Turkey - Seyhan Irrigation Project - Stage Two

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 2747 PROJECT PERFORMANCE AUDIT REPORT TURKEY: SEYHAN IRRIGATION PROJECT STAGE II (Loan 587/Credit 143-TU) November 26, 1979 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: SEYHAN IRRIGATION PROJECT STAGE II (Loan 587/Credit 143-TU) Table of Contents Page No. Preface ...................................................... i Basic Data Sheet .............................................ii Disbursement Table ......i.................................. i Highlights ....................................... .......... iv PROJECT PERFORMANCE AUDIT MEMORANDUM I. Summary ................................ 1 II. Main Issues ...............................5...... 5 The Training and Visit (T&V) System of Extension.... 5 Land Tenure and On-Farm Development ................ 8 Water Management and Environmental Impact........... 9 The Economic Rate of Return.................. 10 PROJECT COMPLETION REPORT I. Background .......... .............................. 11 II. Formulation ... .......................... ....... 12 III. Implementation .................. .... 16 IV. Agricultural Impact ........................ ......25 V. Economic Rate of Return ........................... 33 VI. Institutional Impact.................... o_........ 35 VII. Special Issues............... ............ 41 VIII. Bank Performance ................................... 46 IX. Conclusions ........ ................................ 47 Annexes: I - II - III - IV (with tables) Charts I - II Map This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  PROJECT PERFORMANCE AUDIT REPORT TURKEY: SEYRAN IRRIGATION PROJECT STAGE II (Loan 587/Credit 143-TU) Preface This is a performance audit of the Seyhan Irrigation Project Stage II in Turkey for which Loan 587-TU and Credit 143-TU were approved in February 1969 in the sum of US$12.0 million each (a total of US$24.0 million). The final disbursements in respect of the Loan and Credit were made in December 1977. The audit report consists of an audit memorandum prepared by the Operations Evaluation Department and a Project Completion Report (PCR) of April 1979. The PCR was prepared by the Europe, Middle East and North Africa Regional Office on the basis of a Completion Report prepared by the Government of Turkey and a country visit in October 1978. The audit memorandum is based on a review of the Appraisal Report (No. TO-614a) of February 4, 1969, the President's Report (P-672) of the same date, the Loan and Credit Agreements of February 10, 1969, and the PCR; rorrespondence with the Borrower and 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 July 1979. The mission held discussions with officials of the Ministries of Finance and Agriculture, DSI and Topraksu. Field trips to visit the project area and to meet some participating farmers were undertaken. Discussions were also held with the consultants in Israel. The information obtained during that mission was used to test the validity of the conclusions of the PCR and permitted further elaboration of the extension, land tenure, water management, environmental and rate of return issues, which are discussed in the audit memorandum. A copy of the draft report was sent to the Borrower for comments, but none have been received. The audit finds the PCR comprehensive and accurate with respect to the pioject's principal achievements and shortcomings. The points discussed by the audit have been selected because of their importance to this and to other Bank-assisted projects. The valuable assistance provided by the Government of Turkey, its officials and the farmers met during the preparation of this report is gratefully acknowledged.  - It - PROJECT PERFORMANCE AUDIT REPORT BASIC DATA SHEET TURKEY: SETAM IRRIGATION PROJECT STAGE II (Loan 587/Credit 143-TU) KEY PROJECT DATA Appraisal Actual or Item Expectation Current Eatimate Total Project Cost (US$million) 63.0 74.4 Overrun (2) - 18 Loan/Credit Amount (US$million) - 24.0 (12012) Disbursed - 24.0 Repaid to 09/30/79 .2 Outstanding to - 23.8 Date Physical Components Completion 12/74 12/79 Proportion Completed by Above Date (2) 100 80 Proportion of Tim Overrun (t) - 100 Economic Rate of Return (2) 16 15 OTHER PROJECT DATA Original Actual Item Plan Revisions Current Estimate First Mention in File - - 10/29/65 L Government's Application - - 01/27/66 Negotiations 11/27/67 03/13/68 06/24/68 12/06/68 02/03/69 Board Approval - - 02/25/69 Effectiveness Date 05/29/69 06/17/69 06/17/69 Closing Date 06/30/75 12/31/76 12/31/77 Borrower Government of Turkey Executing Agency State Hydraulic Works (DSI) Land and Water Development (Topraksu) Ministry of Agriculture Fiscal Tear of Borrower March I - February 28 MISSION DATA Month/ No. of No. of Date of Item Year Weeks Persons Staffueeks Report Identification - - . /2 Preparation - - - - /2 Preappraisal - - - - 11/04/65 Appraisal (Bank/IDA) 06/67 2 4 8 02/04/69 Supervision I " 10/69 2 2 4 11/19/69 Supervision II 9-10/70 2 2 4 10/09/70 Supervision III " 02/71 1 1 1 04/28/71 Supervision IV ' 09/71 1 2 2 11/19/71 Supervision V " 12/72 1 1 1 01/24/72 Supervision VI " 06/73 2 1 2 09/07/73 Supervision VII " 11/73 1 3 3 01/22/74 Supervision VIII " 06/74 1 3 3 09/18/74 Supervision IX " 07/75 2 3 6 08/26/75 Supervision I 11/75 1 1 1 12/11/75 Supervision XI 02/76 1 2 2 02/23/76 Supervision III " 07/76 1 4 4 07/12/76 Supervision XIII " 04/77 2 3 6 04/22/77 Supervision XIV " 10/77 2 3 6 10/31/77 Completion 1 06/78 1 2 2 07/11/77 Completion 2 10/78 1 1 1 04/12/79 Total 16 22 48 COUNTRY EXCHANGE RATES Name of Currency (Abbreviation) Turkish Lira (TL) Year: Appraisal Tear Average - 1969 Exchange Rate: US$1 * L 9.000 Intervening Years Average - 1970 US$1 * TL 11.500 - 1971 US$1 IL 14.917 - 1972 US$1 * TL 14.150 - 1973 US31 * TL 14.150 - 1974 US$1 * IL 13.927 - 1975 US$1 - TL 14.442 - 1976 US$1 TL 16.053 Completion Year Average - 1977 US$1 TIL 18.002 Completion Report Year Average - 1978 US$1 * TL 25.000 . Regarding Stage II specifically. Discussions on Seyhan Development as a whole date back to 1949 and on irrigation component (Stages 1, II, III together) to 11/7/61. /2 Project Identification dates to 1949 as part of multipurpose development project. Project preparation reviedwed during supervision of Seyhan I project.  - iii - TURKEY: SEYRAN IRRIGATION PROJECT STAGE II (Loan 587/Credit 143-TU) Disbursement Table US$000, cumulative Appraisal Actual Actual as % Period ending Estimate Disbursement of Estimated 12/31/1969 1.6 4.0 250 12/31/1970 10.2 6.0 58 12/31/1971 16.2 8.3 51 12/31/1972 19.5 11.7 60 12/31/1973 22.0 14.3 65 12/31/1974 24.0 18.7 78 06/30/1976 - 22.7 - 06/30/1977 24.0  - iv - PROJECT PERFORMANCE AUDIT REPORT TURKEY: SEYHAN IRRIGATION PROJECT STAGE II (Loan 587/Credit 143-TU) Highlights The project, which was the third Bank participation in the develop- ment of the Seyhan River and the second for its irrigation component, aimed to increase agricultural production and exploit the full potential of the region. Construction lasted almost twice as long as anticipated, due to the addition of 5,900 ha omitted under the preliminary design, renegotiation of all contracts in 1974 as a result of a sudden increase in inflation and a serious delay in on-farm development, due to the reluctance of farmers to participate. Project costs were US$74.4 million, compared with the appraisal estimate of US$63 million. The project's success in promoting agricultural development was mixed: in the short-term greater increases in crop yields were achieved than expected at appraisal; however, these yields were not sustained. The long-term goal of diversification has not been achieved. The economic rate of return is estimated at 15%, similar to the appraisal estimate of 16%. The following points may be of special interest: - this was the first Bank-assisted project to include an exten- sion component based on the Training and Visit (T&V) System; after initial success, many problems occurred (PPAM paras. 17-26; PCR paras. 6.03-6.07); - insufficient attention was given to land tenure problems and other sociological issues (PPAM paras. 27-31; PCR paras. 7.11-7.16); - in spite of only partial recovery of O&M costs, farmers are not fully utilizing the available irrigation water (PPAM para. 31; PCR para. 4.14); and - the project's environmental impact has increased the difficulty of pest control and contributed to the severity of the recent malaria epidemic (PPAM paras. 32 and 33; PCR paras. 7.08 and 7.10).  PROJECT PERFORMANCE AUDIT MEMORANDUM TURKEY: SEYHAN IRRIGATION PROJECT STAGE II (Loan 587/Credit 143-TU) I. SUMMARYV! 1. The Seyhan Irrigation Project Stage II is part of a multipurpose development scheme in the Adana Plain, one of the richest agricultural regions of Turkey. The project, the third Bank participation in this scheme and the second for its irrigation component, aimed to increase agricultural production and exploit the full potential of the region. It involved the construction of irrigation and drainage works on 39,250 ha and on-farm devel- opment on this and an additional 9,000 ha of the Stage I area, totalling 48,250 ha. It also provided for: construction of a concrete pipe factory; building of operation and maintenance facilities; procurement of equipment; undertaking of a study of cost recovery; implementation of an extension service program; and overseas training of project personnel. Responsibilities for implementation were under State Hydraulic Works (DSI) for irrigation and under Land and Water Resource Development Directorate (Topraksu) for on-farm development and subsurface drainage works. 2. The Project was substantially implemented without major modification by the Loan/Credit closing date of December 31, 1977, and 45,700 ha (net) have been equipped with irrigation and drainage facilities and on-farm development has been implemented on 44,000 ha with an additional 2,500 ha planned to be completed by 1979. Originally planned for completion in 1975, construction lasted almost twice as long as anticipated. The major causes of delay were the addition of 5,900 ha omitted under the preliminary design, renegotiation of all contracts in 1974 as a result of a sudden increase in inflation, and non-availability of land for on-farm development due to reluctance of farmers to lose a cotton crop under condition of rising cotton prices and yields. Delays are also due to a general slowdown of construction after 1974 when works were about 80% completed. Performance of the consultants, contractors and suppliers was satisfactory. Project reporting was timely and of high quality. Major issues of project implementation were: inadequate recovery of project costs; failure to submit audit reports; less than full expansion and phasing out of the extension service; and inadequate provision of agricultural credit. 3. Total project costs were US$74.4 million (TL 1,039 million) com- pared with the appraisal estimate of US$63.0 million (TL 567 million). Most of the increase is accounted for by increases in quantity of work and local prices. There was no provision made in cost estimates for inflation, a serious omission since the burden of the Government was considerably higher than anticipated. 1/ Adapted from the PCR. - 2 - 4. Measured against appraisal expectations, the project's success in promoting agricultural development was mixed. It met the short-term goal of increasing production by increasing yields of main crops, cotton and wheat, to levels far exceeding appraisal estimates. Whereas cotton yields were forecast to increase from 2.0 tons per ha to 2.4 tons over 15 years, by 1985, they reached 3.8 tons in 1971. However, these levels were not sustained and cotton has been affected by a variety of infestation and other problems. 5. The long-term impact of this project falls short of appraisal expectations. The key to exploiting full potential of the project area's natural resources lay in diversifying crop production. By full development in 1985, the appraisal mission had forecast a reduction in the cotton area from 83% to 48% of the project area, to be replaced by fruit, vegetable and alfalfa areas which would increase from 8% to 40%. Although by 1977 the cotton area had fallen to 52%, horticultural and forage crops still only accounted for 8% of the total area and land shifted out of cotton was put under wheat (without irrigation). Although this development was affected by special circumstances, severe infestation of cotton with white fly and increase in wheat price, farmer response to the extension service's diversification program has generally been disappointing. Most farmers explain that the risks presented by fluctuating prices and the lack of well organized marketing channels discourage them from producing fruits and vegtables. Other factors, such as land tenure, characterized by relatively large holdings (60% of the area is under holdings of over 20 ha, averaging 35 ha) may also have con- strained expansion. 6. In view of this experience and the general phasing down of extension activities with project completion, appraisal expectations with respect to diversification are unlikely to be met. Some progress can be expected because many new crops and their improved production methods have been introduced, and farmers, experiencing increasing problems with cotton, are looking for alternatives. 7. On the basis of revised assumptions taking account of these con- siderations, the rate of return is 15%. This is similar to the appraisal estimate of 16%, an accidental result since yield increases were raised, cropping intensity was reduced, output prices were raised and production costs were increased. 8. The extension service implemented under the project provided an interesting experience for both the Government and the Bank. A special organization was created under the Stage I project in 1967 and continued under this project. Main features of this organization which differed from the existing Government programs were: the use of foremen, village level workers selected among the villagers, backed by a hierarchy of more qualified staff; the coordination of all extension activities undertaken by different Government agencies under a single program implemented by a single "unified" service; and concentration of activities on extension rather than on data collection for the use of the Ministry. Success of the program in attaining - 3 - long-term agricultural development objectives was limited as explained above, but its positive achievements in helping farmers were significant, partic- ularly when measured against the existing national program. 9. Although the benefits of this type of organization were recognized, it was rather little publicized and made little direct institution-building impact on the Government's extension organization. The service on the whole did not enjoy full support of the Central Ministries which led, for example, to difficulties in procuring the equipment required, in achieving full cover- age of 100 villages, etc. Moreover, the Government had contemplated dis- .solving the organization altogether with completion of ' the Bank supported project. This is because as a special unit it was not part of any national program and the Government did not see justification for continuing to allo- cate scarce staff and other resources to an area which had already received priority and where the work of special extension was essentially finished because farmers had been successffully introduced to irrigated agriculture. In addition, as a special unit incorporating many agencies, there was no particular agency which was responsible for it or had interest in supporting it during project implementation or continuing it thereafter. This experience showed the weaknesses of an organization separated from the general Government structure and drawing key staff from different Ministries, and these lessons have influenced the designs of subsequent Bank-supported extension programs in Turkey, particuarly for the proposed Erzurum Rural Development project, intended to test a system more capable of wider replication. Thus, the project had an indirect institution-building impact. 10. Cost recovery was a major issue throughout project implementation, and involved three main points: water charges set for DSI's O&M were inade- quate to cover actual costs, and even these were not collected fully; DSI's capital recovery charges were low; and legislation did not permit recovery of Topraksu investment costs. Although a comprehensive study for the project area was undertaken under the project, and its recommendations including rais- ing of water charges and increasing penalties for delinquency were accepted by DSI, they were not adopted by the Government. However, in 1978, O&M charges were raised fivefold to levels covering O&M costs as Part of the new economic program. Nevertheless, with proposed levels of capital recovery charges and unless collection rates are improved and yearly adjustment made for infla- tion, recovery of project costs is likely to be negligible, a small fraction of annual O&M costs only. This level of subsidy is hard to justify for bene- ficiaries whose average incomes are well above national average. This expe- rience illustrates the difficulty of rationalizing through project lending sector policies which a Government does not feel it wishes or is able to change. 11. Related to cost recovery is the issue of the income distribution impact of this project and the project's justification in terms of sectoral priorities. The recent Agricultural Sector Review has concluded that alloca- tion of public investments had been too weighted in favor of new irrigation schemes with inadequate attention paid to completing old schemes, to farmer support services and to on-farm development, since there is a gross country- wide underutilization of existing irrigation facilities. The allocation of public investments has also been concentrated in the already advanced Mediter- ranean and Aegean areas, widening disparity in incomes between regions. Distribution of income within the project is also highly skewed. Although a sound project in itself, Bank support for a project of this sort is no longer consistent with its overall sectoral recommendations. 12. The project is also associated with an environmental controversy, since resurgence of malaria coincided with irrigation development. However, an in-depth study of the causal relationships between irrigation and malaria undertaken by the Bank in February 1979 concluded that the primary breeding places for mosquitos are in the unreclaimed and urban areas of Adana Plain rather than the project area. 13. The Bank's intervention at the design stage had significant impact on the scheduling of this project to take account of staff and other resources available to the Government, and to meet needs to strengthen extension and increase provisions for drainage. It also raised farm support services, coordination and cost recovery as major issues with Government. However, the assurances it sought with respect to credit and extension were not sufficient and had they been more fully integrated into the project and financed as such by the project as important components, their implementation might have been more successful. The long association the Bank has had with the Seyhan development scheme has made Bank intervention particularly meaningful. Most of the key issues were identified during implementation of Stage I. Bank supervision during Stage II was also based on a close working relationship with the Government. However, continuity was broken by a series of reorgani- zations starting in 1972 and supervisions became less constructive, particu- larly on agricultural extension, a program which was not well defined in the appraisal report. Also, supervision at this stage did not suggest ways of accelerating progress, which, in view of the general tendency of the Govern- ment to concentrate on new schemes, was an important issue. 14. The experience of this project clearly indicates the importance of providing for farm support services and maintenance facilities and looking beyond completion of the project to ensure that the capital investments financed are fully utilized. It also shows that designing institutions which can survive project completion and which can strengthen existing institutions is an important means to achieve this goal. Bank participation in successive phases of development schemes also improves the Bank's ability to identify key issues and contribute to institutional development. 15. Finally, the appraisal, which underestimated yield increases on the one hand and overestimated crop diversification on the other, reveals need for improvement in considering factors affecting farmer decision-making and in institutional changes. An analysis of these factors in relation to identify- ing constraints to diversification should provide useful recommendations for development of the project area and policy guidance for other parts of Turkey. - 5 - II. MAIN ISSUES 16. On the whole, the audit finds the PCR very comprehensive, detailed and act,urate. A few points, however, merit some further elaboration, namely: the Training and Visit (T&V) System of Extension,-Y the effect of land tenure on the on-farm development program, the project's water management and environmental impact and the economic rate of return. The Training and Visit (T&V) System of Extension 17. In October 1971, the Bank agriculturist who had been supervising the Seyhan Project wrote a letter to the Director of Israel's Agricultural Exten- sion Service, Mr. Daniel Benor, who had been employed by the project consul- tants since 1967 to advise and assist in setting up a new type of extension organization in the project area. The letter invited Mr. Benor to come to Bank headquarters to deliver a lecture to the Bank's agriculture projects staff on the new type of extension organization and achievements of the Seyhan Project. I/ Thus began a collaboration which, over the years, has borne fruit in 37 Bank-supported T&V Extension Systems in 15 different borrowing countries; a book and, most recently, a film. 18. In spite of this worldwide enthusiasm for the T&V System, the Borrower, until quite recently, viewed the Seyhan extension unit as a rather interesting ad hoc experiment, which succeeded in achieving its limited initial objective of raising cotton yields and could thus be dismantled and replaced once again by the normal country-wide "county-agent" type of service (which combined a modicum of extension advice and actual contact with farmers with administrative, regulatory and statistical duties, mainly performed in the office) This would be supplemented by a number of specialized services under different agencies.S' In 1975, however, the Bank obtained the agree- ment of the Government to the idea of applying the T&V System in other agri- culture and rural development projects in Turkey, particularly the new Corum Cankiri Rural Development Project and the ongoing Fresh Fruit and Vegetable Export Project. In 1977, the heads of the new T&V extension units in these two 1/ For a description of the T&V System, see Benor, D. and Harrison, J., "Agricultural Extension, the Training and Visit System", World Bank, May 1977. 2/ Although not really a new idea to Bank staff at the time (a number of earlier projects in Africa and Asia incorporated similar extension components), the Seyhan extension unit offered, at the apogee of its achievement in 1971, a particularly striking demonstration of.t.he success of this approach. 3/ Topraksu, livestock, plant, protection, etc. - 6 - projects attended a Bank-sponsored seminar/workshop on the T&V System in Manila, and by 1978, the T&V System had spread to a fourth project, the Ceyhan Aslantas Irrigation Project, and was being planned for a fifth, the new Erzurum Rural Development Project. As recently as 1978, although by then they agreed that a single, unified country-wide extension service should be organized, primarily to advise farmers and free of administrative duties, the Government still commented that, in projects financed from outside sources, "farmer training services are being recommended which are unsuitable to our structure incompatible with legal conventions and which result in contradic- tions..."I/ 19. The Government's concern was directed specifically, not towards the T&V System as such, but at its specific manifestation in the Seyhan Project, where it appeared as a kind of surgical implant, operating in parallel with the existing service and "borrowing" its field staff from other agencies such as Topraksu. The introduction and rapid and successful growth of this alien form, beginning in 1967, triggered a kind of immune reaction in the regular line agencies which together with a period of neglect of the extension com- ponent by Bank supervision missions following reorganization in 1972, nearly resulted in the "rejection" of the implanted organ. 20. At the same time, following the Arab-Israeli War in October 1973, the Israeli experts who had been advising the Seyhan extension unit left Turkey and were not replaced. In the following year, 1974, cotton yields in the Seyhan fell to their lowest level since the introduction of the T&V System in 1967, and in 1975, they fell even lower, to only 2.1 tons per ha. Since then, yields have recovered slightly but, with the exception of 1976 (see para. 22), still remain well below 3 tons, compared with the 1969-73 five-year average of 3.4 tons. Although, as noted in the PCR (para. 4.05), the immediate cause of the fall in yields was adverse weather and pest at- tacks, the withdrawal of the consultants also had an adverse effect on the ability of the Borrower's research, extension and plant protection services to deal effectively with these problems, and environmental complications have arisen-2/ 21. The problem thus seems to lie, not in the T&V System itself, but in the shortcomings of the existing research and plant protection services, which were not strengthened under the project,and the sharp drop in cotton prices in 1974/75, after farmers had already purchased large quantities of pesticides to combat the white fly attack. As a result, many farmers suffered losses, and their confidence in the extension service was shaken. 22. The following year (1976), Seyhan farmers planted only 36,000 ha of cotton, the smallest area since 1970. The white fly did not return, the market price soared from 7.5 TL/kg to 10.10, the yield reached 3.4 tons, and 1/ Memo from Agricultural Extension Committee to the Office of the Minister, February 28, 1978. 2/ See para. 33 below. - 7 - those farmers who had been able to afford the risk of staying on in cotton rather than falling back on wheat made huge profits. Following their example, many poorer farmers again planted cotton instead of wheat a year later, in 1977, when an all-time record area of 57,000 ha was sown. This time, not white fly, but green bollworm and red spider attacked the crop, again forcing the farmers to purchase more pesticides. To make matters worse, the cotton price fell again too, resulting in widespread losses and a sharp reduction in the area planted in 1978, the most recent year for which data are available. 23. Thus, the rapid but stable growth experienced by the Seyhan cotton economy as a result of the project has been replaced since 1973 by a series of very unstable short-cycle fluctuations in which many smaller farmers have decided to cut their losses and either go back to less risky wheat production or out of farming altogether. Another victim of this instability has been the extension program itself, which has lost most of its momentum and initiative with farmers, having no further quick and easy improvements to offer them in cotton growing, and is currently in the position of responding to, and trying to cope with, a series of emergencies. Lagging salaries, lack of transport and shortages of fuel and agricultural chemicals resulting from Turkey's rapid inflation and foreign currency crisis have further aggravated the situation. Plans for dairy development and crop diversification have been postponed, and the Seyhan extension unit is reduced to trying to prevent further losses of cotton production, while finding increasing difficulty in holding on to its staff, budget, vehicles and equipment. 24. Although the Bank's first successful prototype for the T&V System has now come upon hard times, the fault does not appear to be in the system, but in its economic, social, institutional and natural setting in this parti- cular project. Considering the constraints within which it operated, the T&V System itself was successful, and the failure of the Seyhan extension unit to achieve its long-term objectives should not detract from that success. 25. Close attention should be paid, however, to other Bank projects incorporating the T&V System, in Turkey and elsewhere, to see that they do not develop similar problems. The major reason why the system did not take root in Turkey was its lack of integration into existing structures. Supervision reports already indicate that some projects in other countries too are having difficulties getting underway or maintaining their momentum in the face of lack of working-level bureaucratic support and a dearth of new innovations and research to follow up the relatively simple, easy improvements on which their initial impact was based. 26. Although it may be premature to generalize on the basis of a single project, a possible lesson to be derived from this experience is that, in introducing the T&V System in countries with an already established rural bureau6racy, particular care needs to be taken beforehand to obtain the wholehearted support at all levels of the existing agencies. Another lesson is the' apparent relative vulnerability of this highly focussed, full-time extension approach, based on a few simple improvements to existing crops, to a drying up of further new technical innovations or diversification possi- bilities, or to a lack of relevant research to cope with newly emerging pest problems, etc. Land Tenure and On-Farm Development 27. The PCR describes, in para. 3.07 and 7.13, some of the diffi- culties encountered by Topraksu in carrying out on-farm development in the project area and suggests a number of possible explanations for them (high cotton prices and fragmentation) The audit mission learned that one of the main factors impeding the work, particularly the installation of sub-surface drains, is the land tenure system in the area, which, as the PCR hypothesized in paras. 7.15 and 7.16, may also be the main constraint holding back crop diversification. 28. Quite apart from the reluctance of nearly all the farmers to give up one year's cotton crop to permit land levelling of their fields, the tenant farmers in particular, currently occupying 15-20% of the total area VA and representing an even larger percentage of the farmers (including those farmers renting some fields in addition to the land they own), have presented a very serious obstacle to Topraksu's drainage program. Fields are rented for typically no more than one or two years at a time, and, although the same tenant may occupy a particular field much longer if his lease is renewed, there is no guarantee that he will be able to do so. These rented fields are scattered throughout the area, and most of the planned sub-surface drains must cross several of them. The drains are installed in the summer, using heavy machines requiring a wide right-of-way which must be kept clear of crops for the entire season. Unlike land levelling which can proceed on individual farms, if one tenant along a particular drain alignment objects, the whole operation is delayed, and, for a farmer who may have to move to a different rented field the following year, there is little incentive to grant Topraksu the necessary permission. 29. The Bank appraisal missions for both Stage I and Stage II, as was the common practice at the time, completely ignored such sociological pro- blems, even making the erroneous statement that "tenant farming is rare and found almost exclusively on land growing rice".V However, most of the implementation problems plaguing the project are in fact attributable to sociological,!/ rather than technical or economic factors. The lesson is obvious and has been repeated in many project audits. Although the tendency on the part of Bank staff to neglect sociological issues is a persistent one, 1/ Topraksu estimate. 2/ Repeated verbatim in both reports. (See appraisal reports Stage I, para. 29, and Stage II, para. 3.11). A survey of the land tenure situation in 1963, on the other hand, indicated that 53% of the land was owned by absentee landlords and that one third of the farmers were landless. (PCR, para. 7.04). 3/ Including the institutional problems of the extension service, described above. - 9 - much greater attention has been given to these questions during the last two years, and a new Operational Manual Statement on this subject has been cleared by senior management, setting forth procedures to be followed during prepara- tion and appraisal which should help ensure that sociological issues are given due attention. Water Management and Environmental Impact 30. While the PCR contains very little information on water management, the audit learned from Bank files and discussions with the Borrower that the irrigation system is operated efficiently. Conveyance and operating losses account for only about 15% of the water diverted at the project headworks (apraisal estimate, 25%). Field efficiency, however, is estimated at about 60%V, which is low, considering that most of the area is now levelled, and is less than the 70% estimated at appraisal. The overall efficiency of 51% is roughly the same as the 52.5% estimated at appraisal. 31. The actual total diversion of water for irrigation also is about 30-401/ less than estimated at appraisal, due to the serious underutiliza- tion of the irrigation system by farmers. As indicated in the PCR, neither the Government nor the Bank fully understands the reasons for this underuti- lization, which is all the more surprising in view of the fact that farmers in the project area are paying less than 20% of the cost of providing the water. Cost recovery is clearly not the explanation, in any case, since in other irrigated areas in Turkey, where farmers pay even less than they do in Seyhan, the rate of utilization is lower, not higher. Here again, the explanation may be sociological, rather than technical or economic. 32. Some of the undeveloped areas adjacent to the project experience drainage problems, particularly the lower lying areas not yet equipped with irrigation canals (Stage III). Also, the drainage canals within the project area itself are poorly maintained. Apart from the problems of water logging and salinity this situation creates, it also contributed to the recent resur- gence of malaria in the region. A Bank study concluded that the irrigation projects are not solely, nor even primarily, responsible for the problem.3V However, the poor upkeep of the drainage canals, abetted by leached fertil- izer, resulted in the canals becoming choked with vegetation and therefore active foci of mosquito breeding; also, the construction of service roads and flood-control levee pock-marked the Plain with borrow pits, further adding to the extent of man-made breeding foci. Even so, there is a higher numerical incidence of malaria cases in the towns than rural areas, and if allowance is made for the large influx of migrants in the irrigated areas, these can be 1/ Topraksu estimate. There is a surplus of water available for the small presently irrigated area, and farmers can obtain as much as they want; on the other hand, they pay only about one fifth of the cost of the water, so there is little incentive to use it efficiently. 2/ The percentage varies from year to year. In 1978, it was 37%. 3/ Back-to-office Report of March 27, 1979. - 10 - seen to be less malarious than the undeveloped Phase III areas._1 Fortu- nately, the epidemic now appears to be under control, with far fewer cases reported in 1979 than in the two previous years. Still, with many countries presently abandoning or neglecting their malaria control programs, as did Turkey, in spite of the rapid world-wide spread of irrigation, the experience of this project offers an important lesson for the Bank and its Borrowers not to relax their concern with problems of environmental impact. 33. Another environmental problem arising indirectly from the project has been the excessive, uncoordinated and indiscriminate use of insecticides in the area in recent years, which, by eliminating biological controls (e.g. natural predators) and building up tolerance or resistance to the chemicals used, has made much more difficult not only the control of the malaria vector but also the campaigns against the recent attacks by crop pests such as the white fly, green bollworm and red spider. Part of the blame for this problem may be attributable to the weakening of the project's extension service, noted above.!/ The Economic Rate of Return 34. The PCR recalculated the project's rate of return at 15%, sub- stantially lower than the Government's estimate of 21%. As explained in para. 5.05 of the PCR, most of the difference is due to the assumed future cotton yields of 3.5 tons/ha, as against the Government's projection of 4. The Government criticized the PCR's estimate as being too pessimistic, pointing out that good farmers in the area already obtain more than 4 tons/ha. To attain a project wide average yield of 4 tons, however, the best farmers would have to reach much higher yields. In view of the present state of the exten- sion unit and plant protection service and the short supply of fuel and agro-chemicals, however, the audit finds the PCR's estimate of yield to be realistic and supports the 15% rate of return, still quite acceptable for a project of this kind. I/ "Irrigation, Anophelism and Malaria in Adana, Turkey", M.E.C. Giglioli. 2/ See para. 20. - 11 - TURKEY COMPLETION REPORT SEYHAN IRRIGATION PROJECT - STAGE II I. BACKGROUND 1.01 The Second Seyhan Irrigation Project is part of a multipurpose project, the Lower Seyhan Project, consisting of flood control, power and irrigation designed to develop the agricultural and related industrial poten- tial of the Adana Plain. This project is concerned with the second stage of irrigation development, covering about 46,000 ha out of total irrigable area of 170,000 ha. 1.02 The Adana Plain is located in south-eastern Turkey, bordering the Mediterranean Sea to the south, and Adana, the fourth largest city in Turkey is situated on the Seyhan River in its north-eastern section (see map). It covers an area of 220,000 ha well endowed with soil, water and climatic resources. It is one of the major cotton producing areas of Turkey and both agriculture and industry are advanced relative to the rest of the country. Conditions are favorable for production of other industrial, horticultural and forage crops and the region's potential for development of modern agriculture, agroindustries and livestock is considerable. 1.03 Agriculture is an important sector in Turkey, accounting for over half of the country's population, employment and export earnings. However, its share of GDP has declined from 38% in 1963 to 25% in 1975 and grew at 2.5% during the period compared with 6.5% for the economy as a whole. The role of agriculture in the Government's development strategy has been to meet domestic demand for food, to provide raw materials for industry and to earn foreign exchange required for industrialization. With unreliable and scarce rain- fall limiting growth in much of the country, best prospects for growth are considered to be in livestock and in coastal areas with richer soils, possi- bilities for irrigation and cultivation of higher value crops. Thus in its strategy for agricultural development, the Government has relied heavily on irrigation and the subsector has accounted for over one half of total public investment in agriculture under the First, Second and Third Five-Year Plans (1963-1975). 1.04 The performance of agriculture during the last 15 years has been disappointing. The growth rate of 2.5% over the period 1963-1975 falls short of the planned target of about 4%, and areas under public irrigation scheme in 1975, about 1.5 million hectares or 5% of cropland, contributed only 13% to total crop production. Moreover, only about two thirds of this area was estimated to be irrigated and about one third is believed to be cropped to its full development potential. This extreme under-utilization has long been of concern to the Bank 1/ since the 1960's and is attributed to the lack of 1/ The term Bank refers to IBRD and IDA. - 12 - attention paid to on-farm development and to supporting services. The recent Agricultural Sector Survey (1978) identified the following as major sectoral issues: need to give priority in resource allocation to projects with shorter gestation periods and high returns implying rescheduling major new irrigation projects, emphasizing completion and rehabilitation of existing works and improving farmer support services; need to revise price and subsidies po- licies, including irrigation water charges, to improve their impact on produc- tion and income distribution; and need to strengthen the Government's capacity to formulate policies, plan and implement a coherent program. 1.05 Bank lending in agriculture to date has totalled US$378 million for 12 projects and its orientation has evolved over the years. From concentrat- ing on capital intensive investments in the 1950's and 1960's, increasing attention was paid to farm support services starting with its involvement with the extension service for the Seyhan Irrigation Project. These elements now provide the central focus of some of our projects. Emphasis has also shifted away from projects with highest growth potential in already advanced areas, such as the six projects approved in the 1960's which are all entirely or partly located in the Adana Plain or a contiguous area. The five more recent projects have attempted to benefit the less developed regions including rain- fed areas. 1.06 The Seyhan Irrigation Project Stage II was the third project financed in the agricultural sector, the preceding two being the Seyhan Multipurpose Project (Loan 63-TU) and the Seyhan Irrigation Project Stage I (Credit 34-TU). The Adana Plain and the contiguous coastal regions have not only been the focus of Bank operations through these three projects but through three subsequent projects; the Ceyhan Aslantas Project (Loan 883/Credit 360-TU) which aims to develop water resources of the Ceyhan River which flows to the east of the Adana Plain, the Fruit and Vegetable Export Project (Loan 762/ Credit 257-TU) intended to develop citrus production and marketing facilities for fruit and vegetable export in Adana, Icel and Hatay provinces, and the Intensive Dairy Project (Credit 236-TU) which aimed to assist farmers develop intensive commercial dairy enterprises in four main milk consuming areas of Turkey including Adana. II. FORMULATION Origin 2.01 The Second Seyhan Irrigation Project was originally formulated in the 1940's as a part of the multipurpose project but was substantially re- formulated by 1969 when the loan/credit was approved. The Bank played an advisory role in this process through its involvement with the preceding phases of the multipurpose project and had a significant impact on strength- ening such aspects of the project as drainage, on-farm development, farmer extension and on rescheduling construction into three phases. - 13 - 2.02 The Bank's association with the Seyhan project dates back to 1949 when the Turkish Governmentfirst requested Bank support for the multipurpose project as a whole. With a loan in 1952 (63-TU), and a credit (34-TU) in 1963, the Bank helped finance the dam, power and transmission system and completion of the flood control works. By 1956, irrigation development which began in 1940, only covered 1,160 ha and serious drainage problems became apparent. Rather than proceed, the Bank suggested that these and other technical problems be studied and a plan of irrigation and drainage be pre- pared. Such a study was undertaken by a firm of consultants engaged by the State Hydraulic Works and Natural Resources, Ministry of Energy (DSI) and Land and Water Resource Development Directorate of the Ministry of Village Affairs (Topraksu) and was completed in 1961. 2.03 While this 1961 plan included the entire 170,000 ha which could be irrigated, it was divided into three stages which could be implemented sequen- tially, starting with areas of highest priority. This was done at the request of the Bank which was concerned that manpower and financial resources of the Government should not be spread over too large a project. In 1963, the Bank agreed to assist in financing Stage I (53,000 ha) with a credit (38-TU) of US$20 million. At this time, Stage II (43,000 ha) was also tentatively appraised and the Bank indicated willingness to consider financing, and included design of works in the loan. Stage III (74,000 ha) covers the lower lying area with more difficult drainage problems associated with higher costs. Feasibility studies for this stage were included in the second project. The Bank decided subsequently not to finance this stage since Bank funds should be used to develop areas with greater potential. The Government has proceeded to implement this stage using their own resources. Preparation, Appraisal and Agreement 2.04 DSI, assisted by their engineering consultants (Tahal-ECI Inc., con- sortium Israel-USA) completed preparation of Stage II in 1965. The Government requested Bank financing and intended to begin construction in 1966. However, the Bank advised the Covernment to delay construction until further progress with Stage I was made and did not appraise Stage II until June 1967. 2.05 The Bank's main concerns with the progress of Stage I were: (i) lack of coordination among various agencies; (ii) inadequate provisions made for agricultural extension; (iii) insufficient provision of production credit to farmers; (iv) poor progress with on-farm works, (levelling, surface drainage, roads) and need to give responsibility for this to Topraksu since the farmers were unwilling to invest in these improvement works themselves, using credit provided by the Agricultural Bank; and (v) inadequate collection of water charges. - 14 - Agreements having been reached over Government actions to resolve these issues, the project was appraised in June 1967. 2.06 Negotiations began with technical discussions in November 1967 but were not completed until January 1969 due to uncertainties over availability of IDA funds. Major issues discussed were: (a) international competitive bidding (ICB) on civil works contracts; (b) importation of machinery and equipment which compete with those assembled in Turkey; (c) full audit reports to be submitted; and (d) the delay by the Government in providing estimates of agricultural credit requirements. Agreement was reached on these points by September 1968 and IDA replenishment position was clarified by January 1969. A loan of US$12 million and a credit of US$12 million were approved by the Board on February 25, 1969, and were signed on February 28, 1969. Project Description 2.07 The project as agreed at negotiations comprised: (a) the construction of an irrigation and drainage system on about 39,250 ha and on-farm development on about 48,250 ha, including about 9,000 ha provided with irrigation and drainage facilities in Stage I; (b) the construction of a concrete pipe factory; (c) the building of operation and maintenance facilities; (d) the procurement of equipment, materials and supplies; (e) the undertaking of a study on water charges; (f) the implementation of an extension service program; and (g) the agricultural training overseas of personnel of the Borrower. 2.08 During negotiations, the Government also agreed to undertake the following: (a) a study of road requirements and construct most roads not included as part of the project; - 15 - (b) a study of agricultural credit requirements of the Seyhan project area and prepare specific recommenda- tions for amounts to be made available by the Agricul- tural Bank of Turkey (TCZB); and (c) to employ engineering consultants for DSI and Topraksu to assist in project implementation. 2.09 The Government also assured the Bank that: (a) the project coordination organization developed for Stage I would continue to function under Stage II; (b) import licenses would be issued promptly; (c) water charges sufficient to cover operation and maintenance costs and a reasonable amount of capital costs would be levied; and (d) the list of prequalified bidders would be sub- mitted to the Bank for approval, together with the recommendations of its consultants. Project Objectives 2.10 The objectives of the project were to raise agricultural production and in the long term develop the agricultural and industrial potential of the Adana Plain. Irrigation, drainage, on-farm development and introduction of improved farming methods were expected to increase yields and diversify crop production. It was assumed that farmers would: Ci) efficiently utilize water provided; (ii) maintain on-farm facilities; (iii) improve cultivation prac- tices; and (iv) change the cropping pattern by introducing alfalfa and other legumes, vegetables and citrus, by reducing the area under cotton from about 81% in 1969 to 48% at full development in 1985, and increase cropping inten- sity. At least 20,000 ha (42% total area) were expected to be double cropped since the main constraints to this were believed to be poor drainage, land levelling, all-weather roads and effective extension, which were to be removed by the project. 2.11 Based on its experience with Stage I, the Bank was particularly concerned with the need to ensure effective utilization of the drainage and irrigation facilities once they had been constructed. Under Stage I, the extension services provided under existing Government programs were thought insufficiently effective and a special program was formulated with the help of a specialist consultant, Mr. Benor, engaged by the Tahal-ECI consortium. This program was to be continued and enlarged under Stage II. It differed from the existing Government programs in-the following respects: it offered more intensive assistance to farmers; activities of different agencies were coordinated under a single program; it used village level workers, "foremen", selected from among villagers who were trained at regular sessions and made regular visits to farmers, and who were backed up by a hierarchy of special- ists (see Chart); it enjoyed greater autonomy which enabled it to concentrate - 16 - on farmer training activities rather than on data collection for use in the central Ministry. Organization 2.12 The major irrigation and drainage works were to be implemented by DSI and on-farm development by Topraksu. The Extension Service was a special organization established for Seyhan Irrigation Project I with a director from the Ministry of Agriculture assisted by an expatriate advisor and incorporated technical staff from Plant Protection Department, Cotton Research Institute and Extension Service of the Ministry of Agriculture and Topraksu's irrigation foremen. The Agricultural Bank of Turkey had responsibility for providing and financing production and development loans. The activities of these various agencies were coordinated by the Central Coordinating Committee in Ankara chaired by the SPO and by a regional committee (see Organization Chart) estab- lished for Seyhan I in 1966. All three implementing units, DSI, Topraksu and Extension Service, were assisted by consultants, Tahal-ECI consortium, who had also been engaged under Seyhan I. III. IMPLEMENTATION Effectiveness and Start-Up 3.01 Project implementation started expeditiously, particularly since this was continuation of work from Stage I for all agencies involved. Some specifications were prepared and tendering for equipment and civil works were begun prior to Board presentation, during the prolonged delay due to uncer- tainties over IDA allocations for Turkey. The one special condition of effec- tiveness, that DSI and Topraksu engage consultants, was met in time for the effectiveness date, but this date had to be extended by one month because the first legal opinion submitted was found inadequate. Physical Implementation 3.02 All components of the project were implemented without major revi- sions in concept, size or design. The following table shows major categories of works completed or planned compared with appraisal estimates. - 17 - (1) (2) (3) Appraisal Completed Planned for As % Estimate at Closing Completion of (1) S(ha) --------------- --%--- DSI Works Irrigation & Drainage Left Bank (Tarsus Plain, TP) 20,150 26,150 26,150 Right Bank (Yuregit Plain, YP) 19,100 19,550 19,550 Total 39,250 45,700 45,700 116 Topraksu Works On-farn Development Land levelling 45,250 44,000 46,150 96 (Planing or light levelling) (52%) (31%) (Medium levelling) (37%) (52%) (Heavy levelling) (11%) (17%) Tile Drainage 20,900 22,446 26,370 126 -------------(km)------------- --%-- Surface drainage 950 763 /1 763 80 Field roads 240 270 274 113 /1 There are discrepancies between Topraksu's data presented in the draft completion report shown above and those in the quarterly progress reports which show only 518 km out of the planned 763 km to have been completed. Since supervision mission reports are consistent with the draft completion figures, they have been accepted. 3.03 Together with Stage I, approximately 104,000 ha of the Lower Seyhan Project area has been equipped with irrigation. This is compared with the plans for original feasibility studies. Seyhan Project - Net Irrigable Area (ha) 1961 Feasibility 1965 Study Study 1978 Stage I 53,000 62,000 59,000 Stage II 34,000 39,250 46,000 Subtotal 87,000 101,250 104,000 Stage III 83,000 68,750 66,000 Total 170,000 170,000 170,000 - 18 - The difference of about 6,450 ha in the total area developed with irrigation and drainage is largely accounted for by the addition of 5,900 ha in the 2YP area which had been omitted under the preliminary design. This area was served by the main drainage canals constructed under the project. Con- sequently plans were revised to extend two canals, subsidiary drainage and canelette distribution systems in order to serve this additional area within the main drainage system. The reasons why this area was not included in the beginning could not be traced. 3.04 Minor modifications were made in the equipment list, involving a change in the loan agreement (side letter) three times, the most significant among these involving O&M equipment for DSI. An inventory study done in 1970 showed a backlog of drainage maintenance work and larger machinery requirement than was anticipated for both Stages I and II areas. DSI proposed and the Bank agreed that the project would finance a larger number of the more expen- sive items required by their machinery pool while DSI would purchase the remainder. This had the advantage that the DSI machinery pool would be standardized, facilitating repair and spare parts facilities. A list of equipment procured for DSI and Topraksu is in Annex I. 3.05 The extension service did not reach full coverage of 100 villages in Stage I and II areas as planned but remained at 70 villages from 1971 until 1976 when it was increased to 86 (Annex 1). Not all the equipment required for this service was procured and only US$106,000 were spent out of US$250,000 allocated. Although the extension service repeatedly made requests for staff to increase coverage, the Ministries of Agriculture and Village Affairs did not act. In fact, in 1976 they stopped replacing staff who resigned. The reason given for this was that they believed that the task of extension was completed because farmers had successfully been introduced to irrigation practices and therefore Seyhan did not deserve special priority when other areas badly needed staff, nor was it justified to continue a special program. After considerable discussion with the Bank, the Government has now decided to maintain the service at full strength. Regarding equipment, agreement of the central ministries was difficult to obtain since they did not always understand that the equipment was to serve project purposes only. The Minis- try finally made proposals for procuring remaining equipment, the Bank's consideration was delayed and it was found that all remaining project funds were committed. 3.06 Consultant services were retained as anticipated and extended by one year totalling 693 man-months compared with the appraisal estimate of 584 man- months (Annex I). The overseas training program for project staff was first prepared in 1970 but could not be implemented because Government budgeting pro- dures did not allow financing of the expenditures. Prolonged discussions with the Ministry of Finance ensued but it was not until 1976 that this constraint was finally removed. A program prepared with Bank assistance in conjunction with two other irrigation projects (Ceyhan Aslantas and Irrigation Rehabilita- tion and Completion) was implemented in 1977. Three studies were included in the project namely cost recovery, roads requirements and credit. While the first two studies were completed, the latter was not undertaken since a more comprehensive study was included in the Fruit and Vegetable Export Project in 1971. However, the Government and TCZB did not undertake this study either. - 19 - Recommendations of the water charge and roads study were accepted by DSI but were not fully implemented due to political and financing problems (Section VI). Implementation Schedule 3.07 Construction of irrigation and drainage facilities was substantially completed at the end of 1977, four years after the appraisal estimate of end 1973, and on-farm development is expected to be completed by 1979/80, about 5-6 years beyond appraisal estimate of end 1974. In other words, construction period was twice as long as was anticipated. Annex I Table 3 shows the yearly progress of implementation. 3.08 Much of this delay occurred in the latter years when construction was well underway or nearing completion. Government attributes the delay to the following. For DSI works: (i) Extension of project area by 5,900 ha. The contract for this extension was the last of seven to be awarded in 1972 and was the last to be completed in early 1978; (ii) Withdrawal of one contract up for tender. An error was discovered in the specifications before bids were received and about a year was lost in retendering for ICB; (iii) Liquidation of contract in 1974. By decree issued in April 1974, the Government allowed existing civil works contracts to be terminated without penalty because of the unexpected increase in inflation. As a result, five of the existing six contracts were terminated. Two of these were advanced and were completed by DSI under force account while one was renegotiated with a price increase. The remaining two were rebid locally; and (iv) Existing works. Difficulties were encountered in areas where there was already an unlined canal network for irriga- tion and cultivation of rice (1 TP and 5 YP areas); For Topraksu: (i) Increase in cotton prices. Increase in the support price per kilogram of cotton from TL 3.75 in 1972 to TL 6.00 in 1973 and TL 8.00 in 1974 made farmers reluctant to allow works to be done during the cotton growing season. In 1973, when land levelling was expected to increase from 5,000 ha in preceding years to 15,000 ha, actual implementation declined from about 10,500 ha in 1972 to 6,000 ha in 1973 and then further to 2,600 ha'in 1974. Substantial comple- tion delays became unavoidable at this point and the project was rescheduled to end in 1977; and - 20 - (ii) Land Fragmentation. In a situation where consent of farmers is difficult to obtain, existence of scattered and irregular farm plots aggravated the difficulty of organiz- ing operations rationally. While these factors were important, the completion mission believes that they do not account for the delays entirely. For DSI works, at the time that contracts were renegotiated in 1974, all construction was estimated to be completed before end of 1975, two years before the actual completion date for some areas. The extent of rice cultivation and unlined irrigation network was also negligible and were in areas which were completed on schedule, by early 1974. For Topraksu, the working season curtailed by weather and high cotton prices was certainly a constraint, however, implementation did not accelerate in 1976 when the wheat area increased dramatically. Since special factors other than those mentioned by the Government cannot be identified, delays can only be attributed to management. While progress was rapid and on schedule during the first years, there was a general slow-down and accumulated delays during the last years, when completion of works substantially con- structed dragged on. By the end of 1975, the original completion date, all but two of the seven contracts were finished and Topraksu's land levelling was 80% complete. Yet a further 3 years for DSI and 4-5 years for Topraksu works were required before actual completion. Thus appraisal expectations were not unrealistic, and the completion mission feels that Bank supervision could have done more to accelerate completion. Procurement 3.09 Procurement was a major issue throughout negotiations; the Govern- ment resisted ICB for civil works contracts and import of equipment which competed with those assembled in Turkey. The Government had argued at that time (1968) that ICB procedures were very lengthy when complying with the Borrower's bureaucratic requirements and that local contractors were very competitive and most likely to win. Therefore, ICB was an unnecessarily time consuming procedure unlikely to result in cost savings. This indeed turned out to be the case; contrary to the appraisal estimate that about 40% of contracts would be awarded to foreign bidders, all contracts were won by local firms. Bank insistence on ICB may have contributed to the overall delay in implementation in one case. In November 1970, an error was found in the tender documents just prior to the bid submission date and had to be withdrawn. The Government proposed that ICB be waived and local bidding procedures be followed. They argued that calling rebids would cause a year's delay, that one of the two pre-qualified local contractors was most likely to be the low bidder and therefore ICB would be pointless, and would cause a delay of one year. The Bank disagreed and counterproposed that the prequalified bidders should be advised of the change in specifications and asked to rebid. However, the Government in turn rejected this proposal, and ICB procedures had to be reinitiated. An award was made one year later, in November 1971, to a local contractor whose bid, as predicted was 40% below the engineers' estimates. - 21 - Performance of the Borrower, Consultants, Contractors and Suppliers 3.10 The main issues regarding the Government's performance are: failure to submit audit reports (para. 6.20); failure to impose and collect adequate water chargers (para. 6.09-6.18); failure to undertake the credit study (para. 3.06); and the accumulated delay in completion of the work noted above (para. 3.08). The pattern of progress is symptomatic of the general tendency in irrigation development in Turkey to concentrate on new projects, letting com- pletion of old ones lapse. General Government policies which were noted as hindering project implementation included the low salaries scales compared to other employers which led to difficulties in retaining qualified engineers, and the policy to lay off machinery operators during the slack season which resulted in difficulties in recruiting well qualified workers. The Govern- ment's progress reports were of high quality and were submitted regularly. The extension service report however deteriorated in quality considerably from 1974, coinciding with the departure of consultants. 3.11 Quality of construction was generally satisfactory with the excep- tion of some works contracted by Topraksu. This latter is attributed to shortage of supervision personnel, a problem noted above. Services of equip- ment suppliers and consultants were satisfactory to all parties concerned. The consultant's contribution to the extension service was of special signifi- cance (para. 6.03-6.07). Costs 3.12 Final project costs totalled US$74.4 million exceeding appraisal forecasts of US$63.0 million by 18%. However, expressed in the domestic currency, cost overrun was 83%, with actual costs totalling TL 1,039 million compared with the appraisal estimate of TL 567 million. This discrepancy reflects high local inflation relative to international levels and the cor- responding successive devaluation of the Turkish Lira from US$1 equalling TL 9 in 1969 to TL 25 in 1978. 3.13 A comparison of final costs with appraisal estimates is in Annex II, Table I and is summarized below. - 22 - Comparison of Total Project Costs ------TL million-------- ---------US$ million-------- Appraisal Final % Appraisal Final % Estimate Cost Difference Estimate Cost Difference DSI Irrigation & Drainage 335.7 631.9 +88 37.3 45.3 +10 Contingencies @ 20% (45.9) - (5.1) - Equipment & Services 24.3 48.6 +100 2.7 3.4 +26 Contingencies - - - Subtotal 360.0 680.5 +89 40.0 48.7 +22 TOPRAKSU On-farm development 106.2 216.8 - +104 11.8 14.2 +20 Contingencies (16.2) - (1.8) - Equipment & Machinery 63.0 89.4 +42 7.0 6.3 -10 Contingencies - - Subtotal 169.2 306.2 81 18.8 20.5 +9 Consultants & Training 19.8 34.3 +73 2.2 3.2 +46 Interest during Construction 18.0 18.0 0 2.0 2.0 0 Total 567.0 1_03. +83 63.0 74.4 +18 / Taking account of works to be completed in 1978,79. 3.14 DSI's civil works, by far the largest component of the project, accounts for most of the increase in total costs in terms of absolute amounts. DSI's analysis shows that cost increase of TL 296.2 million (88%) can be accounted for as follows: Analysis of Cost Increases Item TL Millions % Total Addition of 5,900 ha 58.0 20 Other changes in quantities 97.8 33 Price escalation 48.8 16 Inclusion of pipe factory costs 23.7 8 Discrepancy between original contract price and estimate 67.9 23 Total 296.2 100 Changes in quantities, including the extension of the project are shown to account for the majority of cost increases, however, the effect of unforeseen inflation is understated. While inflation was negligible at the time of appraisal, it increased gradually and reached 27% in 1974, a year of world wide rise in inflation. In this year, the Government issued a decree allowing contracts to be terminated without penalty. Completion of work under con- tracts thus terminated totalled TL 103.3 or 35% of the total cost overrun. - 23 - This is included in the discrepancy between the original contract price and the appraisal estimate (TL 49.4 million) and in the cost of adding 5,900 ha (TL 33.9 million). This latter amount is large because it was the last contract to be signed, least advanced and therefore most affected by infla- tion. 3.15 Increase in Topraksu's on-farm development costs were also signifi- cant and is attributed by the Government largely to heavier land levelling being required than was anticipated and to the increase in area to be put under tile drainage from 20,900 ha to 22,400 ha. Composition of Land Levelling Works Appraisal Estimate Actual Light levelling or planing 52% 31% Medium levelling 37% 52% Heavy levelling 11% 17% Total 100% 100% Total area 48,250 ha 46,250 ha 3.16 Although much of the cost increase can be attributed to changes in project design, price escalation due to unanticipated inflation and implementa- tion delays was also an important factor. The appraisal cost estimate made no provisions for price contingency since inflation both worldwide and in Turkey was relatively low (below 2% for internationally traded manufactured goods and below 10% for retail prices in Turkey) and contracts were planned to be awarded within one year of loan effectiveness specifically without an escala- tion clause. This experience illustrates the difficulty in projecting inflation and the weakness of relying too heavily on the economic conditions prevailing at the time of appraisal. Disbursements 3.17 Disbursements lagged behind appraisal estimates considerably as shown below: - 24 - Disbursement Schedule Appraisal Actual as % Calendar Estimate Actual of Estimate Year -----(US$ million)-- -- ---%----- 1969 1.6 4.0 25 1970 10.2 5.4 50 1971 16.2 8.3 51 1972 19.3 11.7 61 1973 22.0 14.3 65 1974 24.0 16.8 /1 70 1975 - 20.6 /1 86 1976 - 22.7 /1 85 1977 - 24.0 100 /1 June 30, data for end of year not retrievable. This discrepancy arises from delays not only in construction but also in the Government requesting reimbursement. 3.18 The loan/credit proceeds were reallocated among different categories four times during implementation, first due to changes in O&M equipment for DSI and secondly to increase financing of civil works. However, final dis- bursements did not differ markedly from appraisal allocations as shown below: Amount Appraisal Actual Allocation Disbursements Actual as % Category 2.28.69 12.3.77 of Estimate --- (Us$ Millions)-------------- I. DSI Equipment, materials and supplies 2.50 3.28 132 II. TOPRAKSU Equipment, materials and supplies 6.30 6.39 101 III. Civil Works 9.10 10.10 111 IV. On-farm Development 1.20 1.29 108 V. Consultants' Services and Training outside Turkey of Personnel of the Borrower 0.90 0.94 104 VI. Interest and other charges on the loan 2.00 2.00 VII. Unallocated 2.00 - - Total 24.00 24.00 100 - 25 - IV. AGRICULTURAL IMPACT Objectives and Strategy 4.01 As explained in Section III, the project's objective was to increase agricultural production by providing irrigation, drainage, and an extension service to help farmers raise yields and diversify production. It has been recognized since formulation of this project, dating back to the 1950's, that the key to exploiting the full potential of Adana Plain lay in diversification, involving a shift from monoculture of cotton periodically complemented with wheat to an intensive cultivation of horticultural and fodder crops in rota- tion with cotton involving double cropping and mixed farming. This was desired not only to sustain high soil productivity and yields but also because horticultural crops offered much higher returns per hectare than cotton and the opportunity to diversify Turkey's exports and develop agricultural pro- cessing industries. It was also recognized that diversification would be constrained by marketing problems and by difficulty of training farmers to grow new crops. The extension service adopted the following strategy to meet these goals. In the first instance, to raise cotton yields and increase farm incomes, establish confidence of farmers and gain their interest. Once this was done, it could concentrate on: (i) raising cotton yields further; (ii) increasing vegetable area; (iii) introducing a second crop after wheat; and (iv) improving dairying. 4.02 The longer term goals were to: (i) increase citrus area; (ii) increase vegetable production in conjunction with developing processing and export capacity; (iii) increase fodder crops; and (iv) develop agricultural practices adapted to local conditions. 4.03 While full development was not expected to be realized until 1985, experience so far has been mixed; yields have increased far beyond appraisal expectations, but the cropping pattern has not changed markedly and prospects for the future are uncertain. In addition, there has been under-utilization of the irrigation facilities, falling to 32% in one year and poor maintenance of drainage and land levelling works by DSI and farmers. Monitoring and Evaluation 4.04 The project did not provide specifically for monitoring and evalua- tion activities. However, there was a regular submission of detailed quarterly reports following guidelines provided by the Bank. In addition the extension service was conscious of monitoring the progress of their work and collected data on yields, prices, costs and net returns and analysed farmer response to their recommendations. These are included in detailed annual reports of the extension service. The quality of the reports tended to deteriorate but the data provided substantial resource for evaluatihg the agricultural impact of the project. - 26 - Yields 4.05 Assumptions made by the appraisal mission (1967) on yield increases for main crops are compared with those achieved to date in the table below. The "with project" data are average yields recorded by the extension service in the areas covered, including Stage I as well as Stage II sections. Comparison of Wheat and Cotton Yields Cotton Wheat ---(tons per ha)--- Appraisal Estimates Without project (rainfed) 0.6 1.5 (irrigated) 2.0 - With project, 1985 2.4 2.8 Actual With project (extension, irrigation and drainage) 1967 2.2 - 1968 2.8 - 1969 3.1 2.8 1970 3.5 2.7 1971 3.8 3.9 1972 3.3 3.0 1973 3.3 2.3 1974 2.3 3.4 1975 2.1 3.5 1976 3.4 3.5 1977 2.6 3.5 1978 2.7 3.7 Source: Extension Service Dramatic increases in both cotton and wheat yields are evident. By 1968, the appraisal forecast for full development cotton yield of 2.4 tons per hectare was surpassed and reached 3.8 tons in 1971. A variety of practices recom- mended by the extension service was responsible for this, such as improved timing of irrigation, ridging, irrigation prior to seeding, furrow irrigation, etc. Wheat yields reached full development expectation of 2.8 tons per hectare by 1969 and its maximum of 3.9 tons by 1971. The extension service attributes this increase largely to supplementing rainfall with irrigation - 27 - combined with various improved practices. However, these gains were not sustained beyond 1971. Cotton yields were affected by climatic factors such as low winter rainfall and hot humid summer in 1972-73 and then by the severe white fly infestation of 1974-75. This infestation was brought under control by 1976 and yields recovered but there was another setback in 1978, the cause of which has not been clearly defined and are thought to include a combination of factors such as late sowing due to late spring rainfall, infestation with red spider and resurgence of the white fly coupled with the resistence it has developed to the pesticides used. Low wheat yields in 1972 and 1973 are also largely attributed to low rainfall (288 mm October-April 1971/72 compared to 478 mm for 1970/71). While these climatic and other natural causes were impor- tant, it is commonly acknowledged that the pervasiveness of these problems and failure to realize full potential yields arise from the continued monoculture of cotton. It is also evident that reduction in yield levels coincided with the slackening off of the extension effort and with the increasing poor main- tenance of drainage canals by DSI and of surface drainage and land levelling by farmers. These aspects and their implications for future development of yields are discussed in succeeding paragraphs. Diversification 4.06 The table below shows appraisal estimates of cropping pattern com- pared with the existing pattern in 1977 and a revised forecast for 1985 made by DSI. Further details are in Annex III. Cropping Pattern - Comparison of Estimates Appraisal /2 /2 Pre-Project Forecast Actual - DSI Forescast / (1969)/l (1985)/l (1977) (1985) ha (%) ha (%) ha (%) ha (%) Irrigated Cotton 1,500 (3) 23,200 (48) 23,900 (52) 22,900 (50) Cereals - 3,500 (8) 100 (<1) 11,400 (25) Rice 600 (1) 2,000 (4) 300 (1) 1,800 (4) Citrus 100 (<1) 7,000 (15) 300 (1) 2,800 (6) Fodder Crops - 2,500 (5) 100 (<1) 2,300 (5) Vegetables 200 (<1) 3,500 (7) 100 (<1) 2,700 (6) Melons & Watermelons - 2,000 (4) 2,400 (5) 900 (2) Other - 4,500 (9) 600 (1) 900 (2) Subtotal 2,400 (5) 48,200 (100) 27,800 (61) 45,700 (100) Rainfed Cotton 38,800 (81) - - - Cereals 5,000 (10) - 17,900 (39) - Melons & Watermelons 2,000 (4) - - Subtotal 45,800 (95) - 17,900 (39) - Total 48,200 (100) - 45,700 (100) - Second Crops 0 20,000 (42) 0 9,100 (20) /l Source: Appraisal Report 12 Source: DSI - 28 - 4.07 Although in 1977 the area under cotton was reduced from over 80% to 52% of total area in line with the appraisal forecast of 48% at full develop- ment, the cropping pattern is less intensive in 1977 than in 1969. The shift from cotton was not occasioned by attractiveness of higher value crops and more intensive cropping but by decreasing profitability of cotton which led farmers to increase cereal cultivation. Higher value horticultural crops which were expected to occupy 35% of the total area covered only 6% of the total area a marginal increase from 5%, and are larely accounted for by the continued cultivation of melons and watermelons. Cotton continued to account for over 80% of the irrigated area. 4.08 This development in the cropping pattern can be explained by the following factors: high price and yield of cotton, white fly epidemic, high price of cereals and institutional difficulties affecting vegetable expansion particularly marketing. In the first few years of the extension program in the Seyhan area, starting with Stage I area in 1967, cotton yields increased from 1.7 tons per hectare in 1966 to 3.8 tons in 1971 (para. 4.04) and farm- gate prices from TL 1,950 per ton to TL 3,770 during the same period. This resulted in the spectacular increase in net income per hectare from TL 370 to TL 6,350 which led farmers to continue to put about 75% of their land under cotton, planting wheat only in order to allow levelling to be carried out. Cotton Production 1966-1971 Yield Price Production Net Income Cotton Area Year tons/ha TL/ton Costs TL/ha TL/ha (ha) (% total) /1 1966 1.7 1,950 3,000 370 - - 1967 2.2 1,990 3,450 920 - - 1968 2.8 2,070 3,900 1,900 - - 1969 3.1 2,140 4,380 2,180 - - 1970 3.5 2,700 4,440 5,090 27,100 48 1971 3.8 3,600 7,220 6,350 42,400 75 /1 % of total area under extension, Stage I. 4.09 Having accomplished the first target of increasing cotton yields, the extension service began its crop diversification effort, based on a program prepared in 1969 which concentrated on increasing vegetable production, citrus plantations and mixed farming incorporating intensive dairy enterprise. A series of vegetable demonstrations were carried out introducing new practices, improved seeds, plastic tunnels, and an active research program was undertaken in cooperation with Tarsus Research Station, Agricultural School and the Faculty of Agriculture at Adana. A citrus study was made in 1970 showing that about 7,000 ha in the Stage II area were suited to this crop and arrangements were made to develop this area in cooperation with citrus extension unit created under the Fruit and Vegetable Export Project (Loan 762/Credit 257). Efforts were made to improve feeding and housing of animals and the extension service helped make marketing arrangements with the Adana milk processing plant. - 29 - 4.10 The extension service encountered administrative and other difficul- ties in implementing the program, for example, importation of equipment for demonstrations was not approved by the Government, establishment of 20 demon- stration family dairy farms including import of animals never took place and importation of vegetable seeds was particularly difficult due to the delay in Government issuing permits in spite of Bank insistence. Possibilities for expansion were limited by the lack of adequate marketing facilities to handle a substantial volume of production outside the project area, and the absence of an organized marketing system which guaranteed a price for farmers as there was for cotton and wheat. The risks presented by prices which fluctuate enormously from year to year and during the season discouraged most farmers. The extension service made an effort to encourage out of season production and also proposed that processing factories should be established which would contract with farmers for raw material supply. In 1970, sorghum was intro- duced to supply the government feed mill in Mersin. This showed considerable promise: as a second crop after wheat, 400 ha were planted and yields aver- aged 2.4 tons/ha, reaching 4.5 tons/ha in some fields despite bird damage. However, there was no subsequent expansion because the government mill halved the price of sorghum just prior to the harvest. 4.11 Nevertheless, some gains were made in expanding vegetable produc- tion. Area under Vegetables Year (under extension, Stages I, II) 1970 1,000 ha 1971 1,100 ha 1972 1,700 ha 1973 2,100 ha Little progress was made in introducing double cropping of wheat followed by cotton since the full season cotton crop gave a higher return with the improved yields and prices. Although these achievements were limited in quantitative terms, there was a significant technological change involving improvement of cultivation practices and introduction of new crops. 4.12 Thus until 1973, cotton continued to be the predominant crop but progress was made with diversification. This progress was suddenly disrupted in 1974 when the white fly epidemic severely reduced cotton yields (para. 4.04), increased costs of protection, and reduced average net income from TL 15,290 to TL 1,660 per ha. Many farmers suffered a net loss and not surpris- ingly, the cotton area was also drastically reduced in the following years. The land thus shifted out of cotton was largely put under wheat, a risk- minimizing solution since it was a well-known crop. This was also encouraged by the doubling of its support price from 1,200 in 1973 to TL 2,050 in 1974 and to TL 2,400 in 1975. 4.13 Despite the continuing reliance on cotton and wheat, there are some indications that diversification may still take place in the near future. Farmers, faced with the persistent problems with cotton in the last few years, are becoming more interested in alternative crops and some progressive farmers have started to replace cotton with citrus and other crops. However, as both farmers and extension workers point out, vegetable production is risky because there is no organized marketing system. Success in diversification would also depend on the quality of the extension service provision of inputs, credit and other farmer support services. Land Improvements 4.14 The project developed 45,700 ha (net) with irrigation and drain- age facilities. However, the total area irrigated has been less that the area equipped partly because farmers were obliged to allow land levelling to be carried out in the dry summer months, and partly as a result of decline in profitability of cotton and absence of alternative crops as explained above (paras. 4.09), which led farmers to put land under wheat, without irrigation. The following table shows the progressive increase in areas equipped compared with the areas actually irrigated. Annual Increase in Irrigated Area Actually Cumulative Total Area Irrigated % Area Years Annual Increase Equipped Net Areas Irrigated (ha) (ha) Gross Net 1970 - - - - - 1972 4,200 4,000 4,000 2,960 74 1973 2,400 1,950 5,950 4,510 76 1974 19,900 19,200 25,150 25,490 133 1975 12,400 12,000 37,150 19,740 53 1976 3,300 3,000 40,150 13,030 32 1977 5,600 5,550 45,700 27,770 61 Stage II, Total 47,800 45,700 72 Stage I and II, Total 112,800 104,100 /1 Areas outside the official scheme irrigated with DSI permission. 4.15 Operation and maintenance of the irrigation facilities have been generally satisfactory. However, drainage has been poorly maintained. Due to lack of equipment, DSI can only clear main drains once every six years rather than three years as required. Farmers have tended not to maintain surface drainage, and in some cases plough over them until they have disappeared altogether. As the equipment provided under the project has depreciated, DSI's maintenance capacity has deteriorated. - 31 - 4.16 Research done at the Cukurova University indicates that irrigation development has led to rising water table in some areas, caused by poor maintenance of drainage system, and by the observed excess application of water and too wide a lateral spacing of drains. Conclusive data are not available but these effects have been widely observed in the field. Topraksu acknowledges that spacing of subsurface drains should have been 50 m rather than 100 m. However, a special Bank mission in February 1979 reviewed the operation and maintenance of Bank irrigation projects and concluded that maintenance of drainage has not significantly affected agriculture. Overall Assessment and Prospects for the Future 4.17 Judged against appraisal expectations, the project was more than successful in increasing cotton yields but progress with increasing the intensity of cropping has not been significant. The appraisal mission's yield projections appear to have been overly conservative, particularly considering that in 1967, the extension service operating in Stage I area aimed to raise cotton yields to 3.5 tons per ha within five years. In view of the experience to date, the appraisal mission's projections regarding changes in the cropping pattern appear to have been over-optimistic even though it could not have predicted adverse price policies and epidemics. It is possible that there will be a shift from monoculture of cotton and increase in double cropping, because a large number of new crops have been introduced and the dramatic decline in profitability of cotton appears to have shaken farmers into looking for alternatives. However, alternatives are presently limited by institu- tional factors: marketing, effective extension service and other support services and probably also land tenure. 4.18 The appraisal mission was not unaware of marketing difficulties, however they assumed that since demand was favorable, the organization would be arranged without special intervention. Similar if not more optimistic assumptions were made throughout preparation of previous phases of the Seyhan project (Annex III, Table 4). For example, the appraisal mission for Seyhan Irrigation Project Stage I assumed in 1963 that by 1973, the area under cotton would be reduced to 38%, citrus and vegetables would occupy 14% of the area, and forage crops 21%, and 45% of the land would be double cropped. 4.19 The extension service was extremely successful in achieving its first short-term objective and made progress with its diversification program until 1974. Progress since then has been more limited and there are indica- tions that the service as a whole has weakened. While it continues to aim for diversification its program is less well defined and many of its activities appear to have lapsed. The failure to deal effectively with the white fly infestation partly due to shortage of chemicals and partly due to disagree- ments with farmers over who should finance the control program, is another indication of the service's shortcomings, as is its inability to deal with other problems encountered with cotton since 1974. The link with research also appears to have weakened, and the Tarsus Research Station for example does little work which results in specific recommendations to farmers. In the field of citrus development, the service was not always aware of all the activities of the Fruit and Vegetable Development General Directorate which - 32 - are concentrated in the neighboring province of Icel and cover parts of the project area. As noted above (para. 3.05), the Government's policy on continuation of the service has been uncertain and its staffing has been reduced, the expatriate advisor departed in 1974, and the number of special- ists fell from 14 to 7 in 1977 and in addition, nine out of 14 vehicles allo- cated were transferred to other areas. By this time, the project was sub- stantially completed and almost all the funds were disbursed. The Government has now agreed that the service would be continued but it is unlikely that it would be strengthened. Effective Government assistance is more likely to come from the proposed Second Fruit and Vegetable Project now under preparation which is expected to provide extension, credit and a marketing organization in the project area. 4.20 Another institutional factor which may account for lack of progress with diversification is land tenure. Vegetable production has proliferated in the neighboring area of Icel which is presumably equally affected by marketing problems and probably receives less adequate extension (except in citrus). A significant difference between Adana and this area is the land tenure pattern: Icel is characterized by small holdings of 1 ha or less while Adana is an area of large holdings and absentee land ownership. Average farm size in the proj- ect area is 11 ha and holdings of over 20 ha are estimated to cover 60% of the project area (para. 7.04). Intensive vegetable production is not suited to holdings larger than a family can handle under local conditions. The exten- sion workers noted that smaller farmers were more receptive to diversification because it spread the family labor utilization throughout the year and reduced risks. One farmer visited by the completion mission was operating a mixed farm with hired workers but had his vegetable production undertaken by a share- cropper as a small family operation. While it is beyond the scope of this report to establish a causal relationship between land tenure and cropping pattern, it would be an interesting subject for further investigation. 4.21 In light of the experience described, appraisal forecasts of agri- cultural development should be modified with higher yields and less intensive cropping pattern. The Government, in its reevaluation, made revised projec- tions in the same direction. Projected Cropping Pattern and Yields Appraisal Estimate Government Estimate Yield Area Yield Area Crop (tons/ha) TIT (tons/ha) (%) Cotton 2.4 48 4.0 50 Wheat 2.8 7 4.0 25 Other - 45 - 25 Total 100 - 100 Second Crop 42 - 20 - 33 - The Government believes and the completion mission agrees that this degree of diversification is justified because new concepts, crops and practices have been introduced which can be expected to spread. The completion mission feels that the yield forecast for cotton may be optimistic and might more conservatively be reduced to about 3.5 tons. V. ECONOMIC RATE OF RETURN Rate of Return 5.01 Using actual data for 1969-77 and revised estimates for 1978 on- wards, the rate of return is estimated at 15%, compared with the appraisal estimate of 16% and the Borrower's estimate of 20% (Annex IV). The similarity between the completion and appraisal estimates is accidental since assumptions regarding production, investments, phasing and prices of inputs and outputs were all considerably changed. The accuracy of this revised estimate depends not only on relative prices remaining stable but also in further gains in crop diversification, full utilization of irrigation, and partial recovery of cotton yields. Production 5.02 Incremental production,estimated on the basis of revised yields and cropping pattern (para. 4.20), far exceeds appraisal expectations for tradi- tional crops, cotton and wheat and is greatly reduced for the new crops, vegetables, fruits and forage. Project Area Production - Annual, at Full Development Completion Estimate - Appraisal Estimate - Present Without Project With Project Increment Without Project With Project Increment ------------------------------------'000 tons----------------------------------------- Seed cotton 36.77 93.65 56.80 26.28 55.56 29.28 Cereal 11.00 45.60 34.60 7.50 9.80 2.30 Melon 26.00 28.80 2.80 30.00 137.58 107.58 Rice 0 20.70 20.70 1.65 8.00 6.35 Citrus 0 81.00 81.00 2.00 182.00 180.00 Vegetables 0 49.00 69.00 0 150.00 150.00 Alfalfa 0 23.00 23.00 0 30.00 30.00 Oilseed 0 2.34 2.34 0 3.25 3.25 Dry Beans 0 0 0 0 22.501 22.50 Potatoes 0 42.00 42.00 0 0 0 Fruit 0 20.70 20.70 0 0 0 Maize 0 92.00 92.00 0 0 0 Peanut 0 3.60 3.60 0 0 0 /1 Covers 45,700 ha. /2 Covers 48,150 ha. - 34 - Prices and Costs 5.03 Both output prices and costs of crop production were increased in real terms beyond appraisal estimates. Output Price and Cost Revisions Price Production Cost Estimated in 1968 Constant Prices Appraisal Completion Appraisal Completion Crop --------TL/ton-------- --------TL/ha--------- Seed cotton 1,400 2,100 1,900 3,800 Wheat 500 880 710 3,550 Rice 1,000 1,890 2,000 5,860 Citrus 500 1,000 8,570 11,500 Vegetables 600 810 5,500 7,300 Melons & Watermelons 200 420 2,500 3,840 5.04 The annual net value of production per hectare at full development is now estimated at TL 5,380 and at TL 390 without project compared to appraisal estimates of TL 3,330 and TL 410. The appraisal mission's estimate is based on a project area of 48,150 ha, including 9,000 ha of the Stage I area for which on-farm development was financed under Stage II. The revised estimate included benefits of the Stage II area only, 45,700 ha, since those from the 9,000 ha in the Stage I area could not be easily quantified, however, all on- farm development costs were included. The phasing of both investment expendi- tures and realization of benefits were much slower than was anticipated. Government's Evaluation 5.05 The revised rate of return estimate of 15% is based on the same data as the Turkish Government's estimate of 21% but is lower because the following changes were made. Cotton yield assumed to average 4 tons per hectare at full development was reduced to 3.5 tons per hectare; border prices were used for cotton, wheat and rice which were lower than prevailing local prices, expendi- tures on some items such as Topraksu equipment, and completion of works in 1978 and 1979 were ommitted in the Government's cost stream; and slightly different deflators were used. The revised estimate used actual price, yield and cost data for the period 1972-77. Financial Returns 5.06 The project would result in a more than five-fold increase in net incomes without taking account of double cropping. Income per ha, net of all labor costs but before deducting taxes or water charges would increase from TL 1,210 to TL 6,420. Corresponding figures estimated at appraisal were a seven-fold increase from TL 410 to TL 2,300. - 35 - VI. INSTITUTIONAL IMPACT Institutional Strengthening for Project Implementation 6.01 Organization of project management was one of the major project issues during appraisal and negotiations (para. 2.05). In order to strengthen DSI, Topraksu and the Extension Service, the project provided for employment of consultants and overseas training of their staff. The Bank placed con- siderable importance on consultants, making their appointment a condition of effectiveness and requested that they be included.in the bidders' pre- qualification committee and the committee's report to be sent to the Bank include their comments. Without speculating on what would have happened in the absence of consultants, both Government and Bank staff believe that they made a considerable contribution to project implementation. Progress was satisfactory during their tenure and the beginning of delays coincided with their departure in 1974. The long involvement of the consultants, Tahal-ECI consortium, in the project during Stage I and preparation of Stage II, prob- ably made them particularly effective. Training of personnel overseas took place too late to make a significant impact on project construction but it is generally considered to have been useful. To the extent that DSI feels it has gained sufficient experience to undertake works such as those of this project without external assistance, these efforts contributed not only to project implementation, but to institutional development. 6.02 Coordination between various Government agencies has long been identified by the Bank as a weakness in the Turkish agricultural sector. Coordination committees at Ankara and Adana functioned as anticipated during the first years of implementation but were not active during the latter years and have been discontinued although coordination is still important. It was particularly important to involve TCZB to meet credit requirements of the agricultural development program, but the committee was not successful in this respect. Extension Service 6.03 As explained previously (para. 2.11), a special extension organiza- tion was established under Stage I and was continued under this project. Its performance with respect to agricultural development objectives has been dis- cussed in section IV. This service also provided an important experience for institution building by demonstrating how an effective extension system might be organized and what it can achieve. The experience of this service provides useful lessons for both the Government and the Bank. 6.04 Although the extension service has not fully succeeded in meeting appraisal expectations and objectives, its achievements are considerable when compared with those of existing Government programs. Perhaps the most valid indication of this is the decision of two villages in 1978 to pay for their own foremen when the Government decided not to replace those who resigned. Given these positive achievements, the project demonstrated in particular the benefits of: (a) coordinating all aspects of extension work under a single - 36 - program as opposed to splitting responsibility for different activities among different agencies as under the existing government structure; (b) focussing activities on extension rather than data collection for the Central Ministry; and (c) establishing a system of close communications with farmers so that the extension service would be fully informed of the area's agricultural condi- tions, quickly disseminate advice and arrange to study key problems. 6.05 Although these achievements are well recognized by Government offi- cials involved, they have received little publicity and the Government has made little effort to apply this experience to restructure their existing services. Certain features are being adopted in a limited way, for example, Topraksu is expanding its program to train irrigation foremen, villagers who teach irrigation practices when Topraksu completes irrigation development in an area. The basic principles of this organization however are only being replicated in the three subsequent Bank projects. 6.06 This failure to achieve a direct institution building impact can be related to the design of the extension organization. It was because the many Government agencies involved had conflicting or overlapping activities that a "unified service" responsible for all aspects of farmers' needs was specially organized as an independent unit. However, this unit encountered many diffi- culties precisely because its organization was not consistent with the overall institutional structure of the Government. Its staff were drawn from different agencies and the Director did not have direct authority over all his staff, for example foremen were paid by the Ministry of Village Affairs and the Director by the Ministry of Agriculture. No single agency had responsibility for it so that it did not command support of any central Ministry. Just as no particular agency had interest in strengthening or even implementing the pro- gram, no agency publicizes its achievements. Created under a special project, not part of any agency, the Government views this experience not as its own. This weakness in design was apparent at the early stages of implementation if not at the time of formulation. However, the organization was formulated as a practical solution to meet a pressing need in 1967 when increasing areas came under irrigation but the Government was not providing adequate extension. In those circumstances it is difficult to see what else could have been done. Moreover, institution building was thus not its primary objective. 6.07 However, this project has been valuable experience to the Bank in its efforts to formulate projects to achieve institutional development goals. Modifications have been introduced in the three projects being implemented with extension components (Fruit and Vegetable Export, Ceyhan Aslantas, and Corum-Cankiri Rural Development) and efforts are being made to formulate an organization more capable of being replicated within the Government's existing structure. A national committee has been created to recommend such a system to be tested under the proposed Erzurum Rural Development project and to dis- cuss the future of the special organizations set up under the three existing projects. Other Farmer Support Services 6.08 Inadequate credit facilities under Stage I and arranging for Govern- ment assurance to meet requirements of farmers was a major issue during nego- tiations. After some discussion TCZB made a special allocation on TL 6.5 - 37 - million for the project area and a joint credit/extension program was intro- duced. However this was discontinued after the first few years. The Govern- ment also did not undertake the study of credit needs (para. 3.06). Cost Recovery 6.09 Cost recovery was an important issue during negotiations since levels of water charges prevailing at that time would have covered O&M costs only and the Bank believed that the rates should be raised to recover some investment costs. It was agreed that the Government would undertake a compre- hensive study to determine charges which could be levied in relation to the farmers' ability to pay and make recommendations for implementation. This study, commissioned to the consultants, Tahal-ECI was completed in 1971. The main conclusions and recommendations of this report were the following: (a) O&M costs in 1970 were TL 400 per ha of total area equipped compared to the prevailing average charge of TL 130 per ha. (b) The farmers' ability to pay, based on an assessment of incre- mental income derived from the project, net of taxes, risk and management allowance was estimated at TL 750 per hectare in 1970 prices. Thus TL 350 should be charged to recover capital costs, and this figure should be reviewed every 5 years to adjust for inflation. (c) Existing penalties for non-payment which comprised a 10% sur- charge imposed at the first repayment date should be increased with additional 2% to be charged each month, beginning with the second month of delinquency. Water delivery should be discontinued in case of prolonged failure to pay. (d) The existing system of different charges assessed for dif- ferent crops should be replaced by a single assessment rate for the entire project area except for the area under rice. In addition, water charges should be levied on the entire project area equipped with irrigation rather than only on the area actually cropped. (e) Farmers' irrigation cooperative should be formed throughout the project area with responsibility to distribute water, operate and maintain the distribution system and collect water charges from the members. 6.10 These main recommendations were accepted by DSI except the proposal to levy a uniform charge per hectare regardless of the crop grown because they thought that this would encourage farmers to grow only cotton. They proposed that the O&M charge should be increased gradually to the recommended level. DSI forwarded its comments on the report to the Ministry of Finance, and copied them to the Bank, but the Government never officially adopted any of these proposals. The Bank made repeated enquiries regarding the Government's plans to finalize this study until 1973, but did not obtain a satisfactory reply. After 1974, the Government implemented its own nationwide study. - 38 - 6.11 Cost recovery continued to be the project's major issue throughout implementation and was subject of high level policy discussions with the Government. It resulted not only in this project being rated as a "problem project" but also in the Bank's decision not to consider financing further projects in irrigation until a satisfactory system was instituted. Three main points at issue were: (i) the charges set for DSI's O&M cost recovery did not adequately cover actual costs; (ii) the charges set were not effectively collected; and (iii) there are no legal provisions for recovery of Topraksu's on-farm development costs. 6.12 Although at the time of appraisal, water charges which averaged TL 140 per ha were thought to be sufficient to cover O&M costs, actual assessment fell far short of costs because the tariffs were not increased although high rates of inflation increased costs, and because charges are levied only on actually irrigated areas and during most years, less than 100% of the area equipped utilized water. The table below shows actual yearly assessment com- pared with costs for the Seyhan Project area (including Stage I and II). Water Charges and O&M Costs 1972 1973 1974 1975 1976 Total O&M costs (TL million) 20.91 26.64 31.99 34.07 52.03 Average O&M costs per ha project area (TL) 360 /1 430 380 360 530 Average O&M costs per ha irrigated (Tb) 360 400 390 520 1160 Average water charge per ha (TL) 170 170 180 190 180 Total water charge assessed (TL million) 10.61 11.39 14.85 12.56 8.1 Assessment as % costs (%) 51 43 46 37 16 /1 The discrepancy between this figure and the study figure of TL 400 (para. 6.09) is due to inclusion in the latter of costs of flood control, exten- sion, drainage, maintenance backlog and "recreation" totalling about TL 70. 6.13 Government policy on cost recovery for DSI irrigation schemes is to recover O&M costs and capital costs depending on the farmer's ability to pay. Water charges are set annually by the Council of Ministers varying for dif- ferent irrigation schemes and by crop. During the project implementation period, 1969-1977, the Government did not change these levels significantly although DSI's annual submissions to the Council recommended increases required to cover their costs. 6.14 In 1978, the newly elected Government substantially raised water charges as a part of its general economic program. In the Seyhan area, the new charges were five times the previous levels and averages TL 900 per ha - 39 - compared with the actual 1977 O&M costs of TL 620 per hectare (or TL 760 per ha actually irrigated). Charges exceed previous year's costs not to take account of inflation but because water charges are considered to be an instru- ment of taxation and since Seyhan project area is one of the wealthier regions of the country, it is expected to subsidize more disadvantaged areas. 6.15 Water charges have not only been set at insufficient levels but have not been effectively collected, thus lowering recovery rates further. Collection of Water Charges - Stage II Area Annual Charges Collection Recovery Year O&M Cost Assessment Collected Rate Rate --- -------------TL' 000-------------- ------ % ------- 1972 1,440 490 320 65 22 1973 2,380 750 670 89 28 1974 9,810 4,600 2,270 49 23 1975 19,310 3,670 2,470 67 13 1976 29.310 2,730 3,110 114 11 Average 1972-76 76 19 6.16 This poor record of collection does not reflect poor performance of project administration in particular, in fact, higher collection rates have been achieved in the project area than in the country as a whole. Under the present penalty system, there is little incentive for farmers to pay since a surcharge is imposed only once, at a rate of 10% after the first repayment date so that the longer payment is delayed, the smaller the actual fee becomes. Particularly in conditions of high rate of inflation, this system effectively becomes an easy source of subsidized credit. In the event of persistent delinquency, DSI has the right to confiscate property, but this right is rarely exercised and as noted above, potentially more effective measures such as cutting off water or imposing surcharges on a monthly basis have been rejected. Both the Government's reluctance to enforce collection and the farmer's attitude reflect the political aspects of cost recovery. 6.17 Capital recovery charges are not collected until after completion of works and in 1978, a decree was passed authorizing collection in the Stage II area from 1979. Under this decree, all beneficiaries are required to repay total capital 1/ costs over 40 years with no interest. Under these terms, DSI's planned annual charge of TL 300 per hectare, or TL 13.71 million for the project area would result in a very low level of recovery of real eco- nomic costs as illustrated below. 1/ Actual expenditure, in current prices. -_,40 - Recovery of Capital Costs Present Value in Recovery 1968 Constant TL Index (%) Project cost discounted at: 2% 272.13 - 5% 247.27 - 10% 213.98 - Recovery charge discounted at: 2% 50.13 18 5% 24.22 10 10% 9.08 4 These recovery rates assume that either tariffs would be adjusted annually in line with inflation or that there would be no inflation from 1979 onwards, and that 100% collection would be achieved. These are unrealistically optimistic assumptions in view of past experience and it is likely that any recovery which would take place would be negligible. 6.18 Although the appraisal mission assumed that Topraksu investment costs would be recovered through the same water charges as DSI costs, Turkish law does not provide for imposition of cost recovery charges for Topraksu works. Although in 1978 Topraksu has presented a draft law to remove this constraint, the outcome is unknown and it is not certain that any recovery would take place. 6.19 With these levels of recovery, the effective subsidy amounts to almost the entire investment costs, amounting to TL 640 per ha in 1968 prices for DSI irrigation development or TL 1,080 for the entire project. This level of subsidy is difficult to justify considering that the increase in annual net farm income due to the project is estimated at TL 5,770 in 1968 prices, with- out taking account of double cropping, and that the beneficiaries have average incomes well above the national average since the per capita income net of family labor with project is TL 36,600 in 1975 prices whereas the per capita gross national product of Turkey in 1975 was TL 14,010. With these considera- tions, the Bank informed the Government in 1976 that it would not consider financing irrigation until steps were taken to improve the system. Audit 6.20 Audit reports of project accounts were never received from DSI and Topraksu as required under loan/credit agreements. However, the Bank rarely followed up on this omission and it is not clear whether the reports were not prepared or merely not sent. This is not a problem of project management but of general Government procedures and is experienced in most other agricultural projects in Turkey. Moreover audit reports may be less important for projects of this type which do not involve income-generating authorities. - 41 - VII. SPECIAL ISSUES Distribution of Income 7.01 At the time of appraisal, the impact of this project on the distri- bution of income between regions in Turkey and within the project area were not considered. The project's objective was to increase production rather than to increase incomes. It is more than likely that the project has contri- buted to both increasing regional imbalance in development and to unequal distribution of income within the project area. 7.02 Adana Plain is not only one of the potentially richest areas of the country but was already one of the most advanced areas at the time the project was formulated. Allocation of public investments for agriculture in Turkey during the last 15 years has been biased to this and other coastal areas and about 60% of the area developed under major irrigation schemes is in the Mediterranean and Aegean regions. Bank lending had also, until recently, been concentrated in the Mediterranean region. Favored with natural resources and benefitting from these investments, this region achieved the highest growth rate in agriculture in Turkey during the period 1963/65 to 1972/74, of 5% per annum compared with 2.6% for the country as a whole. On a per capita basis, this was 3.1%, far above the national average of 0.6%. The region has also benefitted from favored access to agricultural credit, which is effectively subsidized since lending rates are well below inflation and loan repayment is poor, e.g. 31% during 1967-69 in the Stage II project area. The Mediterranean region which accounts for 15% of the country's rural population and 17% of the cropped area had 29% of TCZB's outstanding credit. The existing system of water charges also includes a large subsidy element (para. 6.17). 7.03 The distribution of land and income within the project area also presents a significant issue. Historically, the project area has been one of large holdings operated in the past as feudal estates characteristic of south- eastern Turkey. More recently, land holdings have been split due to the general population growth, and sharecropping has been taken over by tenancy, and by operation under hired managers. Still, in 1965, only half of the families involved in farming were owner operators. Population of Project Area and Adana Plain (1963) Seyhan Area Adana Plain Number % Number % Farming families 2,600 72 15,250 81 Owner operators (1,550) (43) (9,580) (51) Sharecroppers & tenants (140) (4) (1,110) (6) Laborers (910) (25) (4,560) (24) Non-farming families 430 12 1,200 6 Total village residents 3,030 84 16,450 87 Absentee owners 590 16 2,400 13 Total families 3,620 100 18,850 100 - 42 - 7.04 A survey in 1963 covering 32,000 ha of the project area (76%) showed that the 590 absentee land lords owned 17,100 ha or 53% of the total area, averaging about 30 ha each but fragmented into 1,520 parcels. On the other hand, over half the families deriving income from farming had less than 7% of the land. Excluding 16 families owning an average holding of 220 ha each, average holding of owner operators was 7.6 ha. Distribution of Land Ownership 1963 Size of Holding No. of Families Area (ha) No. % Total Ha % Total Landless owner operators 1,050 33 0 0 0.1 - 5 960 30 2,140 7 5.1 - 10 290 9 2,060 6 10.1 - 20 165 5 2,280 . 7 over 20 135 4 8,640 27 Absentee landlords 590 19 17,060 53 Total 3,190 100 32,180 100 7.05 Some Government officials feel that the average holding size has decreased as a result of irrigation because many holdings, which were too small under rainfed conditions, became viable and many families who tended to sell such holdings kept them. In addition, efforts to introduce land reform since 1963 led many larger owners to split holdings among family members. Unfortunately, comparable data have not been collected more recently, but an analysis of the impact of the project on land tenure would provide much useful guidance for policy formulation in irrigation development. 7.06 Thus the majority of direct project benefits accrued to the minority of the area population. The majority of smaller farmers and landless people obviously benefitted too, as did a large number of workers who migrate to the region from areas to the south east to pick cotton. Assuming that 20 mandays are required for picking a ton, the incremental employment thus created would be 1,136,000 mandays annually after full development. This would be worth about TL 5,100 per annum for 16,000 workers assuming that the picking season is 70 days. Although such benefits are significant, if a national development objective is to raise incomes of the poorest segments of its population, there may be equal economic alternatives. Moreover, the Bank did not consider this to be an issue and apparently did not discuss it with the Government although the latter was considering implementing land reform. Environmental Impact 7.07 The project's environmental impact is subject to controversy. There was a resurgence of malaria, which was endemic in the project area but was more or less eradicated by the late 1960's as a result of an intensive campaign. - 43 - Reported Malaria Cases, -1968-79 Year Adana Total Cukurova /1 1968 36 37 1969 24 49 1970 72 150 1971 540 980 1972 850 1,340 1973 730 1,390 1974 340 1,830 1975 2,580 5,660 1976 15,350 28,850 1977 59,416 102,000 1978 47,428 64,000 1979 (6 months) 4,581 NA /1 Adana, Hatay, Icel Provinces. Source: WHO 7.08 WHO considers the primary cause to be irrigation development, par- ticularly poor water management and excessive use of chemicals associated with it. In February 1979, the Bank sent a special mission, including a consultant malariologist, to undertake a thorough study of irrigation water management and other on-farm practices and their relationship to the malaria situation in Cukurova. They concluded that the irrigation project is neither solely nor primarily responsible for the resurgence of malaria. Rather, most important breeding places for mosquitos are unreclaimed areas in the plain, and poor drainage and sanitation in urban areas. The breeding places related to irrigation development are limited to (a) the stagnant water in drainage canals due to poor maintenance and weed control; (b) inverted syphons, ponding due to uneven settlements in lined canals and canelettes, insuffi- ciently back-filled trenches under canelettes, undrained borrowpits for roads and flood embankments; and (c) on-farm ditches in poor stage of repair. Of these, only the first is considered to be of significance. Another indirect relationship is the increase in numbers of migrant workers from areas where malaria was not under control due to the increase in crop production which coincided with the termination of malaria control program in 1969. 7.09 As part of the Seyhan Multipurpose Development Scheme, one of the long-term objectives of this project was the development of the Adana region as a whole as agricultural development was expected to stimulate industrial development and vice-versa. In the Turkish Government's view, the project's impact on regional development was considerable. The project brought new industries such as textiles, manufacture of farm implements, etc., and attracted the University of Cukurova and other educational institutions, and created "job opportunities for thousands of unemployed" people of the region. Adana region certainly impresses most visitors as a thriving and growing center of industry and agriculture. Unfortunately it is beyond the scope of this report to attempt to substantiate this picture and the effect the project has had. It should be remembered however that Adana was already a prosperous region with a number of agricultural processing industries before the project began, and that crop diversification is constrained by lack of adequate marketing and processing facilities. On the other hand, technological changes increasing yields and introducing new crops and cultivation practices intro- duced by the project have generally spread throughout the region, although other Government activities and price trends, including subsidized fertilizer prices, have stimulated agricultural development generally. Pest Control 7.10 Effective pest control became urgently needed to combat the white fly epidemic and the resurgence of malaria. However, it is generally recognized that despite the efforts of the extension service, farmers have tended to apply any chemical they were able to purchase because they were not readily available on the market, and to spray too often and at the wrong time. The authorities were not able to identify the required chemicals, to ensure ade- quate supply and to enforce an appropriate regime of application. This has adversely affected both crop production and mosquito control. At the same time, DSI has not made effective use of chemicals to control plant growth in the drainage network. The study of malaria situation (para 7.08) concludes that DSI should establish a research program on aquatic plants and their con- trol. Improvements in pest control would therefore be a significant factor in the long term development of the Adana Plain. Sociological Issues 7.11 Sociological factors which have affected implementation of this project include the participation of farmers in operating and maintaining irrigation and drainage works, and changes required in the farmers' behaviour and local institutions to bring about an improved cropping pattern. 7.12 The poor record of water charge collection described above (paras. 6.15-6.16) can be explained in terms of financial incentives provided under the penalty system and administrative policies of the Government. Poor main- tenance of DSI facilities can in part be attributed to shortage of machinery. However, it is very much in the interest of farmers to have continued main- tenance of irrigation and drainage works and the existing institutional setup does not take advantage of this strong financial incentive. The comprehensive study of water charges recommended an organization of irrigation cooperatives with full responsibilities for distribution, maintenance beyond the point of delivery and collection of charges. Although 68 irrigation groups have been formed in the Stage I and II areas including about 70% of the irrigators, their responsibilities are limited to assisting in cleaning of tertiary canals. These groups are paid 25% of the assessed water charges and use the funds to contract out the required works and for community development. The farmers have neither been involved in planning nor in implementing the project, and although they have benefitted, they place all responsibillty for the works on the Government. - 45 - 7.13 The relative lack of support given by the farmers for on-farm devel- opment works, which are intended to increase yields and farm incomes, has also been surprising. Under the Stage I project, farmers were expected to request and finance land levelling and surface drainage themselves, using credit pro- vided by TCZB. This resulted in little response from farmers, and almost no progress was made. It was decided in 1966 with Bank advice, that Topraksu would plan and carry out the works free of charge. The pace of implementation improved substantially and the project was completed. The initial lack of response was thought to be due to TCZB's cumbersome procedures and farmers' lack of understanding of the benefits of on-farm development, particularly levelling and drainage. Topraksu states that the benefits of these works have now been demonstrated to the farmers and they are anxious to have the work done. This was confirmed by farmers interviewed. However, Topraksu's work program is still constrained by difficulty of getting permission to work in a large block in any given season, and their costs have increased by the scattering of the work. In addition, many farmers have not maintained, and have even destroyed surface drains. The possible explanations for this attitude of the farmers include the following: that on-farm development does not increase net income; that farmers' objective is something other than to maximize net income; or that there are other factors such as severe hardship caused by losing one year's cotton crop or fear of Government interference which makes on-farm development unattractive in the long term. Whatever the reason, it has not received attention in planning or implementing the project. These factors should be explored in order to guide future projects. 7.14 Although special circumstances affecting cotton and wheat yields, prices and costs together with the difficulties of marketing vegetables can explain the continuing monoculture of cotton under irrigation and a shift to wheat when the cotton crop fails, a cropping pattern which utilizes the full potential of the region's climatic soil and water resources is not likely to materialize in the immediate future. Farmer response to opportunities to greatly increase net incomes by cropping more intensively, encouraged by the extension service, has been low even when compared with Bank appraisal and other project planning reports. Although this is generally attributed to lack of marketing opportunities, since diversification holds the key to agricul- tural development of this region, the constraints and ways to remove them should be identified more precisely. 7.15 Attributing the failure to diversify on the absence of adequate marketing facilities and organization merely begs the question why the latter did not occur since it is well known that there is both demand and potential supply for the commodities in question, at least on the domestic market. Farmers in Adana have organized, on private initiative, an impressive network of facilities for marketing cotton. Ginning facilities have been expanded to textile manufacture, exports and provision of some inputs to farmers. Although there are intrinsic difficulties in marketing perishable vegetables, organizations in other parts of Turkey have successfully set up processing facilities for sugarbeet, fruit juice, etc. In the project area sociological factors favoring development of cotton marketing exist. Much of the cotton production has been in the hands of large landowners, often living in cities, - 46 - with financial and other means to invest in processing facilities. By con- trast, many of the vegetable producers of the area are small holders selling by themselves in the Adana markets. 7.16 A study of factors which determine the cropping pattern in the project area must include not only the physical factors affecting production but also the resources available to each farmer. Although operating under similar market conditions, farmers have different resources in family labor, skills, ability to bear risks, access to credit, inputs and markets and may have different objectives. An absentee landlord faces quite different con- straints to expanding vegetable production than a small holder. VIII. BANK PERFORMANCE Project Justification and Objectives 8.01 Although the project was consistent with the Turkish Government's development objectives and the Bank's strategy at the time it was appraised, experience in Turkey over the years and analysis of the agricultural sector as a whole, led the Bank more recently to believe that the Government should reduce allocation of resources to projects of this sort, to meet objectives of both growth and income distribution. Project Content 8.02 The Bank played an important role in scheduling and defining this project to take account of the implementation capability of the Government. The issues raised at appraisal and negotiations, farmer support services and cost recovery continued to be key problems under this project. Although the Bank's involvement in cost recovery is limited by the political nature of the issue, more attention could have been paid to the organization of irrigator groups. The assurances obtained did not prove adequate for credit and if farm support services were planned in detail and included in the project, they might have been more effective. The extension service, for example, would have benefitted from more detailed provisions for staffing and equipment and a separate loan/credit category. Financing of short-term credit was not normal Bank policy at the time the project was appraised but would perhaps have supported extension activities as part of the project was never seriously contemplated. The quantity of O&M equipment required was underestimated. 8.03 Although the cost overrun in foreign exchange was not large and financing was not difficult, the burden on the Government for local costs was larger than expected. The assumption that there would be no cost increases was overly optimistic and the lack of provision made for price contingency was a serious omission on the part of the Bank. Project Administration 8.04 The project was well prepared and the Bank's appraisal was based on a relatively good knowledge of the project area and implementing agencies. -47- However, a better understanding of the project area's farmers and institutions may have led the appraisal mission to make more realistic assumptions regard- ing cropping changes. Effectiveness of project supervision and close rela- tions with the Government appear to have deteriorated after the first few years of implementation due to a series of reorganizations within the Bank which broke the continuity of staff involvement. Subsequent supervision missions tended to lose sight of original project aims, followed up less effectively on points raised in previous communications, and generally became less constructive. For example, specific proposals regarding the diversifica- tion program were not followed up, being a less obviously critical part of project implementation than construction of civil works. Such outcomes probably had most adverse effects on the extension service. While all super- vision missions were highly concerned with the staffing situation and with general state of agricultural development, constructive suggestions are limited to requesting full staffing and there is no mention in the reports, for example, of possibilities to promote citrus development in cooperation with the Bank's Fruit and Vegetable Export Project and follow-up on the initial approaches made, or on the 1971 study. It is also noted that general slowdown in project construction began and delays accumulated when continuity in supervision was broken and supervision reports do not question why comple- tion of works dragged on and whether there were possibilities for speeding up work. IX. CONCLUSIONS 9.01 Construction for irrigation development of Seyhan Irrigation Project Stage II has been substantially completed and about 45,700 ha (net) are equipped with irrigation. The rate of return anticipated at appraisal of 16% is expected to be approximated although this would depend on continued main- tenance of irrigation and drainage facilities and provision of agricultural extension services at existing levels. It is also assumed that there would be improvement in cropping intensity by full development in 1985 and cotton yield levels achieved in 1969-73 would be regained. 9.02 Measured against appraisal objectives, the project was successful in increasing production and in introducing certain technological improvements in agriculture. However, the project's long-term development impact is likely to fall short of appraisal expectation since the key to full exploitation of Adana's development potential lies in diversification of crop production which is unlikely to be achieved to the degree anticipated. It was assumed that farmers, on average, would increase cropping intensity, diversify cropping and utilize irrigation water because on the basis of market price, such changes would bring large financial gain. Market channels were expected to be orga- nized without intervention because demand and supply existed. It was assumed that farmers would maintain their farm roads and surface drains provided by :- -48 - Topraksu because they, brought financial gain. These assumptions had been made throughout development of the overall Seyhan Multipurpose Project dating back to the 1950's but have not materialized to date. The appraisal of Stage I assumed that by 1973, 45% of its area would be double cropped. Today, five years after this date, the area double cropped is negligible. A more systematic review at appraisal of how farmers' behaviour might change and how institutions might develop might have resulted in more realistic forecasts. 9.03 Given that diversification is the key to agricultural development under this project, a systematic analysis to identify constraints to achieving this goal is recommended. This would require a special study of land tenure structure of the project area and factors affecting farmers' production decisions, such as their production objectives, access to credit and inputs and markets, the availability and costing of family labor and other resources, ability to take risks, etc. The private sector activities and incentives for developing marketing organizations and processing facilities should also be considered. 9.04 Although the Bank laid considerable emphasis on provision of farmer support services, it did not become directly involved in formulating these services as part of the project and costing them as such. Although the extension service in the first few years was effective, it could have been made more so if more detailed planning of the services had been done as this would have helped implementation and perhaps continuation after the life of the project. The more recent Bank projects in Turkey have taken these weak- nesses into account. 9.05 The Bank played a significant role, starting under Stage I, of strengthening the extension service in the project area by establishing a special organization. Although agricultural development generally fell short of appraisal expectations and in recent years, the service could not effec- tively deal with problems lowering cotton yields, its positive achievements were considerable in comparison with the general standards of the country. It established a solid system of communication with the farmers, coordinated various extension activities under a single program and in the first years, built an effective link with research. Thus some farmers decided to finance their foremen when they resigned and were not replaced by the Government. Although this experience had considerable demonstration effect, it is not widely publicized and its features are not being adopted in the Government's existing system. It was replicated only in the three subsequent Bank projects. 9.06 While its direct institution building impact was limited,the Seyhan extension service provided many useful lessons for the Bank and the Government in formulating projects aimed to strengthen the extension organization in Turkey. A national committee was established to recommend an organization more suitable for replication throughout the country to be tested under the proposed Erzurum Rural Development Project. The key lessons drawn include the following: (i) setting up a separate unit apart from the main lines of responsibility of existing Government structure led to difficulties in obtain- ing central Ministerial support for its program and for its continuation after - 49 .- completion of the Bank project; (ii) staff drawn from different ministries and organizations led to difficulties in coordination; and (iii) lack of strong and continued link with research reduced their effectiveness in dealing with new problems. 9.07 The project has illustrated the importance of maintenance and effective utilization of capital investments after construction is completed. Although DSI's maintenance and operation of irrigation system has been satis- factory, there has been poor maintenance of drainage and-signs of excess watering and inadequate pest control. Together with the slackening off of the extention effort, this has had adverse impact on agriculture and may also have led to rising water tables and contributed to the resurgence of malaria. 9.08 The project raises significant issues with respect to income distri- bution and allocation of public investments. While the Seyhan Project was and still is a sound one, it supported the Government's agricultural development strategy which was biased to investments in large new irrigation projects in rich areas with potential. A review of the sector as a whole by the Bank (1972 and 1978 Sector Studies) recommends allocating more resources to com- pleting or rehabilitating existing works. The project also helped increase inter-regional and intra-regional disparity in wealth and income. If an objective of improving incomes of poorer population is adopted, projects with similar impact on production may exist which meets this objective better. Cost recovery under this project has been extremely low which, considering the income of beneficiaries is difficult to justify. 9.09 The Bank attempted to assist in rationalizing the public financing of such investments by a study of water charges but was ineffective since this issue had political significance. However, this affected future Bank lending, for the Bank decided not to consider future irrigation projects until improve- ments were made. Subsequently, in 1978, the Government decided to raise water charges. 9.10 The Bank played an important role in scheduling project implementa- tion taking account of the manpower and other resources of the Government and identified issues which proved critical to successful implementation of the project. These were coordination among agencies, farm support services and cost recovery. However, the Bank did not go far enough in designing and financing these support services as parts of the project, as is now being done under other agricultural projects. 9.11 The importance of providing generously for realistic cost increases is shown by this project. 9.12 The Bank's long association with the Seyhan project enabled it to identify critical issues and follow up on them during implementation. However, the Bank's effectiveness was reduced when continuity in staff involvement was broken by a series of reorganizations within the Bank. This experience also suggests involvement in a series of projects forming phases of a single devel- opment plan and offers a possibility for the Bank to assist in strengthening - 50 - recurrent and operating as opposed to investment activities. These include farmer support services, completion of construction, maintenance and operation of existing capital infrastructure, all activities which are critical to institution-building and agricultural development but last beyond the time limit of Bank projects. ga il r4 ;4 iý F4Pligl p F4 ý I P4i lr4 1URE SE,M , IRRIGATION ~PWJE' STAGE Il ~MPLTION HEFOR'! Eaie..1~n LiFt Appraleed Eatiantes tPRÅJvU Cot (CIF Adna) Actual 1 US 1000. No. @ iiemarks Lnd levelling Scraper with Tractor 12-17 cu yd 33 1,419 Scrper with Tractor 7-9 cu yd 29 986 (76 871 Scrapera only Scraper 8 cu yd wIth «heeled tractor 14 392 - - Tractor., bulldo~er. - - 1,177 U-doer blade with control. for 150 lp tractor 8 30 -- u-doer blade with control for 110 lp tractor 8 22 - - Land Plane with 110 bp cwler tractor 4 100 25 528 Crawker Trfetor D4D, D6C only. Land plane with 50-65 hp wh.eled tractor 10 70 24 30 ~eled Traetor John Deer + 14 land plana. Subtotal 3,019 2,936 T1e Draeinge Trencher 5 150 5 11N Cleveland J57 Pront end loader, 1-3/4 ca yd aide dup bket 5 100 2 29 Bray 552 Frot end loader, 1-3/4 cu yd standard buket 2 38 7 91 Nichgavn 55R E~aator dr~glin-, 3/4 cu yd 3 1 29 Rumto Bueyru. 11 22 Du~p Truck 8 Ton pacity 18 180 18 116 BO '1135 Flat bed truck - 3 ton 5 30 - - Grader. eotor patrol, 110-130 lp 4 75 6 95 Aveling Barford NT Scrnig plant 2 2 10 Subtotal 653 511 Surface Drainge Grader 2 38 - - Fara ditcher with tractor 50-60 lp 5 40 ( Fara dltcher with träctor 30-40 lp 5 30 (10 4 DItchr only Ubtotal 108 4 General and Transporttio -o. bed tractor-trallar - 30 ton capcIty 5 290 5 316 Truck tractor DMch I 6096 Nu for 45 persone 7 98 7 91 Self propelled trailer 4 4 l 5 - - Jeep 5 10 - - Trailer Truc5 2 100 5 30 Trailer Bla.hard Pik-Up 2 x4 30 90 30 92 PUck-Up 4 4 13 47 13 45 Station Wegon 2 1 4 4 14 b 15 Station agon. 4 . 4 2 8 2 8 TVo-way radio ror car or pick-up 14 4 - - Central radio tran1ittor l 2 1 16 Subtota1 568 413 Service Equipaeont Mobile -itenanc shop 8 80 8 93 Mobile repair bop 7 248 7 210 Fel tanker 5 1 4 60 4 64 Central Work abop eqapent - 50 - - Other. - - - 108 Include. Tralning Tratler Unit, alecti,c Vulcaniser, Bheel Balancer, etc. Subtotal 435 475 Extention Sercice Vehicle. 18 59 - 9 Station oa6ons Tractorw 4 66 - - Pilkjp. L.S. 88 - - Mi-cellaeoun L.S. 56 - - 1 PritnIng Machine aod 1 .aaf element analywing .quisment Subtotal 275 106 Total E-aip-ent 5,058 6,339 Spare Part5 20% 1,010 Alwo incldes tire. and .ubew Total before . tingencie 6,068 contingencies 15% 910 AL TOPRAKSU 6,978 6,1. - TURKEY SEYHAN IRRIGATION PROJECT STAGE II COMPLETION REPORT Consultant Services 1- By DSi 1969 1970 1971 1972 1973 1974 1969-1974 IRRIGATION AND DRAI'JAGE AORKS: No 1 MA-MO2 No YIN-?,) No MAN-MO No MAN-N0 No MAN-HO No MAN-O No MAN-IdO PROJECT ANAGER (EXPATRIATE)a 1 2 1 7 1 12 1 12 1 6 - - 5 39 CHIEF CONSTRUCTION ENCINWR 1 3 1 12 1 11 1 12 1 12 1 10 6 62 (EXPATHIATE) 3 CONSTRUCTION NGINEE1(EXP4TRIAT): - - 1 2 - - - - - - - - 1 2 CONSTRUCTION ENGINEER(LOCAL) 2 10 2 24 2 24 2 20 2 2!1 2 8 12 110 CANALET SPECIALIST (LOCAL) : - - 1 1/2 1 1 4 , 1 12 1 7 - - - - 3 3 1/.. SURVEYOR (LOCAL) 1 4 1/2 1 12 1 12 1 10 1 12 1 4 6 4 1/2 DESIGN ENGINEER(EXPATRIATE) 3 15 3 24 1/2 1 1'? 1 12 1 6 - - 2 69 1/2 NECRANICAL ENCI-EER (EXPATRIATEI: - - 1 1 1 3 - - - - - - DESIGI EGINEER (LOCAL) _ 3 10 4 27 1 12 1 12 1 6 - - 10 67 SPECIFICATIONS iRITER(EPAMRIATE): - 1 1 - - - - - - - 11 SPECIFICATIONS WRITER(LOCAL) : - - 1 3 1/2 - - - - - - 1 3 1/2 O AND M ADVISOR (EXPATRIATE) S 1 4 1 12 1 12 1 12 1 5 - - 5 45 0 AND M ADVISOR (LOCAL) - - 1 11 1 12 1 12 1 4 - - 4 39 SUIS-TOTAL s 12 50 1/2 19 141 1/2 11 122 10 109 9 75 4 22 65 520 JOM I CONSULTWC SERVICES WAS COWPLETED ON OCTOBER 31, 1974. 0 o Consultant Services BY TuPRAKSU 1 1969 19'io 19'll 1972 19j3 1974 1969 - 1974 AÅN-V:k 2 O MN - rLA-14 o .MN-,iu o V-4-MO No IN-t.- RRIGATION AkD DRAINAGE DVIW8Ri -FPATRIA'I') - - I 7+/½ ~ 12 1 12 1 - - 37-1/2 1rGHiANLCAL EQUIPENT -Ar=.r t!!PATHIAT¯~ - - 1 12 1 1 lu 1 12 - - 2 ELTSJS1u14 I,P!CIALI.TS3 (ExPATRIATE) . i 5 1 12 1 12 1' i $ - - ACillCILTURAL CIAI (XPATIATE) : 1 1 1 12 1 ) 1 , 3 3 12 10 t nU-T·Ti.L 2 6 i 43f1/2 4 11 4 :9 . 1 23 1I,+1/2 ifor'N CZ3SULTIG SERVICES siAS OMPIEÉ'Et u; 00ER 31, 1974 ANNEX 1 - 55 - Table 3 TURKEY SEYHAN IRRIGATION PROJECT STAGE II COMPLETION REPORT Staffing and Coverage of Extension Service Year Specialists Foremen Villages Covered 1967 4 23 25 1968 7 43 43 1969 14 54 54 1970 16 54 61 1971 12 54 70 1972 14 52 70 1973 14 60 70 1974 14 58 77 1975 14 58 70 1976 14 50 86 1^77 7 49 86 1978 7 52 86 1979 8 51 94 TURKr.Y . CCIMPLATIM REPMR BEYAN IRRIGATIOM PROJECT MASE 11 Construction schodule • 1MAK8U Total at Cloing Total Eatimt*d 121 721É 1121 11 lii A9l 11.. 12/m im »7 171231/77 Pind 1978 -979 Apprefseal itimate Land lovelling (ha) • 6,000 6.000 6,000 13,000 15,250 • • 48,250 48,20 TIle drainage <ha) - 3.000 4,500 4.500 4,500 4,400 - • 20,900 20,900 • Surf«e dratnage Nam) 47 190 190 190 190 143 - 950 - ileid road* <b=> 12 48 48 46 48 36 • - 240 240 Actual Schedule - Coepleti" of Conltructtoq Land levelling (ha) 2.100 3,452 10,347 10,523 5,830 2,632 3.954 3,791 &,3u 44, j 46,170 1,661 Til* dranage (ha) - 2,716 4,262 1,748 5,485 2,984 2,489 2.762 1244 26.370 00 3,424 Til* dratnage (km) - 217 296 103 390 238 228 260 - 1,732 2.167 M4 Surfact drain (km) 87 107 102 100 22 Li 251 94 - 763 7M3 (33) (65) (24 or 183) (St* ar 677) Fleld roada (ka) 36 60 38 44 61 32 - 271 274 (33) (13) (71) (26) (279) Actual av *f Aofraeal Betimato (Cutative ~rogreA) Land levelling (.) • 93 132 147 98 72 80 96 91 91 96 Til* drøinage () • 91 93 73 86 82 94 107 107 107 126 Surface drains (t) 185 82 69 64 68 70 100 • 80 (70) (61) (5)(71)L (55)(71)j field roade (7) 300 160 124 114 117 11l 113 113 II3 (107) (94) (109) (120) (115) L Figur*& in brackoto are those reortød le Topraku's quartørly progreas reporte whieb differ fra~ tboe raported by Tonrakam le the ca~lette reort. Thi diocreo*ocy could ~ot be elarif~d by comletian mia9ion. L The aggregatt total for surface drain le *hon in 1977 as 518 hm which le not co~estent with the om of all years which wuld h 477 km. ta likely to be In 1975 or in total. Source: Toprakau. DSI and Draft Co~'letton eport. c' TURKEY SEYHAN IRRIGATION PROJECT STAGE II COMPLETION REPORT Cost Estimates ----- Total Cost (TL millions) ---Total Cost (US$ millions) Final Cost as Final Cost as Appraisal % of Appraisal Appraisal % of Appraisal Estimate Final Cost Estimate Estimate Final Cost Estimate DSI I. Irrigation & Drainage Works Area 1TP, 5YP 51.3 54.8 107 5.7 2TP 89.1 199.3 224 9.9 2YP, 3TF 76.5 235.6 308 8.5 Headworks 6.3 14.0 222 0.7 06M Facilities 6.3 8.6 137 0.7 Contingencies 45.9 - - 5.1 Subtotal 275.4 512.3 186 Right of Way 29.7 38.5 130 3.3 Engineering Administration 30.6 57.4 188 3.4 Subtotal 60.3 95.9 159 6.7 Pipe Factory Construction - 5.4 - Pipe Manufacture - 18.3 Subtotal 0 23.7 - Total Irripation Drainage 335.7 631.9 188 37.3 45.3 110 II. Equipment & Supplies Equipment for Pipe Manufacture 1.8 7.9 439 0.2 0.7 350 Gates, Hoists & Material for Irrigation Drainage Works 6.3) 40.7 181 0.7) 2.7 108 06M Equipment 16.2) 1.8) - 1 Total Equipment Supplies00 TOTAL DSI 360.0 680.5 189 40.0 48.7 111 OQ a- mo M Cost Estimates - Total Cost (TL millions)------------ - Total Cost (US$ millions) Final Cost as Final Cost as Appraisal % of Appraisal Appraisal % of Appraisal Estimate Final Cost Estimate Estimate Final Cost Estimate TOPRAKSU /1 III On-farm Development - Land Levelling 47.7 138.8 291 5.3) Tile Drainage 21.6 50.7 235 2.4) Surface Drainage 8.1) 14.3 106 .9) 13.3 121 Feeder Roads 5.4) .6) Contingencies (20%) 16.2 - - 1.8) Engineering & Administration 7.2 13.0 1 .8 .9 113 Subtotal On-farm Development 106.2 216.8 - 204 11.8 14.2 120 IV Equipment & Machinery 63.0 89.4 142 7.0 6.3 90 TOTAL TOPRAKSU 169.2 306.2 181 18.8 20.5 109 V. Consultants )19.8 33.1 173 2.2 2.5 146 Training ) 1.2 ) 0.7 VI. Interest during Construction 18.0 18.0 100 2.0 2.0 100 Total Project Cost 567.0 1,039.0 183 63.0 74.4 111 /1 Includes works to be completed 1978-80. Expenditures to closing date: Land levelling : TL 11.6 million Tile drainage 45.4 Surface drainage & feeder roads . 14.3 Engineering & design 13.0 fotal TL T9 million TURKEY SEYHAN IRRIGATION PROJECT STAGE II COMPLETION REPORT Analysis of DSI Cost Increases Irrigation and Drainage Works Estimated Cost in Increase due Increase due Difference Appraisal Contract to Changes to Price Total Final in Appraisal Area 2YP - 3 TP L Report Prices In quantities Escalation Cost and Final Cost --------------------------------------------------(TL'000)---- Construction of Section 10 37.369 6.881 3.827 48.077 Construction of Section 11 38.943 3.454 1.582 43.979 Construction of Section 12 38.300 1.823 2.505 42.628 Construction of Additional Area 2YP - 4.904 /5 - .032 4.126 Completion of Additional Area 2YP Construction / 53.946 22.804 20.000 96.750 Subtotal 76 500 172.652 34.962 27.946 235.560 159.060 /1 Refers to the Appraisal Report, Annex 6. /2 In Appraisal Report, estimated costs are given for 2YP and 3TP area as separately. During execution of works, 2YP and 3TP area was combined and contracted in 3 parts. L3 5,900 ha area was not included in Stage II at the time of appraisal. Later on, 2YP area was extended namely Extension of 2YP Area. / Government decree dated April 3, 1974 giving contractors the option to terminate their contracts, a new contract for extension of 2YP area was executed. /5 Here, the contract price indicates the expenses spent up to the termination date of the contract. Analysis of DSI Cost Increases Irrigation and Drainage Works Estimated Cost in increase due Increase due Difference Appraisal Contract to Changes to Price Total Final in Appraisal Area 2TP Report Prices In Quantities Escalation Cost and Final Cost --------------------------------------------------(TL'000)-------------------------------------------------- Construction of Section 7 33.239 9.945 3.492 46.676 Completion of Section 7 19.571 17.728 6.000 43.299 Construction of Section 8 23.578 15.983 3.713 43.274 Completion of Section 8 /1 3.789 - - 3.789 Construction of Section 9 32.356 - 3.854 36.210 Completion of Section 9 /1 26.000 - - 26.000 Subtotal 89.100 138.533 43.656 17.059 199.248 110.148 01 /I These works were carried out through force-account procedure by DSI. 0 P 0 0 . Analysis of DSI Cost Increases Irrigation and Drainage Works Estimated Cost in Increase due Increase due Difference Appraisal Contract to Changes to Price Total Final in Appraisal Report Prices In Quantities Escalation Cost and Final Cost --------------------------------------------(TL'000)------------------------------------------- Construction of Section 9 in 5YP Area and Section 4 in ITP Area 51.300 31.854 19.206 3.754 54.814 3.514 Headworks 6.300 13.858 - .078 13.936 7.636 Operation and Maintenance Facilities 6.300 8.562 /1 - - 8.562 2.262 Pipe Factory /2 - 5.407 - - 5.407 5.407 Pipe Manufacture /2 - 18.338 - - 18.338 18.338 Subtotal 63.900 78.019 19.206 3.832 101.057 37.157 a' Grand Total -3 229.500 389.204 97.824 48.837 535.865 306.365 /1 The contract amount includes the cost of total owrks carried out by force-account and by tender. /2 Pipe manufacture and construction of pipe factory is carried out by force-account. /3 Grand totals include subtotals. OQa TURKEY COMPLETION REPORT SEYHAN IRRIGATION PROJECT PHASE II Comparison of Crop Yields - Actual Without Project 1961 - 1969 Area 1961 1963 1965 1967 1969 Source DSI/ToprakSu Bank Tahal-ELI/DSI Extension Bank Area Stage 1, II, III Stage I Stage II Stage I Stage II Without Project Without Project Without Project With Project Without Prolect crop Rainfed Cotton NA 0.6 0.5 NA 0.6 Cereals NA 1.5 1.3 NA 1.5 Melons & Watermelons NA 15.0 12.5 NA 15.0 Irrigated Cotton NA 0.8 1.5 1,7 2.0 Rice NA 2.3 2.65 NA 3.0 Citrus NA 15.0 21.0 NA 20.0 Fodder Crops NA 16.0 20.0 NA NA Vegetables NA 1.0 16.7 NA 15.0 Dry Beans NA - 1.0 NA NA Weighted Average Cotton 0.5 0 64 0.61 NA 0.65 Cereals 1.5 1.5 1.3 NA 1.68 Rice 2.0 2.3 2.65 NA 3.0 Citrus 20.0 20.0 21.0 NA 20.0 Fodder Crops 12.0 15.0 20.0 HA NA Vegetables 15.0 16.0 16.7 NA 15.0 Dry Beans - 1.0 1.0 NA NA C-z Melons & Watermelons 15.0 15.0 12.5 NA 15.0 X Total Area Irrigated 5% 12% 13% 5% Source- 1961-DSI/Topraksu Lower Seyhan Project Report TURKEY COMPLETION REPORT SEYHAN IRRICATION PROJECT PHASE II Comparison of Yields - Projected Year Projected With Project 1973 With Project 1985 1985 Source DSI/Topraksu 1961 Bank 1963 Tahal-ECI/DSI 1965 Bank 1969 DSI, 1978 Area Stage I, II III Stage I Stae IT Stage I Stage 11 ----- Tons per ha------ Cron Irrigated Cotton 1.75 2.5 2.4 4.0 Cereals 1.75 2.0 2.5 2.8 4.0 Rice 2.0 4.0 4.5 4.0 6.0 Citrus - 26.0 3.0 26.0 30.0 Fodder Crops 27.0 12.0 /1 4.5 12.0 1l 10.0 Vegetables 12.C 19.3 2 4 20.0 - Oilseeds 34.0 1.0 1.0 1.3 - Melons & Watermelons - - 25.0 25.0 32.0 Other Fruits 25.0 - 12.0 22.5 Second Croo Winter Vegetables -19.3 20.0 10.0 Cabbage - 35.0 - -5- Melons & Watermelons - 25.0 25.0 - Dry Beans 1.8 1.8 1.8 - Fodder Crops 0.5 - Corn - 6.0 - 40.0 /1 Alfalfa Sjurce DSI. 'orraksu, 1961, Lower Seyhan Oroject Report. Bank, 1963 Anpraisal Report. Seyhan Irrigation Pro1ect (Stage I). Tahal-ECI for DS1, 1965, Seyhan Irrigacion Project Stage II, Economic Annraisal (Vol 2) banK, 1969 Appraisal Renort, Seyhan Irrigation Pro3ect, Stage TI. DS1, 1978, Draft Comnletion Report. I URR,EY Ltt El I N REPURI SLYIAN IRRII;ATIUN PROJECT PHASF 11 Comi parison of Actual Cropping Pattern pre-Project 1990 - 1969 Year 1952 1961 1963 1ji6 1969 Source (Bank) (DSU'r.praksu) (Bank) (Tahal-ECI. DSI) (Bank) Area Stage 1, II Ill Stage I. 11, Ill Stave I. II Stage - - Stage-11 -----------------------------------ha (% total area)---------------------------------- Cron Rainfed Cot ton - NA 67,100 (77) 22,400 (66) 38,800 (80 5) Cereals - NA 8,700 (10) 6,200 (18 4) 5,000 (10 4) Melons & Watermelons - NA 900 (1) 200 (0.5) 2,000 (4 1) Sesame - NA - 100 (0.2) - Other - NA - 400 (1.2) - Subtotal - 176,700 (95) 76,700 (88) 29,300 (86.3) 45,800 (95) irrigrated Cotton - NA 8,500 (10) 2,700 (8) 1,500 (3 1) Cereals - NA - - - Rice - NA 700 (1) 200 (.5) 600 (1 3) Citrus - NA 400 (0.5) 100 (0.4) 100 ( 2) Fodder Crons - NA - - - Vegetables - NA 300 (- 1,400 (4) 200 ( 4) Dry Beans - NA 400 (0.5) 200 ( 8) - Melons & Watermelons - NA - - Maize - NA - - Other - NA - - Subrotal - 9,300 (5) 10,300 (12) 4,600 (13.7) 2,400 (5) Rainfed and Irrigated Combine' Cotton 60,000 139,500 (75) 75,600 (87) 25,100 (74) 40,300 (83 6) Cereals 60,000 18,600 (10) 8,700 (10) 6,200 (18.4) 5,000 (10 4) Bice - 13,000 (7) 700 ( 8) 200 (.5) 600 (1 3) Citrus 7,000 5,600 (3) 400 ( 5) 100 (0 4) 100 ( 2) Fodder Crops - 400 (,1) - - - Vegetables 7,000 7,400 (4) 300 ( 2) 1,400 (4) 200 ( 4) Dry Beans - 400 ( 5) ( 8) - Sesame 60, 000 - - 100 ( 2) - Melons E. Watermelons - 1,500 (1) 900 (1) 200 ( 5) 2,000 (4 1) Other - - - 400 (1 2) - Total Area Crooned 194,000 186,000 (100) 000 (100) 33,900 '100) 48,200 (100) Total Area Irrigated (7 total) 9,100 (5) 300 (12) 4,600 (11) ",400 (59 Total Area Equioped 16,500 500 NA NA i Area Equnned Irrigatee 56 62 NA EA Sou,rce 1991 - 1SI/Tonraksu, I ke.,r Seyban Bert 111 1952 - Bank, Seyhan Multipurpose lroject Appralal Report 1('l - Bank, Seylian IrrigatLIin Piolect Stage 1, Apriiisal Report Z 1I - fahal-ECI for 1IM sevnan Irrigation Projecl Stage 11 19,' - Bank, Seyban Irrigation Procit Stage II Appraisal Report TURKEN' COMPLETION REPORT SEYHAN TRRTGATION PROJECT PHASE 11 Comparison of Projected Cropoing Pattern Year Projected With Irrigation With Irrigation 1973 With Project 1985 1985 Source DSI (1951) Bank 1952 Bank 1963 -Tahal, DSI 1964 Bank 1969 DSI 1978 Area Stage I, 11, III Stage I. II, III StaRe I Stage 11 Stage 11 Stage H Crop Irrigated Cotton 55,000 (34) 60,000 (31) 20,000 (26) 16,500 (42) 23,200 (34) 22,900 (42) Cereals 20,000 (12) 60,000 (31) 7,800 (10) 2,400 (6) 3,500 (5) 11,400 (20) Rice 8,000 (5) - 3,000 (4) 800 (2) 2,000 (3) 1,800 (3) Citrus 13,000 (8) 7,000 (4) 4,000 (5) 7,000 (18) 7,000 (10) 2,800 (5) Fodder Crops 31,000 (19) - 11,000 (15) 900 (2) 2,500 (4) 2,300 (4) Vegetables 24,000 (15) 7,000 (3) 3,200 (4) 2,000 (5) 3,500 (5) 2,700 (5) Dry Beasa- Melons & Watermelons 1,500 ( 1) - 1,200 (3) 2,000 (3) 900 (2) oilIseeds - 4,000 (5) 600 (2) 2,500 (4) - other 5,000 (3) 7,000 (3) - 2,500 (7) 2,000 (3) 900 (2) Total 157,500 (96) 194,000 (100) 53,000 (69) 33,900 (87) 48,200 (71) 4L 700 (83) Second Cron Fodder Crops 6,300 (4) . - 2 Rice - - 2,700 (5) Melons & Watermelons - 1,000 (1) - 3,500 (5) - Sesame - - 2 ( 7- 1,800 (3) Dry Beans & Peas . - 21,000 (27) 2,300 (6) 12,500 (18) - Winter Vegetables - 2,000 (3) 1,100 (3) 4,000 (6) 1,400 (3) Other _ ______-- 1.600 (4) LI - 3,200 (6) /2 Total 6,300 0 24,000 (31) 5,000 (13) 20,000 (29) 9,100 (17) Area Irrigable 157,500 194,000 53,000 33,900 48,200 45,700 Area CroDoed 163,800 (100) 194,000 (100) 77,000 (100) 38,900 (100) 68,200 (100) 54,800 (100) Cropoing Intens,Ly 1 04 1.0 1.15 1.42 1.20 Ll Sorgh.w. and maize. 12 Peanut (90n) and maize (2,300) d, TURKLY SLYIIAM 1R,(GATION PROJECT STACE II COMPLETION REPORT Economic Benefits Gross Value Production Net Value Total Net Value Yield Price (1968 TL) Production Cost Production Area Production (tons/ha) (TL/ton) (Tb/ha) (TL/ha) (TL/ha) (ha) (TL'000) Without Project Seed Cotton 0.95 2,100 2,000 1,800 200 38,700 7,740 Cereal 2 2 880 1,940 1,480 460 5,000 2,300 Melon & Watermelon 13.0 420 5,460 1,560 3,900 2,000 7 Total Net Value Production per ha 0.390 With Project (Full Development. 1985) Seed Cotton 3.5 2,100 7,350 3,800 3,550 22,900 81,300 Cereal 4.0 880 3,520 2,300 1,220 11,400 13,910 Rice 4.0 1,890 7,560 5,860 1,700 1,800 3,060 Citrus 30.0 1,000 30,000 11,500 18,500 2,700 49,950 Vegetables 25.0 810 20,250 7,300 12,950 1,400 18,130 melon & Watermelon 32 0 420 13,440 3,840 9,600 900 8640 Fodder Crops 10.0 600 6,000 2,200 3,800 2,300 8,740 Potatoes 30.0 680 20,400 5,850 14,550 1,400 20,370 Other Fruit 23.0 800 18,400 6,800 11,600 900 1 Subtotal 45,700 214,540 Second Crop Rice 5.0 1,890 9,450 5,260 4,190 2,700 11,310 Sesame 1.3 5,760 7,S00 3,870 3,620 1,800 6,520 Vegetable 10 0 810 8,100 2,910 5,190 1,400 7,270 Peanut t 1' 3,000 12,000 6,060 5,940 900 5,350 Maize 4.0 610 2,440 2,100 340 2,0 780 Subtotal 9,100 31,230 Total Net Value P'rodctLion per ha 5.380 Production loss, Land Lost for 740 Works (1,900 ha) Net BenetlL 245,030 Nut enet I~i5,360 TURKEY SEYHAN IRRIGATION PROJECT STAGE II COMPLETION REPORT Schedule of Annual Investment Expenditure (TL millions, in current prices) Irrigation DSI Drainage Equipment On-Farm Topraksu Consultants Total in Total in Constant Year Investments (O&M & Pipes) Works Equipment & Training Current Prices 1968 Prices 1969 32.29 0 5.12 0 1.59 39.00 38.39 1970 28.95 7.87 11.24 0 6.22 54.28 46.27 1971 73.41 35.07 22.14 0 7.86 138.48 99.77 1972 91.71 0 20.53 72.92 7.84 193.00 121.16 1973 80.19 0.09 28.59 3.76 5.57 118.20 61.28 1974 55.53 5.55 19.09 6.33 4.00 90.50 36.91 1975 51.75 0 27.53 4.74 0 84.04 29.47 1976 46.54 0 33.35 0.64 0 80.53 24.11 1977 100.49 0 8.69 1.04 1.18 111.40 26.65 1978 25.54 0 5.25 0 0 30.79 4.91 1979 0 0 22.20 0 0 22.20 3.51 Totals 593.00 49.58 203.75 89.43 33.08 492.43 Consultants & 1.18 training TURKEY SEYHAN IRRIGATION PROJECT STAGE II COMPLETION REPORT Farmgate Prices 1968 - 1985 1968 1972 1973 1974 1975 1976 1977 1978-2018 (Appraisal) (Projected in 1968 TL) Cotton -(Financial) 1,400 3,750 8,950 7,200 7,250 10,460 10,620 3,140 -(Economic) 1,520 3,200 5,800 6,000 5,500 8,300 6,700 2,100 Wheat -(Financial) 500 1,000 1,350 2,250 2,350 2,800 2,880 840 -(Economic) 500 1,210 2,500 3,490 3,140 2,870 2,810 880 Rice -(Financial) 1,000 2,750 3,900 4,300 4,750 4,960 9,000 1,490 -(Economic) 1,000 2,500 5,940 9,060 6,290 4,890 5,830 1,890 Citrus 500 1,100 1,800 2,950 3,000 3,000 3,650 - co Vegetables 600 850 1,350 2,500 2,600 2,400 3,500 810 Melons & Watermelons 200 - 1,250 1,300 1,350 1,400 1,890 420 Forage Crops 150 900 1,250 1,400 1,700 2,000 1,300 600 Potatoes - 900 1,500 1,700 1,850 2,250 - 680 Other Fruit 600 1,200 2,100 2,750 2,750 2,650 4,730 800 Sesame - 7,000 11,500 19,000 18,000 19,200 31,840 5,760 Maize -(Financial) - 1,150 1,800 2,850 2,900 3,250 3,560 150 -(Economic) - 950 1,660 2,210 2,080 2,160 2,480 610 Peanut 1,300 4,500 6,000 6,250 7,500 9,940 li,450 2,980 Dry Beans 1,400 3,500 4,500 6,500 7,800 8,250 - 2,480 M X ANNEX IV - 69 - Table-g- TURKEY SEYHAN IRRIGATION PROJECT STAGE II COHPLETION REPORT Price Deflators Used Year Index 1968 100 1969 101.6 1970 11.7.3 1971 138.8 1972 159.3 1973 192.9 1974 245.2 1975 285.2 1976 334.0 1977 418.0 1978 627.0 TURKEY SEYRAN IRRIGATION PROJECT STAGE II COMPLETION REPORT Economic Rate of Return Calculation - Comparison Revised Completion Estimate Appraisal Estimate Turkish Governmat Estimate Incremental Investment Net /1 Incremental Year Investments O&M Production Costs Benefit-- Investments 06M Production -----------------------------------------TL millions (constant 1968 prices)------------------------------------------ 1 (1969) 74.39 - - 122 9 47.82 - 2 46.27 - - 144 15 30.44 - 3 99.77 - - 146 24 62.23 - - 4 121.16 0.95 5.57 120 40 61.68 4.52 8.55 5 61.28 1.46 23.00 68 56 49.89 5.57 14.64 6 36.91 4.41 26.10 21 67 46.29 6.50 80.22 7 29.47 5.17 20.45 7 77 25.95 7.01 36.84 8 24.11 7.34 53.21 0 84 22.56 7.47 34.26 D 9 26.65 7.71 25.53 0 91 26.79 8.00 51.24 10 4.91 7.71 54.33 0 100 6.18 8.13 68.63 11 3.51 7.71 54.33 0 107 - 8.13 87.25 12 0 7.71 83.13 0 124 - 8.13 112.30 13 0 7.71 111.93 0 139 - 8.13 137.47 14 0 7.71 140.73 0 147 - 8.13 171.22 15 0 7.71 169.53 0 154 - 8.13 210.74 16 0 7.71 198.33 0 162 - 8.13 225.51 17 0 7.71 227.19 0 167 - 8.13 283.85 18 0 7.71 227.19 0 169 - 8.13 283.85 19 0 7.71 227.19 0 169 - 8.13 283.85 20 0 7.71 227.19 0 169 - 8.13 283.85 21 0 7.71 227.19 0 169 - 8.13 283.85 50 0 7.71 227.19 0 169 - 8.13 283.85 /l Net value production less annual O&M costs Economic Rate of Return- 15% 16 21%t LOWER SEYHAN PROJECT ORGANISATION SCHEME OF 'EXTENSION SERVICE DRCTOR DEPUTY DIRECTOR AND FOREIGN SECIALIST INFORMATION SPECIALST FIRST TEAM SECOND THIRO TEAM FOTH TEAM -5- 5 FO EM N 5 1 FRjEMN j5 FOREMEN 15 FOREMEN 2VILLAGlES 2] VILLAGE S21 VILLAGES 21 VkILLAGES r5 3 TURKEY: SEYHAN IRRIGATION PROJECT ORGANIZATION CHART -STAGE U CENTRAL COORDINATION COMMITTEE - ANKARA CHIEF COORDINATOR OF THE STATE PLANNING ORGANIZATION GENERAL DIRECTOR - D S.I., MINISTRY OF ENERGY AND NATURAL RESOURCES GENERAL DIRECTOR - TOPRAKSU, MINISTRY OF VILLAGE AFFAIRS REPRESENTATIVE OF THE AGRICULTURAL BANK REPRESENTATIVE OF THE EXTENSION SERVICE, MINISTRY OF AGRICULTURE SEYHAN PROJECT COORDINATION COMMITTEE -ADANA PROJECT COORDINATOR (Senior Official of the Stole Planning Organization) REGIONAL DIRECTOR-D.S.I.* REGIONAL DIRECTOR-TOPRAKSU* EXTENSION SERVICE* AGRICULT.RAL BANK Responsible for planning,design, Responsible for on-form Responsible for providing technical Responsible for supplying credits construction, operation and development in the project area. assistance and services to farmers to formers in the project area. maintenance of the project. in the protect area. PLANT COTTON PROTECTION RESEARCH INSTITUTE INSTITUTE A foreign engineering consulting firm would assist 0.S.I. in all irrigation arid drainage works and would provide technical assistance to TOPRAKSU and the Extension Services. 方 JANUARY 1969 IBR亂）ZIS4k