreements with the PCR ....... 12 Figure I: Energy and Agricultural Self-Sufficiency Ratios for Selected Countries, 1976 .............................. 13 Annex 1: Borrower Comments .................... 14 ATTACHMENT: PROJECT COMPLETION REPORT I. BACKGROUND ......o....................................... 16 II. FORMULATION ............................................... 17 III. IMPLEMENTATION ........................................... ....19 IV. AGRICULTURAL IMPACT ......................................27 * V. ECONOMIC ANALYSIS .....o................................... 39 VI. INSTITUTIONAL IMPACT ..................................... 42 VII. SPECIAL ISSUES ........................................... .....50 VIII. BANK PERFORMANCE ...................................... 53 IX. CONCLUSIONS ............................................... 55 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 Worid Bank authorization. TABLE OF CONTENTS (continued) ANNEXES Annex I - Table 1: Equipment List Annex I - Table 2: Construction Schedule - TOPRAKSU Annex II - Table 1: Cost Estimates Annex II - Table 2: Analysis of DSI Cost Increases Annex III - Table 1: Cropping Patterns and Yields 1973-79 - Koprucay Annex III - Table 2: Cropping Patterns and Yields 1973-79 - Silifke Annex III - Table 3: Cropping Patterns and Yields 1973-79 - Tokat Annex IV - Table 1: Schedule of Annual Investment Expenditure Annex IV - Table 2: Farmgate Prices Used 1971-1985 Annex IV - Table 3: Price Deflators Used Annex IV - Table 4: Economic Rate of Return Calculations - Koprucay Annex IV - Table 5: Economic Rate of Return Calculations - Silifke Annex IV - Table 6: Economic Rate of Return Calculations - Tokat Annex V - Table 1: Water Charges and O&M Costs Chart I Organization Chart Maps Koprucay Project Area Silifke Project Area Tokat Project Area PROJECT PERFORMANCE AUDIT REPORT TURKEY IRRIGATION REHABILITATION AND COMPLETION PROJECT (CREDIT 281-TU) PREFACE This is a performance audit of the Irrigation Rehabilitation and Completion Project in Turkey for which Credit 281-TU was approved in January 1972 in the sum of US$18 million. The final disbursement in respect of this loan was made on August 13, 1979. The audit report consists of an audit memorandum prepared by the Operations Evaluation Department and a Project Completion Report (PCR). 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 (No. PA- 108a) dated January 4, 1972, the President's Report (P-1011) of January 6, 1972, the Credit Agreement dated January 25, 1972, and the PCR. Internal Bank memoranda on project issues as contained in relevant Bank files have also been consulted and Bank staff associated with the project have been interviewed. An OED mission visited Turkey in November 1980. The mission held discussions with officials of the Ministry of Agriculture, the State Hydrau- lics Works (DSI), and the Soil Conservation Agency (TOPRAKSU) the executing agencies for the different components of the project. The information ob- tained during that mission was used to test the validity of the conclusions of the PCR. A copy of the draft report was sent to the Borrower for comments on April 7, 1981; their comments are reproduced as Annex 1 to the PPAM. The audit finds that the PCR covers adequately the projects salient features and the PPAM generally agrees with the conclusions. The PPAM focuses principally on issues -of recovering operations and maintenance costs of the irrigation system, and whether national water use criteria should be deter- mined by maximizing hydro power generation (Turkey has almost no other indi- genous power sources) or irrigated agricultural production (in which field it has to compete in the world markets with other, sometimes better placed, producers. The valuable assistance provided by the Government of Turkey and staff of the executing agencies met during the preparation of this report is gratefully acknowledged. - ii - PROJECT PERFORMANCE AUDIT REPORT BASIC DATA SHEET TURKEY IRRIGATION REHABILITATION AND COMPLETION PROJECT (CREDIT 281-TU) KEY PROJECT DATA Appraisal Actual or Item Estimate Current Estimate Total. Project Cost (US$ million) 36.6 (includes 10% phy- 40.2 sical and 5% per year price contin- gencies) Overrun 0.0% 10% Credit Amount (US$ million) 18.0 Disbursed 18.0 Cancelled 0.0 Repaid to - Outstanding to 18.0 Date Physical Components Completed 12/31/76 Est. Dec. 81 Proportion Completed by Above Date 100% 90% Proportion of Time Overrun - 100% Economic Rate of Return Silifke - 18% Silifke - 33% Koprucay - 19% Koprucay - 23% Tokat - 12% Tokat - 42% Cumulative Estimated and Actual Disbursements (US$ million) FY 72 FY73 FY74 FY75 FY76 FY77 FY78 FY79 Estimated 0.5 5.9 11.5 15.0 16.9 18.0 - - Actual - 1.7 4.4 19.6 13.3 15.3 16.8 18.0 Actual/Estimated % - 29 38 65 79 85 93 100 OTHER PROJECT DATA Original Actual or Item Plan Revisions Est. Actual First Mention in Files or Timetable 01/71 Government's Application 04/71 Negotiations 11/3-10/71 11/3-16/71 Board Approval 01/18/72 01/18/72 Credit Agreement Date 01/25/72 01/25/72 Effectiveness Date 04/27/72 04/27/72 Closing Date (Credit) 06/30/77 12/31/77 & 12/31/78 04/30/79 Borrower Government of Turkey Executing Agencies State Hydraulic Works (DSI): Irrigation and Drainage Work Soil Conservation Agency (TOPRAKSU): On-farm Development Extension Service (Ministry of Food Agriculture and Livestock): Extension Services Agricultural Bank of Turkey (TCZB): Agricul- tural Credit Fiscal Year of Borrower March 1 to last day of February Follow-on Project No follow-on project MISSION DATA Returned to Headqtrs. No. of No. of Man- Date of Sent by Month/Year WeeksL Persons Weeks/e Report Identification/ Preparation Bank 03/71 0.5 2 1.0 04/21/71 Preparationi/c Appraisal/d Bank 06/71 4.0 5 20.0 01/04/72 Supervision I Bank/IDA 05/72 1.0 2 2.0 07/06/72 Supervision II Bank/IDA 04/73 3.8 1 3.8 06/15/73 Supervision III Bank/IDA 11/73 1.0 3 3.0 02/05/74 Supervision IV Bank/IDA 12/74 2.0 2 4.0 01/14/75 Supervision V Bank/IDA 10/75 2.4 2 4.8 11/25/75 Supervision VI Bank/IDA 06/76 2.6 2 5.2 06/30/76 Supervision VII Bank/IDA 05/77 2.1 2 4.2 06/10/77 Supervision VIII Bank/IDA 12/77 1.5 2 3.0 01/06/78 Supervision IV and Preparation for Completion Bank/IDA 08/78 2.0 1 2.0 09/13/78 Completion Bank/IDA 03/80 1.0 3 3.0 10/20/80 Total 23.9 56.0 COUNTRY EXCHANGE RATES Name of Currency (Abbreviation) - Turkish Lira (TL) Appraisal Year Average - 1971 Exchange Rate: US$1 = TL 14.917 Intervening Years Average - 1972 US$1 = TL 14.150 - 1973 US$1 = TL 14.150 - 1974 US$1 = TL 13.927 - 1975 US$1 = TL 14.442 - 1976 US$1 = TL 16.053 - 1977 US$1 = TL 18.002 Completion Year Average - 1978 US$1 = TL 24.280 Completion Report Year Average - 1979 US$1 = TL 36.400 /a In some cases, the number of weeks was estimated when the mission had multiple responsibilities. /b Project was identified by State Hydraulic Works (DSI). /c Preparation consisted of presentation of various data by Government of Turkey. No separate feasibility studies were done. Data was presented at various times, but completed by the end of April 1971. /d The appraisal mission was combined with final preparation. /e Field work only. - iv - PROJECT PERFORMANCE AUDIT REPORT TURKEY IRRIGATION REHABILITATION AND COMPLETION PROJECT (CREDIT 281-TU) HIGHLIGHTS The project aimed to rehabilitate and improve the existing irriga- tion and drainage works on about 26,000 ha and to finance the construction of new works on a further 24,000 ha. Land levelling and strengthening of the agricultural extension system was to be carried out over the entire area; surface drainage was to be supplemented with tile drains wherever necessary. The credit was closed nearly two years later than expected at appraisal. However, irrigation and drainage works will not be completed in one of the areas until well into 1981; in another area works were completed in late 1980, about 18 months after the credit was closed. Land levelling and on-farm development suffered the most delays. Yields and cropping intensities have increased faster, and have already reached a higher level, than projected at appraisal. This, as well as the higher (in real terms) farmgate prices have more than balanced cost over- runs, and the economic rate of return of the three sub-areas is now estimated to average 33%, about twice the appraisal estimate of 17%. The following points are of special interest: - since wheat was expected to be the major crop, the water delivery system was not designed to meet the higher water requirements of newly introduced high value crops which, at full development, will not permit maximum area utilization although they will still exceed appraisal estimates (PPAM, paras. 9 and 22; PCR, paras. 4.23 and 4.27 - 4.30); fine land levelling and on-farm development in the largest sub- project area cannot be completed during the short single fallow period allowed by cotton. Farmers are not offered compensation for crops foregone while their land is being developed and the Soil Conservation Department has no automatic right of entry to carry out the work (PPAM, para. 4 and PCR, paras. 3.07, 7.09, 7.10 and 9.06); - capital costs, to be collected over 50 years after project comple- tion with a ten-year grace period, (and no interest charges) are not adjusted to replacement values and therefore are unrealistic. O&M costs, as assessed, rarely cover actual expenses and the rate of actual collection is low (PPAM, paras. 14 to 16 and PCR, paras. 6.13 - 6.24); national priorities on water use need to be re-examined in the light of greatly increased energy costs, wherever irrigation competes with hydroelectric power generation (PPAM, paras. 26 - 32); - common grazing lands with irrigation potential were not included in the irrigated area (PPAM, para. 23); and - implications of the Bank's withdrawal from lending for irrigation since 1976 because of non-recovery of full O&M costs (PPAM., paras. 17 - 18; PCR, para. 9.04). PROJECT PERFORMANCE AUDIT MEMORANDUM TURKEY IRRIGATION REHABILITATION AND COMPLETION PROJECT (CREDIT 281-TU) I. SUMMARY 1. The Irrigation Rehabilitation and Completion Project aimed at increasing agricultural production in three areas--Silifke and Koprucay on the Mediterranean Coast, and Tokat on the Anatolia plateau--by improving and extending existing irrigation and drainage systems and carrying out on-farm development works. Major project works were the rehabilitation and improve- ment of existing irrigation and drainage systems serving about 26,000 ha and construction of new systems to serve an additional 24,000 ha. The proj- ect would also finance land levelling, tile and surface drainage, improvement of village roads, and the improvement of agricultural extension activities in the three areas. The State Hydraulic Works Department (DSI) would construct the irrigation and drainage works, the Land and Water Resource Development Directorate (TOPRAKSU) of the Ministry of Village Affairs would carry out on-farm development, and the Ministry of Food, Agriculture and Livestock, would be responsible for the improved extension services. 2. Bank contribution to project preparation and design was minimal. Full feasibility studies of the three project areas were not carried out. After a review had shown that the basic data were available, the Bank agreed to the Government's suggestion to limit project preparation to DSI's presen- tation of the basic data. Because the Bank did not insist on the Government carrying out full feasibility studies, the implementing agencies were not fully committed to the project. 3. During appraisal in May/June 971 the agricultural extension and operation and maintenance of the irrigation and drainage components were not set out in sufficient detail. In the case of extension it was understood that the evident success of the Seyhan Extension Servicel/ would ensure its adoption in the Irrigation Rehabilitation and Completion Project. However, more detailed provisions for these components, and more integration into the project, would have made implementation more successful. 4. The original Credit closing date was June 1977; this schedule could not be met. In 1979 project works were stopped because of the country's economic disarray caused severe budgetary constraints for the implementing agencies. The Credit was eventually closed in April 1979, but irrigation and on-farm development works at Tokat were not completed until later that year. In 1980, finishing touches still had to be added to irrigation works in the Silifke area. At Koprucay, works by DSI are expected to be completed only in 1/ For further discussion, see OED, PPAR on Turkey Seyhan Irrigation (Stage II), Loan 587/Credit 143. Report No. 2747. November 26, 1979. -2- 1981. On-farm developments works at Koprucay are hampered because farmers are not legally obliged to allow TOPRAKSU onto their land to carry out the works. Political considerations make it unlikely that appropriate legislation to force farmers to allow access to their land could be introduced. 5. Similar difficulties in getting cotton farmers to release their lands for on-farm development works were experienced in the Seyhan Irrigation Project!/. Farmers in Koprucay and Seyhan, which are one-crop or predomi- nantly one-crop areas, must be compensated if they are to forego one crop to allow on-farm works to be carried out. Neither this nor the Seyhan project paid sufficient attention to this factor. 6. Performance of contractors and suppliers, and quality of DSI and TOPRAKSU works, were satisfactory. However, consultants were not employed to strengthen the Extension Service and TOPRAKSU in spite of a provision in the Credit Agreement. The Government also failed to impose and collect adequate water charges and to support the Extension Service adequately. Audit reports were not submitted and the quality and timeliness of reporting by the implementing agencies deteriorated over time. In the later years of project implementation, poor quality and infrequent Extension Service reports handi- capped both the Government and the Bank in their monitoring and evaluation of the project's agricultural impact. 7. Supervision by the Bank took longer than usual, because three separate areas had to be visited. Several reorganizations of the Bank and the Regional Office led to poor staff continuity. Long intervals occurred between some of the earlier visits, compounding initial difficulties and start-up problems. Supervision activities tended to concentrate more on the implemen- tation of physical works than on aspects of agricultural development. 8. Actual project costs were US$40.2 million (TL 677 million), exclud- ing the cost of the remaining 50% of Koprucay on-farm development works which were not completed by the closing date. Project works (new and improvements) were carried out on a total of 50,000 ha. Hence, average cost of works per hectare is about US$800. The appraisal estimate was US$36.6 million (TL 547 million), giving a cost overrun of 10% (or 24% in local currency). However, a more realistic assessment requires that comparable appraisal estimates be adjusted downward to take into account the non-appointment of consultants and the partial completion of on-farm development works at Koprucay. The adjusted appraisal project cost would thus be US$35.0 million (TL 523 million). Cost overruns therefore amount to .15% (30% in local currency), which are mostly acceptable. Much of the increase is accounted for by increases in quantity of work and local prices, and by implementation delays. 9. The pro,ct was successful in increasing yields and cropping inten- sities in Silifke and Tokat. On the whole, appraisal projections for yield increases and cropping pattern changes were conservative. For many crops, yield increases projected to be attained only at full development (1979 in the 1/ Ibid. -3- case of Silifke and 1981 for Tokat) were achieved by 1978. Movements into higher value crops were also more marked than expected at appraisal. A large part of these can be explained by changes in relative prices which could not have been predicted. In Koprucay, progress made in achieving yield increases was less satisfactory and little, if any, headway was made in increasing cropping intensity. Reasons for the relatively poorer performance are: incomplete on-farm development, slower pace of DSI works, ineffective exten- sion service, and fewer alternative sources of agricultural advice than in the other two areas. 10. Farmgate prices increased far beyond appraisal expectations. As a consequence, so have nominal income gains by project beneficiaries in Tokat and Silifke. Tokat experienced the greatest increase: in 1979, income per hectare was TL 21,240 (TL 3,186 in 1971 constant prices) compared to its 1971 pre-project level of TL 1,890. Comparable figures for Silifke are TL 39,680 (TL 5,953 in constant prices) and TL 3,950. Real income gains in Tokat and Silifke well exceed the 15% without-project inc:ease anticipated at appraisal: 70% over pre-project levels in Tokat, and 50% i. Silifke. In Koprucay, real income per hectare declined by 35% over its pre-project levels; from TL 3,512 to TL 2,267 in 1971 constant prices. This is in o9rt a reflection of low cotton yields and prices in later years and the slowe pace of project imple- mentation. 11. Economic rates of return for the three project areas are currently estimated to be: Koprucay 23%, Silifke 33% and Tokat 42%. Appraisal esti- mates were 19%, 18% and 12%, respectively. The completion rate of return estimates reflect differences in relative production increase and project investment costs between the three areas; the Tokat project experienced the greatest gains in production and least cost overruns in relative terms; Koprucay had the smallest production gains and largest cost overruns. 12. The impact of the project already exceeds appraisal expectations. In the longer term, the impact should be sustained provided marketing arrange- ments in the three areas are satisfactory. No special attention was paid to these at appraisal becaust it was assumed that the project's incremental production would not be large enough. Now, however, farmers are growing high value crops on larger areas and obtaining considerably higher yields than expected at appraisal. 13. The project's contribution to institutional development was limited. Both DSI's and TOPRAKSU's capabilities were strengthened by the project expe- rience. However, little institutional development of the Extension Service was achieved because neither were the consultants employed, nor did the Government provide the necessary support. Efforts to coordinate the activi- ties of the implementing agencies involved by establishing coordinating committees in each of the project areas had limited success. The coordinating committee met less and less frequently and finally not at all. -4- 14. Cost recovery, as under the Seyhan Irrigation Project, was a major issue throughout project implementation. Despite assurances obtained from Government at negotiations to set water charges at a level sufficient to cover O&M costs, this has rarely been achieved. In many years, assessments did not cover even 50% of O&M costs. More seriously, actual collection rates averged only 62% of the assessed total between 1973 and 1978. Consequently, actual collections have only once exceeded 50% of actual costs and averaged 30% for the 1973-1978 period. DSI's capital costs are partially recovered from farmers but there is still no provision for recovery of TOPRAKSU"s on-farm development costs. 15. In 1978, the new Government took an important step towards improving cost recovery by substantially increasing water charges. In the three project areas, water charge assessments were increased fivefold over their previous year's levels and they then covered O&M costs. However, this improvement was not sustained in 1979: that year water charge assessments were less in two of the three areas than in 1978, while 0&M costs had increased significantly. The subsidy thus enjoyed by project beneficiaries is difficult to justify since their incomes are well above the national average. II. THE ISSUES A. Cost Recovery 16. Turkish law explicitly acknowledges the obligation of irrigators to pay full capital costs (over 50 years, with no interest) and operation and maintenance costs of irrigation projects built by the Government. The former are computed after project completion and beneficiaries are billed annually; the latter are estimated each year by DSI and submitted for Cabinet approval. However, the Government does not fully recover any of these costs: (i) infla- tion erodes capital amortizations, which are fixed in nominal terms and never adjusted to compensate for inflation losses; (ii) except in 1978, water charges have not been set at levels that would cover O&M costs in full; (see para. 15 above) (iii) collection rates average less than two-thirds of assess- ments; and (iv) there is no legislation permitting recovery of on-farm devel- opment works carried out by TOPRAKSU. This situation affects all irrigation projects. The performance audit report on Seyhan II project discussed the water charges issue and the factors affecting cost recoveryl. 17. For several years, the Bank made progressively stronger representa- tions to the Government on the full recovery of O&M costs and recovery of "as much as practicable" of capital costs, both of which had been a stipulation of the Credit Agreement (DCA Section 4.07). But to no avail. Eventually, in 1975, the Bank decided that there would be no further lending for irrigation projects in Turkey, unless steps were taken to comply with the DCA. As already noted above (see para 16) the Government increased water charges in 1978, aiming fully to cover estimated O&M costs for that year - an increase, 1/ OED, PPAR on Seyhan Irrigation Project.. .op. cit. -5- in nominal terms, of 500% above then prevailing rates. While the increase was rapidly eroded by increases in the Consumer Price Index (CPI)L/, it was, politically, indeed a bold decision. It would, therefore, appear understand- able that, for socio-political reasons, the Government would not agree to increase water charges again in 1979 in line with accelerating rates of inflation. However, since this did lead to a reduction in water charges in fiscal terms, on top of which a reduction was also made in nominal terms, the position taken by the Bank under DCA Section 4.07, that the Government of Turkey remained in default of this covenant, was justified2/. 18. Suspending lending for irrigation, especially since the Bank was the only external agency involved in the sector, was indeed a serious decision. Irrigation is a crucial sector for agricultural development, particularly for fully exploiting its potential for high value, labor intensive crops such as fruits and vegetables. Covering 7% of the crop area, irrigated agriculture generates 40% of agriculture's gross production value and 25% of all agricul- tural exports. At the same time, the Government"s (DSI's and TOPRAKSU's) capability to complete irrigation works and hand them over to farmers has been declining rapidly from about 60,000 ha per year ten years ago to less than 20,000 ha per year today. The main factors leading to this decline are bud- getary constraints, insufficient equipment, and shortage of spare parts. These are precisely the kinds of constraints Bank-supported projects can help to remove. Instead, with about 0.75 million hectares requiring irrigation rehabilitation and on-farm development, and despite the possibility of a sizeable increase of the irrigable area, a substantial lending pipeline is closed because of non-collection of capital and O&M costs in rehabilitated projects. 19. No tangible results have emerged from the Bank's action so far, but there are indications that an imminent change is now likely. The present Government's declared policy is to reduce or eliminate subsidies and it is likely that O&M charges for 1981 will be assessed at a realistic level. However, the system of determining O&M charges is somewhat inflexible and therefore there is no assurance that, even if the 1981 assessments are adequate, they will not be eroded by inflation, as it had happened in 1978, or that actual collection rates at the time the charges are due, will more closely approximate assessment figures than in the past (see para 16 above). 1/ The CPI varied as follows: 1972, 28; 1978, 100; 1979, 160; 1980, 320; the estimate for 1981 is 512. 2/ In 1981, DSI submitted its proposal for water charges to the Council of Ministers, which would have fully recovered O&M costs for 1980. However, the Council has reduced this assessment by 50%. The increase over the previous assessment, in real terms, was thus only 10%, in view of the 40% inflation rate in 1980 (see Annex 1). - 6 - 20. While faced with the fact that the Government of Turkey had not complied with the cost recovery covenants, and the Bank had little choice but to withdraw from further irrigation lending, the question can be raised whether all avenues of cost recovery had been explored. The Bank took the line in this project that farmers' payments for the appreciation of their land and for the operation of the irrigation system must be expressed in terms of the irrigated area. In other countries, however, either at the suggestion of the Bank or at the initiative of the Borrower, other avenues have been con- sidered, for instance, a levy on export or industrial crops, a form of income tax on the farmers!! (e.g. "zakhat" in Malaysia and Yemen) or a procurement price for farm output reduced by the amount of cost recovery payment due. Since almost all irrigation departments, including DSI, operate on budgets allocated by the central treasury, the transfers of taxes, levies, urban subsidies etc. presents no administrative problems. In the audit's view, the Bank should have constructively advised on different ways of collecting appropriate charges to enable resumption of lending for irrigation projects. B. The Limited Area Expansion 21. The three areas covered by the Irrigation Rehabilitation and Comple- tion Project showed limited farmer interest in taking up irrigation, in this respect duplicating the reaction of farmers in the Seyhan II project. The audit mission made a special effort to try to ascertain why farmers were not behaving as the Turkish and Bank technicians had assumed they would. Several factors were mentioned which, although they do not fully resolve the issue, shed additional light on system operations and farmer behavior. 22. Non-irrigation of wheat. Farmers grow wheat - at least for subsis- tence - in all four areas (Seyhan included). If spring rains are adequate, wheat is grown as an entirely rainfed crop; if they are not, supplementary irrigation water may be applied once or twice to the crop during the growing period. Attempts to grow wheat as an irrigated crop in Tokat ended in fail- ure: local varieties grew too tall under irrigation and lodged. Farmers are not yet fully aware of the higher yields and resistance to lodging of the new, dwarf high-yielding varieties (HYVs). Furthermore, even though HYV seed supplies are rapidly increasing, they are not yet available in adequate quantities in Turkey. Information for the Tokat area shows overall wheat seed supplies of 40 tons in 1978, 50 tons in 1979, and 110 tons in 1980, compared with total requirements of some 1,200-1,500 tons. 1/ In 1981, the Government has introduced a withholding tax of 5% for all crops marketed and has "suspended" all subsidies on agricultural inputs. In this way they would substantially increase farmers" contribution to general revenue from which the Irrigation Department's O&M budget is financed (Annex 1). There appears to be a slight problem in semantics. Annex 1 uses the phrase "suspended all of the subventions which were recognized for most of the agricultural inputs". The EMENA regional office advises that although subsidies have been drastically reduced on inputs, they have not been completely eliminated. - 7 - 23. Common pasture lands. There are village-owned common pasture areas in all project areas. Exact figures could not be obtained but Turkish offi- cials estimated them to account for some 5-15% of the total project area. These areas, which have low output because of overgrazing, lack of pasture management, and general neglect, remain unirrigated because there is no formal organization to manage and tend them. The pasture lands offer substantial potential for increased livestock production, but this development would require the assistance of Government policy and local leadership. DSI would like to see these areas distributed among poor farmers and its O&M Department is currently preparing some basic studies on development possibilities. 24. Incomplete/insufficient development of the project areas. The project did indeed rehabilitate most of the project areas, but not all of it as it was meant to do. The officials estimate that the project raised the maximum irrigable area: whereas only 40-50% of the project areas could have been irrigated prior to the project, some 70-80% can be irrigated now. 25. The immediate causes for less than full development of the project areas are insufficient and inadequate installations. There are several reasons for this: (i) there was little coordination between the departments in charge of design and construction, and those responsible for operations and maintenance. As a result, tertiary canals were planned at 500 m intervals, without prior surveys to determine whether this was the appropriate distance. The O&M department discovered belatedly that a tertiary canal was required every 250 m. Only some of the addi- tional tertiaries have been built so far. Also, when the O&M department actually took over responsibilites for operation and maintenance they found several deficiencies which hampered proper O&M. Most of these were corrected under the project, but not all. (ii) Canal capacities were necessarily designed according to requirements of the then current or forecast cropping patterns. However, during project implementation a strong demand had developed for crops with higher water requirements: areas under sugar cane, sugar beets and alfalfa expanded rapidly, but had to be served by canals originally designed for wheat, that has much lower water requirement. Conse- quently, part of the area served by each canal had to remain dry. Some of these bottlenecks are being removed - but, again, not yet all of them. (iii) DSI constructs only the main and secondary drainage systems; TOPRAKSU is in charge of the field drains. Due to TOPRAKSU's shortage of staff and equipment, this work has usually been delayed. (iv) the project agencies were unable to convince farmers - and are legally not empowered to force them - to accept land consolidation, thus making system rehabilitation and completion much more diffi- cult. Thus, original rehabilitation and completion plans could not be fully implemented. - 8 - (v) Other factors mentioned to the audit mission as preventing full utilization of the project's irrigation potential by farmers in- clude: lack of farmer training; shortage of extension staff; lack of extension mobility (the project did provide vehicles for extension staff, but operating funds were inadequate); insufficient farm mechanization; lack of good marketing facilities; and lack of price incentives to producers. C. Trade-off Between Power Generation and Irrigated Agricultural Development 26. The ten-fold increase in oil prices and the prospect for a continu- ing increase of all energy prices have aroused concern about the effect of more expensive energy on agricultural development. This has focused on the increasing cost of fuel, nitrogenous fertilizers, pesticides, etc. and on the possibilities for switching agricultural commodities - in addition to residues - from food and fiber markets to energy markets!/. This project faces an even more direct confrontation over the use of a crucial agricultural input: water. 27. The specific case is the Tokat sub-project area. The Yesilirmak River waters are stored at the Almus Dam and released (mostly through a power house) back into the river bed, to be partly diverted some 10 km downstream at the Tokmakkaya Regulator, to irrigate the Omala plains (the uppermost of the three plains constituting the Tokat project area), and partly at the Gumenek Regulator, some 13 km further downstream, into the main Right Bank and Left Bank canals. 28. Water supplies are more than enough to irrigate all irrigable areas in the Tokat valley. Further, the Yesilirmak River bed between the Almus and Tokmakkaya Dams runs some 430 m higher than its confluent, the Kelkit Cayi River, running just 10 km north of it. Therefore, a trans-basin hydroelectric project was identified and is currently under construction. It consists of: (i) a new diversion dam, between the Almus and Tokmakkaya dams, the Kizildere dam; (ii) a trans-basin tunnel, the Kuvvet Tunnel, 6.7 km long, and (iii) the Kokluce hydro-electricz power house, on the bottom of the Kelkit Cayi. 29. DSI claims - and so it was understood by the Bank at appraisal - that only surplu.s water in the Tokat area will be diverted into Kokluce. DSI's claim is based on the then existing policy that, after domestic needs, irrigation has the highest priority for water use. However today, oil imports alone cost more than the total amount of foreign exchange brought in by all Turkish exports. While Turkey is largely deficient in power (it produces domestically only 40% of the total energy it requioes), it is more than 1/ cf., for example, The World Bank. Alcohol Production from Biomass in the Developing Countries, September 1980. - 9 - self-sufficient in agricultural production by some 10%1/. Therefore, since 1978, power generation, and not irrigation for agriculture, is the second highest priority water user. 30. The possible effect of such a policy change on agricultural develop- ment in the Tokat area can be estimated as follows: - average river runoff 758 Mm3/year - Almus dam's live storage capacity 812 Mm3 - Tokat area's irrigation requirements 196 Mm3/year - Kuvvet/Kokluce system's diversion - maximum capacity (25m3/sec) 788 Mm3/year - average diversion (20m3/sec) 630 Mm3/year - minimum diversion (18m3/sec required to generate Kokluce's firm yield) 568 Mm3/year On the basis of these figures: (i) the average planned diversion would leave only 128 Mm3/year (758m3- 630m3) available for irrigation, i.e., only 65% of estimated requirements, (ii) even the minimum possible diversion would leave only 190 Mm3/year, still under the required 196M m3/year, (iii) if Kokluce is operated at full capacity, no water would be left to irrigate the Tokat area. This is a realistic possibility four years from now, when Kokluce is com- pleted. In such an event the needs of the farmers in Tokat would be in competition with the needs of large urban centers for power. 31. Taking current electricity and farm prices into consideration, the 196Mm3 of water would produce US$7m worth of electricity but,used on the currently irrigated crops, US$16M worth of farm products. However, as with changing prices the balance shifts towards power generation, the following steps would have to be seriously considered, either separately or in conjunc- tion: (i) allocating irrigation water only to crops that respond well to it; (ii) progressively developing the groundwater resources of the Tokat area, currently estimated at about 28Mm3 sustainable yield per annum; (iii) identifying and evaluating the costs and benefits of more water regulating structures on the Yesilirmak River or its boundaries. 1/ See Figure 1: Energy and Agricultural Self-Sufficiency Ratios for Selected Countries 1976. - 10 - D. Information on Reimbursements 32. The Bank regularly informs the borrowers on the dates their reim- bursement claims have been honored and - once the relevant exchange rate has been utilized - the amounts involved. The Bank does not, however, inform the executing agency if different from the borrower. In the case of TOPRAKSU this practice led to problems. TOPRAKSU incurred project expenses, for which they were entitled to reimbursement. However, since they did not know if and when the Bank had reimbursed the Government, they felt their efforts to claim the funds from the Treasury were not as effective as they could have been. TOPRAKSU suggested to the audit mission that it would facilitate project execution if a copy of the note sent by the Bank to the Government informing of the reimbursements be sent to the executing agency. III. WATER MANAGEMENT IN THE TOKAT PROJECT AREA 33. Each year farmers must inform DSI what they are planning to plant and the date they want water deliveries to start. Requests must be submitted by the end of March; if the request is submitted late, a 10% fine over normal charges is collected. DSI can raise no objection or introduce no change to the farmers' requests. It just prepares requirement sheets (demand cards) for each farmer, which each farmer and the respective village chief sign. Tndi- vidual farmer requests are then aggregated by DSI and checked against overall water availabilities (no problem has occurred yet). A schedule of water deliveries into each main, secondary and tertiary canal is then prepared. 34. There is a hydro-electric power generation plant at the foot of the Almus dam. However,. no need to coordinate dam water releases for power generation and for irrigation has arisen. Power releases (60 m3/sec) are much larger than irrigation requirements; similarly, live storage capacity at the dam is four times as large as total irrigation requirements. Schedules for water release from each dam in Turkey for power generation and irrigation are centrally prepared in Ankara jointly by DSI and TEK, the Turkish electric concern. 35. DSIs O&M Department has prepared a standard operational manual for all irrigation projects in Turkey. This fits properly all of Tokat's needs; no special manual for Tokat is thus required. 36. Staffing. The Tokat area is operated by a chief operation engineer, three operation engineers and one maintenance engineer. Two diversion dam operators and 32 water distribution technicians (one per approximately each 1,000 ha) work at the operation engineers' command. Overall supervision and control is exercised by the chief operation engineer, together with DSIfs chief district engineer. - 11 - 37. Conveyance efficiency. No regular measurements are undertaken over the whole system. Diversion into the main canals, and from the main into the secondary canals are regularly measured and recorded; diversion into the tertiary canals is measured only from time to time. Water deliveries to farmers are made through farm outlets which can be set at three positions: 10; 20; and 30 1/sec. Based on the figures thus available, DSI estimates conveyance losses at between 10-15%. This would be a satisfactory figure, which can be achieved because all system canals were lined under the project. 38. Water efficiency at the farm level. There are no figures available on this matter. - 12 - Attachment. SOME MINOR DISAGREEMENTS WITH THE PCR. Some statements or conclusions in the PCR were investigated by the audit and found unwarranted. (i) Tokat"s project area. The appraisal report refers to a total project area of 22,900 ha; the PCR to 21,530 ha, without explaining the difference. The figure quoted by DSI to the audit mission was 21,350 ha. The difference would appear to be due to the expansion of Tokat town and its associated industrial suburbs, covering some previously irrigated areas, and to the fact that appraisal estimates were based on 1:25,000 maps whereas the figures given to the audit mission were based on the more accurate 1:5,000 maps. (ii) Tokat's delivery capacity. The appraisal report refers to a 0.95 1/sec/ha flow rate (actually, DSI states that it was 1.0); the PCR puts it at 0.8 1/sec/ha and states that this is 15% below the requirements according to the appraisal report, which could cause some constraints in future years. The PCIrs figure was found to be wrong. Current flow rate is: Tokmakkaya Dam 3 m3/sec Gumenek Dam, left bank 9 m3/sec Gumenek Dam, right bank 7 m3/sec f9 m3/sec = 0.89 1/sec/ha (iii) Extent of Delays. The PCR states that "..... construction of project works went according to schedule after the (start-up) delay" (para. 3.05). This statement is not correct, for it is expected to take until 1981 to have most project works completed, (PPAM, para. 4), as compared to the five-year schedule envisaged at appraisal. Thus, the project suffered an average of about 40% time overrun after the one-year start-up delay (PCR, paras. 3.01-3.03). Total delays, including the latter, amount to approximately 60%. - 13 - FIGURE 1 ENERGY AND AGRICULTURAL SELF-SUFFICIENCY RATIOS FOR SELECTED COUNTRIES, 1976 Australia 1.8 * Brazil C1 1.6 - Argentina *Thailand South *Colombia 1.4 Africa 0 0 Philippines 0 United * Sudan States 1.2- Canada TURKEY 0 g Burma France India 1.0 Pakistan Peru Mexico S Bangladesh P 0 *Zaire Nigeria OSpain Poland USSR Korea, Rep. of Egypt .8 Japan Venezuela *Italy - .6- Germany, Fed. Rep. of United Kingdom .4- " CD) .2- I I I IIIII .2 .4 .6 .8 1.0 1.2 1.4 1.6 1.8 Energy Deficit Energy Surplus Energy Self-Sufficiency Source: Developed by Dr. N. Rask from FAO and World Bank Data World Bank - 22515 Cable Addrss - DOI -"14 - TELEX NO: 42305 DSIM TR REPUBLIC OF TURKEY ANNEX I MINISTRY OF POWER AND NATURAL RESOURCES Page 1 GENERAL DIRECTORATE OF STATE HYDRAULIC WORKS (DS) CHIEF ADVISOR ON FOREIGN CREDITS AND FOREIGN RELATIONS ANKARA - TURKEY Our Ref No.D.K.-04-93/81 3une 1,1981 Mr. Shiv S. Kapur Director Operations Evaluation Department The World Bank 1818 H. Street, N.W Washington, D.C., 20433 USA Dear Mre Kapur, Re : Project Performance Audit Report on Turkey Irrigation Rehabili- tation and Completion Project (Credit 281-TU, Draft Report). Thank you for the draft copy of the Audit Report where all of the data on three projects have been compiled and processed diligently and delivered to us in such a short time. The Report has been studied by the authors of our draft Project Comple- tion Report which was submitted to the Bank, fbund the Report agreeable with all the data that were supplied by us. In addition, we would like to make the following remarks on the report, which have been' occured since the date the Report had been submitted: 1- The Credit was closed in April 1979 but the activities gf DSI and TOPRAKSU at Tokat were not completed until later part of 1979. In 1980, finishing touches went on the drainage works in Silifke Project. At KbprOgay DSI and on-farm development Works are going on and will be completed in 1981. The farm development works at KbprOgay have been rather slow so far, because of farmers' (mostly the renters or crop shares) uncooperative attitude. Because, the farmers are not forced legally to allow access to the TOPRAKSU onto their lands to carry out on-farm,development works. Presently, TOPRAKSU has taken this matter to the Council of Minis- ters for a legislation on this and the land consolidation issues. 2- The Audit Report covers the project activities performed up to the end of 1978. There have been some delays during the implementation phase and the succeeding years,of the projects due to the following factors : 1974 and the following years constitute an unlucky period where country past through asocial and economic turmoil. The energy crises which had embraced the whole World hit Turkey even harder. This eventually effected the project activities of DSI, TOPRAKSU and Agricultural Extension Service, adversely. Because of the galloping inflation,most of the constractors were granted withdrawal from their oblilations and OSI and TOPRAKSU had to complete some of the urgent projects on force-account basis. - 15 - General Directorate Of State Hydraulic Works (D81) Chief Advisor On Foreign Credits And Foreign Relations ANNEX I Page 2 However, since September 1980 the Country rejoined to peace and reactivated He resources with full potential. DSI's annual invest- ment budget jumped from 17 Billion TL. in 1978, to 91 Billions in 1981. 3- Because of the political considerations, until 1978 the cost recovery of 0 and M cost and recovery of capital costs were not high enough to meet the full investments. But in 1978, the Government of Turkey then had increased the irrigation water charges 500 % for full cover- age. But as this was also noted in the Audit Report, it has been rapidly eroded by the galloping inflation which ruled over the Country until the end of 1980.In 1981, DSI had prepared a new tariff for the full coverage of water charges and recovery. of capital costs and submitted it to the approval of the Council of Ministers but only assessment of 50 % of 0 and M costs for 1980 has been approved . This is a little over then 1980 overall inflation rate which was as high as 40 %. However, the present Government partly compensated these costs by levying new taxes on farm products by witholding 5 % of each crop mar- keted by farmers. The Government, on the other hand, increased the sale taxes on consumer goods and the property taxes. In addition to these, the Government has suspended all of the subventions which were recog- nized for most of the agricultural inputs. This means the farmers who benefited from irrigation have became subject to repay, implicitly the unpaid part of water charges and the capital costs. This way there have been more revenue sources created for Government's enterprise to develop more soil and water sources to the services of people living in the rural areas. May I lake this opportunity, Mr. Kopur also to thank you and Mr. Olivares for his keen observations in the Tokat Project Area and his guick writey up in evaluting all the data that were provided to him during his visit to Turkey. Please, don't hesitate to call on us if you need any more information or clarification in regards to the completion of the Audit Report. With best regards. Very Truly Yours General Director DSI - 16 - TURKEY IRRIGATION REHABILITATION AND COMPLETION PROJECT. COMPLETION REPORT I. BACKGROUND 1.01 The Irrigation Rehabilitation and Completion Project was designed to improve and complete three irrigation projects in Turkey. Two of the projects, Silifke and Koprucay, are on the Mediterranean Coast, and the third, Tokat, is in the Anatolia Plateau about 350 km northeast of Ankara. 1.02 Silifke (about 5,000 ba) is primarily a wheat, rice, groundnut and vegetable producing area, which had an-existing irrigation and drainage system. Koprucay (about 22,000 ha) is a cotton and wheat producing area with a small area of citrus and vegetables. Irrigation and drainage works had been essentially completed prior to the project on about 40% of the area. In Tokat (about 23,000 ha) main crops are winter wheat, sugar beets and orchard, with small areas in vegetables, maize and vineyards. In 1971, the irrigation and drainage system was only about 50% complete. 1.03 In the past, Turkey had made large investments in irrigation projects but they were seriously underutilized. This situation resulted from favoring construction of major works at the expense of providing distribution and drainage systems and on-farm development and supporting research and extension. A Bank mission that visited Turkey in 1970 to review general economic conditions recommended that existing irrigation systems be completed and fully utilized. This mission estimated that Turkey should be able to bring each year some 60,000-80,000 ha of project lands under full irrigation and intensive agricultural development and complete work, over the next 10-13 years, on the entire 800,000 ha for which major works had then been or were being constructed. The command area of public irrigation schemes in Turkey now comprises about 1.5 million ha, of which at least half require on-farm development before the land can be cropped to its full potential. 1.04 The Irrigation Rehabilitation and Completion Project was expected to be the first of a number of similar Bank operations in Turkey. 1/ The project would complete irrigation and drainage facilities on three projects so that 50,000 ha of land would be suitable for intensive irrigated agri- culture in contrast to the 26,000 ha then receiving only partial service. The project would also complete the on-farm develbpment by leveling the lands, installing tile and surfacr drainage and improving village roads, and would improve extension activities in the three areas. 1.05 Previous Bank Group operations in the Turkish agricultural sector had focused on the Seyban Irrigation projects which were being executed 1/ The term "Bank" refers to IBRD and IDA. - 17 - satisfactorily. In addition, one credit and one credit and loan, not yet effective, had been made to finance a Dairy Development project and a Fruit and Vegetable Export Project. Five Bank operations were planned for 1972. In addition to the rehabilitation project, loans or credits were also contemplated for livestock development, a steel plant, a fertilizer plant and urbanization. 1.06 Bank/IDA lending in Turkey to date has now reached US$2,440 mil- lion for 57 projects. Agriculture accounts for 30% of funds lent. From concentrating on capital intensive investments in the 1950s and 1960s, increasing attention is now being paid to farm support services commencing with the agricultural extension components-of the Seyhan II Project. Beginning with the Corum-Cankiri Rural Development Project, Livestock III and Livestock IV Projects, emphasis has begun to shift away from investing in the more agriculturally and industrially developed areas to less developed regions including rainfed areas. II. FORMULATION Preparation, Appraisal and Agreement 2.01 The Government of Turkey and the Bank were agreed on the impor- tance-of bringing existing irrigation systems to completion and full utilization. However, DSI's engineering consultants, Tahal-ECI Inc., were certain that detailed feasibility studies of proposed projects would be absolutely necessary. .2.02 In order to eliminate the long time required to recruit consul- tants and carry out feasibility studies, the Bank suggested that DSI invite assistance from FAO/IBRD Cooperative Programme and prepare background data and information in lieu of full feasibility studies. 2.03 Discussions between the Bank and Government representatives on the subject of the Irrigation Rehabilitation and Completion Project were held in DSI headquarters in Ankara, Turkey in February 1971. The Government was reluctant to undertake full feasibility studies for the three-project areas as it wished to commence work on the project as soon as possible. The'Bank therefore reviewed the available data and felt it adequate. Hence, the .Bank agreed to the Government's proposal to limit project preparation to the presentation of necessary basic data by DSI in the form of tables and estimates. Had the Bank insisted that the Government carry out full feasibility studies, the Government's commitment to the Project may have increased, with consequent beneficial effects on project implementation. 2.04 The three project areas, Koprucay, Silifke and Tokat, were chosen by the Government from a much longer list as being projects of high priority, where work was actively in progress and sufficient data was available to form a basis for seeking financing. These projects had also -18- been endorsed by the 1970 Bank mission as being of high priority. The project was appraised in May/June 1971. Project Description 2.05 The project as agreed at negotiations comprised: a. rehabilitation and improvement of about 26,000 ha of existing irrigation and drainage systems; b. construction of new irrigation and drainage systems to serve an additional 24,000 ha; c. provision of on-farm improvements in the form of land leveling, surface drainage and tile drainage; d. procurement of equipment, materials and supplies; e. consultancy services to assist TOPRAKSU and the Extension Service in project implementation; and f. training abroad for government staff. 2.06 During negotiations, which took place in November 1971, the following agreements were reached: a. the Government would follow international competitive bidding procedures.; b.-- DSI and TOPRAKSU would submit a schedule within six months of the signing of the Credit Agreement to show specific civil works to be carried out by contract and those by force account; c. the Government would issue import licenses promptly; d. separate coordinating committees for each of the projects would be established; e. an extension coordinator would be appointed for each project and extension activities would be strengthened; f. TOPRAKSU and;the Extension Service would employ consultants in project execution; g. the Government would ensure that adequate short- and medium-term agricultural credit was available; h. the Government would assess and collect water charges sufficient to cover operation and maintenance costs and to recover a reason- able portion of investment costs; and i. audited project accounts would be submitted to the Bank within four months after the closing date of the Turkish fiscal year. - 19 - 2.07 A credit of US$18 million was approved by the Board on January 18, 1972. The Credit Agreement was signed on January 25, 1972 and became effective on April 27, 1972. Project Objectives 2.08 The project's objective was to make some 50,000 ha fully suitable for intensive irrigated agriculture through the rehabilitation and improve- ment of existing irrigation and drainage systems in three separate areas-- -Koprucay, Silifke and Tokat. Prior to the project, about 26,000 ha received only partial service. The irrigation and drainage works, together with the completion of on-farm developmen, increased extension activities and adequate credit facilities were expected to result in increased yields and cropping intensities, and a shift in cropping patterns toward higher value crops in all three project areas. Organization 2.09 The three principal agencies involved in implementation of the project were; a. DSI - which was responsible for constructing all project irrigation and drainage works down to the farm unit level; b. TOPRAKSU - which was responsible for carrying out all on-farm development works including land leveling, installing tile and surface drains, improving village roads and amending soil conditions; and c. the Extension Service - which was responsibile for educating farmers in improved methods of cultivation and in inducing the required change in cropping patterns by expanding the areas in higher value crops. In addition, TCZB, the Agricultural Bank of Turkey which is the major institutional source of credit to farmers and farmers' cooperatives in Turkey, was responsible for supplying agricultural credit in the project areas. To coordinate the activities of these agencies, a committee, consisting of the top local official of each agency, was to be formed in each project area (Chart I). III. IMPLEMENTATION Effectiveness and Start-up 3.01 The Credit Agreement became effective on April 27, 1972, but project implementation did not make an auspicious start. - 20 - 3.02 Negotiations were held in November 1971, and Board approval was January 1972; yet in May 1972, the 'staff of the three agencies primarily responsible for project implementation, DSI, TOPRAKSU and the Extension Service, were not aware of the terms and obligations of the Credit Agree- ment. Consequently, no steps had been taken to implement the provisions of the side letters and the Agreement. In particular, no moves had been made to appoint the coordinating committee for each project area, to prepare the construction and on-farm development work schedules for submission to the Bank, and to make provisions in the 1972 budget to finance the project work scheduled for that year. In addition, no agency had started preparation of specifications and bid documents for ordering equipment either locally or overseas because of lack of information as to the equipment and vehicles to be ordered and, for local equipment, lack of provision in the current budget. 3.03 In 1972, the budgetary allocation for construction work by DSI amounted to TL 17.5 million compared with the TL 66 million envisaged in the appraisal report. TOPRAKSU had made no plans to carry out any drain- age, road work or soil amendment in 1972. The equipment and vehicles needed by TOPRAKSU for on-farm development had not been ordered and therefore could not be on the site until 1973 at the earliest. Hence,,the implementation schedule of both DSI and TOPRAKSU had to be amended, but no change in the closing date of the credit was envisaged. 3.04 According to the Credit Agreement, consultants should have been employed by TOPRAKSU and the Extension Service by July 1, 1972. However, no steps had been taken to fulfill that obligation by either agency because Turkish laws and regulations existing at that time prevented TOPRAKSU and the Extension Service from entering into any contract of greater than one year's duration. To contract for longer periods, a decree had to be obtained from the Council of Ministers which, in Turkey, can take an extremely long time. This should have been made known at the time of nego- tiations so that the Bank could have appropriately modified the condition of the Credit Agreement requiring the recruitment of consultants. The outcome of this legal impediment was that it was impossible for TOPRAKSU or the Extension Service to hire consultants by the appointed date. Both agencies applied for what was to be the first of several extensions, despite which, consultants were never employed. Physical Implementation 3.05 In general, construction of project works went according to sched- ule after the delay in start-up mentioned above, which was caused by the implementing agencies not being prepared and by inadequate budgetary allo- cations. By early 1980 DSI works were virtually complete. TOPRAKSU had finished its on-farm development works in Silifke and Tokat. However, its program in the Koprucay area was only about 50% complete for reasons partly beyond its control (Annex I, Table 2 and para 3.07). 3.06 The project completion and credit closing dateowas extended from June 1977 in the original plan to April 1979. The supervision mission at - 21 - the end of 1977 reported no progress in DSI works since May of that year; but on-farm works, except for those at Koprucay, were practically com- pleted. In February 1930, DSI had still to put the finishing touches on irrigation-an& drainage pumping stations in the Silifke and Koprucay areas. In part because of the country's economic disarray and severe budgetary constraints on the implementing agencies, project works ir the Koprucay areas had come to a standstill in 1979 which resulted in 2,000 ha not being drained properly and a further 2,000 ha being irrigated either inefficiently or not at all. 3.07 On-farm development works in the Koprucay area have been only half completed because high cotton prices in the early years of project imple- mentation resulted in the farmers not allowing TOPRAKSU to level their land and construct farm drains during the cotton growing season and they could not legally be obliged to do so (paras 7.09 and 7.10). Programmed works for the other two main components, the tile drains and farm roads, have fairly well attained their completion objective (around 70%). Programmed objectives for heavy land leveling have been met. Although medium and light leveling is not completed, practically all the area is irrigable, with unleveled land requiring extra water.- In general, the irrigation system has the capacity to supply this extra water although some farmers with unleveled land in pumped areas complain of not receiving sufficient water. It is expected that, in the future, farmers with chronic supply constraints will gradually have the leveling done and accept foregoing the income from one summer crop. Procurement' 3.08 Procurement was a major issue during negotiations. The Government was most reluctant to accept international competitive bidding (ICB) pro- cedures for civil works. In particular, the Government did not agree to the Bank's proposal for TOPRAKSU to contract out on-farm development works on the grounds that: (i) a formal commitment of that nature was a depar- ture from previous Bank requirements under the Seyban Project (Loan 587/Credit 143-TU); (ii) the procedure had in fact already been tried at Seyhan without attracting suitable bids; and (iii) legislation would be needed to enable TOPRAKSU to hire out equipment. 3.09 It was finally agreed that DSI and TOPRAKSU would submit to the Bank within six months from the date of signing of the Credit Agreement a schedule of work to be carried out on each of the project areas. The schedules would give recommendations for specific work to be undertaken by each agency either by contract (under ICB procedures) or by force acc6unt, giving reasons for such recommendations. The Bank recognized that it would probably be necessary for DSI to carry out civil works by force account in existing irrigated areas due to the need to maintain water services. But in all other areas it was expected that civil works would be carried out by contract according to ICB procedures. TOPRAKSU had traditionally carried out all work by force account in the past, in view of lack of contractor interest and capability to perform on-farm improvement works. However, the Bank felt that local contractors should be encouraged and recommended that - 22 - TOPRAKSU prepare tender documents to enable ICB and to invite local con- tractors to submit bids for a block of at least 1,000 ha of land leveling and tile drainage in each of the project areas. These blocks would be discussed at the time of the review of the work schedule. 3.10 Work schedules were submitted by DSI and TOPRAKSU in July 1972. Both agencies proposed to carry out all works either by force account or local. contract only on the grounds that the work involved was of small size and scattered in nature, and that there was a need to coordinate construc- tion activities with the delivery of irrigation water. The Bank agreed that for Silifke, the smallest project area, all civil and on-farm develop- ment works could be carried out by force a7ccount, as proposed by DSI and TOPRAKSU. However, in Koprucay and Tokat, the Bank reiterated that certain civil works were suitable for ICB, and that for on-farm development, TOPRAKSU should put to ICB at least 1,000 ha of land leveling and tile drainage. The Bank also suggested that TOPRAKSU could make the land leveling and tile drainage contracts more attractive by including feeder roads and surface drainage in these contracts and/or obtaining government permission to enter into contracts of greater than one year's duration (para. 3.04). 3.11 A Bank supervision mission in October 1972 again discussed the work schedules with the Government, as a result of which -the Bank accepted the Government's position that project works in general were not attractive for ICB, in view of their small and scattered nature 1/ and the one-year limitation on TOPRAKSU contracts. Only the Yesirlimak Channel Improvement at Tokat was to be reserved for ICB, 3.12 The final outcome was that most project works were carried out by force account with limited participation by local contractors. Even the Yesirlimak Channel Improvement was put to local tendering following a proposal by DSI in 1974 (to which the Bank agreed) that the small size of the civil works made it suitable for local tendering only. Greater participation by local contractors, particularly for on-farm development work, may have reduced the delays in project implementation. But project works were less attractive to local contractors because of the Government's one-year limitation on TOPRAKSU contracts. This remained in force throughout the implementing period of the project and the Bank was unable to persuade the Government to remove the limitation. 3.13 Equipment procured by DSI and TOPRAKSU is shown in Annex I, Tables 1 and 2. TOPRAKSU ordered its equipment early and avoided major cost over- runs. But DSI's slowness in taking procurement action--the bulk of its equipment was not ordered until 1975--resulted in considerable cost over- runs. Although DSI made minor revisions to the equipment list and deleted 1/ This was particularly true of land-leveling works. In the Tokat area in 1972, TOPRAKSU had leveled 647 ha of land spread over 2,000 parcels of land and grouped in 349 operations. - 23 - those items which were already available in its machinery pool,. the cost of the revised list exceeded allocation by over US$1.2 million (or 50%). Consultants 3.14 The Credit Agreement provided for two consultants for TOPRAKSU to be in place for two years and two to be employed by the Extension Service for at least five years. The TOPRAKSU consultants were to assist TOPRAKSU in organizing the on-farm development work and ensuring that the heavy equipment was used to its potential efficiency. The consultants to the Extension Service would organize and coordinate extension services in the field. Government limitations on the ability of TOPRAKSU and the Extension Service to enter into contracts of more than one year caused both agencies to request extensions of the contract date by which consultants were to be employed from the original date of July 1, 1972 to October 1, 1972, December 1, 1972 and March 1, 1973. These extensions proved meaningless for no consultant was ever employed. 3.15 The Bank continued to write and cable the Government urging that action be taken to recruit the consultants. At various times, the Bank proposed alternative arrangements whereby consultants could be employed without contravening the Government's constraints on TOPRAKSU's and the Extension Service's contractual capabilities. But this was all to no avail. It was clear that TOPRAKSU and the Extension Service were merely going through the motions and had no real plans to hire consultants. In mid-1976, the Bank proposed to the Government that the US$400,000 allocated for consultants be cancelled since it would have made no sense to hire con- sultants when the closing date of the Credit was (at that time) June 30, 1977. However, the Government subsequently requested and the Bank agreed to the transferring of these funds to the unallocated category. 3.16 It can be argued whether consultants would have been hired had the Bank been willing to take strong measures in the early years of the project to insist on the Government's compliance with the Credit Agreement (such as a possible suspension of disbursements). Nevertheless, the completion mission believes that the failure to hire consultants contributed in large part both to the slowness which characterized TOPRAKSU's leveling and drainage program and to the weaknesses in agricultural extension. Performance of the Borrower, Consultants, Contractors and Suppliers 3.17 The main issues regarding the Government's performance are: failure to employ consultants (para 3.16), failure to impose and collect adequate water charges (para 6.15), failure to submit audit reports (para 6.23), failure to provide adequate support to the Extension Service (para 6.10), and delays in completion of the work (para 3.06). The quality and timeliness of the Government's progress reports varied between agencies (the Extension Service reports were the least informative) and over time. Reporting by DSI, TOPRAKSU and the Extension Service tended to be superior in the early years of the project. The Government's failure to submit - 24 - audit reports and to impose and collect adequate water charges as well as accumulated delays in completion of works were problems also encountered in the Seyhan Project and are believed symptomatic of the Government's prefer- ence to concentrate on new projects, letting the completion of old ones lapse. 3.18 Services of equipment suppliers were satisfactory both to the Bank and to the Government. Consultants were not employed, so their services cannot be evaluated. Quality of DSI and TOPRAKSU construction works was also satisfactory. However, the planning and pace of implementation of TOPRAKSU works would have been improved had consultants been employed as anticipated (para 3.15). 3.19 The performance of the Extension Service did not meet appraisal expectations. Its efficiency was impaired by lack of Government support and by the nonappointment of consultants. These issues are discussed in greater detail in paras 6.07-6.11. Costs 1/ 3.20 Final project costs including estimated costs of work to be com- pleted in 1980 and 1981 are US$40.2 million. However, this figure does not include the cost of consultants nor the cost of leveling about 10,000 ha of land at Koprucay. In this project area, TOPRAKSU intends to complete only 8,000 ha of the 18,000 ha of land leveling envisaged at appraisal because of farmer resistance to the program. All the land can now be irrigated, however, despite its not being completely leveled. At appraisal, total project cost was estimated to be US$36.6 million. When compared to this appraisal estimate, cost overruns on project works are 10%. Also of interest is the comparison between final project costs and appraisal costs adjusted to take into account the non-appointment of consultants and the reduced area of land leveling at Koprucay (US$35.0 million). In this case, cost overruns amount to 15%, still relatively small. 3.21 Expressed in local currency, total project cost is TL 677.4 mil- lion which amounts to a cost overrun of 24% when compared to the original appraisal estimate (TL 547.6 million) and an overrun of 30% when compared to the adjusted appraisal estimate (TL 522.7 million). The discrepancy between US dollar and local currency cost overruns is a reflection of h.igh 1/ In some cases, cost estimates for the same component of project works given in three sources of project cost data--the Government's draft completion, quarterly reports submitted to the Bank by the Government's implementing agencies, and Bank supervision reports--are widely disparate. For example, the draft completion report estimates the cost of completed DSI works at Tokat to be only TL 148 million; whereas DSI quarterly reports estimate the cost of these works to be TL 185 million at the end of 1977, a time when DSI work at Tokat was not completed. The completion mission and the Government have attempted to reconcile such discrepancies, but in some cases; the issue could only be resolved using best estimates. -25- local inflation relative to international levels and the corresponding suc- cessive devaluations of the Turkish Lira from US$l equalling TL 14 in 1971 to TL 36 in 1979. 3.22 A summary analysis of appraised and final costs is given below. Table 1 in Annex II gives a more detailed breakdown of project costs. Coparison of Total Project Costs (1 (2) (3) -() (2) (3) Adjusted Adjusted Appraisal Appraisal Final % Difference % Difference Appraisal Appraisal Final % Difference % Difference Estinate Estimates Cost (1)& (2) & (3) Estimates Estimates Cost (1) & (3) (2) & (3) -------------------------------------------------------(SMilo)-------- Civil Works Irrigation and Drainane 199.0 -199.0 359.7 81% 81% 13.3 13.3 22.2 67% 67% On-Farm Develop* ment 122.6 105.2 180.5 147% 72% 8.2 7.1 10.0 22% a% Equii!ent 131.4 131.4 135.6 3% 3% 8.8 8.8 7.9 -10% -104 Consultants 7.5 - - - 0.5 - - TraInIM 1.5 1.5 1.6 7% 7% 0.1 0.1 0.1 . Contingencies 8 -2-2 -1 -- Total 22.7 677.1. .16. "_o 4 0.2 _0f_ 3.23 The irrigation and drainage works carried out by DSI are the largest project component and account for most of the increase in total costs in absolute terms. . The TL 160 million increase is accounted for by inclusion of a canalette manufacturing plant in Koprucay (TL 58 million), price escalation (M 55 million), discrepancies between estimates 'and original contract prices (TL 29 million) and quantity changes (TL 18 million). Details are; given in Table 2, Annex II. Aside from the cost increase due to the building of a canalette factory which had not been anticipated at appraisal, cost overruns for both DSI and TOPRAKSU works are mainly due to underestimated price contingencies and implementation delays. The appraisal provision for contingencies was only 5% per year, which proved quite inadequate in the face of accelerating rates of infla- tion experienced since the end of 1973 both within Turkey and inter- nationally. 3.24 When final costs are compared to appraisal estimates on a project area basis, the following table shows that cost overruns were greatest at - 26 - Koprucay (44% or TL 106 million). These differences can be explained largely by the slow pace of implementation at Koprucay and the absence of any pumping irrigation works at Tokat. The cost overruns on construction of pump buildings carried out in Koprucay and Silifke were high, being particularly high at Koprucay (see Annex II, Table 2). Total Project Costs by Project Area Koprucay Silifke Tokat Total . ----.--------(Current Prices - TL Million)----- Appraisal Estimate - 239.4 . 59.3 248.8 547.5 Final Cost 345.8 70.9 260.7 677.4 Difference 44% 20% 5% 24% Disbursements 3.25 The slow pace of disbursements was an issue during implementa- tion. By May 31, 1975, three years after credit effectiveness, only US5.3 million (or 30%) of the credit proceeds had been disbursed. Two factors were responsible--delays in construction and delays in the Government requesting reimbursement. DSI's explanation for their slowness in sub- mitting withdrawal applications was that it took a long time to obtain the necessary clearances from the different governmental agencies and the Central Bank. A Bank disbursement officer visited Turkey in 1975 to assist the authorities in expediting the preparation and submission of withdrawal applications. The Bank also followed up on this matter through cables, letters and supervision missions. As a result, the pace of disbursements improved. Disbursement Schedule Appraisal Actual as % Calendar Year Estimate Actual of Estimate --- (US$ Million) - 1972 0.5 - 1973 5.9 1.7 29 1974 11.5 4.4 38 1975 15.0 19.6 65 1976 16.9 13.3 79 1977 18.0 15.3 85 1978 16.8 93 1979 18.0 100 3.26 The credit closing date was extended three times. The first extension, from June 30, 1977, to December 31, 1977, was granted because of increased construction costs and the need for additional time to utilize the proceeds of the credit. The'second extension, from December 31, 1977, - 27 - to December 31, 1978, followed a review of project status by a supervision mission in late 1977. At the same time, the Government agreed to complete the works'using its own funds since TOPRAKSU was unlikely to complete its on-farm development works by 1981, much less accelerate the work to permit completion even earlier. The final extension, from December 31, 1978, to April 30, 1979, was necessitated by DSI's slowness in taking procurement action. 3.27 The US$400,000 provision for consultants in Category III of the Credit was reallocated to the Unallocated category in 1977. The US$1.9 million then in the Unallocated category was later transferred to Cate- gories I and II enabling the credit funds to be expended before the extended closing date. Final disbursements as compared to appraisal allocations are shown below. Appraisal Actual Allocation Disbursements Actual as % Category 1-25-72 8-13-79 of Estimate ------ (US$ Millions) ---- (%) I. Irrigation and drainage works 9.9 10.64 108 II. Equipment, materials and vehicles .6.1 7.28 119 III. Consultants 0.4 - IV. Training 0.1 0.08 80. V. Unallocated 1.5 - - Total 18.0 18.00 100 IV. AGRICULTURAL IMPACT Objectives and Strategy 4.01 The project's objective was to increase agricultural production in all three areas by improving and extending existing irrigation and drainage systems and by carrying out on-farm development works. Cropping intensi- ties were expected to increase from their preproject levels of between 80% and 90% to 100% in Tokat and to 130% in Koprucay and Silifke. Cropping patterns would change as the areas in higher value crops expanded. In order to achieve these targets, assurances were obtained from the Govern- ment that adequate credit to farmers would be provided so that proper inputs of seeds, fertilizers and pesticides could be used. The extension service was to be strengthened so that it could effectively carry out its responsibility of educating farmers in modern and improved methods of cultivation. The Extension Service would also be responsible for inducing the required change in cropping pattern by expanding the areas in higher value crops on suitable lands. 4.02 Full development was not expected to be reached until 1979 in Silifke and 1981 in Koprucay and Tokat. Delays in the implementation of project works iti all three areas meant that full development is now - 28 - expected to be attained by 1981 in Silifke and by 1985 in Tokat. For Koprucay, DSI works are expected to be complete by 1981 but it is unlikely that TOPRAKSU will complete on-farm works even by 1985. Progress in achieving the project's objectives in each of the three project areas has been mixed. In Koprucay, appraisal expectations for yield increases were not met in all cases and, contrary to appraisal expectations, there has been a movement away from cotton production. In Silifke however, yields of all major crops have either equalled or exceeded appraisal expectations. The dropping pattern is in line with that expected at appraisal although more diversified. In Tokat, yields have generally increased far beyond appraisal expectations and the movement away from cereal production toward higher value cash crops has been much greater than anticipated at appraisal. Irrigation facilities have never been fully utilized since large areas of cereal are grown in all three areas and this is not normally irrigated except in very dry years. Monitoring and Evaluation 4.03 The extension service in each area was required to submit detailed quarterly reports following guidelines provided by the Bank. This was rarely done; generally, reports were infrequent and uninformative. The Bank took few steps to rectify the situation until 1977 when a supervision mission reported that serious efforts should be made to improve the quality and frequency of the extension servide reporting. However, this recom- mendation came much too late as the project closing date was then expected to be December 1978. Poor follow up on this matter early on in the project was a serious omission on the part of the Bank for it meant that for almost the entire period of project implementation, no real monitoring and evalua- tion of the project's agricultural impact could be or was carried out. The consequences of this omission became particularly evident at the time of completion report writing, both for the Bank and for the Turkish Govern- ment. Discrepancies and inadequacies in data abound and in many cases these could only be resolved through the use of best estimates. It is essential that throughout project implementation detailed reports should be submitted to the Bank on a regular basis and it is equally important that the Bank should take appropriate steps, both at headquarters and in the field, to maintain the quality of these reports and their usefulness as a tool of monitoring and-evaluation. Koprucay 4.04 In Koprucay, incremental benefits were expected to result largely from increased cotton production due to the completion of the irrigation and drainage works and increased yields as a result of better agronomic practices. Cotton area would expand from 10,500 to 12,200 ha and produc- tion was estimated to increase by 65% over preproject levels to reach 43,000 tons. A new crop, tomatoes, would be introduced; and -at full development, 1,600 ha of tomatoes would be grown. 4.05 The following table compares appraisal and completion estimates of cropping pattern and yields. Cropped areas and yields over the period 1973 to 1979 are shown in greater detail in Annex II, Table 1. 29- Koprucay: Cropping Pattern and Yields With Project With Project With Project Full Development Full Development Completion Mission Without Project Actual A.ppraisal ZZ: =.L;w"tv EbLA..:LC (1971) (1979) (1981) (1985) (1985) Crop Area Yield Area Yield Area Yield Area Yield Area Yield Ha Tons/Ha Ha Tons/Ha Ha Tons/Ha Ha Tons/Ha Na Tons/Ha Irrigated Cereal - - 1,819 4.0 6,600 4.25 3,092 4.4 3,700 4.0 Cotton 8,700 2.9 8,186 2.7 12,200 3.5 12,588 3.75 9,500 3.5 Rice 700- 3.0 257 4.0 1,100 4.7 1,546 6.0 650 5.0 Sesame - - 1,926 0.9 1,600 0.6 883 1.2 2,400 1.0 melon 70 25.0 337 25.0 700 30.0 442 25.0 1.050 25.0 Citrus /2 - - 186 28.0 - - 1,546 30.0 720 28.0 Vegetables- 70 30.0 754 30.0 1,600 40.0 2,871 27.0 1,280 35.0 Peanuts - - 5,000 2.2 - - - Others 70 160 Subtotal 9,610 13,625 28,800 22,968 19,030 Non-Irrigated Cereal 8,200 2.65 2,623 1.8 - - 2,000 2.0 Sesame 800 0.4 1,015 0.8 - Cottc 1,r00 0.6 3,2;9 1.9 - Vegetables - - 253 17.5 - Citrus N/A 85 16.0 - Others 85 - Fallow 1,790 1,100 1,055 Subtotal 12,590 8,460 - 3,055 Total 22,200 22,0851- 28.800- 22,96&,- 22,085 Cropping Intensity 92% 95% 130% 104% 95% /1 Both DSI and the extension service estimates of rice growing area in 1971 are much lower - at around 60 ha - and appear to be more reliable than the appraisal estimate. /2 Mostly tomatoes. T3 DSI estimate with which the coapletion mission agrees. Appraisal mission overestimated net project area. 4 Includes 6,600 ha double cropping (sesame and peanuts). 5 Includes 883 ha double cropping (vegetables). - 30-- Judged against the preproject situation, only minor progress was made in achieving increases in yields. Between 1973 and 1977, yields for irrigated cotton exceeded 3 tons/ha, but a white fly epidemic in 1978 caused a dra- matic drop to 2.3 tons/ha and in 1979, production was'just beginning to recover. Rice yields between 1972 and 1979 averaged 4 tons/ha, with 4.5 tons/ha being obtained in 1973 and 1976. Vegetable yields remained rela- tively constant at 25-26 tons/ha until 1978 when yields rose to 30 tons/ha. Cereal has been grown predomi-antly on nonirrigated land where yields have averaged a low 1.8 tons/ha. Only sesame and citrus production have shown sizeable gains in yields. Until 1978, around 90% of all sesame was grown on nonirrigated lands. The situation changed in 1978 when 832 ha (or 39%) of the total area under sesame was irrigated and in 1979, 1,015 ha (55%) of sesame production was irrigated. -Yields for irrigated sesame have risen from 0.6 tons/ha in 1972 to 0.9 tons/ha in 1979. Citrus growing has shown a similar although much less dramatic increase in irrigated area. In 1979, 70% (or 186 ha) of the citrus area was irrigated compared to its 1972 level of 55% (or 112 ha). Yields for irrigated citrus have increased from 17 tons/ha in 1972 to 28 tons/ha in 1979. 4.06 Three factors can help explain the poor performance in achieving yield increases. The quality and coverage of the services provided by the extension service was always inadequate, the pace of implementation of DSI works was slow and on-farm development works lagged far behind schedule. By 1979, less than 40% of the targeted land leveling had been completed. Surface drainage works had come to a halt in 1975 after only 20% of target had been completed (see Table 2, Annex I). The negative impact on yield increases of the poor extension service and slow pace of on-farm and DSI development works would have been even more serious had it not been for the following mitigating influences. Farmers in the area were also provided with extension services by TOPRAKSU and plant protection facilities were supplied by the Ministry of Agriculture's Plant Protection and Quarantine Service and also by the Antalya Union, the large cotton cooperative in the area. 4.07 The changes in cropping pattern forecast by the appraisal mission and DSI are not likely to be achieved in the opinion of the completion mission. The most important difference between the forecasts, which are shown in the table on page 25, concerns cotton areas. Both the appraisal mission and DSI expect that cotton areas at full development-will have in- creased by around 15% over preproject levels but the completion mission estimates that cotton areas will have then declined to around 90% of pre- project levels. Between 1972 and 1978, the area planted to cotton ranged between 12,604 ha and 14,434 ha. In 1979 plantings were reduced some'38% to 8,136 ha. Farmers found it less profitable to grow cotton because of reduced cotton prices in later years and sharply increased labor and input costs. This trend may continue unless cotton growing becomes less labor intensive through mechanical picking, if this can substitute economically for manual picking, and/or cotton prices rise. The impact oi rising labor costs on cotton production will be particularly marked in those large areas where cotton is grown on unleveled land. Man-day requirements for cultiva- tion on unleveled land are considerably higher than those for leveled -31- land. Another difference in cropping pattern forecasts is that peanuts are included in the appraisal's forecasts but not in DSI's forecast cropping pattern nor in the completion mission's forecast. The reasons for the inclusion.of 5,000 ha of peanut growing in the appraisal forecast are not known although land and climate conditions are suitable for this crop. However, given the absence of both an effective extension service and marketing arrangements for peanuts in the area, it is unlikely that peanuts will be grown by farmers in the near future. Hence, both DSI's and the completion mission's forecasts for cropping pattern at full-development exclude this crop. 4.08 Notable features of the actual cropping pattern in 1979 are the very large increases in sesame area (from around 800 ha in 1971 to 2,941 ha in 1979), in rice growing area (from only 60 ha in 1972, if the DSI/Exten- sion Service data is accepted, to 257 fa in 1979), in melon area (70 ha in 1971 to 337 ha in 1979) and in vegetable area (70 ha in 1971 to 1,007 ha in 1979). Cereal growing has almost halved over this period (from 8,200 ha in 1971 to 4,442 ha in 1979) as the movement toward higher value cash crops has accelerated. The increase in sesame production has already exceeded appraisal expectations. At full development, areas under melon (1,050 ha) and citrus (720 ha) will also be considerably larger than those anticipated at appraisal forecasts. 4.09 At appraisal, it was anticipated that cropping intensity would rise to 130%. The Koprucay area undoubtedly has the agricultural potential to reach this level. However, under the present conditions with an inef- fective extension service and absence of reliable marketing arrangements for fruits and vegetables, the completion mission believes that cropping intensity will be less than 100%. The extension service has always been understaffed with insufficient transport, farmers have been faced with per- sistent problems with cotton in the last few years, vegetable growing is risky because there is no organized marketing system so that prices fluc- tuate enormously from year to year, and on-farm development works remain, uncompleted. 4.10 In 1971, the appraisal mission estimated the equipped area at 9,400 ha. However, from .information supplied to the completion mission by DSI and the extension service, it seems that the appraisal estimate did not include a rather large area (about 2,000 ha) which was irrigated by farmers using their own pumps, etc. By 1979, 20,000 ha (net) had been equipped with irrigation and drainage facilities. Works are outstanding on about 2,200 ha of land, comprised of the Stage II pumping area which is not yet complete and some low-lying land which needs drainage by pumping. These areas should be complete by the end of 1981 bringing total equipped area to 22,200 ha. Area irrigated by farmers reached a peak of 15,732 ha in 1977 but declined in the following years as the declining profitability of cotton led farmers to put land under wheat, without irrigation. -32- Silifke 4.11 In Silifke, all of the net area was under irrigation prior to the project,- although the drainage system was inadequate. Improvements to the irrigation and drainage works carried out under the project were expected to permit a more intensive agriculture. Incremental benefits would result from a reduction in the area devoted to winter cereals and a shift to tomatoes and other vegetables. 4.12 At appraisal, the net irrigated area was estimated to be 5,000 ha. However, according to DSI maps and extension service estimates, original project area was around 6,000 ha, and the completion mission has accepted this estimate as the more accurate one, The following table com- pares estimates of cropping pattern and yields. Further details of cropped areas and yields for the period 1973-79 are given in Annex II, Table 2. Silifke: Cropping Pattern and Yields With Project With Project Full Development Full Development Without Project Actual Appraisal Estimate Completion Estimate (1972)L (1979) (1979) (1982) Crop Area Yield Area field Area Yield Area Yield Ha Tons/Ha Ha Tons/Ha Ha Tons/Ra Ra Tong/Ba Peanuts/2 - 1,440 1.8 1,840 1.85 1,500 2.4 1,506 2.2 Sesame 640 0.4 480 0.7 200 0.5 538 1.0 Rice /2 640 3.0 1,040 3.5 1,200 4.0 1,261 5.0 Vegetables- (mostly - tomatoes) - 800 30.0 800 35.0 1,800 40.0 1,316 40.0 Citrus 280 30.0 510 41.0 200 32.0 439 41.0 Other fruit 109 23.0 Cotton 300 3.0 274 3.5 Melon 200 10.0 q. Seedlings 240 . 335 Cereal (dry) 2,800 2.0 2,880 4.0 1,000 3.6 2,468 4.0 Beans 240 8.0 280 5.0 - - Other 300 404 /3 Total 7,280 8,165- 6,500 8,315 Net Project Area 6,005 6,400 5,000 6,400 Cropping Intensity 121% 128% 130% 130% /1 Given the inaccuracy of the appraisal mission's project area estimate, the fact that DSI and TOPRAKSU works did not really get underway until 1973, and that,amongst the available data sources, only the extension service provided a complete and reasonably accurate picture of agricultural development in Silifke over the years 1972-1979, the without project situation used is the extension service's estimate'of cropping pattern and yields in 1972. /2 Includes'second cropping. 7i Includes completed (1977) Stage II pumping area of 195 ha and rice areas (about 200 ha) lying outside the lower project boundary but benefiting from the irrigation water supply, so that net project area is 6,400 ha. 33- Yields of all major crops have equalled or exceeded appraisal expecta- tions. Cereal and citrus yields have probably peaked but vegetable, peanut, sesame and rice yields will increase further before full develop- ment is reached in 1981. 4.13 Significant factors in achieving these yield increases were the variety of services provided to farmers in the area as a result of the close coordination between extension service and agricultural organizations in the area. Peanut growers received credit from Yerfiskobirlik, the local peanut cooperative. Vegetable seeds and citrus plants developed by plant research institutes in the area were distributed to farmers. The Soil and Water Research Institute at Tarsus provided information on improved irriga- tion methods to the extension service which then'transmitted this informa- tion to farmers. In addition, TOPRAKStJ extension staff were working in part of the project area. Hence, although the close working relationship between the extension service at Silifke and that at Seyhan did not materialize as anticipated at appraisal, this did not appear to have unduly damaged the quality of extension services provided to the farmers. 4.14 At full developm.ent in 1982, the cropping pattern projected at appraisal is generally in agreement with that projected by the completion mission. However, cereal production will not decline to the extent expected as high yields have enabled *farmers to maintain reasonable returns from this crop. Peanut production has leveled off although it will remain a major summer crop partly because there exist well-organized marketing arran,gements (through the peanut cooperative). Considerable diversification into fruit and vegetable growing will continue in the future. In recent years, strawberries and peaches have been grown by the more innovative farmers and these have proved to be highly profitable crops. The Silifke area is fortunate in being able to produce very early fruit and vegetables and hence can take advantage of the high prices at start-of-season. Marketing arrangements for these crops are also good. A local fruit and vegetable cooperative is active and any surplus production is bought by merchants from nearby Mersin, the regional center for fruit and vegetable marketing. 4.15 Double cropping was already practiced in 1972 and this has gradually increased over the years so that by 1982 cropping intensity is expected to reach 130%, the level anticipated at appraisal. Total equipped area rose by 195 ha over its 1972 level to reach 6,400 ha in 1978 and an additional 200 ha of rice growing area lying outside the lower project boundary benefits from the irrigation water supply. Actual irrigated'area, however, has been less than equipped area because cereal is not irrigated except in years of very low rainfall. .Tokat 4.16 In the Tokat area, the appraisal mission assumed that incremental benefits would occur primarily through an increase in vegetable production and the introduction of alfalfa in the cropping pattern. The vegetable - 34 - growing.area would increase from its appraisal level of 1,000 ha .to 2,600 ha. Alfalfa would be grown on 1,800 ha which would allow improvement of the considerable livestock population in the area. In addition, there would be a slight increase in fruit tree area. The growing area under sugar beet, already a major crop, would not change, although production was expected to increase. 4.17- The table below compares appraisal and completion estimates of cropping pattern and yields, and details of cropping areas and yields for the years 1973 to 1979 are given in Annex III, Table 3. Tokat: Cropping Pattern and Yields With Project With Project Full Development Full Development Without Project Actual Appraisal Estimate DSI Estimate (1971) (1979) (1981) (1985) Crop Area Yield Area Yield Area Yield Area Yield Ha Tons/Ha Ha Tons/Ha Ha Tons/Ha- Ha Tons/Ha Irrigated Cereal 6,080 2.3 3,805 3.2 13,500 3.0 4,697 3.5 Sugarbeet 2,910 40.0 4,117 60.0 3,210 50.0 4,270 60.0 Watermelon 490 20.0 124 37.0.- Orchard 1,150 8:0 1,278 22.0 1,560 12.0' 3,415 25.0 Alfalfa 298 8.0 1,840 12.0 1,068 10.0 Vegetables 1,891 31.0 2,620 N/A 4,697 32.0 Beans /1 1,585 3.5 1,068 3.5 Other 1,360- . 1,285 170 2,135 Subtotal 11,990 14,383 22,900 21,350 Nouirrigated Cereal 6,900 1.5 3,694 1.8 Sugarbeet - 340 21.0 697 30.0 Vineyard 450 4.0 418/3 4.3 Other 270 778- Fallow 2,950 1,380 Subtotal 10,910 - 6,967 ~ 22.900 21,35LM. 22jM0135 Cropping Intensity 87% 94% 100% 100% /1 Includes vegetables. The appraisal mission considerably underestimated the yield from fruit trees. For 1972, both DSI and the Extension Service estimate this yield to be 19 tons/ha. 3 Including 139 ha of alfalfa. 4 DSI estimate with which completion mission concurs. - 35 - As the table shows, all expected yield increases, with the exception of alfalfa production, were either realized or surpassed by 1979. Yields are expected to increase further before full development is reached in 1985. Particularly significant are the yield increases obtained for sugar beet (from a preproject level of 40 tons/ha to 60 tons/ha in 1979) and vegetable (from 18 tons/ha in 1972 to 31 tons/ha in 1979). 4.18 The extension service can claim only partial responsibility for achieving these yield increases. Other factors appear to have been more important. Farmers in the area received extension services from TOPRAKSU whose staff are generally more effective than the extension service person- nel (para. 6.10). The local sugar beet factory continued to provide sugar beet growers with on-the-job training, advice on planting, cultivation and harvesting, and assistance with seed and fertilizer requirements. Other agricultural institutions in the area, such as the TOPRAKSU Research Station, have been carrying out research on improved agricultural practices and livestock husbandry. These institutions run courses for local farmers and carry out on-farm demonstrations. 4.19 The changes in cropping pattern anticipated at appraisal were not generally fulfilled. Cropped area under cereal fell from 12,980 ha (65% of total cropped area) in 1971 to only 7,499 ha (38% of total cropped area) in 1979. At full development, cereal growing is expected to decline even fur- ther to 4,697 ha (22% of cultivated area). On the other hand, sugar beet area in 1979 considerably exceeded appraisal expectations and vegetable and orchard areas at full development will be double those anticipated. Area under alfalfa grew much slower than expected. In 1979, only 437 ha of alfalfa was grown and DSI's forecast for full development of 1,068 under alfalfa cultivation may not be realized. 4.20 The switch away from cereal growing towards sugar beet, fruit and vegetables can be explained by the following factors. Cereal has become an only marginally profitable crop as prices have remained low in real terms and costs of inputs such as fuel, fertilizer and machinery have increased. However, there is a lower limit below which cereal cultivation will not fall which is dictated by the needs of home consumption. For sugar beet, although prices remained relatively constant in real terms, high yields have enabled farmers to achieve significant increases in the net value of production. In addition, sugar beet growing is attractive to farmers . because of the extension services they receive from the sugar factory, the absence of marketing problems (all beet is bought directly by the factory) and the added side benefit the farmers receive in the form of a free entitlement to beet residues which are used as a supplementary animal feed. Vegetable and fruit production h.ve been stimulated by high yields and, although prices fluctuate (there are no official support prices for these crops), net returns to farmers have been good. However, the achieve- ment of the increased volume of fruit and vegetable p%oduction anticipated at full development will depend upon the development of a much more efficient marketing system than presently exists. Alfalfa production - 36 - has failed to reach appraisal estimates. Yield increases have not been as large as expected and there is not a strong demand for alfalfa in the area, the farmers preferring to feed their animals on cheaper alternatives, e.g., beet residue. 4.21 No double cropping was foreseen in the Tokat area and, in fact, little double cropping would be possible due to the late harvest of winter wheat. The cropping intensity in 1979 was 94% compared to the preproject appraisal estimate of 87% and will reach 100% by 1985. - 4.22 By developing irrigation and drainage facilities, the project increased net irrigable area by 78% from 12,000 ha in 1971 to 21,350 ha in 1979. However, total area irrigated has always been less than that equipped because farmers were obliged to permit land leveling work to be carried out in the dry summer months and because the farmers do not irrigate cereal growing areas in years when there is sufficient rainfall. In 1977, when most of the project area was equipped, less than half the project area was irrigated as rainfall was adequate for most of the cereal crop. Overall Assessment and Prospects for the Future 4.23 Comparisons between actual performance and appraisal expectations in the three project areas show that the project was quite successful in increasing yields and cropping intensities in Silifke and Tokat; but in Koprucay, progress made in achieving yield increases was less satisfactory and little,--if any, headway was made in increasing cropping intensity. On the whole, the appraisal mission's projections both for yield increases and cropping pattern changes were overly conservative. -For many crops, yield increases projected to be attained only by full development were achieved by 1979. The actual movements into higher value crops as reflected in cropping pattern changes were also more marked than appraisal expecta- tions. A large part of these changes can be explained by changes in relative prices (hence changes in relative net production values) which could not have been predicted at the time of appraisal. 4.24 Given that high value crops are being grown on larger areas and are returning higher yields than was anticipated, farmers may face market- ing difficulties in the future, particularly in the Koprucay and Tokat . areas. At Koprucay, lack of farmer support has meant that the only exist- ing fruit and vegetable cooperative is not effective. Most of the fruits and vegetables produced are bought by private traders. There are no offi- cial support prices. Hence, although the potential rewards from fruit and vegetable growing are very good, they are also risky as prices fluctuate widely from year to year depending on the quality and quantity of the harvest. In 1978, for example, a bumper crop of tomatoes forced the price down at one stage to a ridiculously low 0.20 TL/kg and many farmers left the crop to rot in their fields. There is a tomato paste factory in the area, but this did not help some farmers because they were not growing the tomato variety desired by the factory. Unless marketing arrangements - 37 - improve in future years, the areas projected to be grown under vegetables at full development are not likely to be attained. At Tokat, most of the fruit and vegetables grown are sold .in local markets either for domestic consumption in the project area, in the Samsun region or for processing. Factories 'in the area include a tomato paste factory which has an annual capacity of 15,000 tons, a sugar factory, and a winery and juice bottling factory. Good markets for fruit and vegetables from Tokat can also be found in the Eastern Black Sea region, but inadequate transportation arrangements and periodic fuel shortages handicap the ability of Tokat farmers to enter these markets. Silifke is fortunate in already having well-established marketing arrangements and in its proximity to the whole- sale fruit and vegetable center at Mersin. 4.25 The success of the extension service in achieving its objectives of improving agronomic practices and expanding areas under higher value crops is difficult to evaluate. Certainly, the early vision of extension services modelled along the lines of the Seyhan extension service (para 6.08) did not materialize. Consultants were not hired for Koprucay and Tokat, nor did the Silifke extension service take advantage of the consul- tants serving the Seyhan Project. Hiring the consultants may well have changed conditions and accelerated agricultural development in at least one of the project areas, namely Koprucay (paras 6.10 and 6.11). 4.26 Consultants would also have helped coordinate the various forms of assistance received by farmers from other agencies in the three areas. In Koprucay, the Plant Protection and Quarantine Service is particularly important for cotton pest control. In Silifke, assistance was received from the Tarsus Research Station on plant varieties and cultivation prac- tices. In the Tokat area, the Turhal sugar factory provides inputs on credit and extension services to the many beet farmers and the TOPRAKSU Research Station provides extension services for deciduous fruits. In all three areas, TOPRAKSU had its own extension personnel to provide advice in irrigated areas on the use of water and fertilizer. However, TOPRAKSU personnel were more effective in Tokat and Silifke than in Koprucay. It is likely that the various agricultural institutions, plus TOPRAKSU staff, partially compensated for the lack of extension services provided by the Ministry of Agriculture. Nevertheless, so long as the extension service does not receive adequate support from the Government, its effectiveness will continue to be hindered by insufficient vehicles and a lack of trained personnel. The transportation problem is particularly serious since extension services in the three areas do not-have the advantage of agents living in the villages and keeping in close contact with day-to-day affairs as was the case in the Seyhan extension system. Efficiency of the Irrigation System 4.27 Judging from interviews with farmers and field staff of government agencies, the irrigation supply kept up with farmers'-demand. Irrigated crops occupied up to 80% of the equipped area in Silifke, 70% in Koprucay and 60% in Tokat. Complaints about supply limitations were few and - 38 - localized. It was not generally felt that further increases in irrigated crop areas would endanger yields due to underirrigation or would unduly burden irrigation systems, services and farmers. 4.28 The following table gives details of water supply and demand during 1978 and 1979 in the three project areas; Water Supply and Demand in McM Irrig. Seasonal Area Year area Month Total Ha Apr. May June July Aug. Sept. Oct. Silifke 78 4,464 5.9 6.1 10.1 12.6 8.2 5.8 4.9 53.6 79 4,801 3.8 6.3 9.6 13.7 13.7 3.9 2.9 54.0 Koprucay 78 15,000 - 14.5 29.6 52.5 52.5 43.9 13.4 206.4 79 15,200 13.9 22.6 28.0 44.4 44.7 38.2 16.0 207.8 Tokat 78 13,800 - 19.0 35.7 38.0 33.2 22.8 9.7 158.4 79 14,400 - 28.9 35.3 32.5 23.1 16.4 4.6 140.8 Seasonal water requirements exceeded the values estimated at appraisal by 20% on average. Estimated and actual quantities on a per hectare basis were 11,600 m3 and 12,900 m3 for Silifke, 9,300 m3 and 13,800 m3 for Koprucay and 8,600 m3 and 11,700 m3 for the Tokat area (average of - 1978 and 1979). The main cause of actual supply to exceed estimates was waste of irrigation water during the nighttime. Farmers do not irrigate during the night if it is not strictly necessary. The irrigation service tries to oblige by providing the farmers enough time to irrigate during the day. During the night, DSI evacuates excess water in the main canals via escapes; or the farmers, by not taking up their allotment, let it spill over the end of the tertiaries. 4.29 In months of maximum demand (July and August), waste is' reduced as farmers do irrigate more during the night because of capacity limita- tions of the systems, which are designed for.operating on a 24-hour basis. During July '78 and '79, irrigation efficiency in all three projects was about equal to or exceeded values estimated at appraisal. In the Silifke project, which had the highest irrigated crop intensity (80%), the two year's average for actual irrigation efficiency was 76% as compared to an estimated 54%. For Koprucay, the two year's average was 54% (51% appraisal estimate) and for Tokat (where irrigated crop intensity was lowest), 44% (40% appraisal estimate). 4.30 River supply is more than adequate for meeting present and future increased needs of the three projects. The capacity of the - 39 - irrigation systems seems to be tight but sufficient. At diversion point, the Silifke scheme has a large overcapacity because of planned extensions in the future, the Koprucay scheme has a capacity of-about 1.2 and Tokat 0.8 1/s/ha. The latter's capacity is 15% below the requirements according to the appraisal report, which assumes a quantity of 0.95 1/s/ha. This undercapacity could cause some constraints in future years if farmers in the Tokat area decide to bring the irrigated crop intensity nearer to 100% of the equipped area. Operation and Maintenance 4.31 Operation of the irrigation systems is based on controlled demand. On the farmers' request, the irrigation-service allots water according to demand, if possible, and bundles supply in off-peak months so as to keep tertiaries either fully supplied or turned off. When supply is bountiful, the service takes into account the farmers unwillingness to irrigate during the night by closing the head gates of the tertiaries and letting excess water drain via escapes or by stopping pumping. Farmers with land supplied by one tertiary distribute the water by rotating the full head on a part-time basis; water is taken out of the tertiaries with plastic syphons. The irrigation season starts in March/April and closes in September or October depending on rainfall and farmers' needs. 4.32 Irrigation canals are cleared and maintained on a yearly basis by mechanical and chemical means. In general, maintenance of these canals is adequate. Drainage canals are mechanically desilted and cleared once every 4-5 years and weed growth is controlled chemically once every 2 years. However, this system for drainage canal maintenance does not pro- vide sufficient control of certain weeds as can be observed in all three project areas. Lack of herbicides due to the import problems faced by Turkey in recent years has also contributed to the problem. The equipment DSI uses for desilting is too heavy and expensive for use in weed control. DSI did try lighter equipment (such as tractors with weed cutting imple- ments) in the Tokat area but was not satisfied with it. 4.33 In the Koprucay area, responsibility for operation and mainte- nance of secondary and tertiary subsystems has been given by DSI to groups of users or irrigation unions. For performing these services) DSI reduces their water charges by 15%-20%. Water users in 24 of the project area's 34 villages have been so organized. The unions appear to carry out operation and maintenance tasks satisfactorily. In the Tokat area, DSI will start the same system this year. V. ECONOMIC ANALYSIS Production 5.01 The success of the project in increasing production, when measured against appraisal estimates, varies according to crop grown and 40 - project area. For crops, notable are the changes in cereal and vegetable production and the poor performance in introducing fodder cropping. When comparing production increases by project area, Tokat is clearly the most successful, followed by Silifke and then Koprucay. The table below com- pares appraisal and completion estimates of incremental production at full development. Completion estimates are based on revised yields and cropping patterns and these are discussed in greater detail in Agricultural Impact (paras 4.05, 4.12 and 4.17). Incremental Production at Full Development for Major Crops - Annual KOPRUCAY SILIFRE TOKAT Appraisal Completion Appraisal Completion Appraisal Completion Crop Estimate Estimate CroD Estimate Estimate Crop Estimate Estimate ('000 tons) ----- ('000 tons) --- -- ('000 tons) -- Cereal 6.1 -2.9 Cereal 0.0 4.3 Cereal 17.0 -7.9 Cotton 16.4 6.9 Rice 1.2 4.4 Sugarbeet 37.0 132.7 Rice 2.3 3.1 Sesame 0.0 0.3 Orchard 9.5 63.5 Sesame 6.4 2.1 Citrus 1.8 9.6 Alfalfa Vegetables 63.8 67.2 Peanuts 2.0 0.7 (new crop) 22.1 10.7 Fodder not Vegetables N/A 107.6 (new Crop) 3.6 -* grown Vegetables '18.0 28.6 Prices and Production Costs 5.02 -Increases in farmgate prices and production costs in real terms have greatly exceeded appraisal estimates. Production costs in particular have increased sharply over the period 1971-79. Projections for farmgate prices and production costs are compared to appraisal estimates in the following table. Where a crop is grown in more than one area, the value shown is an average. Farmgate Prices and Production Costs at Full Development Farmgate Price Production Cost ---------Estimated in 1971 Constant Prices--------- Crops Appraisal Completion Appraisal Completion ------- (TL/ton)-------- --------(TL/ha)-------- Cereal 900 1,500 2,910 3,200 Seed Cotton 2,700 4,400 2,385 9,000 Sugar beet 200 220 3,400 6,000 Orchards 1,030 1,300 6,500 26,000 Sesame 3,100 9,000 410 4,800 Peanuts 3,250 4,200 1,090 7,000 Vegetables N/A 1,010 6,280 16,000 - 41 - Financial Returns 5.03 The following table compares appraisal and completion estimates of income per hectare in the three project areas, net-of-labor costs but before deducting taxes or water charges. Nominal income gains experienced by project beneficiaries greatly exceed appraisal estimates. 'Tokat has experienced the greatest increase. In 1979, income per hectare at Tokat was TL 21,240 (TL 3,186 in 1971 constant prices) compared to its 1971 pre- project level of TL 1,890. Comparable figures for Silifke are TL 39,680 (TL 5,953 in constant prices) and TL 3,950; and for Koprucay, TL 15,110 (TL 2,267 in constant prices) and TL 3,512. High rates of inflation in Turkey in recent years have meant that none of the three project areas achieved the doubling in real income per hectare that was expected at appraisal. However, despite the negative impact on real incomes that high rates of inflation have had, real income gains in Silifke and Tokat well exceed the 15% without project increase anticipated at appraisal. Real income per hectare rose by 70% over its preproject levels in Tokat, and by 50% in Silifke. In Koprucay, real income per hectare declined by 35% over its preproject levels, in part a reflection of low cotton yields and prices in latter years and the relatively slower pace of project implementation at Koprucay. Income Per Hectare: Comparison of Appraisal and Completion Estimates Appraisal Estimates Completion Estimates With Without Project Ratio Without With Ratio Project Project (Full With/Without Project Project With/Without Area (1971) Development) Project Income (1971) (1979) Project Income Koprucay 3,512 7,925 2.3 6,920 15,100 2.2 Silifke /1 3,950 7,643 1.9 7,300 39,680 5.4 Tokat 1,890 3,935 2.1 2,830 21,240 7.5 /1 Includes double cropping. Economic Rates of Return 5.04 On the basis of actual data for the years 1972 to 1979, and revised estimates for 1980 onwards, economic rates of return for the three project areas are estimated at Koprucay 23%, Silifke 33% and Tokat 42%. These rates reflect differeaces in relative production increases and proj- ect investment costs between the three areas; the Tokat project experienced the greatest gains in production and least cost overruns in relative terms; conversely, Koprucay had the smallest production gains and largest cost overruns (paras 3.24 and 5.01 and Tables 4, 5 and 6 in Annex IV. The accuracy of the revised rates of return depends not only on relative prices remaining stable but also on achieving further gains in crop diversifica- tion, full utilization of irrigation and partial recovery of cotton yields. - 42 - 5.05 The appraisal report estimated economic rates of return to be Koprucay 19%, Silifke 18%, and Tokat 12%. The Turkish Government's esti- mates are 29%, 35% and 20%, respectively. Appraisal, completion and Government estimates of economic rates of return thus differ considerably although the completion estimates are somewhat closer to those of the Government than to appraisal estimates. All three assessments are based on significantly different underlying assumptions regarding production, investments, phasing and prices of inputs and outputs. 5.06 Any similarities that do exist between the completion and the Government's estimates are purely coincidental. While completion estimates are based on actual data for the years 1972 to 1979 in each area, the Government used more than.two years' actual data for Silifke only and then just to 1977. For Tokat and Koprucay, only two years actual observations were used, the intermediate years beifig interpolated. The Government also includes, as part of project investment costs, the costs of past irrigation works in the three areas. In 1971 prices, these costs amounted to TL 226 million for Koprucay, TL 66 million for Silifke and a very high TL 333 million in the case of Tokat. An opportunity cost of capital (OCC) of 9.5% (completion estimates are based on an OCC of 12%) is another instance where Government's and completion estimates differ. 5.07 Actual project performance has been quite the inverse of that anticipated at appraisal for a variety of reasons described in this report and is reflected in the completion report's ranking of Koprucay 23%, Silifke 33% and Tokat 42% as compared to appraisal values of 19%, 18%, and 12%. VI. INSTITUTIONAL IMPACT Institutional Strengthening for Project Implementation 6.01 The project gave particular emphasis to the strengthening of TOPRAKSU and the Extension Service. Both agencies were to receive help from consultants. TOPRAKSU would 'employ two consultants to assist it in organizing the on-farm works and in ensuring the operation of heavy equip- ment to its potential efficiency. The Extension Service also was in need of assistance in organizing and coordinating extension services in the field. Two consultants would be employed by the Extension Service, one to be stationed at Koprucay and the other at Tokat. The Silifke Extension Service would be assisted by the consultants then serving the Seyhan-Proj- ect. DSI assured the Bank during negotiations that it had gained suffi- cient experience to undertake the project works without outside assis- tance. Hence, DSI would not receive outside assistance unless both the Bank and the Government thought it necessary. The project also provided for training abroad for the staff of all three agencies. 6.02 The failure of both TOPRAKSU and the Extension Service to hire consultants has been discussed earlier (para 3.16). The impact of this - 43 - failure on project implementation is necessarily speculative. However, the opinion of the successive supervision missions and that of the completion mission is that consultants for TOPRAKSU could have helped considerably to speed up the pace of land leveling and drainage works and improve the effi- ciency of equipment usage, and that consultants for the Extension Service would have significantly improved the efficiency of extension operations, particularly at Koprucay (para 6.11). The consultants might also have assisted in establishing better cooperation between the agencies involved in implementation, especially between TOPRAKSU and the Extension Service, although effective coordination may not be possible between two such agen- cies with overlapping responsibilities in extension and research work. The training overseas of DSI and TOPRAKSU staff occurred in 1976-77 which was too late to have any significant inflijence on project construction. Exten- sion Service personnel received tiaining at the same time. How useful this training was in aiding project activities is not known as no evaluation was made to determine project staff reaction. 6.03 The extent to which the project contributed to institutional development has been limited, particularly by the failure to hire consul- tants. However, DSI's capabilities were strengthened by its successful experience in undertaking works such as those of this project without out- side assistance. TOPRAKSU is in a position to apply the lessons learned as a result of its experience with this-project to future on-farm development works. But the project resulted in little institutional development of the Extension Service because consultants were not hired nor did the Government provide the necessary support. So long as Government support is not forth- coming, the Extension Service in the project areas will continue to be weak and relatively ineffective. Coordination Between Agencies 6.04 The Bank's efforts to coordinate the activities of the major agen- cies involved in the project was at best only partially successful. The Credit Agreement required that a committee consisting of the top local official from each of DSI, TOPRAKSU, the Extension Service and TCZB would be formed in each project area within three months of the date of effec- tiveness of the Credit. The committees were to meet at least every three months in order to discuss past accomplishments, plan futureiimplementation programs and coordinate respective activities. Summaries of the meetings' minutes were to be forwarded to the Bank. 6.05 The committees were set up in July 1972, and TOPRAKSU was given the responsibility for calling meetings and forwarding minutes to the Bank. Supervision missions reported conflicting evidence of these meetings. Apparently, the committees did not become operational until 1974. It quickly became apparent that the committees were set up merely to meet the requirements of the Credit Agreement. None of the agencies in- volved seemed convinced of the necessity and efficacy of the committees. The result was that meetings were held infrequently, attendance was dele- gated to junior staff, no topics of substance were discussed and minutes of meetings were never forwarded to the Bank . From 1975 onward, the situation deteriorated despite repeated reminders by the Bank of the Government's. -44- obligations under the Credit Agreement and of the benefits that could be derived by the project through effective functioning of these committees. Coordinating committees in two of the project areas ceased to operate and committee.meetings in the third project area were held only sporadically. The committees were finally disbanded. Outside of the committees, DSI and TOPRAKSU coordinated their activities in Tokat and Silifke; but little coordination between these agencies was evident in Koprucay. 6.06 Successful implementation of the project in three geographically and agriculturally distinct areas would (and does) require effective co- ordination of all four agencies' actions. That this never eventuated obviously had detrimental effects on project implementation. One require- ment for the effective functioning of-coordinating committees is that the committee chairman and members be the highest possible level officials with, preferably, the chairman being appointed from outside the implement- ing agencies. In the Corum-Cankiri Rural Development Project (Ln 1130-TU), for example, project committees are chaired by the provincial Governor. For the lead agency in the coordinating committees of this project, DSI rather than TOPRAKSU, should have been the obvious choice. The consultants to TOPRAKSU and the Extension Service, had they been hired, would have helped considerably to improve inter-agency coordination. Extension Service 6.07 As mentioned earlier, existing extension services in the three project areas were to be strengthened under the project. An extension coordinator would be appointed in each area to select and train village agents. Two consultant extension specialists were to assist in organizing and coordinating extension services at Koprucay and Tokat. The Silifke project was to be assisted by the consultants then serving the Seyban Proj- ect. However, other than for the appointment of coordinators and employ- ment of consultants, the way in which extension activities would be strengthened was not specified in the appraisal report, as it was agreed during appraisal that a system similar to that of the Seyhan Project would be set up by the extension service in all three areas. 6.08 Subsequently, in May 1972, the then Director General of Extension proposed to a Bank supervision mission that the Extension Service in each project area be organized along such lines. Each area would have a full- time team comprising a field crop specialist, a crop protection specialist and an irrigation specialist, all assisted by agricultural technicians and foremen. Although the Bank endorsed this proposal and urged the Government to complete this organization as soon as possible, it did not eventuate. 6,09 Extension coordinators were appointed, but they were usually the extension directors for the region with other responsibilities. Therefore, they could'not devote their full time to training and supervising extension personnel. Despite repeated requests by the Bank for the Government to provide the necessary support to the Extension Service, transportation and equipment were always inadequate and staffing was never-complete. Con- sultants were never employed (para 3.24) nor did the Silifke Extension 45 - Service on any occasion take advantage of the consultants on the Seyhan Project while they were still in post as had been agreed at appraisal. 6.10 It seems likely that hiring the consultants would have changed conditions and accelerated agricultural development in the three areas, particularly at Koprucay, where agricultural productivity has been less than anticipated. In 1975 the extension services hired the maximum number of extension agents, called technicians, in the project areas. They were backed by 11 supervisory staff. In all, 22 technicians assisted 11,540 farm families in the project areas. However, the ratio of technicians in the villages to farms varied considerably: Number . Number of of Area Technicians Farms Ratio Tokat 15 6,055 403 Silifke 4 1,655 413 Koprucay 3 3,830 1,277 Clearly, the extension service in Koprucay needed to service about three times as many farmers as the other areas and were only supported by super- visory extension personnel in Antalya on a part-time basis. The situation in Koprucay was further aggravated by the absence of a Regional Director of Agriculture in Antalya during 1973-75. One can argue that with the consul- tants greater attention would have been given, especially in the Koprucay area, to hiring more village extension agents and providing better training and supervision. One mitigating factor was that TOPRAKSU had its own extension personnel to provide advice in irrigated areas on the use of water and fertilizer. Their staff appeared to be better trained and received better guidance from their supervisors. TOPRAKSU also had 21 extension agents (called foremen) backed by four supervisors in the three project areas. The Extension Service and TOPRAKSU carried out their activities independently and usually in separate villages in the project areas. Unfortunately, coordination of efforts was at a minimum. Coordina- tion may well have been improved had consultants been employed. 6.11 Farmers also received assistance from other sources. In Koprucay, farmers benefitted from the Plant Protection and Quarantine Service, the Cotton Cooperative and several research stations in the area including the Mediterranean Regional Agricultural Research Institute, the Horticultural Research Station, the Cotton Research Institute and the Citrus Research Station. Silifke farmers received advice on varieties and cultivation practices from the Tarsus Research Station. In Tokat, farmers were served by the Orchard Nursery Station, Kazova Livestock Breeding Directorate, Veterinary Organization, Chamber of Agriculture, Turhal Sugar Factory Directorate and the TOPRAKSU Research Station and State Farm. The role that these institutions played in furthering agricultural development in the three areas is also discussed in para 4.26. The completion mission believes that these institutions, plus TOPRAKSU staff, partially com- pensated for the lack of extension agents provided by the Ministry of - 46 - Agriculture, especially in the Koprucay area. Nevertheless, the objective of developing a viable extension service for the Ministry of Agriculture along the lines initiated in the Seyhan Project was not attained in the three project areas. At least one of the project areas would have been helped by consultants in organizing and aiding farmers. Consultants would also have improved coordination between the various sources of extension services available to farmers. The Bank, however, was unwilling to take strong measures to insist on compliance with the Credit Agreement. Other Farmer Support Services 6.12 The Government had provided assurances to the Bank during negotia- tions that adequate credit facilities-would be made available to farmers in the project areas. During 1973, Bank supervision missions reported that there were shortages of short-term credit in the project areas. The Government was reminded of its obligations by the Bank and thereafter the situation apparently improved. There were also other important sources of credit for farmers in the project areas, such as the cotton cooperative at Koprucay, the peanut cooperative at Silifke, and the sugar beet factory at Tokat. Cost Recovery 6.13 At the time of negotiations, water charges imposed were insuffi- cient to meet project operation and maintenance costs. 1/ The Government was at that time considering a report on a study of appropriate levels of water charges for the Seyhan Project and it was expected that a future policy on such charges would evolve. The Government assured the Bank that in the three project areas water charges would be assessed and collected to cover operation and maintenance costs and recover a reasonable portion of the investment cost with due regard to the farmer's ability to pay. Such policy as would evolve from the Seyhan study was to be applied in setting the charges which would be subject to periodic -review between the Govern- ment and the Bank. 6.14 The study of water charges for the Seyhan Project was completed in 1971. The major recommendations, including raising water charges and increasing penalties for delinquency, were accepted by DSI who forwarded its comments on the report to the Ministry of Finance and the Bank. The Government, however, never officially adopted any of these proposals. After 1974, the Government implemented its own nationwide study under, the leadership of DSI. 6.15 For DSI irrigation schemes, Government policy is to recover O&M costs and capital costs depending on the farmer's ability to pay. Water charges are set each year by the Council of Ministers and vary according to 1/ For Silifke and Koprucay projects, water charges imposed in 1970 had covered 82% and 35% respectively of their operation and maintenance costs. -47- the particular irrigation scheme and crop grown. From 1972 until 1978, there were no significant changes in DSI water charges; charges levied fell increasingly short of levels required to recover even operation and main- tenance expenses let alone to begin to recover capital costs. The Bank repeatedly expressed its concern to the Government about this situation which constituted a default of the Credit Agreement. At the same time, cost recovery under the Seyhan Project was a major issue throughout imple- mentation of that project. Three main points were involved: water charges set for DSI's O&M were inadequate to cover actual costs, DSI's capital recovery charges were low, and legislation did not permit recovery of TOPRAKSU investment costs. Cost recovery under the Seyhan Project was also the subject of high-level 'policy discussions with the Government. It resulted in the Bank informing the Government in 1976 that it would not consider further irrigation projects until steps were taken to improve the cost recDvery system. 6.16 In 1978, the newly elected Government substantially increased water charges as part of its general economic program. In all three proj- ect areas, new charges were five times the previous levels. In Koprucay, the average charge for gravity irrigation in 1978 was TL 960 per ha com- pared with actual 1977 O&M costs of TL 740 per ha (or TL 980 per ha actually irrigated). For Silifke, comparable figures are TL 1,530 per ha compared to TL 810 per ha (or TL 1,040 per ha actually irrigated) and for Tokat, TL 670 1/ compared to TL 600 2/ per ha (or TL 1,240 2/ per ha actually irrigated). However, available data for water charges in 1979 suggest that this improvement in cost recovery has not been sustained. In Koprucay and Silifke, water charges for gravity irrigated areas in 1979 were less than their 1978 levels, while operation and maintenance costs had increased significantly. Recovery of project costs remains unsatisfactory, with O&M costs not being fully recovered, let alone the capital costs of DSI and TOPRAKSU works. 6.17 Cost recovery under the Irrigation Rehabilitation and Completion Project has been more successful than under the Seyhan Project but assess- ment, collection and recovery r-ates are still unacceptably low. Water charges have never been sufficient to cover O&M costs. There have been many years when charges have not even covered 50% of O&M costs (1975-77 in Koprucay, 1976-77 in Silifke and 1974-77 in Tokat). Collection rates have also been poor with the result that recovery rates have been lowered further. Recovery rates between 1973 and 1978 have only once exceeded .50%--for Koprucay in 1973--and are generally much lower. The average collection and recovery rates over the years 1973-1978 for all three proj- ect areas were 62% and 30% respectively. Table 1 in Annex V gives further details. 1/ Gravity irrigation only. 2/ Average gravity and pumped irrigation. -48 - 6.18 Although these rates are unacceptably low, the performance of project administration in collecting water charges in the project areas is significantly better than in the country as a whole. In 1978, the average collection rate in the three areas was 59% compared to the country-wide average of 39%. Poor collection rates can be.attributed largely to the Government's ineffective penalty system for late payment of water charges. The penalty system imposes a surcharge of 10% at the first repayment date but this does not induce prompt payment since no further penalties are imposed. In times of high rates of inflation, the result is that farmers effectively obtain cheap credit with late payment. DSI has the right to confiscate property in the event of persistent delinquency but it is doubt- ful whether this right has'ever been exercised. In an effort to institute an effective penalty system, DSI has proposed to the Ministry of Finance a system of increasing penalties or interest rates to be imposed periodically and, in cases of persistent delinquency, to discontinue delivery of water. To date, these recommendations have been rejected by the Government reflecting the largely political nature of the cost recovery question. 6.19 In addition to the charge intended to cover O&M costs of DSI, there is a further charge levied for recovery of capital costs of DSI irri- gation projects. DSI capital recovery charges are not collected until after completion of works. Charges are set on an area basis and are designed to recover the original project costs by equal installments over a period of 50 years at zero interest with a 10-year grace period. Under these terms, recovery of real economic capital costs are extremely low. No attempt has. been made to revise charges in the light of increased costs of projects or of high inflation rates. 6.20 -In the Koprucay project area, no capital cost recovery charges are levied. At Silifke, charges have been levied since 1972 at a rate of TL 110 per hectare; and at Tokat, a charge of TL 100 per hectare has been levied since 1973. It appears the charges are designed to recover costs of irrigation works that were completed prior to the project but the comple- tion mission was unable to clarify this matter. Silifke Capital Recovery Charges Charge Assessment Collection Collection Rate TL Per Ha '------(Million TL) ------- (%) 1972 110 0.34 0.18 55 1973 110 0.46 0.24 52 1974 110 0.52 0.26 50 1975 110 0.46 0.28 62 1976 110 0.43 0.28 65 -1977 110 0.48 0.35 72 1978 110 0.50 0.30 60 - 49 - Tokat Capital Recovery Charges Charge Assessment Collection Collection Rate TL Per Ha ------(Million TL)------- (%) 1973 100 0.95 0.32 34 1974 100 1.05 0.36 34 1975 100 1.02 0.40 39 1976 100 1.16 0.48 41 1977 100 0.87 0.37 43 1978 100 .1.14- 0.56 49 6.21 There are no legal provisions-for recovery of TOPRAKSU's on-farm development costs. In 1978, TOPRAKSU presented to the State Planning Office draft legislation to introduce cost recovery charges, recommending recovery of 100% of land leveling costs and 50% of drainage costs, over 20 years with interest. It was anticipated that action would have been taken on this by the end of 1978. However, queries from the Bank on the status of this or other measures taken towards cost recovery charges for TOPRAKSU works have yet to be answered by the Government and it appears the matter has been in abeyance. 6.22 Irrigation water is becoming an increasingly costly input for raising agricultural output. The development costs and operation and maintenance costs of irrigation projects tend to be high compared with many other forms of agricultural investment and other agricultural inputs. Yet the high levels of subsidy enjoyed by project beneficiaries as a result of these low levels of cost recovery are difficult to justify, especially when their average incomes are well above the national average. The Bank's decision not to consider financing further projects in the irrigation sector in Turkey will remain until the Government takes appropriate steps to significantly improve the cost recovery situation. Audit 6.23 Audit reports of project accounts which were to be furnished to the Bank under the terms of the Credit Agreement were never received and the Bank seldom followed up on this omission. The problem is not peculiar to this project but is experienced in most other agricultural projects in Turkey. Noncompliance with auditing covenants was a matter which was brought up during the Turkey-Bank discussions of problems of project - implementation which were held in Washington in mid-1978. The Turkish delegation promised to recommend to the Minister of Public Enterprises a solution to late and inadequate audits on Bank-financed projects. 1/ To date, there is no evidence that the Turkish delegation's recommendation has been adopted. 1/ Draft notes of understandings reached during Turkey-Bank discussions June 26-30, 1978. - 50 - 6.24 To be of value, audit reports must be concise but adequate. The Bank should therefore specify, to the auditing bodies concerned, the information needed by the Bank and the format in which it should be reported.- For projects like the Irrigation Rehabilitation and Completion Project, where the requirement is for the audit of project accounts only, the Bank should be satisfied with a certified statement of project expendi- tures detailed to show items broken down by credit or loan category as well as the total of Bank disbursements for each category. Such a report would be of-use to Projects Departments in determining whether project expendi- tures are in accordance with terms of the Agreement and to the Controllers Department in reconciling Bank disbursements with the audited figures. VII. SPECIAL ISSUES Environmental Impact 7.01 At appraisal it was thought possible that the return flow from increased irrigated areas may have caused river pollution and adversely effected existing offshore and riverine fisheries in the three project areas. At Tokat and Silifke, there was no evidence that this had occurred. But at Koprucay, local fishermen have been complaining to the Extension Service that their catches.are suffering due to the pesticide spraying of the large cotton areas. But this may be due to the greater number of fishermen working in the area rather than to the cotton spray- ing. If cotton spraying has had an adverse effect on offshore fishing at Roprucay, then this cannot be attributed to the project since cotton spray- ing was also carried out, and on large areas, prior to the start of the project. Regional Impact 7.02 The Turkish authorities strongly believe that the project has had a significant impact in promoting regional development and in stemming the urban drift, particularly at Tokat. The rehabilitation and improvement of irrigated areas has increased the potential profitability of farming on a given area so that villagers are now able to earn a living by farming rather than being forced to go to the cities in search of work. Unemploy- ment has fallen significantly, as additional employment opportunities have been generated by the construction of project works, the growth of existing industries and the establishment of new industries in the project areas. While it is likely that these changes have occurred, an attempt to sub- stantiate them is beyond the scope of this report. In addition, it is not possible to separate the stimulus to regional development provided by the project from that provided by the-Government through its agricultural policies including input price subsidies. Land Consolidation 7.03 During negotiations, land consolidation was a major issue. The Bank had felt there was a need to establish'legal means and procedures in - 51 - order to implement a program for the consolidation of fragmented landhold- ings in agricultural areas of Turkey. A covenant had been included in the Draft Credit Agreement requiring the Government to carry out a study on ways and means to effect land consolidation. Discussions at negotiations centered on the scope of land consolidation (whether limited to the project areas or extended to the whole of Turkey) and the nature of the tools needed for achieving land consolidation. 7.04 The Turkish delegation stated that (a) legislation enabling land consolidation to be carried out already existed, (b) a program of land con- solidation was unacceptable at that time in view of the land reform pro- posals under consideration by the Government, and (c) in many ways a land consolidation program was not feasible before cadastral surveys and regis- tration of land titles had been completed. The Bank therefore agreed to delete the covenant relating to the land consolidation study from the Draft Agreement. However, although the Government was no longer legally obliged to carry out the study, the Bank urged the Government to undertake the study in order to identify problems that the Government would face in undertaking land consolidation and the benefits that would be derived therefrom in the form of lower construction and operation costs for irriga- tion systems and lower production costs to individual farmers. 7.05 The Government's draft completion report states that the Govern- ment has initiated some demonstration land consolidation programs which have included the construction or improvement of farm roads, irrigation and drainage systems and reallocation of scattered farm plots. TOPRAKSU has apparently also carried out some land consolidation programs in various irrigation projects and found that for these projects there were signifi- cant cost savings in both land development and irrigation works. Benefits were also higher due to greater farming efficiency and smaller losses of farm land. However, land consolidation efforts have been very small scale. Between 1961 and 1970, only 29 land consolidation projects were. carried out covering a total of 173,177 hectares (less than 3/4 of 1% of the total arable area in 1970). 7.06 The Government believes the major problems encountered in effect- ing land consolidation on a large scale are: (a) inadequate cadastral surveys, (b) initial apathy and/or reluctance on the part of farmers to participate in land consolidation programs, (c) some farmers do not have officially recorded land titles, (d) transfers of land where ownership is not officially recorded cause disputes, (e) planning personnel do not have the authority to settle disputes, and - 52 - (f) there is currently no legal means to prevent future land sub- divisions, without which land consolidation efforts are doomed to failure. Some Government officials believe that irrigation projects have helped sustain increased land fragmentation because many holdings which were too small and dispersed to be economic under rainfed conditions became viable under irrigation. 7.07 In 1963, the average farm size in Turkey was 6.8 hectares (a decline from its 1952 level of 7.7 hectares). Nearly 50% of farm families owned six or more parcels of land. Land fragmentation increases over time as farmers redivide their holdings amongst -their heirs. Average farm sizes in 1971 were 6, 3 and 4 hectares for Koprucay, Silifke and Tokat projects respectively. Fragmentation in all pr6ject areas was common, with individual ownerships averaging four plots. Unfortunately, comparable data for more recent years is not available. 7.08 In order to achieve the greatest economic benefits from irrigation projects, such as the Irrigation Rehabilitation Project, action needs to be taken wherever possible to consolidate fragmented landholdings provided it is possible to prevent further fragmentation of consolidation holdings. Hence, there is still a need for a study to be conducted to quantify addi- tional capital costs incurred in the irrigation distribution system by serving such fragmented holdings, and-the additional operating costs incurred by individual farmers through farming several small plots instead of one contiguous plot. Lack of Farmer Support in Koprucay for On-farm Development Works 7.09 Although on-farm development works were completed in Silifke and Tokat without major delays, this was not the case at Roprucay. By the end of 1978, less than 6,500 ha of land leveling or one-third of the target of 18,000 ha had been carried out, surface drainage works had come to a halt in 1975 after only 20% of the target had been achieved, and 40% of the tile drainage work was still outstanding. TCPRAKSU's explanation for its unsatisfactory performance at Koprucay is that the area is dominated by cotton growing (unlike Silifke and Tokat) and high cotton prices in the early years of project implementation made farmers reluctant .to allow on-farm development works to be carried out during the cotton-growing season. Thus, the available working season for TOPRAKSU was constrained to less than three months per year and the pace of work was further handi- capped by the scattered nature of the work involved. Similar difficulties in carrying out on-farm development works had previously been encountered by TOPRAKSU at Seyhan which is also a cotton-growing area. 7.10 The scattered nature of on-farm development works and short working season were constraints not unique to Koprucay. At Silifke and - 53 - Tokat, the work was also fragmented and the working season short, partic- ularly at Tokat where the wheat is harvested late and the onset of incle- ment winter weather is comparatively early. But Koprucay, unlike Silifke and Tokat, is a predominantly one-crop area. Land release by farmers in a one-crop or predominantly one-crop system is very difficult to secure as these farmers are dependent for most or all of their household income on that single crop. Such farmers must therefore be given some financial incentive if they are to be expected to forego income from one season's crop in order to allow on-farm development works to be carried out. Neither this Project nor the Seyhan Project paid sufficient attention to this constraint to development. 7.11 One possible incentive to farmers would be provided if the Government were to compensate farmers for their income losses but later recover, over a fixed period, the compensation payments as, perhaps, a separate charge in addition to the water charges and DSI capital recovery charges already paid by farmers. The Bank might consider making such a suggestion to the Government. Demonstration of the potential benefits to project beneficiaries is also an important factor in the successful imple- mentation of on-farm development works but one which was not carried out in either this or the Seyhan Project. Future Bank projects in this area could therefore usefully incorporate into project design a set of activities to be undertaken by the Extension Service or other appropriate implementing agency that would demonstrate to the project beneficiaries the benefits of on-farm development. 7.12 In the Seyhan Project area, the land tenure situation, with 15-20% tenant farmers, presented another serious obstacle to TOPRAKSU's work pro- gram. Tenant farmers were particularly reluctant to allow on-farm develop- ment works to be carried out as they feared their leases would not be renewed and they would not, therefore, reap the benefits of the works. In the project areas of the Irrigation Rehabilitation and Completion Project, tenant farming is much less common and did not present a significant obstacle to the implementation of TOPRAKSU work. VIII. BANK PERFORMANCE Project Justification and Objectives 8.01 The project was consistent with the Government's objectives and the Bank's strategy at the time it was appraised. The 1978 Agricultural Sector Review concluded that existing irrigation facilities were still grossly underutilized. Public investments have been allocated to new irri- gation schemes with too little attention paid to completing old schemes, to farmer support services and to on-farm development. Of the 1.5 million ha command area of public irrigation schemes, at least half requires on-farm development before the land can be cropped to its full potential. However, the Bank is not willing to finance further rehabilitation and completion - 54 - projects until the problem of inadequate cost recovery by the Government is resolved. Project Content 8.02 The Government resisted carrying out separate feasibility studies as it wanted to commence work on the project without delay. The Bank reviewed the basic data available and concluded that it was adequate. The Bank therefore agreed to the Government's proposal to limit project preparation to the presentation of the necessary basic data. The subproject areas and project content were largely determined by the Government. However, the Bank recomm-ended, and the Government agreed to, the exclusion of a fourth subproject at Aksu. This was done primarily because preparation of the Aksu project lagged far behind that of the other three projects and would not have been ready at the time of appraisal. 8.03 Cost recovery, the employment of consultants and the role of pri- vate contractors in on-farm development works were issues raised during negotiations and continued to be key problems during project implementa- tion. The resolution of the cost recovery problem is a political matter and lies outside the scope of the Bank's involvement. However, it was unfortunate that neither the appraisal nor the negotiating teams identified the legal impediments faced by TOPRAKSU and the Extension Service in enter- ing into contracts of greater than one year's duration. Contractual limitations were a major contributing factor to implementation delays caused by the non-employment of consultants and the lack of private con- tractor interest in on-farm development works. Had these problems been identified at appraisal and/or negotiations, the Credit documents could have been appropriately modified to circumvent many of the obstacles to hiring consultants or using private contractors that subsequently arose. At appraisal, provisions made for the Extension Service were not well defined. Even if consultants had been hired, the Extension Service would still have benefitted from more detailed staffing and equipment provisions being established at the time of appraisal. Inadequate attention at appraisal was paid to the operation and maintenance of irrigation and drainage facilities. It was assumed that farmers, as in other projects, could be relied upon to maintain on-farm works such as leveling, tile and surface drainage. Drainage facilities, however, have sometimes not been maintained satisfactorily by farmers because adequate funds have not been available for the purpose or farmers have n6t fully accepted this respon- .sibility. 8.04 Both physical (10%) and price (5%) contingencies were provided for at appraisal but proved to be quite inadequate. The cost overrun in foreign exchange was US$3.6 million (or 10%) and in local currency was TL 130 million (24%). While the worldwide acceleration in inflation from 1973 onwards could not have been foreseen, the price contingency allowed for was low considering the rates of inflation prevailing at the time of appraisal. - 55 - Project Administration 8.05 Supervision in the field was more time-consuming than usual. Three quite separate areas had to be visited, but it was not always possible to devote sufficient time to each subproject. Excluding one preparation for completion mission, the project was supervised eight times. Thirteen individuals (Bank staff members and consultants) partic- ipated in these missions, two of whom participated in two missions and one participated in three. Continuity of staff involvement was handicapped by the series of Bank reorganizations which occurred in the early years of the project. Supervision missions in the first few years of the project should have been more frequent. The first mission (May 1972) was not followed until nearly one year later (in April 1973). Six months later, in November 1973, the project was again supervised. But another year elapsed before a fourth mission was sent (December 1974). Thereafter, the average length of time between missions improved slightly to eight months from its previous level of more than nine months. Clearly, more frequent super- vision missions, however short, in the first 1-2 years of project imple- mentation would have been extremely useful in overcoming initial difficul- ties and resolving start-up problems. During such supervision missions, continuity of Bank staff is most important. 8.06 Bank supervision missions generally concentrated.more on the implementation of physical works than on the general state of agricultural development. Supervision missions were also very much concerned with the staffing, transportation and organizational inadequacies of the Extension Service. But measures taken by Bank staff to improve the quality and use- fulness of Extension Service reporting were too few and too late. This was a serious omission by the Bank. Through frank discussions with Government and demonstration of acceptable progress reports, the Bank should have communicated more strongly to'the agencies concerned the importance of timely and responsible progress reports. IX. CONCLUSIONS 9.01 The Project was consistent with the Government's objectives and the Bank's strategy at the time it was appraised. These have remained un- changed. Irrigation facilities in Turkey are still grossly underutilized. Of the 1.6 million ha command area of public irrigation schemes, between 600,000 to 900,000 ha needs complementary works and/or on-farm development before the land can be cropped to its full potential. 9.02 The Project did not receive the full commitment of the implement- ing agencies. This had detrimental effects on project implementation as evidenced by DSI's resistance to carrying out full feasibility studies in the project areas, ineffectiveness of the coordinating committees, and of the Extension Service, non-appointment of consultants and delays in DSI and TOPRAKSU works. -56 - 9.03 The success of the Project in spite of these obstacles should enhance the Government's commitment to future irrigation and rehabilitation projects. As a result of the 1980 Agricultural Projects Identification Mission, agreement has been reached between the Government and the Bank on the major potential for completing and rehabilitating existing irrigation projects. 9.04 Before the Bank is likely to consider financing further irrigation rehabilitation and completion projects, however, the Government must take significant steps toward improving cost recovery. Irrigation water is not a free good and is becoming an increasingly costly input for raising agricultural output.. The development costs and operation and maintenance costs of irrigation projects tend to be high compared with many other forms of agricultural investment and other agricultural inputs. At appraisal, the Government had given assurances that in the three project areas water charges would be assessed and collected to cover operation and maintenance costs and recover a reasonable portion of the investment costs. Despite these assurances, the level of total project costs recovered since 1973 has only once exceeded 50% of O&M costs. DSI's capital costs are partially recovered from farmers but there is no provision for recovery of TOPRAKSU's on-farm development costs. Given the low levels of capital recovery charges for DSI works, the lack of provision for cost recovery of TOPRAKSU works, poor collection rates and no yearly adjustment for inflation, recovery of project costs is likely to remain unsatisfactorily low, its O&M costs not being fully recovered, let alone the capital costs of DSI and TOPRAKSU works. This is difficult to justify, given that project beneficiaries have average income well above the national average. 9.05 The Bank has long been concerned at the low cost recovery levels for irrigation projects which have been a feature of not only this project but also the Seyhan Irrigation Project (Loan 587/Credit 143-TU). Bank attempts over the years to encourage the Government to improve the situation may have influenced the Government's decision in 1978 to raise 0&M charges fivefold. O&M charges then covered O&M costs. Unfortunately, this improvement has not been sustained. Water charges in 1979 in two of the three project areas were less than their 1978 levels. Experience with both projects clearly illustrates the difficulty faced by the Bank of rationalizing, through its project lending, sector policies which a Government does not feel it wishes or is able to change. However, the Bank should continue to urge the Government to take steps to rectify the present unsatisfactory cost recovery situation. 9.06 Successful implementation of on-farm development works requires the timely release by farmers of the land reauired. Land release by farmers in a one-crop or predominantly one-crop system, such as at Koprucay, is very difficult to secure. These farmers are dependent for most or all of their household income on that single crop and must there- fore be given some financial incentive if they are to be expected to forego income from one season's crop in order to allow on-farm development works to be carried out on their land. Consequently, consideration should be - 57 - given to persuading the Government to compensate farmers for their finan- cial losses incurred as a result of the on-farm works with the Government later recovering compensation payments from the farmers over a fixed period of years. Demonstration of the potential benefits of on-farm development work to potential beneficiaries should also be explicitly incorporated into project design as a set of activities to be undertaken by the Extension Service or other appropri- ate implementing agency. 9.07 The Bank should take the necessary steps, throughout the period of project implementation, to maintain the timeliness and utility of reporting by the implementing agencies, as the quality of these reports has a direct impact on the Bank's and the Borrower's ability to monitor and evaluate project progress. 9.08 Provisions for organization and management of the Extension Service should have been made explicit-at appraisal and fully integrated into the project. This was not done at appraisal as it was understood that the evident success of the Seyhan Extension Service would be sufficient to ensure its adoption by the Government in the Irrigation Rehabilitation and Completion Project. Clearly, this understanding was not sufficient. As a result, subsequent Bank-financed projects in the Turkey agricultural sector, such as the Corum-Cankiri Rural Development Project (Loan 1130-TU), have incorporated detailed provisions for the Extension Service with beneficial effects. Most recently, the Bank's efforts to assist the Government in building an effective Extension Service throughout Turkey are being continued through the proposed Erzurum Rural Development Project where provisions for the Extension Service take into account lessons learned from Bank experience to date. 9.09 With the Irrigation Rehabilitation and Completion Project, the relative gains in agricultural development achieved in each of the three areas have been the inverse of appraisal expectations, in that Koprucay has turned out to be the least successful and Tokat the most successful. The Project Completion Report for the Seyhan Project also concluded that appraisal predictions for agricultural development in the project area were rather inaccurate. It recommended that the Bank's understanding of the factors affecting farmer decision making be improved by the Bank undertaking research in this area. This report reiterates that recommendation. Inaccuracies in appraisal predictions for agricultural development are likely to continue so long as inadequacies-exist in the Bank's understanding of the factors affecting farmer decision making. 9.10 The consultants, had they been employed as anticipated at appraisal, would have benefitted project implementation considerably by helping to speed up the pace of project works, improving coordination between the implementing agencies and accelerating agricultural development in the three project areas. The Bank should therefore have given greater consideration to taking firmer steps, such as a possible suspension of disbursements, to ensure the Government employed the consultants. - 58 - 9.11 In view of the Government's resistance to carrying out full feasibility studies of the three project areas, the need for expediency, and because the Bank believed that adequate data was readily available, the Bank agreed to the Government's limiting project preparation to the presentation by DSI of the necessary basic data. Had the Bank insisted on the Government carrying out the studies, the implementing agencies might have been more committed to the project with consequent beneficial effects on project implementation. 9.12 A necessary requirement for the effective functioning of coordinating committees is that their chairman and members be the highest possible level officials, with preferably the chairman not being an official of one of the implementing agencies. This requirement was not met in the Irrigation Rehabilitation and Completion Project. However, the lesson learned has been incorporated into the design of the Corum-Cankiri Project where provincial project coordinating committees are chaired by the provincial Governor. 9.13 There are many expected advantages from an agricultural standpoint that would result from consolidation of fragmented landholdings. The Bank should therefore continu.e to urge the Government to carry out a study to quantitatively assess the costs and benefits of land consolidation. The study could be financed under a future Bank project in the agricultural sector. T11UYY IRRIGATION RE11ABLITATION AID COMPLETION PROJECT Equipment List - DSI SILIFKE KOPnUCAY TOKAT TOTAL Price Appraised Estimates Actual Appraised EstLimatce Actual Appraised Estimates Actual Appraised Estimates Actual 1 ~ I . ~a. as0. as S .) CSt( 0K(N.) CODS0. CostoUS '00ostostUsr'0rm CosttUS4-5 '00) N o stUs 'cost (Ust(US '0006 (osNUo.00) Cost(U,S,;'0005U Der For operation and maintenancet Motor grader (110/125 hp) 20 1 20 1 39 2 10 2 78 3 60 3 116 6 120 6 233 Cravler excarator (1 a.y. 35 1 35 1 37 3 105 3 111 4 140 4 147 8 280 8 295 Crawler tractor (60/75hp 1b 1 15 1 25 2 30 2 * 50 2 30 2 51 5 75 3 126 Crawler tractor (110/130 hp) 30 2 60 2 84 1 30 1 42 3 90 3 127 6 180 6 253 Grade-all_(whee tI ) 55 1 55 1 51 3 165 3 155 4 220 4 206 8 440 8 412 In"der (1-3/h a.y.7 20 1 20 1 24 2 4o 2 49 3 60 3 74 6 120 6 147 Trailer (30 ton) 60 - - - - - - - - 1 40 1 45 1 ho 1 45 Dump truck (10 ton) 11 7 77 7 151 14 154 14 302 20 220 20 431 41 451 41 8864 Station vagon (4 vd) 5 1 5 1 9 1 5 1 9 1 5 1 9 3 15 3 27 Pick-up (4 v) 3 4 12 4 23 10 30 10 57 10 30 10 57 24 72 24 137 Compressor (125 c.f.a.) 14 1 14 - - 1 4- - 1 4 - - 3 12 - - Sprayer (boom type) 10 1 10 - - 1 10 * - 1 10 - * 3 30 * * ater stage recorder 1 1 1 - - 3 3 - - 5 5 - 9 9 - Current-meter 1 1 1 - - 3 3 - - 2 2 - - 6 6 - Spare parts (20%) * 174 124 343 184 509 371 1,026 Subtotal 378 743 1,100 2,221 P=ps and motors A -190 84 - 190 84 Ln Radio system - 30 - . . 30 - Ttal caf Cast 61p rc u r441 tociDadLy procured 늬 TURKEY IRRIGATION REHABILITATION AND COMPLETION PROJECT COMPLEMON REPORT Construction Schedule - TOPRAKSU Actual as Subproject Appraisal % of Appraisal ire& Work Item 1972 -7-3 1974 1975 127-6 1977 1278 Estimate Estimate Koprucay Land levelling 80OLl- 1,576 2,434 3,663 4,795 5078 6,370 18,00OZ2 35 Tile drainage (he) - - 300 11295 2 , 95 3095 3P995 69900 58 Surface drainage (ha) - IP500 3050 39550 39550 3,550 3,550 18,000 2 Ll Feeder roads (km) - 6 12 32 52 72 75 108 69 Soil amendment (he) - - - - - 34 134 84o L Silifke Land levelling (he) N/A 427 lt208 lt845 2,040 3,14o 39300 MOO 100 Tile drainage (he) - 520 1,130 1,705 29385 29385 2,350 101 Surface drainage (he) - 19300 19930 3p120 39300 3000 100 Feeder roads (km) 5 15 19 30 30 25 120 Tokat Land levelling (ha) 2.,6ooZl 4,16o 6,366 8,215 lo,46o 12,742 l5vOOO 15,000 100 Tile drainage (he) - 300 goo 2,100 3,130 3,330 3,390 99 Surface drainage (he) N/A 1057 4,226 6#326 9,026 13p926 15,000 15,000 100 Feeder roads (km) N/A 7.5 23.5 38.5 6o 99 101 go 112 Includes 500 he (Koprucay) and 2,200 he (Tokat) of land levelling which were carried out prior to credit effectiveness. 2 The Appraisal Report maybave overestimated the amount of land levelling necessary at Koprucay io view of the effective irrigation practices in operati6n"_&h*c'ejrta'in areas of non-levelled land. In 1975 in the project area., the Bank suggested that the consultant who was to have been employed by TOPRAKSU review the land levelling estimate. However, as the consultant was never employed, the review was not carried out. /3 No surface drains were constructed,after 1974. The farmers did not feel they were advantageous and refused TOPRAKSU permission to carry out the work on their lands. The soil improvement program was hampered by the lack of suitable gypsum in the area, Target was reduced from 3,860 he to 3,36o he. Sourcei Quarterly Progress Reports, TOPRAKSU. PåIGATIOff REILuITATIO! An coMiro0n POJECT Cot utlinaten OmUMAT A . mILum flnal dofta r~ cILM, Wlr! de Appraisal sa %. Appraleal as - Appraisl ua i ZEtimtte Final Coft 4 ~ Eutinate Final Coat Apraial etieate Final Coa An 1a1 -- L -----(IL ~Ine;--- -b~lang)- W. CIvn.wa . CIVIL VOMS 1. CIL VOnKS Irr1 ation and PraleaC Vorke Irrigation and Drainage worka Irrigatinn änd pranae Woka a) night mank R igbjt Bank 26.12 33.88 1 Drainac works la re 11 L-ft Bank ConstrueLlon 15.00 27.90 presenti irriCad 2.964 6.8.. 1) Irrinatjon and Draina e 39.33 30.06 Engincering 2.l5 h.1 i1) Cpletion of gravity aea 17.20 ph.1 - i) Jake XR% laprovemnt 3.10 32.16 land Acquiittion l2.0 111) P"p1nC Area, Ctage 1 22.38 1.67 111) tesirlirmak Channel Dubtotal Irri- Y) rurpIng Ara, quo 22 9.7 26.13 Iffrovement .50 36.55 2 . gatio and 15.75 31.12 18 v) Ako rppbig arwa 2.93 7- Bubtota. . T3.63 DMIEAge lort pacn 7.56 20.12 Eögineering (,5%) 11.11 19.60 r elfiång, .1.50 18.85 . [and Aequilition _1.87 _. d) Canalette anaraet - 7. Bubtetal Irrlgation and g"totai Drainage 10.03 155.68 156 Magtneertag (1%) .61 30"n land ACEelttan .16 3. Bubtotal Irriat1« .*nd Drainage 80.21 336.861 TOPRAKSU i0PRAU Ca-rar Develop~et Ot-fare Developenot On-farm Developent Sand levcllff 28.m 55. land levlling 23.01 50.63 Iand levelling 12.55 Tita raMge 11.71 29. TLIe drainage . 7.16 11.82 Tile dralnage . 6.68 Gott en ent 3.78 1.97 Noll muendgent 1.13 - furface drsine and Surface fraise ad pa4e 10.01 3.62 Burface draina 11.57 6.88 roada 1.S5 1.58 Dgneertg (0%) M .nglneering (10%) .9 691 ngineering (10%) 1.16 2.08 Stotat er än-frm gubtotal or O-farm ubtotal of Om. Develoat 59.30 99.51 16 Deveio~nb 50.66 76.06 1 •rarm Develop. ment 12.77 22.89 179 n.II*ern éduI02ST n. EQuIFnerrf N1i 17.10 26.11 DSi 17.10 37.03 DSr 5.85 12.96 lo^110 62. 26.80 TOPpASU 35.60 21.97 TOPAsu 1.65 7.9 z~tenulon Servicse 0.109 Qatenslon Services 0. 1.23 rten9ion service* 0.21 1. Dtsta 3 WU 89 Dubtotal 1 W_g.27 116h subtotal I.SI1 119 us.~st OMmmCsHI 0TIMIMCIES •ni. COMTIM CIEs tedhleal (10%) 13.96 - Tuohnica (10%) 15.05 - Technieal (101) 3.15 - Price (51 couennfd) 21.76 Prie (5% CoGpounded) 26.07 - Pris (5% etpaded) I.1 - 9bt,ta - gubotal 1ubtotal total imj19etJ1 28.1.56 M2 MWf ~11 MaGldes merk ta be empletad to 1980 and 191 et a cttnated got of'% 8.78 milon. I IMalee work to be aplett& ta 1980 at m fotimatec acst er T& 3 mlta. -63 - ANNEX II Table 2 TURKEY IRRIGATION REHABILITATION AND COMPLETION PROJECT COMPLETION REPORT Analysis of DSI Cost Increases Irrigation and Drainage Works Estimated Cost Increase Due Increase Due Total In Appraisal Contract to Changes in to Price Final Work Items Report Price Quantities Escalation Cost ------------------------- (TL '000)------------------------ KOPRUCAY (a) Right Bank (i) Drainage works 2.94 6.30 -1.25 1.43 6.48 (ii) Completion of gravity area 17.20 19.21 - 5.20 24.41 (iii) Pumping area, Stage I 22.38 34.38 1.66 5.63 41.67 (iv) Pumping area, Stage II 9.57 16.29 0.58 7.48 24.35 (v) Aksu pumping area 2.93 6.50 0.95 - 7.45 (b) Left Bank () Pumping area, Stage I 5.74 5.79 - - 5.79 (ii) Pumping area, Stage II 1.80 6.12 1.02 7.19 14.33 (c) Construction of Pump Buildings - 7.76 2.04 7.56 17.36 (d) Office and Services Buildings 1.50 1.50 - - 1.50 (e) Canalette Manufacture - N/A N/A - 57.78 Subtotal 64.06 201.12 SILIFKE (i) Improvement of existing and proposed works . 15.00 14.30 7.90 - 22.20 (ii) Construction of drainage pump building - 5.70 - 5.70 Subtotal 15.00 27.90 TOKAT (i) Irrigation and drainage works at right bank 26.12 33.88 - - 33.88 (ii) Irrigation and drainage works at left bank 39.33 30.04 - - 30.04 (iii) Lake Kaz improvement 3.10 22.96 5.18 4.02 32.16 (iv) Yesilirmak Channel improvement 5.50 17.85 - 16.70 34.55 Subtotal 74.05 130.63 Grand Total 153.11 359.64 11URKEY IRRIGATION REHABIL1TATION AND COMPLETION PROJECT COMPLETION REPORT Koprucay - Cropping Patterns and Yields 1973-79 1985 1973 1976 1975 1976 1977 1978 1979 (Projected) Area Yield Area Yield Area Yield Area Yield Area Yield Area Yield Area Yield Area Yield 1TW Tons/IS =a1i To-n-s7Ul 7a Tons/Ila 7-i Tons/lia Th onelila ThI Tqns/11aa ~Tns/Ra Ha Tb-n-sFr-a Irrigated Helon 133 17.50 62 23.0 253 23.0 369 25.0 211 20.00 688 20.0 337 25.00 1,050 25.0 Cotton 13,777 3.25 14,276 3.3 13,229 3.1 12,911 3.0 11i,134 3.50 13,138 2.3 8,186 2.75 9,500 3.5 sesame 28 0.60 46 0.7 472 0.7 80 0.8 93 0.90 510 0.8 1,926 0.90 2,400 1.0 Rice 56 4.50 229 l4.o 205 11.0 465 4.5 223 1.c" 11 1.3 257 1.00 650 5.0 Fruit 112 17.50 68 16.0 105 21.0 116 26.0 150 25.00 166 26.0 186 28.00 720 28.0 Vegetables 202 25.00 213 26.0 239 26.0 465 21.0 431 27.50 1121 30.0 754 30.00 1,280 35.0 Cereal 42 2.50 6 2.6 22 3.0 171 3.6 114 3.75 197 4.0 1,819 4.oo 3,700 4.0 Others 22 - - 76 - . 160 - Irrigated 11,372 11,930 11,525 14,1430 15,732 15,260 13,625 19,030- Nonirrigated Cotton 3,008 1.75 2,801 1.8 2,065 1.80 2,693 1.80 2,1478 1.85 2,705 1.9 3,299 1.9 Vegetables 231 17.50 215 16.o 159 16.50 207 16.70 191 17.00 208 17.2 253 17.5 Cereal 2,391. 1.75 2,226 1.7 1,641 1.75 2,141 1.80 1,969 1.80 2,150 1.8 2,623 1.8 2,000 2.0 Sesame 926 0.70 862 0.7 635 0.70 829 0.75 762 0.80 832 0.8 1,015 0.8 Pelon 77 15.00 72 15.0 53 15.90 69 15.50 63 15.50 69 15.8 85 16.0 Fruit 77 15.00 72 15.0 15.50 69 15.80 63 16.00 66 16.0 8 16.0 Total Nonirrigated 6,710 6,248 1,606 6,oo8 5,526 6,033 7,360 2,000 Fallo 1003 2,954 1,647 827 792 1,100 1,055 Grand TotWl 222 22,035 20522852,5228522,085 2.2,085Q 21-0-2 EAMa E-'-8 TURKEY IRRIGATION REHABILITATION AND COMPLETION PROJECT COMPLETION REPORT Silifke - Cropping Patterns and Yields 1973-79 1982 1973 1974 1975 1976 197 1978 1979 (Projected) Crop Area Yield a Yield Area Yield Area Yield Area Yield Area Yield Area Yield Area Yield Ha Tons/Ha Ia Tons/Ha Ha Tons/1a Ha Tfns/Ha Ha Tons/Ha Ha Tons/Ha Ha TonHaH& Ha Tons/H& Peanuts 1,440 1.9 1,44o 1.7 1,240 1.75 1,600 1.80 1,81io 1.75 1,760 1.80 1,840 1.85 1,506 2.2 Sesame 640 0.6 440 0.5 680 o.60 o 0.80 360 0.80 Wo0 0.70 480 0.70 538 1.0 Rice 64o 4.0 1,080 4.0 1,000 3.30 800 4.00 1,o4o 3.80 1,130 4.50 1,040 3.50 1,261 5.0 Vegetables 800 30.0 600 30.0 560 36.00 720 33.00 760 36.00 815 30.00 800 35.00 1,316 40.0 Fruit 280 35.0 280 29.0 280 35.00 310 35.00 310 35.00 350 39.00 510 41.00 439 41.0 Cotton - - 200 1.5 - - - - - - - - - - 274 3.5 Melon 200 10.0 - - Beans 240 7.0 240 6.0 520 6.50 480 8.00 480 8.00 1o 7.00 280 5.00 - - O Others 240 440 530 510 510 480 335 513 cereal (Dry) 2,800 3.5 2,800 3.0 2,720 3.75 2,760 3.75 2M0 2.90, 2,880 2.25 2,880 4.00 2 4.0 Total Cropped Area ifa) 6" L&28 71520 7,580- 7"70821816 IRRIGATION REHABILITATION AND COMPLETION PROJECT COMPLETION REPORT Tokat - Cropping Patterns and Yields 1973-79 1985 1973 1974 1975 1976 1977 1978 1979 (Projected) Cro Area Yield Area Yield Area Yield Area Yield Area Yield Area Yield Area Yield , Area Yield Ha Tons/ft Ha TonH/fa Ha Tons/a To/Hlf Ha Tons/ft Ha Tono/ Ha Irrigated. Cereal 2,758 2.85 3,o48 2.55 3,318 2.5 3,520 2.7 1,590 2.7 3,626 3.1 3,805 3.2 4,697 3.5 Beans 953 1.65 625 2.30 595 2.8 735 3.3 756 3.5 968 3.5 1,585 3.5 1,o68 3.5 curarbeet 2,948 51.00 3,019 56.oo 3,824 57.0 3,737 58.0 3,903 60.0 3,808 60.0 4,117 60.0 6,270 60.o Altalfa 175 6.00 215 6.oo 247 7.0 215 7.0 214 8.0 226 8.0 298 8.0 1,068 10.0 Fruit 1,146 19.oo 1,139 19.00 1,104 19.0 1,160 21.0 1,193 22.0 1,228 23.0 1,278 23.0 3,415 25.0 Tobacco 188 1.00 211 1.00 118 1.0 365 0.9 450 1.0 486 1.1 329 1.1 N/A Ve.etableo 1,821 22.50 2,324 24.00. 1,506 28.0 1,783 31.0 1,571 29.0 2,510 31.0 1,891 31.0 4,697 32.0 Sunflower 259 2.60 1o 2.50 53 3.1 329 3.4 235 3.7 210 2.6 679 2.3 1,281 3.5 V:elon 202 20.00 251 18.00 117 19.0 - 164 23.0 126 24.0 399 35.0 124 37.0 427 37.0 Others 25 425 284 230 292 277 4 127 71tal Irrigated 10,939 11,387 11,207 12,292 10,268 13,753 14,383 21,350 0' Iflonirrigated 0\ Cereal 5,517 1.75 4,728 1.80 5,170 1.8 4,053 1.8 5,874 1.8 4,026 1.8 3,694 1.8 Suarbeet 1,041 30.00 892 30.00 975 31.0 765 31.0 1,108 30.0 760 30.0 695 31.0 Alfalfa 208 3.00 178 3.00 195 3.0 153 3.0 220 3.0 152 3.0 139 3.0 Orchard 220 17.00 200 18.00 200 18.0 160, 17.0 210 17.0 155 18.0 145 18.0 Vegetables 250 12.00 225 12.00 210 12.0 175 12.0 240 12.0 170 13.0 148 13.0 .elon 520 8.00 446 8.00 488 8.0 382 8.0 554 9.0 380 9.0 348 8.0 Vineyard 718 4.oo 625 4.00 -im 4.0 _ 4.0 48 4.0 460 4.0 418 4.0 Total Honirrigated 8,381 7,204 7,823 6,146 8,871 6,103 5,587 eFalow 2,0302752,22,1221149138 Grand Total 21,350 2UL35 21350 21,350 LA 21L35 21,50 Q~ IRRIWT1m REABILItATIOm ND C TRfPT11N RJ ehejMLfnnuk nestwent Cxpendttura (tt. Milloffs, la Current prife) Extenfoft hrigsttag and prattlp service 6 Ire*tmentf DI Eautyment On-farm Vorkf ToRAKStV rautpeeuit Irinan Total In Currgnt Frico Total i 1971 Prtces Ituvey lIfkg Tok31 TotAU ILoecy MMg 1_iki Ta Total Kopruca StIkea Tokat Total Kopruca 5111ko lkat Total kopruca silljfke Teks Total Korry547 å1ttfke ikat TPtal 1971 - . - - - - - - 1.39 - 2.94 4.33 - - - 1.39 - 2.94 A.33 1.39 - 2.94 4.33 1972 9.11 3.90 0.53 13.36 1.19 - 1.19 1.02 0.20 4.24 5.36 - • • 0.09 11.32 4.09 4.79 20.20 9.71 3.51 4.11 17.33 1973 3.42 6.40 12.62 7 - - - • 2.91 1.41 3.539 9.91 11.21 2.77 10.07 24.05 0:70 67,58 10.81 28.78 107.18 47.73 7.64 20.32 75.69 1974 44.28 3.80 12.61 72.67 3.69 1.23 4.92 9.85 4.96 2.93 7.58 13.47 12.14 4.03 9.01 25.18 - 67.05 11.99 44.12 123.16 37.42 4.69 14.62 68.73 1975 12.66 4.30 33.33 50.31 1.12 0.37 1.68 3.36 9.35 3.99 7.37 20.71 2.28 0.87 1.95 5.10 1.18 25.88 9.73 45.04 80.67 12.42 4.67 21.63 38.73 1976 24.85 3.35 25.32 33.67 8.03 4.94 12.58 23.56 11.39 2.27 16.00 29.86 0.05 0.01 0.04 0.10 1.30 44.99 12.88 54.61 112.48 18.30 5.24 22.22 41.76 1977 16.66 2.30 16.94 33.90 0.16 0.16 0.16 0.49 20.34 10.11 16.79 47.44 0.12 0.04 0.06 0.24 0.94 37.76 13.05 34.21 85.02 12.42 4.29 11.25 27.94 1978 14.54 1.90 12.43 28.87 1.14 0.38 1.69 3.40 9.95 - 8.63 18.38 1.00 0.23 0.84 2.07 .3 26.63 2.86 23.39 53.06 6.39 0.69 5.66 12.74 1979 2J.2, . s:_ .3 023 "29 . 4 " .4 10 32. 2 _ 25.7 - -. • - . . 63.1 .4 22.3 91.27 9.484 3.39 13.69 Total PÅMU w LUJA' 2WI5 &bu8 MU L-B 1022a ä& LL U 21h 74 la na~ &L2 2%L ZJ J 3.5161 Iefldes amt et wrke te haeoutlet9d ta 1980. t.elde~ emst of ffike ta bo Comploted I 1980 ämd I981. -68- ANNEX IV Table.2 TURKEY IRRIGATION REHABILITATION AND COMPLETION PROJECT COMPLETION REPORT Farmgate Prices 1971 - 1985 1971 1973 1974 1975 1976 1977 1978 1979 1980-2021 (Appraisal) - - TL/Ton in Current Prices ------- ---- (Projected in 1971 TL) Cotton: Financial 3,000 8,950 7,200 7,250 10,450 10,500 14,650 24,500 4,400 Economic 2,500 5,800 6,000 5,500 8,300 8,600 11,080 18,630 2,900 Rice: Financial 1,600 4,000 5,250 5,500 7,000 8,750 15,000 20,000 2,100 Economic 1,000 5,940 9,060 6,290 4,890 5,830 2,620 Cereal 900 1,550 2,700 2,350 2,600 2,500 3.000 5,250 1,500 Sesame 3,100 11,150 19,250 18,000 16,400 20,000 35,500 70,000 9,000 Melon 400 900 1,200 1,350 1,500 2,050 3,150 3,300 600 Peanuts 3,250 4,750 6,750 8,000 9,000 11,000 16,000 30,000 4,200 Beans N/A 5,600 6,800 7,350 8,100 10,000 21,000 30,000 5,000 Sugarbeet 200 300 400 500 690 630 820 1,300 230 Fruit 800 1,150 1,450 1,750 2,470 3,440 7,600 7,500 1,300 Vegetables: Tokat N/A 900 1,150 1,350 1,700 1,850 2,040 2,500 550 Silifke 400 1,400 1,500 1,750 2,420 3,050 4,240 5,000 1,100 Koprucay 400 1,800 2,500 2,800 4,200 4,400 5,740 6,000 1,400 Sunflower N/A 2,950 4,250 3,500 7,000 9,000 10,430 16,000 3,000 Tobacco 7,000 14,500 21,450 34,000 46,540 47,790 56,610 65,000 14,000 -69 - ANNEX IV Table 3 TURKEY IRRIGATION REHABILITATION AND COMPLETION PROJECT COMPLETION REPORT Price Deflators Used Year Index 1971 100.0 1972 116.6 1973 141.6 1974 179.2 1975 208.3 1976 245.8 1977 304.1 1978 416.6 1979 666.6 - 70 - ANNEX IV Table 4 TURKEY IRRIGATION REHABILITATION AND COMPLETION PROJECT COMPLETION REPORT Economic Rate of Return Calculation for Koprucay - Comparison Revised Completion Turkish Government Estimate Estimate Appraisal Estimate Incremental Incremental Incremental Incremental Incremental Incremental Year Costs Benefits Costs Benefits Costs Benefits 1972 11.10 - - - 51.7 - 1973 47.82 30.67 226.34 50.99 77.5 7.2 1974 38.42 22.04 23.78 52.51 54.7 13.9 1975 13.72 9.12 18.13 54.08 31.4 20.5 1976 21.50 2.44 28.14 57.54 20.7 27.0 1977 15.82 25.30 20.30 61.80 5.6 44.01 1978 10.49 22.311 9.72 63.20 5.6 42.1 1979 14.18 15.27 7.86 71.40 5.6 50.7 1980 4.70 22.86 7.88 77.96 5.6 59.3 1981 4.70 30.44 7.89 86.05 5.6 67.9 1982 4.70 37.82 7.91 94.59 5.6 67.9 1983 4.70 45.40 7.93 103.62 5.6 67.9 1984 4.70 52.98 7.95 113.15 5.6 67.9 1985 4.70 60.36 7.97 123.16 5.6 67.9 1986 4.70 160.36 7.99 133.68 5.6 67.9 1987-1996 4.70 60.36 8.01 144.82 5.6 67.9 1997 4.70 60.36 8.01 144.82 7.2 67.9 1998 4.70 60.36 8.01 144.82 5.6 67.9 1999 4.70 60.36 8.01 144.82 6.6 67.9 2000 4.70 60.36 8.01 144.82 6.2 67.9 2001-2021 4.70 60.36 8.01 144.82 5.6 67.9 Economic Rate of Return 23% 29% 19% /1 Includes the salvage value of TOPRAKSU equipment: Completion estimate TL 10.7 million Appraisal estimate TL 10.5 million - 71 - ANNEX IV Table 5 TURKEY IRRIGATION REHABILITATION AND CONPLETION PROJECT COMPLETION REPORT Economic Rate of Return Calculation for Silifke - Comparison Revised Completion Turkish Government Estimate Estimate Appraisal Estimate Incremental Incremental Incremental Incremental Incremental Incremental Year Costs Benefits Costs Benefits Costs Benefits 1972 3.51 - - - 11.0 - 1973 7.84 3.90 66.61 12.83 21.8 1.9 1974 6.89 1.86 7.02 18.20 17.9 3.9 1975 4.77 5.88 7.26 18.65 5.9 6.0 1976 5.84 7.12 6.47 23.63 5.9 8.1 1977 5.19 6.22 5.98 28.75 2.0 13. 41 1978 2.09 8.85L1 2.72 30.09 2.0 12.6 1979 2.22 2.32 2.27 31.46 2.0 15.1 1980 1.40 9.02 2.28 32.86 2.0 15.1 1981 1.40 15.72 2.29 34.30 2.0 15.1 1982 1.40 22.41 2.30 35.77 2.0 15.1 1983 1.40 22.41 2.31 37.28 2.0 15.1 1984 1.40 22.41 2.32 38.82 2.0 15.1 1985 1.40 22.41 2.33 40.40 2.0 15.1 1986 1.40 22.41 2.34 42.02 2.0 15.1 1987-2021 1.40 22.41 2.35 43.67 2.0 15.1 Economic Rate of Return 33% 35% 18% /1 Including salvage value of TOPRAKSU equipment: Completion estimate TL 3.2 million Appraisal estimate TL 3.3 million - 72 - ANNEX IV Table 6 TURKEY IRRIGATION REHABILITATION AND COWPLETION PROJECT COMPLETION REPORT A Economic Rate of Return Calculation for Tokat - Comparison Revised Completion Turkish Government Estimate Estimate Appraisal Estimate Incremental Incremental Incremental Incremental Incremental Incremental Year Costs Benefits Costs Benefits Costs Benefits 1972 7.05 - - - 43.3 - 1973 20.92 8.93 332.83 46.85 66.8 4.2 1974 25.22 14.63 33.72 55.94 54.8 8.5 1975 22.33 18.75 30.08 54.83 41.4 14.0 1976 23.02 34.10 29.72 67.90 41.4 19.8 1977 13.15 22.86 22.62 53.79 4.6 35.8/1 1978 7.66 52.92 L 16.07 73.86 4.6 33.1 1979 5.59 37.12 11.58 83.36 4.6 36.7 1980 2.20 40.62 11.62 89.86 4.6 38.8 1981 2.20 47.51 11.65 96.81 4.6 39.3 1982 2.20 54.40 11.69 102.56 4.6 39.3 1983 2.20 61.29 11.73 112.17 4.6 39.3 1984 2.20 68.18 11.77 120.64 4.6 39.3 1985 2.20 75.06 11.81 129.68 4.6 39.3 1986 2.20 75.06 11.85 139.32 4.6 39.3 1987 2.20 75.06 11.89 152.60 4.6 39.3 1988-2021 2.20 75.06 - - 4.6 39.3 Economic Rate of Return 42% 20% 12% /1 Includes salvage value of TOPRAKSU equipment: Completion estimate TL 8.8 million Appraisal estimate TL 8.8 million 긔 TURKEY IRRIGATION REHABILITATION AND CO IPLETION PROJECT ORGANIZATION CHART Project Coordination Committee Chairman: Appointed by Government from membershin Members: Regional. Directory DSI Regionol Director; Topraksu Provincial Director, mt. Serv. Director, A ,r- Bank Local Dr. Local Coordinator o Extension DS-1 TOPRAKSU EXTENSION SERVICE AGRICULTURAL BMK Responsible for.planning, Responsible for on-fa'rm Responsible for providing ReRponsible for supplying design, construction, opera development works* technical assistance and credits to farmers tion and maintenance. services to farmers Each project was to have its own organization. IRRIG MAN CANAL P4 P3 P2 Kayaburnu nmenkö BULGARä*,' 8 1 A C K. S E A V.$,5AR 4,4 Lunll T ULu R 1E y Anoy ROJECT Adl •._^lAREAMe a M E D T ERR NEA T~ ~ U R Q<E ' MED)/TERNEAJN EY R l AR SEA AUGUST 1971 TURKEY IGATION REHABILITATION & COMPLETION PROJECT KOPRUCAY PROJECT Existing Main Canal Secondary Canal Proposed Secondary Canal Existing Drain Proposed Drc-1 'Ito" Proposed Deep Drain Tunnel W Existing Pumping Plant IA 0 Proposed Pumping Plant Kl e Diversion Dam 3' Project Areas-Now Irrigated Project Areas -Proposed 2 Uj,K6prVqoy Diversion r Dee DriAe Dam Dam ~ L . ]Deep Drain Area Roads Towns & Villages LEFT BANK IRRIGATION ;P1 MAIN CANAL Belkis osil ~C2'Serik 40 LEFT BANK PUMPING MAIN CANAL LEFT BANK IRRIGATION MAIN CANAL (EAN S EA IBRD 3525 -l BULGARhý',,ý 8 A CK S E A U S. 5. R., PROJECT AREA u, ANKARA 1RRIGA T U R K E y AM.Jy. Ad- I R' A- Q EDITERPANEAN 5 Y R I A Turhal u GUmbstop TURHAL B.Yildiz Yazibogi KAldiz KAZOVA N Coriksiz Akbugdoy 0 ecip Katkoy Koloycik S, Dokmetepe OTatl,ak . . . . . . .......... ovoykift B Baglar ome ese -T..t*., zo 0 Erksl& Toslik 0 AUGUST 1971 Gokdere TURKEY IGATION REHABILITATION & COMPLETION PROJECT TOKAT PROJECT - comag Kizilkoy Bula 0 O YayladahOcoregibuyuk BYld Emrsey O Yolodol EKomec OgU uk d Gumenek Dom OMAL A -.'- TCanals (Main) Cansals (Secondary) * \ E s6ngExisting c.ng rci N.s KAZOVA .•••• »•••• For Improvement """"""" For Improvement cA New New * ulas Drains (Main) Drains (Secondary) 111 Existing Existing Fora Im rvnin Gulpinar ••••• For Improvement For Improvement - -- New ----- New , Bogor >1r4i- - Railroad 0 Towns & Villages +++++++++ Channel Improvement i( Diversion Dam I Pumping Plant Project Area Levees 0 1 2 3 4 5 6 7 8 9 10 Rod Roads Kilometers BRD 3531