Turkey - State and Provincial Roads Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 18038 IMPLEMENTATION COMPLETION REPORT TURKEY STATE AND PROVINCIAL ROADS PROJECT Loan No. 3324-TU June 15, 1998 Infrastructure Sector Unit Europe and Central Asia Region This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EQUIVALENTS (as of June 1998) Currency Unit Turkish Lira (TL) 1 TL = US$0.000004 US$1 250,000 AVERAGE EXCHANGE RATES (per US$1) Jan 1991 Jan 1992 Jan 1993 Jan 1994 July 1996 July 1997 2750 5,113 8,628 14,000 83,000 175,000 WEIGHTS AND MEASURES Metric System ABBREVIATIONS AND ACRONYMS IBRD - International Bank for Reconstruction and Development SAR - Staff Appraisal Report PIU - Project Implementation Unit SOE - Statement of Expenditures MOF - Ministry of Finance ICR - Implementation Completion Report ERR - Economic Rate of Return MOPW - Ministry of Public Works and Settlements KGM - General Directorate of Highways TURKEY'S FISCAL YEAR January 1 to December 31 Vice President: Johannes F. Linn, ECA Country Director: Ajay Chhibber, ECCO6 Sector Director: Ricardo A. Halperin, ECSIN Sector Leader: Eva Molnar, ECSIN Team Leader: Mirtha Pokorny, Senior Operations Officer, ECSIN  FOR OFFICIAL USE ONLY IMPLEMENTATION COMPLETION REPORT TURKEY STATE AND PROVINCIAL ROADS PROJECT (Loan 3324-TU) Contents Preface ...................i Evaluation Summary .................................... ii Part I. Project Implementation Assessment .....................................1 A. Project Origin and Objectives .1 B. Achievement of Project Objectives .3 C. Implementation Record and Major Factors Affecting the Project .5 D. Project Sustainability .12 E. Bank Performance .12 F. Borrower's Performance .13 G. Assessment of Performance and Outcome .13 H. Future Operations .14 I. Key Lessons Learned .14 Part II. Statistical Tables ..........15 TABLE 1: SUMMARY OF ASSESSMENTS .1 5 TABLE 2: RELATED BANK LOANS/CREDITS ....................................................................17 TABLE 3: PROJECT TIMETABLE .17 TABLE 4: LOAN DISBURSEMENTS: CUMIJLATIVE ESTIMATED AND ACTUAL .18 TABLE 5: KEY INDICATORS FOR PROJECT IMPLEMENTATION .18 TABLE 6: KEY INDICATORS FOR PROJECT OPERATION. 1 8 TABLE 7: STUDIES INCLUDED IN PROJECT .18 TABLE 8A: PROJECT COSTS ................................ 19 TABLE 8B: PROJECT FINANCING ......................................... 19 TABLE 9: ECONOMIC COSTS AND BENEFITS ......................................... 20 TABLE 10: STATUS OF LEGAL COVENANTS; ...................................... 21 TABLE 11: COMPLIANCE WITH OPERATIONAL MANUAL STATEMENTS ....................... ...................... 21 TABLE 12: BANK RESOURCES: STAFF INPIJTS .......................................................22 TABLE 13: BANK RESOURCES: MISSIONS ......................................... 22 Appendices: A. Civil Works Bid Evaluation .............................................. 23 B. Borrower's Contribution to the ICR .............................................. 24 Map IBRD 29618 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.  IMPLEMENTATION COMPLETION REPORT TURKEY STATE AND PROVINCIAL ROADS PROJECT (Loan 3324-TU) Preface This is the Implementation Completion Report (ICR) for the State and Provincial Roads Project in Turkey, for which Loan 3324-TU in the amount of US$300.0 million equivalent was approved on May 14, 1991 and made effective on November 8, 1991. US$50.0 million of the loan amount was canceled on August 31, 1993. The loan was closed on December 31, 1997, compared with the original closing date of June 30, 1997, and is fully disbursed. Cofinancing for the project of US$15 million was provided by the Government of Switzerland, and US$14,987,021.39 of this amount has been disbursed. The ICR was prepared by Robert H. Nooter, Consultant for ECSIN, and reviewed by Mirtha Pokorny, Senior Operations Officer, and Eva Molnar, Transport Sector Leader for ECSIN. The borrower provided comments that are included as Appendix B to the ICR. Preparation of this ICR is based on mraterial in the project file. The borrower contributed to preparation of the ICR by providing summary data including current estimated Economic Rates of Return, and submitted comments contained in Appendix B.  - 11 - IMPLEMENTATION COMPLETION REPORT TURKEY STATE AND PROVINCIAL ROADS PROJECT (Loan 3324-TU) Evaluation Summary Introduction 1. The State and Provincial Roads Project, supported with a World Bank loan of US300.0 million, was the third World Bank project directed to the improvement of Turkey's national road network and the institutional strengthening of the General Directorate of Highways (KGM). Preparation was started in 1990, based on an agreed perception that improvement in Turkey's road network was critical to its economic development. Due to macro-economic pressures, budget financing for road maintenance and upgrading had fallen seriously behind necessary levels. The project was prepared in eleven months, and was approved by the World Bank Board of Executive Directors on May 14, 1991. Project Objectives 2. The project objectives were to (i) keep transport costs low in Turkey by ensuring adequate renewal and maintenance of the road network; (ii) reduce the backlog of road strengthening; (iii) continue the downsizing, modernization and improvement of the management of KGM's equipment fleet; (iv) strengthen and modernize the planning and design capabilities of KGM; and (v) improve the management and safety of the road system through reforms, modernization and training. 3. The project objectives were realistic in terms of the borrower's capabilities, although the first objective was stated more broadly than was consistent with the work to be carried out. The strategy underlying the project design was supportive of Turkish development efforts in view of the importance of the road network in meeting Turkey's transport requirements. 4. Most of the project objectives were met. Under (i) and (ii) above, more than 950 km of roads were improved, thereby assisting in keeping transport costs low and reducing the backlog of road strengthening. The economic rate of return of 14.0 percent is considerably less than the 30 percent estimated during project preparation, but is still satisfactory. For (iii), the purchase of new equipment under the project made possible the retirement of old, outdated machines. For (iv), the purchase of modern design software and research and survey equipment improved KGM's design and planning capabilities. Object (v), however, was only partially achieved due to the delay in the training component and the absence of any effective activities to improve  KGM's management systems, while some progress was made in road safety through the provision of materials and equipment for improved road markings, traffic signs and guard rails. Implementation Experience and Results 5. Overall, the project successfully achieved its objectives. More than the originally planned 950 km of road improvements were iraplemented, equipment needed to upgrade KGM's maintenance equipment and to improve research, design and training facilities were procured and put into service, and some road safety improvements were made. 6. 31 civil works contracts for road upgrading were let, some of which are still under implementation, for a total bid value of US$451.2 million compared to an estimated US$312 million included in the Staff Appraisal Report. The original contracts were increased during implementation to a total estimated cost of US$660.7 million for improving 1844 km of roads, which is a 46 percent increase over the original contract prices. The increases in the original contracts were the result of extensions of the length of the roads being improved, the addition of by-passes, connecting roads and other works, and the upgrading of some roads from two to four lanes. US$381.9 million had been spent on the 31 contracts as of project closing, of which the Bank loan has financed US$157,853,351.30, including US$25 million provided for financing works started in a previous IBRD-funded highway project. Thus, US$278.8 million additional work must be financed from KGM's own or from other resources in order for all 31 contracts to be completed, based on current cost estimates. 7. The estimate of the cost of road upgrading at appraisal was US$360,000 per km, compared to an estimated cost of US$358,300 per km at project closing, based on KGM's estimated cost to complete all 31 contracts. The estimated average economic rate of return of 14.0 percent calculated for the 31 contracts for 1815 km of road improvements, based on the estimated cost to complete the unfinished works, is significantly less that the 30 percent return estimated at project appraisal but still acceptable. This can be attributed in part to a reduction in vehicle operating costs related to a reduction in fuel costs. Of the 31 contracts, 13 with a cost of US$209 million for improving 633 km of roads have actually been completed or are nearly completed. For these 13 contracts, the average economic rate of return is estimated to be 15.1 percent. These rates of return and average costs per km are not readily comparable, however, as the final works in many cases differ substantially from the original design. 8. Bidding for procurement of US$50 million of bitumen to be used for the road upgrading resulted in only one responsive bidder due to restrictive bidding specifications insisted on by KGM at the time the bidding documents were prepared. KGM wanted to reject the responsive bid and rebid on the basis of revised specifications. However, the Bank advised KGM that, in fairness to the successful bidder, the responsive bid should be accepted or else the US$50 million allocated in the loan for bitumen procurement would be canceled on the basis of misprocurement. Given these alternatives, KGM opted for the reduction in the loan amount. 9. A review of the bid prices received for the civil works contracts revealed that during one period of bidding, the spread between the highest and lowest bid varied from 52 to 67 percent of KGM's estimated price, whereas during another period, bids were from 87 to 95 percent of the  - iv - estimated price. A possible explanation for this difference, which had a substantial impact on the cost of the works, was that a large amount of civil works was put out to bid during the second period, so that contractors were not eager to bid low prices. 10. Project conditionality and legal covenants were generally adhered to satisfactorily. Progress reports and audits were submitted on time and in good order, and project administration was excellent. Bank staff also generally performed well, although in retrospect it is not clear why the Bank staff finally accepted the restrictive shipping requirements for bitumen procurement proposed by KGM, which ultimately contributed to the misprocurement referred to in para. 8. The only other major problem related to delays in the availability of Government counterpart funding during the later years of the project, which delayed the work of some contractors, and can be explained by overall macroeconomic problems. KGM staff performed well, although the amount of civil works included in the project appears to be excessive for the funds available and the additions to the original contracts are high compared to good procurement practice. Summary of Findings, Future Operations, and Key Lessons Learned 11. The overall project outcome is rated as satisfactory, in spite of the problems of bitumen procurement, occasional delays in counterpart funding, and the reduced economic rate of return. The cost of the civil works component was in line, on a unit basis, with the appraisal estimate. The estimated average 14.0 percent economic rate of return at project closing is acceptable. However, the amount of road improvements added to the project could lead to delays in completing the civil works contracts now that Bank funding is no longer available. Project sustainability seems assured since KGM is a competent and dedicated organization with long experience in carrying out road improvement activities. 12. Some lessons that can be leamed from this project are: (a) Highway contracting authorities should be aware of road construction market conditions, and to the extent possible let a level volume of work in each year to avoid either flooding the market, resulting in a high price level for contracted work, or a shortage of work that will make it hard even for good contractors to survive; (b) It is generally preferable to let the civil works contractors be responsible for all of the inputs required, in order to avoid divided responsibility for quality and delivery; (c) Care should be taken to avoid restrictive bidding specifications that will result in lack of effective competition; and (d) Additions to civil works contracts should be limited to a reasonable percentage of the original cost; major additions should be subject to economic analysis before deciding to proceed, and should be put out for bid as separate contracts.  IMPLEMENTATION COMPLETION REPORT TURKEY STATE AND PROVINCIAL ROAD PROJECT (Loan 3324-TU) Part I. Project Implementation Assessment A. Project Origin and Objectives 1. Transport is vital for the economic development and integration of Turkey. Its freight transport demand includes not only domestic fireight and import/export traffic, but also transit traffic between Europe and the Middle East. The backbone of Turkey's transport system is its 360,000 km of roads, 60,000 of which are state and provincial roads constructed and maintained by the central government, and 300,000 of rural roads. Road transport, operated entirely by the private sector, handles over 80 percent of the fieight and 95 percent of the passenger demand. The Ministry of Public Works and Settlements (MOPW), through its Directorate of Roads (KGM), is responsible for the development and maintenance of the state and provincial roads. 2. Due to macro-economic pressures, budget allocations for state and provincial roads decreased continuously in real terms between 1984 and 1988, and were only two-thirds of the level of the early 1980s by the end of that period. As a result, strengthening and upgrading programs fell seriously behind, and the need for increased maintenance became acute. Also, many road sections designed in the 1950s and 1 960s were not designed for today's axle loads. The Government recognized the importance of increasing allotments for roads, both by increasing budget allocations and by requesting World Bank assistance for its road program. 3. KGM was already a mature organization at project inception. It had carried out two previous World Bank-financed projects satisfactorily. The Project Completion Report for the Second Highway Project notes: "Overall, the project is rated as satisfactory, its sustainability as likely, and its institutional impact as substantial". By the time of implementation of the State and Provincial Roads Project, KGM had developed methods for evaluating the economic feasibility of road investment projects on the basis of the 'World Bank HDM model, and was successfully performing its own design and supervision for road construction and maintenance work. 4. Road safety was still a serious problem in Turkey in 1991, in spite of road safety components in both of the previous Highway Projects. In 1989, there were 104,000 accidents injuring 80,000 people and causing 6,300 deaths. The incidence of fatalities on Turkish highways was about 6 to 10 times higher than in other European countries. The fatality rate on state and provincial roads, however, was falling, and in 1989 was less than half of its level in the early 1970s. 5. With this background in mind, the project objectives were to (i) keep transport costs low by ensuring adequate renewal and maintenance of the road network; (ii) reduce the backlog of  -2 - road strengthening; (iii) continue the downsizing, modernization and improvement of the management of KGM's equipment fleet; (iv) strengthen and modernize the planning and design capabilities of KGM; and (v) improve the management and safety of the road system through reforms, modernization and training. 6. The project description included the following components: (a) strengthening about 650 km of high priority sections of existing state roads; (b) improvement of about 300 km of selected provincial roads; (c) procurement of equipment and materials for road maintenance, research, planning, survey and design, reorganization, computerization and training; (d) improvement of road safety; (e) training for KGM's staff; and (f) consulting services. 7. Project conditionality included: (i) criteria for selecting the road sections to be improved under the project, based on minimum traffic load for state roads and minimum economic rates of return; (ii) appropriate feasibility studies for road sections to be improved under Bank financing would be submitted to the Bank for review before construction bids are invited; (iii) KGM would extend traffic counting to provincial roads in order to improve the quality of its planning decisions and enhance the technical and cost effectiveness of its design work; (iv) agreement on procurement procedures and bidding documents; and (v) provision of counterpart funding as required to carry out the project. This was considered to be a sector project, since the project included financing for a multi-year program of works with only the first year program of road strengthening specifically identified at the time of appraisal. However, none of the conditionality usually applied to sector loans, such as annual reviews of the sector investment program, were specified. 8. The project objectives were realistic in terms of the borrower's capabilities. However, the first objective, to keep transport costs low by ensuring adequate renewal and maintenance of the road network, was not fully reflected in the more limited itemized project components and conditionality, which only addressed work to be carried out on a limited portion of the state and provincial road network. 9. The strategy underlying the project design was supportive of Turkish development efforts, although the Staff Appraisal Report (SAR) did not directly relate the strategy to the Bank's Country Assistance Strategy of that time. The prospects for satisfactory project implementation were considered good considering KGM's past implementation experience with World Bank projects. The only identified risk, which subsequently did materialize, was the possible delay in availability of counterpart funds, although past performance in this respect was observed to be quite good.  - 3 - B. Achievemeint of Project Objectives 10. Overall, the project successfully carried out the project components identified in the SAR, and generally met the project objectives. The specific objectives are discussed below. To Keep Transport Costs Low 11. The first objective, to keep transporft costs low by ensuring adequate renewal and maintenance of the road network, was too broadly stated to be measured during project implementation or at completion. For example, there was no reporting on the overall level of maintenance of the national road network, condition of the overall road network that would indicate possible transport costs, and no conditionality relating to the relationship between spending on maintenance and for new investments. The planned improvements of 950 km of state and provincial roads outlined in the project description were in fact carried out and exceeded, and no doubt helped to keep transport costs lower than they would have been without these improvements However, this does not in itself indicate whether the objective as stated was met. This is more a shortcoming of the wording of this objective, however, than a shortcoming of the project design or implementation. To Reduce the Backlog of Road Strengthening 12. The components relating to this objective called for the strengthening of about 650 km of high priority state roads and 300 km of provincial roads, with an estimated average Economic Rate of Return of 30 percent. The cost of these improvements was estimated at US$390,000 per km for strengthening of state roads and US$290,000 per km for improving provincial roads, for an overall average of US$360,000 per km. 13. 31 civil works multi-year contracts were let for improvements of state and provincial roads at an original bid value of US$451.2 million compared to US$312 million of such works included in the Staff Appraisal Report. During execution of the contracts, additions were added to these contracts, including extending the length of the roads, constructing city by-passes, the addition of connecting roads, and upgrading some road sections from two to four lanes. At project closing, nine contracts had been totally completed and four more were nearly complete. At the end of 1997, KGM calculated the estimated cost to complete all 31 contracts to be US$660.7 million, or 46 percent more than the original bid value, which included improvements for 1875 km of roads. A total of US$381.9 million had been expended on the work carried out as of the end of CY1997, leaving an estimated US$278.8 million required to complete the 31 contracts. 14. The overall average cost per km, based on KGM's estimated cost to complete the 1875 km, is US$352,400 per km, very close to the original estimate in spite of the substantial cost increases over the original bid prices. The project appears to have met its objective of reducing the backlog of road maintenance, but the final accounting must await the full completion of all 31 contracts to know exactly what was accomplished and at what actual cost.  - 4 - 15. The estimated ERR at time of project approval was an average of 30 percent for the civil works component of the project. KGM has made an estimate of the ERRs for the 31 civil works contracts, as shown in Table 9, covering 1875 km of road improvements, based on the estimated cost to complete the contracts and updated traffic figures. The weighted average of the ERRs for the 31 contracts is 14.0 percent, considerably less than the 30 percent estimate made at the time of appraisal but still acceptable. This is attributable in part to a reduction in vehicle operating costs related primarily to a reduction in fuel costs, and to the substantial changes made to the contracts during execution. 16. A review of the results for the 13 completed or nearly completed contracts shows an economic rate of return of 15.1 percent, also lower than the appraisal estimate but still acceptable. Three of these contracts have ERRs ranging from 0 to 7.2 percent, in part due to increases in the original contract prices and in part the result of lower than anticipated traffic increases. These rates of return and average costs per km are not readily comparable, however, as the final works in many cases differ substantially from the original design. Continue the Downsizing, Modernization and Improvement of the Management of KGM's Equipment Fleet 17. New equipment for routine and emergency maintenance purchased with loan funds has improved the efficiency of KGM's operations, and has enabled KGM to dispose of large quantities of unserviceable equipment or machines that were being kept well beyond their useful life, thus reducing maintenance costs. The new equipment has also improved the effectiveness of KGM's road maintenance services, including winter maintenance. The training program carried out by the equipment suppliers in connection with the purchase of the new equipment supported these objectives and assured that the staff was adequately trained for its use. Therefore, this objective was achieved. Strengthen and Modernize the Planning and Design Capabilities of KGM 18. The loan financed computer software, including state of the art CAD applications, which has greatly improved local road design capabilities. The purchase of 50 "total stations" (theodolites) have improved the volume, accuracy and speed with which topographic data can be fed electronically into the CAD program. Also, research and survey equipment financed from the loan has expanded KGM's capacity to carry out technical research and road surveys (including the surveys of the provincial road network). Therefore, this objective was also achieved. To Improve the Management and Safety of the Road System Through Reforms, Modernization and Training 19. The project's principal contribution to road safety was through the provision of materials needed to improve accident black spots, install guard rails, install portable weighing devices and improve traffic signs and road markings. The attention to road safety included in the project did, however, stimulate KGM to establish a working group that has prepared a detailed traffic safety action plan as part of the preparation of a follow-on project (Road Improvement and Traffic  - 5 - Safety, financed by IBRD Loans 4048/49-TU). The plan is very comprehensive, addressing the issues of physical improvements, enforcement, and education. A bill to create an interministerial traffic safety council to coordinate the implementation of the plan has recently been approved by the Parliament. 20. Training of KGM staff planned for the State and Provincial Roads Project was delayed due to Government restrictions on travel abroad, and will now be carried out under the new project. While institutional strengthening under the State and Provincial Roads Project was limited to the provision of physical inputs, the project did lead to the more comprehensive institutional strengthening included in the new project. Nevertheless, the objective of improving management can not be said to have been achieved under this project. C. Implementation Record and Major Factors Affecting the Project 21. The overall implementation record was good, with excellent project management, timely reporting and audit submissions, a good dislbursement record, and an acceptable quality of works. The project's three problem areas include the availability of counterpart funding from about the mid-point of the project onward, a major procurement problem relating to the purchase of a large quantity of bitumen, which resulted in the cancellation of US50 million of the loan amount, and the excessively large additions made in the civil works contracts during contract implementation. Civil Works 22. The original loan allocation included an amount of US$150 million for Civil Works. This amount was intended to finance 50 percent of the contracts for works to improve 650 km of state roads and 300 km of provincial roads. Contractors were selected on the basis of competitive bidding using the ICB procedures outlined in the World Bank Guidelines for Procurement. Contractors were to provide all of the inputs necessary to carry out the work except the bitumen, US$50 million of which was to be financed from the loan and provided by KGM to the contractors. Selection of the sections of roads to be put out to bid during the first year was agreed at appraisal, and the criteria for selection of additional road sections were agreed, including an adequate Economic Rate of Return. Contracts were expected to be an average of about US$10 million each, and to take three to four years to complete. Selection of the successful bidder was to be made by a Bidding Commission established for this purpose, and in each case the qualified low bidder was selected. Initially, there were some delays while KGM and the Bank sought, and eventually reached, agreement on procurement documents and an inflation adjustment clause. While ICB bidding procedures were used, very few foreign contractors bid on the works, and all contracts except one were awarded to Turkish contractors. 23. Design of road improvements was carried out by KGM or by local design firms financed under the loan. Design included improving the pavements and base courses, minor improvements in alignments, widening of the roadways in some cases, improvements in drainage, widening or replacement of bridges and structures and intersection redesign where appropriate. Supervision of the contractors' performance was carried out by KGM with its own staff.  - 6 - 24. US$25 million of the Civil Works allocation was used to finance roads under retroactive financing provisions of the loan for works let under a previous Bank project. Also, Swiss Government financing of US$15 million was made available in support of the civil works component, and financed 50 percent of the contracts for improvement of provincial road sections in lower income regions. The Swiss Government financing was administered and disbursed by the Bank. By 1995, budgetary pressures in Turkey made it difficult for KGM to meet its commitments to finance its 50 percent share of the civil works contracts, and the Government requested that the loan percentage be increased to 70 percent. Instead, the Bank agreed, in a letter amendment dated January 22, 1996, and accepted on February 16, 1996, to increase the loan percentage to 60 percent. 25. In view of Turkey's high inflation rate, the civil works multi-year contracts contained an inflation adjustment clause that provided for an upward adjustment in the contract unit prices at the beginning of each year. This clause was agreed between the Government and the Bank, and was considered fair to both the contractors and KGM. There were some reports, however, that some contractors were reluctant to carry out work in the last half of the year when costs had risen, preferring to wait for the following year's adjustment before proceeding. KGM now makes inflation adjustments regularly throughout the year, which should resolve this problem. 26. At project closing, 13 of the 31 civil works contracts had been completed at a cost of US$209 million for improving 633 km of roads, although the total value of work carried out on all 31 contracts as of December 31, 1997 was US$381.9 million. US$157,853,351.30 of the Bank's loan (less the US$25 million used for retroactive financing of the previous highway project) was disbursed from the Civil Works category for financing part of this amount. 27. The overall cost to bring all 31 contracts to completion was estimated by KGM in February 1998 to be US$660.7 million, a 46 percent increase over the US$451.2 million original bid value. The Government requested and the Bank agreed that up to US$25 million of the financing available from Loans 4048/49-TU could be used for financing incomplete portions of Loan 3324-TU, including goods, works and training. Even if all of the US$25 million were used for this purpose, there would still be an estimated US$253.8 million required to complete the unfinished work. This may be possible since KGM's total budget for 1997 was over US$1 billion equivalent, but the difficulties that KGM had in providing the counterpart for its share of project financing during the past several years raises some doubt as to the feasibility of this solution. KGM has assured the Bank that it now has the resources needed to finance the completion of these contracts. 28. An analysis of the bid prices for the 18 contract bids that were available in the files revealed some anomalies that KGM might want to explore further, since it could lead to cost savings in future contract work of this kind. As shown in Appendix A, all of the six bids received from July 1992 to June 21, 1993, showed a wide diversity of prices, with a spread between the highest and lowest price being from 43 to 81 percent. During this period, the lowest price was from 52 to 67 percent of the KGM estimated price for the work as designed. From June 25, 1993 to February 1994, the spread between the highest and lowest of the ten bids received were between 3.8 and 10.6 percent, and the lowest bid was from 87 to 95 percent of the  -7 - KGM estimated cost. The final two bids, in May 1994, once again had a significant spread in bid prices and a significant discount from the KGM estimate. Thus the prices received in the earlier period were an average of 62 percent of KGM's estimated cost, compared to 91 percent for the June 1993 to February 1994 period. 29. One possible explanation for this bidding behavior is that a large amount of civil works was put out for bid during the June 1993-February 1994 period, so that contractors were not eager to bid low prices. If this is the case, KC;M may want to give more attention to spreading its work more evenly from one year to another. Bitumen 30. The original project design provided for the loan to finance 100 percent of the foreign exchange cost of US$50 million of bitumen, which was listed as a separate line item in the Disbursement Schedule. The bitumen was to be purchased by KGM and made available to the contractors carrying out the civil works contracts. Bidding documents were approved during project preparation on the basis of using imported bitumen since KGM did not at that time consider local bitumen to be of satisfactory quality. 31. On July 26, 1991, approximately one month after the loan signing, the World Bank Task Manager sent a Fax to KGM inquiring whether locally produced bitumen might be acceptable. He informed KGM that he had been led to believe that the local Government-owned refinery Tupras might be able to produce an acceptable quality of bitumen with a significant saving in transport costs. KGM replied with a request to proceed with the procurement of imported bitumen. On August 21, 1991, the Task Manager again raised the question of whether it was necessary to use imported bitumen since local bitumen should result in lower costs, and also whether it was necessary to purchase the entire four year supply in a single contract. He also questioned the specification that required the supplier to have a vessel on "time charter" for four years continuously. 32. In a reply received September 17, 1991, KGM stated that "Bitumen supplied by Tupras complies with the ASTM test methods used in our research laboratories and Tupras can also participate in this tender in the coverage of World Bank regulations". It further stated that a better price could be obtained in a single large tender, separate tenders might disrupt the delivery schedule, and the supplier could use the same ships for other jobs as long as the KGM monthly program was met. The actual specifications wvere prepared on the basis of "imported bitumen" in spite of the stated intention in the letter of September 17, 1991 to allow local supplier Tupras to compete, and required delivery in nine ports over a four year period. The specifications also required that the bidder own, or have on charter, special asphalt ships of sufficient quantity and capacity to transport the bitumen to the nine specified ports, and that it had successfully transported at least 100,000 metric tons of asphalt in programmed deliveries to various ports in Turkey under one contract. KGM insisted that these shipping requirements were necessary to assure timely delivery. 33. KGM proceeded to seek bids on the basis of the specifications outlined above. Only three bids were received, one of which was disqualified because it did not provide a proper letter  - 8 - of guarantee. A second bid, from an American supplier, met all of the specifications at a price of US$61 million. A third bid from Tupras was received at a price of US$36 million on the basis of using bitumen from its local refinery. 34. On the basis of the bids received, KGM requested a "no objection" for an award to Tupras in a Fax dated July 31, 1992, as KGM had concluded that Tupras could in fact meet the material specifications required for the civil works contracts. However, as the specifications clearly stated that imported bitumen was required, and as Tupras did not meet the stringent shipping requirements, the Bank could not agree to the KGM recommendation. 35. After a lengthy review, the Bank wrote to the Government on January 27, 1993, stating that "a failure to sign a contract with the responsive bidder would constitute misprocurement, and would lead to a cancellation of the funds allocated under this category of the Loan Agreement". The Bank's letter explained its decision as follows: "The present bids were invited under a set of documents prepared by KGM, under the Bank's International Competitive Bidding guidelines. This exercise must be properly concluded. "Paragraph 2.60 of the Procurement Guidelines specifies the conditions under which all bids can be rejected and requires consultation with the Bank before doing so. The paragraph permits cancellation and rebidding if the lowest bid exceeds the cost estimate by a substantial margin. However, the estimates shown in the Staff Appraisal Report (SAR, page 18) show a budget for bitumen of $62.6 million including contingencies of $6.6 million. Therefore, rejection of the lowest evaluated responsive bidder and cancellation of bids on these grounds, would not be permissible. "Exceptionally, where there has been evidence of a clear misunderstanding between the Bank and the Borrower, declarations of misprocurement without cancellation of funds have occasionally occurred. However, this exception does not apply because we have been unable to identify any evidence of misunderstanding regarding the intent of KGM at the time of the bid. "In consultation with our Legal Department and procurement advisers, we have carefully examined this issue and have concluded that KGM intended this bid to apply to foreign bidders and to imported bitumen only, from the time that the new technical specification was first discussed in 1986. This conclusion is based on the following facts: "KGM did not accept the Bank's suggestions to use local bitumen in the first instance, backed up by imports if and when required, but provided firm technical arguments to support its wish to proceed with a bid for imported bitumen; "In addition to the requirements of the 1986 Technical Specifications for bitumen, KGM added to the 1992 document, very specific and demanding language requiring extensive previous experience of bitumen shipment, and either evidence of ownership or long term charter for specialized bitumen delivery tankers (Clauses 6.2 and 6.4). KGM argued at the time that these were essential performance criteria to assure satisfactory supplies and these restrictive criteria were, therefore, accepted by the Bank.  -9 - "The Loan Agreement which specifies "imported bitumen" and specifies reimbursement of 100% of foreign exchange cost was reviewed in detail by the KGM delegation at loan negotiations and again when changes were made to the Loan Agreement in 1992. We have concluded that this wording reflects fully the b,orrowers intention at the time." 36. In a reply dated August 27, 1993, the Government reiterated its preference for either amending the Loan Agreement so as to enable local bidders to participate in the bidding, or to transfer the amount of the bitumen to the civil works category. However, in view of the Bank's rejection of these alternatives, the Government reluctantly chose the cancellation of the US$50 million rather than place the order with the one responsive bidder. It also agreed to provide the bitumen for the project from its own resources. The Turkish Treasury held an inquiry into the matter, and the head of KGM's Asphalt Division was removed from his post. Local bitumen was procured by KGM, and presumably met the required specifications. The loss of funding, however, added to the difficulties of financing the civil works contracts when budgets came under pressure in 1995 and 1996. 37. In a letter dated November 6, 1996, the Ministry of Public Works and Settlement wrote to the President of the World Bank in response to the President's message to all of the World Bank's clients at the time he took office. The Ministry's letter reviewed the background of the bitumen misprocurement, and requested that the Bank reconsider its "unjust decision" and restore the US$50 million to the Loan. On December 6, 1996, the Director for the Southeastern Europe Department replied that it was not possible to reverse the cancellation, but "when the Project Completion Report comes to be written for this project in 12-18 months, that the Bank's actions in this incident (would) be satisfactorily explained". Review of the Bitumen Misprocurement Issue 38. The paragraphs above set out the facts in this case as they are known to the Bank. The relevant paragraph of the Bank's Procurement Guidelines dated May 1985, which were in effect at the time of the bitumen procurement, are reproduced below to give a full sense of their meaning.: "2.60 Bidding documents usually provide that Borrowers may reject all bids. However, the Borrower should consult with the Bank beifore holding any negotiations, or rejecting all bids or soliciting new bids. All bids shall not be rejected and new bids invited on the same specifications solely for the purpose of obtaining lower prices, except in cases where the lowest evaluated bid exceeds the cost estimates by a substantial amount. In such cases, the Borrower may, as an alternative to rebidding, negotiate with the lowest evaluated bidder (or failing a satisfactory response, with the next lowest bidder) to try to obtain a satisfactory contract. Rejection of all bids is also justified when bids are not substantially responsive or there is lack of effective competition. If all bids are rejected the Borrower should review the causes justifying the rejection and consider making either revisions in the specifications or modifications in the project or both before inviting new bids." 39. As is obvious from the above, the Guidlelines require interpretation and judgment. The Bank's reasoning in this case is outlined in the letter of January 23, 1993, quoted above. The  - 10- Bank was correct to find that the Tupras bid was unresponsive to the bidding specifications since Tupras was not offering imported bitumen and did not meet the shipping requirements. In retrospect, however, it is unclear why the Bank approved the original bidding documents with shipping requirements so restrictive that few suppliers (indeed, perhaps only one supplier) could meet them, and resulted in bidding that produced only one responsive bidder. The Bank had an opportunity to remedy this by agreeing to the KGM request to reject all bids and to retender on the basis of revised specifications, as contemplated in para. 2.60 above, since bidding resulting in only one responsive bidder would seem to justify a finding of "lack of effective competition". However, in accordance with its consistent practice, the Bank decided that it was not fair to the one responsive bidder to change the rules of the game after bidding had taken place, especially in view of KGM's insistence on restricting the bidding to imported bitumen when the specifications were prepared, and when the Bank was not convinced that the local bitumen was of adequate quality in view of KGM's earlier rejection of this option. 40. At this stage, the only course is to leam from the experience for the next time. Some possible lessons include: - It is generally preferable to let contractors provide all of the materials required to carry out their work, in order to avoid divided responsibility for both quality and delivery. - Specifications should indicate the quality standards required rather than the source (i.e. "imported"). - Specification should not be more restrictive than necessary since the resulting lack of competition can be costly. 41. In its contribution to the ICR (see Appendix B), KGM requested that the Bank revise its decision regarding the cancellation of US$50 million of the loan amount. In a letter dated June 1, 1998, Mr. Ricardo Halperin, Director of ECSIN, responded to this request as follows: "While the draft ICR suggests that the Bank may have had some options in terms of the decisions that could have been taken, the ICR also cites KGM's stand on the quality of the local bitumen as a factor in the decision that was eventually taken by the Bank with respect to rebidding. The Bank's decision was heavily influenced by the information provided to the Bank by KGM during loan negotiations that only imported bitumen could meet the specifications and that the restrictive shipping specifications were necessary to ensure delivery of a satisfactory product. This led to agreement on these specifications during the loan negotiations and for the Loan Agreement to provide only for the financing of imported bitumen. "Subsequently, the Task Manager on two occasions prior to bidding sent Faxes to KGM raising the possibility that local bitumen might be acceptable and should result in lower costs (see Para. 31), but in each case KGM preferred to retain the option of relying on imported bitumen. Thus, when KGM reversed its position with respect to local bitumen after the bidding and approached the Bank for a revision of the specifications, the Bank had no grounds to believe that local bitumen could meet the requirements. I trust that you will agree that under these circumstances, allowing for rebidding would have been inappropriate."  Equipment 42. The Loan Agreement allocated US$85 million for the procurement of equipment comprising 300 multi-purpose maintenance trucks, 100 dump trucks, 50 front end loaders, 30 dozers, 16 asphalt distributors, and one slurry seal plant. This allocation also included equipment for technical research, monitoring and testing the condition of pavements, photogrammetric plotters and surveying equipment, and computers and software for pavement and bridge design and management. Also, equipment for KGM's Training Center, including laboratory equipment and visual aids, was included. During implementation, the Equipment category was increased to US$95 million by a transfer from the Unallocated category, but the final amount actually utilized was US$88,263,251.49. 43. Generally, the equipment procurement proceeded as planned and without difficulties. A total amount of US$75,015,063.47 was used to purchase equipment for the Department of Equipment and Supply, and the balance was used for the other purposes outlined above. Consultants Services and Training 44. Local consultants were engaged to carry out the design and prepare the bidding documents for the civil works component of the project. US$3,883,397.21 was used for this purpose, compared to the initial allocation of US$2.0 million. Performance of the design consultants was satisfactory. 45. An extensive training program was prepared, to be carried out at a U.S. university, to train trainers for the Training Center. However, due to fiscal constraints, the Government prohibited all overseas training and the program had to be canceled. As a result, the project's only contribution to training, other than the training provided by the equipment suppliers, was the equipment purchased for the Training Center. Project Conditionality 46. Project conditionality was met satisfactorily. In some cases, there were extended debates between the Bank and KGM regarding items such as the inflation adjustment clause and whether acceptance of the lowest qualified bid should be the basis for the award of civil works contracts. However, these issues were ultimately worked out consistent with World Bank procedures and without project delays. Progress reports and audit reports were submitted on time and in good order. The only serious problem related to the lack of Government counterpart funding during the later years of the project, which delayed the work of some contractors. This may be explained by the macroeconomic problems that Turkey was facing at that time, which required tight control over Government expenditures.  - 12- Co-Financing 47. The original project did not contemplate co-financing. However, when Swiss Government financing of US$15.0 million became available, the Government and the Bank agreed to accommodate it within the project, as described in para. 21. This financing was administered by the Bank, and was used to finance 50 percent of several road sections under the same conditions and procurement procedures applying to the Bank-financed sections. This work was executed with no difficulties, and allowed an expansion of the project accordingly. As of this date, US$14,987,021.39 of the US$15.0 million has been disbursed. D. Project Sustainability 48. KGM is a competent and dedicated organization that carries out its responsibilities in a serious and conscientious manner. Thus the institutional changes supported by physical inputs financed through the project, such as improved planning, design, economic evaluation and establishment of the Training Center can be expected to be sustained in future years, just as the benefits of the earlier two highway projects have been sustained during the present project period. 49. The physical improvements made in upgrading state and provincial roads will have a long lasting effect since maintaining the upgraded roads will be less expensive than the cost of the initial upgrading. Again, the sustainability of these changes will be enhanced because KGM is an organization that can be expected to provide the routine and periodic maintenance that are needed to get maximum economic benefit from these improvements. E. Bank Performance 50. In general, the Bank's performance was satisfactory. Preparation was accomplished in a minimum amount of time, assisted by the fact that this was a follow-on project. During implementation, prompt attention was given to the many issues that needed to be resolved regarding procurement. In retrospect, the Bank should not have agreed to the restrictive shipping requirements for the bitumen procurement. 51. The Bank's commitment of staff and funding for supervision, however, appears to be less than adequate, given the size and complexity of the project. The approximately eleven and one half staff weeks per year for supervision is considerably under the Bank's norm. While the project was supervised by a highly competent highway engineer, he had very little assistance during the supervision of this sizable project. For example, of the eleven supervision missions that were carried out during the six year implementation period, all but two were carried out by a single individual, usually for less than ten days. 52. As a sector loan, the project conditionality should have given more attention to sector issues such as the sector investment program, the overall level of Government spending for the sector, and the relationship between maintenance and new investment, among others. In reality, it functioned as an investment project, but with some of the investments agreed after project startup on the basis of agreed criteria. Also, the Bank's decision in regard to the bitumen  - 13 - procurement issue appears to be more pumitive than necessary to be consistent with World Bank procurement procedures, and left feelings of resentment on the part of the client. F. Borrower's Performance 53. KGM's performance was excellent in regard to project administration. Progress reports were submitted on time, as were good and timely audit reports. Technical performance was also good, with relatively little delay in contractor performance, and with few problems of quality of the work. There was some resistance to the use of Bank procurement procedures in some cases, mainly because of resistance to change from long standing Turkish practices. However, most of these differences were ultimately worked out satisfactorily. 54. The expansion of the project scope to include 1875 km of road improvements compared to the original 950 km included in the appraisal document led to a considerably higher total project cost than contemplated at appraisal. This appears to be the result of KGM wanting to accomplish as much as possible under the project. However, it has the potential for leaving contracts partially unfinished and contractors with possible costly claims against KGM. Also, the additions to the original contracts resulting in increased contract costs of 46 percent appear excessive and no doubt contributed to the lower than expected economic rates of return for the completed civil works contracts. Some of the additions for which unit prices would not apply, such as by-passes, should have been subjiected to economic analysis before proceeding with the works, and might have been handled at lower cost under new bidding rather than as additions to the existing contracts. Also, the inflation clause in the civil works contracts should be examined to determine if the substantial increase in contract costs was caused in part by a price adjustment formula that more than compensated for actual cost increases. 55. As pointed out in the Staff Appraisal Report for the Road Improvement and Traffic Safety Project, the contribution of the Bank to the sector was seen by the Government largely in financial terms, rather than as a source of relevant experience for institutional strengthening and policy reform. A more open relationship) not only with the Bank but with highway experts available through international consultants and other lending organizations would likely benefit KGM through access to new ideas. G. Assessment of Performance and Outcome 56. The overall project outcome is ralted as satisfactory, in spite of the problems described above. The project objectives were generally achieved, as stated in Section B. The specified amount of road improvements were carried out at approximately the estimated cost per km. The economic rate of return was considerably lower than the 30 percent estimated at appraisal, but is still acceptable (see para. 15). Some institutional improvements were made, and the prospect for sustainability of project benefits is good. On the negative side, an excessive amount of civil works was included in the original bidding documents, as well as in the substantial additions made to the contracts during execution, resulting in a substantial total cost overrun that must be financed by KGM.  - 14- H. Future Operations 57. A follow-on project, financed by Loan Nos. 4048/49-TU, has already been put in place and is under implementation. This project, supported by US$250 million of Bank financing, will continue the road improvements for 900 kin of roads, support the borrower in implementing its Road Traffic Safety Action Plan, and continue the institutional strengthening of KGM. Thus it is a logical continuation of the State and Provincial Roads Project. The lessons learned from the State and Provincial Roads Project as given below have been addressed and corrective measures included in the follow-on project. I. Key Lessons Learned 58. There are a number of lessons to be learned from this project, as follows: (a) Highway contracting authorities should be aware of road construction market conditions, and to the extent possible let a level volume of work in each year to avoid either flooding the market, resulting in a high price level for contracted work, or a shortage of work that will make it hard even for good contractors to survive; (b) It is generally preferable to let the civil works contractors be responsible for all of the inputs required, in order to avoid divided responsibility for quality and delivery; (c) Care should be taken to avoid overly restrictive specifications that will result in lack of effective competition; and (d) Additions to civil works contracts should be limited to a reasonable percentage of the original cost; major additions should be subject to economic analysis before deciding to proceed, and should be put out to bid as separate contracts.  - 15 - Part II. Statistical Tables Table 1: Summary of Assessment Table 2: Related Bank Loans/Credits Table 3: Project Timetable Table 4: Loan/Credit Disbursements: Cumulative Estimated and Actual Table 5: Key Indicators for Project Implementation Table 6: Key Indicators for Project Operation Table 7: Studies Included in Project Table 8A: Project Costs Table 8B: Project Financing Table 9: Economic Costs and Benefits, Table 10: Status of Legal Covenants Table 11: Compliance with Operational Manual Statements Table 12: Bank Resources: Staff Inputs Table 13: Bank Resources: Missions Table 14: Civil Works Bid Evaluation Table 1: Summary of Assessments A. Achievement of Objectives Substantial Partial Negigble Notapplicable Macro Policies O O C] C Sector Policies El O 0 0x Financial Objectives El a El 0 Institutional Development C] El 0 C] Physical Objectives ER O O al Poverty Reduction El El E El Gender Issues E] O El [E Other Social Objectives El El 11 0 Environmental Objectives C] C] E 01 Public Sector Management C] 3 E El Private Sector [l 0x El E Development Other (specify) O E El 0  - 16- B. Project Sustainability Likely Unlikely Uncertain 03 0 0 Highly C. Bank Performance satisfactory Satisfactory Deficient (1) V) (V) Identification O 0 O Preparation Assistance Cl 0 0 Appraisal Cl E[ 0 Supervision O 0 El Highly D. Borrower Performance satisfactory Satisfactory Deficient Preparation 0 E3 a Implementation 0 0x3 a Covenant Compliance 0 0 E Operation (if applicable) El 0 E Highly Highly E. Assessment of Outcome satisfactory Satisfactory Unsatisfactory unsatisfactory E' ) (E/) (/)  - 17- Table 2: Related Bank Loans/Credits Loan/credit title Purpose Year of approval Status Preceding operations 1. Highway Highway FY1982 Complete Rehabilitation Improvements 2. Second Highway Same FY1984 Complete Following operations 1. Road Improvement Highway FY1996 Implementation and Traffic Safety Improvements Table 3: Project Timetable Steps in Project Cycle Date Planned Date Actual! Latest Estimate Identification (Executive Project Summary) June 1990 June 1990 Preparation July-Sept 1990 July-Sept 1990 Appraisal 10/23/90 10/23/90 Negotiations April 1991 April 1991 Board Presentation 5/14/91 514/91 Signing 6/28/91 6/28/91 Effectiveness 9/28/91 11/8/91 Project Completion 12/31/96 6/30/97 Loan Closing 6/30/97 12/31/97  - 18- Table 4: Loan Disbursements: Cumulative Estimated and Actual (US$ thousands) FY92 FY93 FY94 FY95 FY96 FY97 FY98 Appraisal estimate 15 50 100 180 270 300 Revised Amount 15 26.4 90 154 218 250 250 Actual 7.96 26.43 122.84 162.5 196.0 239.7 250 Actual as % of revised 53% 100% 136% 106% 90% 96% 100% estimate Date of final Prior to disbursement 12/31 197 Table 5: Key Indicators for Project Implementation Estimated Actual Key implementation indicators in SAR: Key implementation indicators were not used for this project. Table 6: Key Indicators for Project Operation Estimated Actual Key operating indicators in SAR: Key operating indicators were not used for this project. Table 7: Studies Included in Project Study Purpose as defined at Status Impact of study appraisal/redefined None.  - 19 - Table 8A: Project Costs Appraisal estimate Actual/latest estimates (U-S$ million) (US$ million) Item Local Foreign Total Local Foreign Total costs costs costs costs I.Civil Works 138.0 138.0 276.0 330.3 330.4 660.7 2.Bitumen 11.2 44.8 56.0 0 0 0 3.Equipment 7.6 68.4 76.0 0 75.0 75.0 4.Road Safety 1.3 11.7 13.0 0 11.3 11.3 5.Cons. Services and Training 0 2.0 2.0 3.9 0 3.9 6.Contingencies 21.9 35.1 57.0 0 0 0 Total 180.0 300.0 480.0 334.2 416.7 750.9 Table 8B: Project Financing Appraisal estimate Actual/latest estimates (US$ million) (US$ million) Item Local Foreign Total Local Foreign Total costs costs costs costs I.Government 180.0 0 180.0 334.2 151.7 485.9 2.IBRD 0 300.0 300.0 0 250.0 250.0 3.Switzerland 0 0 0 0 15.0 15.0 Total 180.0 300.0 480.0 334.2 416.7 750.9  - 20 - Table 9: Economic Costs and Benefits Contract Original Orig. Est. Act.Est.Cost Est. IRR Actual Est. IRR- % No. Km Cost-US$m -US$m % CWA-1 60 26 25 19.1 11.75 CWA-4 103 41 29 23.6 12.70 CWA-2 62 22 28. 18.8 13.52 CWA-11 82 45 80 13.5 13.00 CWA-5 76 32 20 42.6 27.78 CWA-3* 31 12 27 57.9 23.03 CWA-13 98 36 23 42.9 13.88 CWA-10 49 19 14 33.5 20.71 CWA-12* 68 28 13 17.9 44.60 CWA-15 110 51 31 67.3 32.34 CWA-14 71 29 21 25.8 19.14 CWA-7 60 24 30 42.8 24.44 CWA-6 66 43 50 17.6 15.17 CWA-19* 35 5 11 28.2 31.39 CWA-21 124 38 145 31.6 2.08 CWA-8* 51 8 13 28.6 29.11 CWA-9* 32 6 9 16.0 32.74 CWA-16* 47 9 14 70.1 12.49 CWA-17* 40 27 43 39.7 0.43 CWA-18* 90 18 29 45.5 7.71 CWA-5a 76 32 19 42.6 27.78 CWY-1 87 40 30 60.3 34.13 CWY-11* 111 46 2 11.0 10.43 CWY-12 41 9 18 29.3 16.16 CWY-3* 37 11 17 39.1 -0.09 CWY-2* 81 25 20 12.0 12.01 CWY-5* 47 24 9 27.6 27.77 CWY-9 45 21 64 41.8 15.44 CWY-13 38 30 56 28.7 1.92 CWY-4* 25 8 2 19.7 7.15 CWY-6 84 26 10 16.5 8.05 * Completed or nearly completed Note: Estimated Rates of Return are not readily comparable as the final works in many cases differ substantially from the original design.  -21 - Table 10: Statuis of Legal Covenants Turkey State and Provincial Roads Project Covenant Present Original Revised Description of Covenant Comments Agreement Section Type Status Fulfillment Fulfillment Date Date Loan 2.01(b) 02 C . Open Special Account Opened May 20, 1992 No.3324 3.01(a) 04 NC . Provide adequate counterpart This was a problem for the funds last several years of the project. 3.01(b) 05 C Appoint Project Director and steering committee 3.01(c) 12 C Select road sections according to Complied with. criteria agreed with Bank 3.01(c) 02 C Fumish feasibility studies for each All submitted road section prior to bidding. 3.01(c) 02 C Furnish 6-monthly progress Submitted on time. reports. 4.01 01 C Maintain audit accounts and have Submitted on time. them audited-submit to Bank by June 30. Sch. 1 05 C June 30, June 30, All civil works contracts to be let Completed within the para.4 1993 19094 by the agreed date. revised schedule. Sch. 1 05 C Retroactive financing up to US$25 Done. para.3 million Sch.1 03 C Disbursements for Civil Works Change effective January increased from 50% to 60% 22, 1996. Covenant types: Present Status: 1. = Accounts/audits 8. = Indigenous people C = covenant complied with 2. = Financial performance/revenue generation 9. = Monitoring, review, and reporting CD = complied with after from beneficiaries 10. = Project implementation not delay 3. = Flow and utilization of project funds covered by categories 1-9 CP = complied with partially 4. = Counterpart funding 11. = Sectoral or cross-sectoral NC = not complied with 5. = Management aspects of the project or budgetary or other resource executing agency allocation 6. = Environmental covenants 12. = Sectoral or cross-sectoral policy/ 7. = Involuntary resettlement regulatory/institutional action 13. Other Table 11: Compliance with Operational Manual Statements Statement number and title DescribeW and comment on lack of compliance n.a. n.a.  -22 - Table 12: Bank Resources: Staff Inputs Planned Revised Actual Stage of project cycle Weeks US$ Weeks US$ Weeks US$ Preparation to - - - - 26.7 61.2 appraisal Appraisal - - - - 24.3 48.2 Negotiations through - - - - 7.3 10.9 Board approval Supervision - - - - 74.3 208,700 Completion 4 11,000 4 11,000 nc nc Total nc nc Table 13: Bank Resources: Missions Performance RatingT Stage of project cycle Month/ No. of Days in Specialization Implem. Developm Pbl f3 Year Persons Field status objectives robems Through appraisal 6190 2 12 HE, E Appraisal through Board 10/90 2 15 HE, E approval Supervision 11/91 2 5 HE, E - - 10/92 1 10 HE 2 1 P&T 3/93 1 10 HE 2 1 P&T 9/93 2 10 HE,E I I 2/94 1 9 HE HS HS 11/94 1 5 HE HS HS 2/95 1 5 HE HS HS CF 11/95 1 4 HE HS HS CF 2/96 1 5 HE S S CF 3/97 1 8 P S S CW 10/97 1 6 HE - Completion n.a. Total I - Key to Specialized staff skills: 2 - Key to Performance Ratings: 3 - Key to Types of Problems: HE-Highway Engineer 1-Highly Satisfactory P-Procurement E-Economist 2-Satisfactory T-Training P-Procurement Specialist CF-Counterpart Funding HS-Highly Satisfactory CW- Civil Works Overruns S-Satisfactory U-Unsatisfactory  -23 - Appendix A: Cuivil Works Bid Evaluation Contract No. Date No. of Bidders Spread* % of Estimate CWY-3 July 92 13 43% 64% CWY-2 9/7/92 26 55% 62% CWA-1 1 12/25/92 13 n.a. 63% CWY-1 7/9/92 10 49% 64% CWY-l1 6/18/93 7 49% 67% CWY-5 6/21/93 31 81% 52% CWA-12 6/25/93 5 8.3 89% CWA-13 9/17/93 15 5.9 89% CWA-19 Sept 93 10 3.8% 93% CWA-10 Oct 93 12 5.7% 92% CWY-12 11/17/93 12 5.1% 92% CWY-9 Dec 93 7 4.3% 94% CWA-14 12/24/93 12 10.6% 87% CWA-15 12/24/93 9 6.6% 89% CWY-13 12/31/93 8 4.0% 95% CWA-21 2/17/94 18 5.0% 93% CWA-22 5/6/94 6 33% 74% CWY-6 5/6/94 28 80% 50% Spread between the highest and lowest bidder.  Appendix B: Borrower Contribution to the ICR TURKiYE CUMHURiYETi REPUBLIC OF TURKEY BAYINDIRLIK VE iSKAN BAKANLIGI i MINISTRY OF PUBLIC WORKS AND SETTLEMENT KARAYOLLARI GENEL MUODURL0L0 GENERAL DIRECTORATE OF HIGHWAYS .Akjum, TURXZY B091TCRo1404-01-750/0275 07 th May, 1998 3324 S-TU IMPLEMENTATION COMPLETION REPORT OF THE ' "STATE AND PROVINCIAL ROADS PROJECT " KMG'S (GENERAL DIRECTORATE OF HIGHWAYS') REQUESTS AND VIEWS ACKNOWLEDGEMENTS We wish to express our gratitude to all World Bark Staff both current, retired, now working for other loans or countries and alsc to all those who have since passed away, in Washington and in Ankara who contributed little or much to the success of our Project, apart from the ones who caused the cancellation of US $ 50 M., allocated for the purchase of bizuminous material in the Loan Agreement nu..bered 3324 S-TU, in soite of the existence of Clauses 2.59, 2.60 and 2.61 of the Procurement Guideline and Ar-icles of 83 and 85 of the Operaticnal Directive and also their being in effective, at the time of ourchase of bituminc s material bid. The ones wh_ caused cancellation argued that the imple-.enting agency's pre-bid cost esti.ate was not exceeded by a substantial amount, and there existed an effective competition in the bid, - while the cost estimate was exceeded by at least $ 1. M.(61-50=11, excluding price escalation value) according to the tender value of B 61 M. of the unique responsive bidder, - in addition, the allocated amount for purchase of bituminous material was only $ 50 M. and the amount that could be added to this allocated amount from the "unallocated category" was only S 2,265 M, - furthermore, the cost estimate was exceeded by $ 34,735 M. (87-50-2,265, including price escalation value, excluding VAT) according to the real comparison base and - finally, there was only one (1) responsive bidder according to the World Bank. -1- 0 a  TORKiYE CUMHURiYETi jS REPUBLIC OF TURKEY BAYINDIRLIK VE iSKAN BAKANULii ' - MINISTRY OF PUBLIC WORKS AND SETTLEMENT KARAYOLLARI GENEL MODORL0)40 GENERAL 0[RECTORCl 6GHWAYS Ankara. TURKEY 3324 S-TU (ICR) A- We want to begin from the cancellation matter of US $ 50 Million allocated for purchase of bituminous material, which is the most important topic, from our point of view. The summary of the World Bank's arguments and opinions pertaining to the purchase of bituminous material bid, which have been stated in the paragraphs of the Part prepared by the World Bank of the Implementation Completion Report, are as follows: Paragraph Nu.8 of E.S. (Evaluation Summary): "Bidding for procurement of US $ 50 Million of bitumen to be used for the road upgrading, resulted in only one responsive bidder, due to restrictive bidding specifications" Paragraph Nu.32: "The specifications also required that; The bidder own, or have on charter, special asphalt ships of sufficient quantity and capacity to transport the bitumen, It had succesfully transported at least 100.000 metric tons of asphalt in programmed deliveries under one contract. KGM insisted that these shipping requirements were necessary to assure timely delivery." Paragraph Nu.33: " Only three bids were received, one of which was disqualified because it did not provide a proper letter of guarantee. A second bid, from an American supplier, met all of the specifications at a price of US $ 61 Million. A third bid from TUpras was received at a price of US $ 36 Million on the basis of using bitumen from its local refinery" Paragraph Nu.34: "However, as TUpra* did not meet the stringent shipping requirements, the Bank could not agree to the KGM recommendation" Paragraph Nu.35: " In addition to the requirements of the 1986 Technical Specifications for bitumen, KGM added to the 1992 document, very specific and demanding language requiring extensive previous experience of bitumen shipment, and either evidence of ownership or long term charter for specialized bitumen delivery tankers (Clauses 6.2 and 6.4). KGM argued at the time that these were essential performance criteria to assure satisfactory supplies and these restrictive criteria were, therefore, accepted by the Bank." -2- .. z .Man Addissa Kara7olIars Cenci '#1id6i&IUj. 86100 YIic,aek. Aukang. TURKE'y  TORKiYE CUMHURiYET F REPUBLIC OF- TURKEY BAYINDIRLIK VE iSKAN BAKANLIdI MINISTRY OF PUBLIC WORKS AND SETTLEMENT KARAYOLLARI GENEL MUDURL00 60 GENERAL DIRECTORATE OF HIGHWAYS AnkSra. TUFRJCY Paragraph Nu.43: "Certainly, KGM bares a share of the responsibility for insisting on the extremely restrictive shipping specifications. KGM management was slow in realizing that the original specifications were too restrictive to result in effective competition" For the time being, please be informed that all we can tell you is that various investigators received and recorded the statements of the persons related with the preparation of the specifications of this purchase and of the other related persons. Though the investigators, KGM and the Turkish Court, continue trying to find the faulty institutions and persons, neither the results of the investigators' reports, nor the decision of the Turkish Court, should not affect the legal rights of KGM, which demanded either cancellation of the bid ard then rebid again; or 7_ reallocation of this $ 50 Million to other categories, under its and World Bank's rights and responsibilities given in the Articles of 2.59, 2.60, 2.61 of the Procurement Guideline and Articles 83 and 85 of the Operational Directive. After the World Bank's approval of technical and administrative specifications prepared by KGM (General Directorate of Highways of TUrkiye', only a total of three tenders were received for this bid, pertaining to the ourchase of bituminous material. The features of these tenders are given below: 1- KGM's Bidding Commission re-turned the internal .ender envolope without opening it, to its firm, as this firm dlidn't provide a proper letter of guarantee to be put in the external envelope. 2- The firm A whose tender was considered by KGM as va^id, but invalid by the World Bank, proposed a bid value of US $ 36 Million in its tender. 3- The firm B whose tender value was considered by 4KGM as too high (with repect to normal market prices avai.lable budget and also with respect to the proposal of US 5 36 Million of the firm A) proposed a bid value of US S 61 Million excluding VAT and price escalation value in its tender envelcpe. KGM's Bidding Commission, proposed to the Bank, to asard the contract to the firm A, to the owner of the lowest tender of US $ 36 Million. But since the World Bank wanted KGM to award the cont-act to the firm B which proposed a tender value of US $ 61 M. (US $ 25 M. bigger than the tender value of the firm A) and insisted on this; KGM proposed the Bank two alternatives below, considering KGM's and Bank's rights stated in the Articles 2.59 and 2.60 of the Guideline for Procurement: -3- a. z E .~~~~~~~~Mail Addre.s: K.avayottar, Cenel :NlIIdurIiLhi. X6 eo Yiactlepe. .Anku. TURKEY  TORKiYE CUMHURiYETi j REPUBLIC OF TURKEY BAYINDIRLIK VE iSKAN BAKANLIGI < MINISTRY OF PUBLIC WORKS AND SETTLEMENT KARAYOLLARI GENEL MODURLUdJ A3P GENERAL DIRECTORATE OF HIGHWAYS Ankan. TURKEY 1- a) To cancel this bid, b) to amend the specifications (deleting, correcting the items in the existing specifications preventing competition such as Article 6.4: "The bidder must have successfully completed delivery and transportation of AT LEAST 100.000 METRIC TONS OF ASPHALT to the total of various ports WITH MONTHLY PROGRAMS, UNDER A SINGLE CONTRACT, WITHIN THE LAST FIVE YEARS" and Article 6.3: The bidder SHALL HAVE THE TANKERS that have the capacity and the quantity or shall have hired (time charter) them for at least 4 (four) years during transportation period, on long term duration etc. and then c) to rebid this purchase of bituminous material work or 2- To cancel this bid and spend the total allocated amount of US S 50 M. of this category, in any other category the World Bank will determine. (but preferably in the category of civil works according to KGM) While the total allocated amount for this category in the Loan Agreement is only US $ 50 M., the proposed value by the Bidder B, is US $ 61 M. (excluding 12 % VAT amount according to the last sentence of the Article 14 of the specification, and also excluding price escalation amount and haulage costs). The cost with the price escalation amount would be about 61x1,436= US $ 87 M., if we take the price escalation percentage of 43,6 % in the previous contract, into consideration, also for this contract. So the real contract cost of the bidder B with the related costs (excluding the haulage cost of 487 500 tons of bituminous material, which would be brought to the Ports, from the ports to the various sites, which are 60-500 kms away from the Ports) would be equal to at least 61x1,436x1,12= $ 98 M. In tne articles 2.59 and 2.60 of the Procurement Guidelines, it is said "Rejection of all bids is justified when there is lack of effective competition." Since according to the World Bank there is only one valid bid, in this bidding and since the total allocated amount of $ 50 M. for this purchase in the loan agreement, is exceeded by a substantial amount, as 61-50= 5 11 M., with the tender value of $ 61 M. (If we make our comparison taking the total costs, then the difference becomes even bigger as 87-62,4= $ 24,6 M. even if the haulage costs that would be paid from local expenditure part of the Turkish Budget, is excluded). It is never possible for the Bank to finance and it was not the intention of KGM, to request the Bank to finance all of the related cost of bitumen. The allocated amount for purchase of bituminous material was only $ 50 M. and KGM wanted to buy only that much. So in fact comparisons for finding whether the budget was exceeded by the real cost of tendered value of $ 87 M. (including also $ 26 M. for price escalation value) should have been done with respect to $ 52,265 M. ($ 2,265 M. is the maximum amount that can be provided for the purchase of bituminous material from the unallocated category of the Loan.) -4- 3 tMaU Addeta,,% K2aruyatxr, Genet NWtdUrfii4ii 06 tOO YCcetepc. Ankara. TURKEY  TORKIYE CUMHURiYETi *X REPUBLIC OF TURKEY BAYINDIRLIK VE iSKAN BAKANLIGI pw MINISTRY OF PUBLIC WORKS AND SETTLEMENT KARAYOLLARI GENEL MODORLOOO 'GENERAL DIRECTORATE OF HIGHWAYS Anka. TLURJEY In spite of these obvious facts, the proposal of the Bank to KGM in the form of: "Either KGM awards the contract to the firm B proposing $ 61 M. or $ 50 M., the allocated amount for this purchase will be cancelled." was very unjust and wrong. But unfortunately this amount was cancelled and the loan amount was reduced from $ 300 M. to $ 250 M. Because of this decrease, 18 road contracts could not be completed by the closing date 31.12.1997 of the Loan. (Widening of some road sections and/or constructing road bed and/or construction of pavement works and/or paving with asphaltic concrete etc. of various road sections of the 18 contracts could not be completed.) This is a formal request and the Bank, should correct this faulty, unjust decision as soon as possible taking the clear, open statements of the articles 2.59, 2.60, and 2.61 of Procurement Guideliness of the Bank and articles 83 and 85 of the Operational Directive of the World Bank (OD 11.00) stating: "When there are legitimate differences in judgement between the Bank and the Borrower" "Rejection of all bids is justified when there is a lack of effective competition" "If the lowest evaluated responsiLve bid, exceeds the Borrower's pre-bid cost estimates by a substantial margin, the Borrower consider requesting new bids." "Every effort should be made to avoid misprocurement, by providing comments and advice in the review of bidding documents and bid evaluation. If, however, misprocurement can not be avoided, the country director should be kept informed and individual cases involving large cancellations should be brought to the attention of the Senior Vice President, Operations, the Regional Vice President, and the Chief of CODPR." KGM believes that the Bank shoulcd have agreed to rebidding of the bitumen with revised specifications instead of its clecision to cancel the US $ 50 M. and that the Procurement Guidelines provide enough justification for a different decision than the one taken by the Bank. It was not the intention of KGM and the Bank to finance all the cost related to bitumen, as haulage costs from ports to sites, 12 % VAT etc. but to finance foreign expenditure cost of the 487.500 tons of bitumen including its price escalation value. Total maximum amount which may be available from the Loan was about 50+2,265=$ 52,265 M. including the amount of $ 2,3 M. which may be allocated from "unallocated category" of the Loan. While the corresponding cost according to the tenderer B was about 61+26= $ 87 M. Thus, the responsive bid was 87-52,265= US $ 34,735 M. above the cost estimate for the quantity of bitumen at issue, which would seem to justify a finding of a "substantial amount" Rebidding with revised specifications was fully possible when we consider the Guideline. In our opinion the Bank took a decision that did not have the interest of Turkey as the main priority and that idea will remain in KGM until the Bank revises 2 its position. -5- pi .. 0 I dMail Addr,..: Karuyallrt, Cernc l&,dUrtlUlAi, 06 100 Yuctlepe. AnIkaLr. TURKEY -.-. .. ._ ... *.... S.fl.. _v ....... ia$*ttA l,~W. tf,lA