Sri Lanka - Third Roads Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 19375 IMPLEMENTATION COMPLETION REPORT SRI LANKA THIRD ROADS PROJECT CREDIT No. 2183 - CE June 21, 1999 Infrastructure Sector Unit South 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.  ii CURRENCY EQUIVALENTS Currency Unit - Sri Lanka Rupee (Rs) Appraisal year, 1990 US$1.00 = Rs 40.0 (at appraisal) Mid term review, 1994 US$1.00 = Rs 49.4 (annual average) Completion year, 1998 US$1.00 = Rs 64.6 (annual average) WEIGHTS AND MEASURES Metric System ABBREVIATIONS AND ACRONYMS ADB - Asian Development Bank CATB - Cabinet Appointed Tender Board CM&C - Construction Management and Contracts EDCF - Economic Development Cooperation Fund (Korea) GOSL - Government of Sri Lanka ICTAD - Institute for Construction Training and Development INC-CPT Inter-Ministerial Committee for Coordination and Planning of Transport MHIH&SS - Ministry of Health, Highways, and Social Services MOF - Ministry of Finance MPPI - Ministry of Planning and Policy Implementation MTH - Ministry of Transport and Highways NTC - National Transport Commission NMT - Non-Motorized Transport OECF - Overseas Economic Cooperation Fund (Japan) ODA - Overseas Development Administration PC - Provincial Council PMMU - Project Management and Monitoring Unit (of RDA) RCDC - Road Construction and Development Company RDA - Road Development Authority RDC - Resources Development Consultants Ltd. SEC - State Engineering Corporation SD&CC - State Development and Construction Corporation SLR - Sri Lanka Railways SWKP - Scott, Wilson, Kirkpatrick and Partners TEC - Technical Evaluation Comunittee TSPC - Transport Studies and Planning Centre TSSS - Transport Sector Strategy Study FISCAL YEAR OF BORROWER January 1 - December 31 Vice President Mieko Nishimizu Country Director Mariana Todorova Sector Director Frannie Humplick Team Leader Juan Gaviria Task Team Leader Jaswant Channe  iii FOR OFFICIAL USE ONLY TABLE OF CONTENTS PREFACE ................................................................1 EVALUATION SUMMARY ................................................................2 PART I - PROJECT IMPLEMENTATION ASSESSMENT ................................................................ 6 A - BACKGROUND ................................................................ 7 B - PROJECT OBJECTIVES ................................................................8 C - ACHIEVEMENT OF OBJECTIVES ............................................................... 9 D - MAJOR FACTORS AFFECTING THE PROJECT ............................................................... 12 E - PROJECT SUSTAINABILITY ...............................................................1 2 F - IDA PERFORMANCE ............................................................... 14 G - BORROWER PERFORMANCE ............................................................... 16 H - ASSESSMENT OF OUTCOME ............................................................... 17 I - FUTURE OPERATION ............................................................... 18 J - KEY LESSONS LEARNED ............................................................... 19 PART II - STATISTICAL ANNEX ............................................................... 20 TABLE 1: SUMMARY OF ASSESSMENTS ............................................................... 21 TABLE 2: RELATED BANK LOANS/CREDITS ............................................................... 22 TABLE 3: PROJECT TIMETABLE ............................................................... 24 TABLE 4: LOAN DISBURSEMENTS: CUMULATIVE ESTIMATED AND ACTUAL ............................................. 25 TABLES 5 AND 6: KEY INDICATORS FOR PROJECT IMPLEMENTATION AND OPERATION .............................. 26 TABLE 7: STUDIES INCLUDED IN PROJECT ............................................................... 27 TABLE 8A, PART 1 - PROJECT COSTS IN US $ ............................................................... 28 TABLE 8A, PART 2 - PROJECT COSTS IN SL Rs ............................................................... 29 TABLE 8B: PROJECT FINANCING ............................................................... 30 TABLE 9: ECONOMIC COSTS AND BENEFITS ............................................................... 31 TABLE 10: STATUS OF LEGAL COVENANTS ............................................................... 33 TABLE I 1: COMPLIANCE WITH OPERATIONAL MANUAL STATEMENTS ....................................................... 34 TABLE 12: BANK RESOURCES: STAFF INPUTS ............................................................... 35 TABLE 13: BANK RESOURCES: MISSIONS ............................................................... 36 ANNEXES A. MISSION'S AIDE MEMOIRE B. ROAD DEVELOPMENT AUTHORITY'S CONTRIBUTION TO THE ICR, (INCLUDING OPERATIONAL PLAN) C. ECONOMIC EVALUATION D. MAP IBRD NO. 30311 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 SRI LANKA THIRD ROADS PROJECT (Credit 2183 - CE) PREFACE This is the Implementation Completion Report (ICR) for the Third Roads Project in Sri Lanka, for which credit 2183-CE in the amount of SDR 30.6 million (US$42.5 million) received Board approval on November 6, 1990 and was made effective on February 21, 1991. The credit was closed on December 31, 1998, compared with the original closing date of June 30, 1998. Final disbursement was made on June 2, 1999. A balance of approximately SDR 235,000 (US$316,095.14) is remaining in the special account and will be refunded. After this refund is received a balance of approximately SDR 3.5 million (US$4.7 million), or 11.4% of the original amount, is expected to be canceled. The ICR was prepared by William Denning for the South Asia Region and reviewed by Juan Gaviria, Team Leader, and Jaswant Channe, Task Leader. The Borrower provided comments that were incorporated and are included as Annex B to the ICR. Preparation of this ICR was started during IDA' s final supervision/completion mission, November 11-24, 1998. It is also based on material in the project file, discussions with IDA staff, including input from Brendan Kennedy, Consultant, and periodic management reports prepared by the Borrower. The Borrower contributed to the preparation of the ICR by: (a) providing views reflected in the completion mission's Aide M6moire (Annex A), (b) preparing their own evaluation of the project's execution (Annex B), (c) preparing an economic re-evaluation (Annex C), (d) commenting on the draft ICR, and (e) participating in a stakeholders' workshop held on May 19, 1999 in Colombo.  2 IMPLEMENTATION COMPLETION REPORT SRI LANKA THIRD ROADS PROJECT (Credit 2183 - CE) EVALUATION SUMMARY Introduction 1. The Third Roads Project was a follow up to previous IDA lending operations in the sector. The project funded road rehabilitation and bridge reconstruction civil works which were a priority to fill in gaps in the strategic highway network. The project was a traditional road rehabilitation operation; it focused mainly on physical achievements with some institutional development objectives. Project Obiectives 2. The project's objectives were to: (a) reduce road transport cost and delays for passengers and goods by restoring major trunk roads to better operational condition, (b) help restore priority road infrastructure damaged by flooding and landslides in May/June 1989, (c) enhance institutional capabilities to increase the quality and extent of road maintenance and rehabilitation, and (d) upgrade the Road Development Authority's (RDA) capacity to supervise and execute rehabilitation works and to identify actions needed to develop the local road contracting industry. 3. The project's overall scope took into account the availability of resources for counterpart financing and the limited absorptive capacity of the implementing agency. The benefits of the project were expected to be distributed widely among all segments of the population. The project was expected to support private sector involvement in the road sector through assistance to the local contracting industry. The objectives were straightforward, in line with the country's needs, and supported IDA strategy in the highway subsector at appraisal. The project components were directly linked to the project objectives. Implementation Experience and Results 4. The overall rating of the project is satisfactory. The objectives to reduce road transport costs and delays, and to help restore road infrastructure damaged by flooding were met by the physical achievements of the project. The main physical objectives were substantially met after some delays. Of 417 km of roads initially appraised, 387 km were contracted and 310 km were completed. Of this 310 kni, 92 km were completed with funding from Economic Development and Cooperation Fund (EDCF - Korea). From a total of 19 bridges, 13 were completed. RDA is currently completing the outstanding works using its own resources. Even though the completed construction was less than appraised and less than awarded, the economic rates of return expected by the road investments were achieved or exceeded. 5. The project's institutional development objectives (paragraph 2 above) were only partially achieved. These included: * strengthening of RDA's administration of projects, * improvement of RDA's highway and bridge design standards, manuals, and procedures, and * support of the local consulting and contracting industries through their direct involvement in the project.  3 improvement of the technical skills of some of RDA's engineers through secondment to supervision consultants, 6. RDA's project management and administrative capabilities were improved under the project through the establishment of a Project Management and Monitoring Unit (PMMU) in RDA initially staffed with consultants. PMMU has been successfully operating with only RDA staff since April 1997. RDA has now expanded the responsibilities of the PMMU to include managing and administering all foreign-aided development projects in the highway subsector. 7. The project's Technical Assistance component was focused on improving road user revenues, the local construction industry, and maintenance standards and design. A Road User Charges Study investigated the benefits received by, and charges currently levied upon, the different types of road users and recomnmended increases in user fees for two types of user. Because of their macroeconomic implications, the study recommendations were referred to MOF/Planning, but implementation of the recommendations has not taken place, and the matter still rests with MOF. 8. A domestic Construction Industry Study was satisfactorily completed by the Institute for Construction Training and Development (ICTAD). The recommendations on resolving contracting difficulties in Sri Lanka were endorsed by the Government of Sri Lanka (GOSL) and implemented by RDA and ICTAD with the assistance of the Contractors' Association. The local civil engineering and construction industry has benefited as a result. A Materials, Highway and Bridge Design Study resulted in RDA publishing and disseminating manuals for: highway geometric and pavement design appropriate to Sri Lankan conditions; bridge design; bridge construction; bridge maintenance; and standards for road construction; rehabilitation; and maintenance, including amendments to legal standards. 9. The development of the local contracting industry received a boost under the project even though some local contractors did not perform well. Following the completion of the construction industry study, agreement was reached among RDA, ICTAD and IDA on the participation of local contractors in the road and bridge rehabilitation program. Most of these contractors satisfactorily completed their works and gained considerable experience in future contracting work. 10. Project implementation was affected by three major factors. (1) Timeliness of decision making. The lengthy GOSL procedures for clearance of standard contract documentation, tender evaluation, and contract award caused major delays in project implementation. There were also delays in obtaining the necessary quarrying licenses despite the active support of RDA. (2) Inexperience of local contractors. Mobilization by some local contractors was slowed by their inadequate financial management and experience. Advance payments were not always managed well, leading to cash flow problems in the early stages of the contract. (3) Inexperience of international contractor. One contractor and one contract accounted for the bulk of the problems in this project. This may have been exacerbated by the contractor's inexperience in road construction relative to other types of civil engineering. Weak pre-qualification screening may also have contributed to this situation. Summary of Findin2s, Future Operation, and Key Lessons Learned 11. As stated above, the overall outcome of the project was satisfactory. This rating is based on the satisfactory, if partial, achievement of the physical objectives; achieving or surpassing the economic rates of return (ERRs) expected at appraisal; on the improved institutional capacity of the RDA; and the development of the local contracting industry. The overall ERR expected at appraisal was 33%; the re-evaluated overall ERR is 52%. The project would have been highly  4 satisfactory if better project management for road contract WB3/3 had resulted in problems being acted upon in time to have an alternate contractor complete the proposed works. 12. Overall sustainability is uncertain. GOSL has recognized the benefits of better road maintenance and has been increasing the appropriate budget (RDA's budget is supplied from GOSL general revenue). Also on the positive side, the RDA' s commitment of resources for maintenance has improved during the course of the project. There has been a steady trend to higher expenditures on maintenance in real terms. On the other hand, it is still not clear that spending is sufficient to guarantee sustainability of the road system. The fact that recommendations for road user charges have not been implemented (paragraph 7 above) is a further impediment to sustainable financing of road maintenance. 13. IDA performance in project identification was found to be satisfactory. The project was a continuation of earlier road development projects supported by IDA. Preparation assistance was satisfactory. During project preparation an effort was made to make better use of local consultants, backed up by international consultants, to establish the details of the investments required. The appraisal work was satisfactory. IDA supervision of the project and progress reporting was satisfactory, and the performance ratings were appropriate. The slow progress in the execution of the contracts and the control of costs were the main concerns of IDA's supervision missions. The delays in the execution resulted mainly from the unsatisfactory performance of one of the two expatriate contractors and from the inexperience of most of the local contractors. 14. Overall the Borrower's project identification and preparation activities were satisfactory. The performance of the Borrower in implementing the project was generally satisfactory. Expatriate consultants in joint venture with a local consultant assisted RDA in the supervision of road works. Their performance was less than satisfactory on occasion. Local consultants were involved in the design of road works and performed satisfactorily. 15. Compliance with project covenants was satisfactory. The Borrower's actions in project operation have been satisfactory. RDA has supported PMMU with adequate staffing and resources since April 1997, and PMMU is, in its present form, sustainable. 16. In support of IDA's Transport Sector Strategy, future IDA involvement in the sector should address institutional change and policy reform of RDA. This should assist RDA to move from a public oriented to a mixed model agency, to improve the efficiency of management of strategic needs and improve human resource development. RDA has developed a priority investment plan for the next decade which requires substantial resources for development and maintenance. District Plans are under development. Since GOSL will not be able to substantially increase the budget for roads, it will be necessary to (a) increase the efficiency of expenditures, (b) identify new sources of financing, (c) attract private investment to complement public sector investment, and (d) increase attention to sustainable and effective maintenance. 17. Future investments in the highway sector must balance maintenance with upgrading of strategic roads to provide additional road capacity. Past IDA projects have been concerned mainly with the physical rehabilitation of road pavements only. No realignment improvements or new alignments have been built, and no substantial increase in road capacity has been provided. 18. With the recent devolution of 15,300 km of roads to the Provincial level and the gradual emergence of road agencies for these subnational levels of government, there is an opportunity for RDA and external funding agencies to assist the newer agencies in developing their maintenance* capability, including identification of sustainable maintenance funding sources. The devolution allows RDA to focus more clearly on the strategic network and increases the responsibility of the Provinces for the local road network. If well managed, devolution will enhance efficiency of expenditures.  5 19. The key lessons learned from the project are as follows (the paragraph numbers below refer to Part I - Project Implementation Assessment): * Institutional reform is needed to enable RDA to move from a public oriented agency to a mixed agency and should: (a) improve efficiency through outsourcing of maintenance to contractors, (b) incorporate information technology in road management, (c) create an owner/provider model, (d) develop environmental and social safeguards in road development and maintenance, and (e) streamline management and decision-making. External financing agencies will play a major role in supporting RDA in the reform process as part of road sector lending. The availability of large external aided support, with limited conditionality, for new road development is a factor to take into consideration. Any future road sector lending should be preceded by a dialogue on reform with GOSL, RDA, and other major donors (paragraphs 28, 30, 56, 66). * GOSL should empower the Tender Committee and a revise the clearance procedures for the approval of bid documents to ensure timely contract awards. This could minimize project delays and improve decision-making (paragraphs 30, 56, 60). * During project design greater attention should be placed on: (a) planning and prioritization of investments, (b) holistic design to address safety issues, environment, and social safeguards, (c) monitoring and evaluation systems including performance indicators, and (d) sustainability, including funding and mechanisms for road safety. Appropriate design and standards, including attention to resettlement, environment, and alignment considerations, are critical to achieve efficient and effective results in construction and maintenance (paragraphs 42, 60, 69, 70, 71). * Support for local contractors needs to be improved, especially when fostering development of a new set of private sector companies. The lack of adequate working capital resulting from inexperience in financial management delayed several contractors (paragraphs 27, 31). Consideration should be given to a more comprehensive development program for the private construction sector as a whole, including more than simply updating government standards. There is also a need to have an adequate stream of on-going construction contracting, to provide the newly created industry with a steady supply of work (paragraph 40). * Decentralization of responsibility for the road network - if not well orchestrated - is likely to have an adverse effect on the sustainability of the strategic road network and the smooth functioning of road transport (paragraphs 38, 39, 68). * To ensure sustainable financing of road maintenance, it is critical to implement reforms in the road user charging system (paragraphs 22, 41, 62). * Bridge opening delays have a major effect on users by creating prolonged traffic bottlenecks at key locations. The economic analysis done at appraisal should explicitly consider the costs of delayed bridge construction in its risk assessment. The appraisal should also consider the potential for bridge construction delays in its treatment of risk mitigation and should include alternative designs (paragraph 14).  IMPLEMENTATION COMPLETION REPORT SRI LANKA THIRD ROADS PROJECT CREDIT 2183 - CE PART I - PROJECT IMPLEMENTATION ASSESSMENT  7 SRI LANKA THIRD ROADS PROJECT (CREDIT 2183 - CE) PART 1: PROJECT IMPLEMENTATION ASSESSMENT A - Background 1. GOSL' s development strategy has focused on export-oriented and private-sector led growth, without sacrificing the environment or quality of life. The country's growth strategies have also sought to ensure that the poor are included by providing access to services and employment. The transport sector has traditionally had a critical role to play in achieving these goals, especially the road system, which is the main artery of the island's economy. 2. The country has an extensive and well-developed road network which evolved during the last century to serve the island's plantations. Almost every part of the country is accessible by a road system of 97,000 km. The national trunk road system (Class A and B roads), under the responsibility of RDA, consists of 11,000 km of paved roads (11 % of the total). Provincial Councils account for another 15,300 km (16%), while local governments and specialized agencies account for the remaining 71,000 km (73%). A road condition survey carried out in 1996 indicates only 10% of the national trunk roads had a roughness of less than 5,000 mm/km (i.e. fair quality). This compares to 51% in a similar survey carried out in 1986, indicating a substantial decline in quality over this period. The road sector is by far the country's most important mode of transport and accounts for over 95% of annual passenger km and for 95% of freight tonne-km. 3. The road sector faces significant problems. Although the network covers the country, it is in a state of serious disrepair. Historically, inadequate maintenance expenditures have been made worse in the face of high population growth, rapid urbanization and slow economic development. The short-term concern is to reduce the maintenance backlog by increasing the recurrent maintenance allocations, improving expenditure efficiency, reducing project completion time, and increasingly outsourcing road maintenance to contractors. 4. Development of an efficient transport sector has long been a GOSL objective, which IDA has encouraged. IDA has had five operations in the road sector. The first credit for a highway project was approved by IDA in 1968, but was cancelled, as GOSL wanted to change the project approach and scope. The next lending operation was in 1979, when IDA approved the first Road Maintenance Project (Credit 900-CE), which became the turning point in IDA's dialogue with GOSL on highways. In March 1980, IDA approved the Road Passenger Transport Project (Credit 994-CE). 5. The Second Roads Project (Loan 2517-CE, approved April 1985) built on the work of the earlier two projects. Its objectives were to lower vehicle operating costs through a targeted investment program for priority road sections and selected bridges. The loan also envisioned strengthening the transport infrastructure institutions in the areas of transport planning, coordination and road maintenance. The Emergency Reconstruction and Rehabilitation Project (Credit 1883-CE, approved March 1988) financed road reconstruction in selected areas of the north and northeast, which had been damaged during the country's civil unrest. To follow up on the Second Roads Project, the general purpose of the Third Roads Project was to assist in rehabilitating and upgrading road infrastructure, promote better and more sustainable road maintenance, and reduce the high operating costs of selected road segments.  8 6. The on-going Colombo Urban Transport Project (CUTP) (Credit 2495-CE, approved May 1993) has three components: (a) an Inmediate Action Plan including, a transport system management program and selected road improvements, (b) a long-term Urban Transport Planning component, and (c) an institutional strengthening component, which provides consultancy services in transport regulation, technical assistance to initiate a process of reform in Sri Lanka Railways, and an Air Quality Monitoring and Management Study. 7. During the implementation period of the Third Roads Project Sri Lanka: (a) continued to suffer a major military conflict against secessionist rebels, (b) implemented a partial decentralization program, and (c) underwent a change of government in which the party that had governed for 17 years was replaced by the opposition. B - Proiect Obiectives 8. The specific project objectives were to: (a) reduce road transport cost and delays for passengers and goods by restoring major trunk roads to better operational condition, (b) help restore priority road infrastructure damaged by flooding and landslides in May/June 1989, (c) introduce institutional improvements in the planning, design, programmning and implementation of works, and the administration and maintenance of the road system, and upgrade technical standards and introduce efficient use of available materials to achieve greater cost-effectiveness, and (d) upgrade RDA's capacity to supervise and execute rehabilitation works and to identify actions needed to develop the local road contracting industry. 9. The project was to be implemented over an eight-year period and was to focus on: (a) rehabilitation of about 417 km of primary and secondary trunk routes (Class A and B roads), (b) repair, rehabilitation and replacement of 24 bridges, (c) repair and rehabilitation of road sections and associated structures damaged as a consequence of floods and landslides, (d) procurement of vehicles, equipment and supplies to strengthen contract admninistration and management, and (e) consultant services, technical assistance, and on-the-job training to improve contract monitoring and cost control capabilities. The consultancy services/technical assistance component included: project design, project supervision, and three studies on: (i) the local road contracting industry, (ii) construction materials and design standards, and (iii) road user charges. 10. The location of the road sections proposed for rehabilitation and of the bridges proposed for repair and reconstruction, were selected from a larger program proposed by RDA based on criteria of obvious risk of failure, filling network gaps, and traffic volumes (paragraph 43). At the time of appraisal there was no national road plan or priority investment. The overall scope of the project was limited by scarcity of resources and by the implementing institution's absorptive capacity. 11. The transport savings made possible by the project were expected to be passed on to the users in the form of lower operating costs and increased product availability. The project's benefits were expected to be widely distributed among all segments of the population. The flood rehabilitation component was expected to directly assist rural low income families by restoring access for agricultural inputs/outputs. The project would support private sector involvement in the road sector through assistance to the local contracting industry and it was expected to generate labor-intensive employment opportunities for residents in the vicinity of project works. 12. The objectives were straightforward, in line with the country needs, and supported IDA strategy in the highway subsector. They were appropriate in the context of the situation in the highway sector when the project was formulated. The project's components were directly linked to its objectives.  9 13. A predominant risk was the possible breakdown of the fragile political stability in the country, which could have affected implementation and costs. Flexibility was incorporated into the implementation schedule to allow for the initial contracts to be in areas unaffected by political unrest. Works were to be initially awarded at annual intervals in order to monitor the situation and enable the timing of the later contracts to be adjusted to prevailing conditions. Other risks identified were the possible weakening of GOSL's commitment to the project and over-extension of RDA's implementation capacity. C - Achievement of Obiectives 14. The objectives to reduce road transport costs and delays, and to help restore road infrastructure damaged by flooding were met by the physical achievements of the project. The main physical objectives were substantially met after some delays. Of the 417 km initially appraised, 387 km were contracted, and 310 km of roads were completed. Of this 310 km, 92 km were completed with funding from EDCF. From a total of 19 bridges, 13 were completed. RDA is currently completing outstanding works using its own resources. The greatest negative impacts are the lingering bottlenecks that will persist until the bridge works are completed. The ERRs expected by the road investments were achieved. The detailed status of the project road and bridge components as of March 31, 1999 is shown as an attachment to the Supervision/Completion Mission's Aide Memoire (Annex A). 15. Road Contracts. Five contracts for roads covering 298 km have been completed. An additional 12 km has been completed on contract WB3/3 (out of a total length of 88 kin). This contract had major problems and was not completed even with the extended Credit closing date of December 31, 1998. Bridge Contracts. Five contracts (comprising 13 bridges) out of a total eight contracts (comprising total 19 bridges) are complete. The balance of work remaining after the original and the extended credit closing dates is expected to be completed by RDA from its own resources. RDA advised IDA that it is going to recover the damages and losses to GOSL caused by contractor delays by enforcing appropriate measures in the conditions of contract. 16. Institutional change in RDA is taking place within a context of trends affecting the road sector in all countries. These global trends include: * Growing service demands leading to accelerated deterioration of an aging infrastructure, * Overall budget tightening, cost-cutting and calls for staff reductions, * Institutional focus on users as the customers, * Increasing concern with environmental and safety issues. 17. In Sri Lanka, RDA currently remains largely in a traditional public agency orientation. The regulatory environment is complicated by a multitude of processes and units. The organizational structure includes a large overall size with many operations performed by their own work force ("force account") or through other government agencies. The organization tends to focus on technical and functional areas and is still lacking a long-term strategy. There is little use of integrated modern computer systems and information processing concepts. Financing is provided through government funding, and limited attention is paid to environmental and social safeguards. 18. In the context of current global trends and RDA's current institutional capacity, the project had specific institutional development objectives including: * introduce institutional improvements in the planning, design, programming and implementation of works, and the administration and maintenance of the road system, * upgrade technical standards and introduce efficient use of available materials to achieve greater cost-effectiveness,  10 * upgrade RDA's capacity to supervise and execute rehabilitation works, * identify actions needed to develop the local road contracting industry. 19. Even though these institutional development objectives were limited, the project only partially achieved these objectives, including: * strengthened administration of projects, * improved understanding of road users, improved understanding of and support for the local construction industry, and improved design standards and construction methods, * support of the local consulting and contracting industries through their direct involvement in the project. * improvement of the technical skills of some of RDA engineers' through secondment to supervision consultants, 20. RDA's project management and administrative capability was substantially improved under the project through the establishment and organization of PMMU in RDA. Senior road and bridge engineers were assigned to PMMU, and the relevant expertise was provided to develop and operate the systems and procedures needed. Computerized programs for monitoring and reporting on the financial and physical progress of the project's road and bridge components were set up and RDA staff trained in their use. In addition, PMMU satisfactorily prepared the many road and bridge contract documents and managed the bid evaluation process. It had a temporary lapse following the departure of the expatriate expert, but additional training was given to RDA staff, and its operations were revived. PMMDvU has been successfully operating with only RDA staff since April 1997. RDA has expanded the work of PMMU to include managing and administering all foreign-aided development projects in the highway subsector. 21. The project's Technical Assistance component focused on three areas of institutional improvement: road user charges, design, and construction and maintenance. The results of these efforts are discussed below. 22. A Road User Charges Study was satisfactorily completed by Transport Studies and Planning Centre (TSPC), assisted by a consulting transport economist, in October 1992. A draft final report was published in January 1993. The objective of the study was to investigate the benefits received by, and charges currently levied upon, the different types of road user. The study recommended increases in user fees for two types of user, and because of their macro-economic implications, the study recommendations were referred to MOF/Planning. The findings and recommendations of the study were discussed at a Workshop on Transport Sector Policy organized in Colombo by GOSL and IDA in July 1996. However, the implementation of the recommendations has not taken place, and the matter still rests with MOF. 23. A domestic Construction Industry Study was satisfactorily completed by Institute of Contractor Training (ICTAD) in September 1992. The Study findings highlighted the difficulties of contracting in Sri Lanka and contained recommendations to resolve these difficulties. The findings were endorsed by the GOSL and implemented by RDA and by ICTAD with the assistance of the Contractors' Association. The local civil engineering and construction industry has benefited as a result (paragraph 26). Recently, the Contractors Association has achieved recognition of construction work as an industry in its right. GOSL just passed a Construction Industry Act to give construction, as an industry, the same access to the same development incentives as other industrial sectors. 24. A Materials. Highway and Bridge Design Study was carried out by RDA in-house with the assistance of experts on highway alignment and pavement design, and an expert on bridge design provided through international consultants. The study with the international consultants was  11 completed in August 1996. RDA is now using the findings and recommendations in its work. Among the study's recommendations were to: (a) relate designs more closely to the availability and use of local materials, (b) upgrade and enhance the skills of the local consultants in highway engineering as well as improve RDA's engineers capability in reviewing designs, and (c) upgrade and modernize the RDA bridge design methodology. To achieve this the consultants prepared manuals for highway geometric and pavement design and for bridge design. This material was circulated for discussion and further adapted to Sri Lankan requirements. 25. The following documents have since been issued by RDA to strengthen design and supervision: * Structural Design of Roads Under Sri Lankan Conditions (Geometric Design Standard of Roads), * Manual and Unified Set of Standards for Road Construction. Rehabilitation, and Maintenance, * Bridge Design Manual, * Bridge Construction Manual, * Bridge Maintenance Manual, * Inventory of Road Construction Materials, e Amendments to Standard Specifications for Road Construction and Maintenance for Roads and Bridges (1989) - Pavement Section 26. The development of the local contracting industry (for construction activities only) started under IDA's Second Roads Project and received a major boost under the Third Roads Project. Following the successful completion of a study on the construction industry done by ICTAD, agreement was reached among RDA, ICTAD and IDA on the participation of local contractors, appropriately prequalified according to their capacity and competence, in the road and bridge rehabilitation program. Most of these contractors satisfactorily completed their works and gained considerable experience for future contracting work. Two local design consultants, one local supervision consultant, and five local contractors were involved in implementing the project. Local firms accounted for 35% of IDA disbursements on construction. 27. A small number of local contractors failed to complete their contracts, and RDA made arrangements with the Contractors' Association, with IDA approval, to complete the contracts. The main cause of local contractor failure was the lack of adequate working capital resulting from inexperience in financial management. This appeared to be particularly the case with the use/misuse of the contract mobilization advance. This problem should be addressed in future contracts by better prequalification and perhaps by more flexible disbursement mechanisms. Overall partial results were achieved in private sector development. 28. Several institutional practices in RDA need further strengthening and were only partly addressed by this project. Examples include: * Competitive tendering of construction and maintenance (for works funded by IDA only), * Increased use of information technology (is now starting), * Creation of owner/provider model (i.e. road agency, not yet applied), * Innovative financing arrangements (not yet applied), * Certification program for contractors(not yet applied), * Performance management and measurement framework (not applied during project), and * Need to incorporate social assessment and resettlement safeguards into project design and implementation. (For example, greater attention and specific guidelines developed by the Borrower and IDA during and after project approval).  12 29. In a broader institutional context, more attention will need to be given in future road sector projects to: * Regulatory context - RDA needs to become more aware of external stakeholders, their interests, and valued responses, * Structural Characteristics - improving service delivery effectiveness requires streamlining certain processes and carrying out maintenance through contractors. Restructuring will require a high level of political support and strong external support (ADB, IDA, OECF), * Processes - technical assistance for process reengineering is required to assist in building organizational capability through joint client-consultant teams, * Systems -there is a need to integrate information technology awareness and its significance into all aspects of change management work (not only to discrete units), and * Financing mechanisms - road sector financing requires initiatives that provide additionality to Central Government funding. Benchmarking study tours could be sponsored to instruct and reorient leaders and financial personnel to more commercial practices and the culture of private investment. D - Maior Factors Affecting the Proiect 30. Timeliness of decision making. The slow and lengthy GOSL clearance procedures for standard contract documentation, tender evaluation, and contract award caused major delays in project implementation. Delays of 8 to 12 months in civil works procurement were not uncommnon. These delays increased costs by triggering contract price escalation clauses. There were also considerable delays in obtaining the necessary quarrying licenses despite the active support of RDA. 31. Inexperienced local contractors. Mobilization by some local contractors was slowed by their inadequate financial management and experience. Advance payments were not always managed well, leading to cash flow problems in the early stages of the contract. Poor performance of some local contractors was encountered. Some contractors alleged failure to find local materials expected in the vicinity of the project sites, which indicates poor pre-construction surveys and experience. 32. Inexperienced international contractor. One contractor and one contract represented the bulk of the substantive implementation problems in this project. This may have been exacerbated by the contractor's relative inexperience in road construction in contrast to other types of civil engineering. Weak prequalification screening may also have contributed to this situation. E - Proiect Sustainability 33. Sustainability. The project's sustainability is rated uncertain. A positive development is GOSL's recognition that the benefits of improved maintenance of existing roads should make it a spending priority. GOSL contributions to RDA have increased from 3.2% of the Public Investment Programme (PIP) over the 1985-89 period (or 33% of all contributions to transport) to 10.8% of the PIP over the 1995-98 period (or 49% of all contributions to transport). The Sri Lanka Transport Sector Strategy Study (TSSS, Report No. 16269-CE, page 21) published in January 1997 estimates (based on existing conditions and unit costs) that this trend will continue, with highways rising to 65% of all contributions to transport beyond the year 2000. While these macro-level allocations do not break out the proportion spent on maintenance, the RDA budget (next paragraph) provides this. RDA' s budget is supplied from GOSL general revenue.  13 34. Also on the positive side, RDA's commnitment of resources for maintenance has improved during the course of the project. There has been a trend to higher expenditures on mnaintenance in real terms. Table 1 (below) summarizes the RDA's maintenance budget and expenditure. Table 1, RDA Maintenance Budget (Rs millions, annual average for period shown) Time period Budget Actual Actual as Actual in Km of Actual amount amount % of constant roads under (constant)/km, spent budget (1990) value RDA (Rs thousands) 1985-1989 309.3 213.3 69.0 % 273.0 25,600 10.7 1990-1994 545.5 543.2 99.6 % 446.9 10,400 42.8 1995-1998 1068.7 946.6 88.6% 521.8 11,133 46.9 1999-2004 1416.2 ... ... 588.7 11,152 52.8 _______________________ ___ __ __ ___ __ _ _ (budgeted) _ _I Sources: SAR, RDA (Annex B, routine plus periodic maintenance). 35. On the other hand, it is still not clear if the increases in spending are sufficient to guarantee sustainability of the strategic road system. Until 1990, RDA was responsible for 25,600 km of roads. Since 1990, with the devolution of 15,300 km of roads to Provincial Councils (C, D, & E roads) and other minor adjustments, RDA has been responsible for approximately 11,100 km of roads. 36. Recent information for RDA shows that during the 1995 to 1998 period, expenditures for maintenance were Rs 946.6 million per year (Rs 521.8 million constant 1990 value). The budgeted amounts for 1999 to 2004 are approximately Rs 1,416.2 million per year (Rs 588.7 mnillion constant 1990 value). 37. The predominant form of periodic maintenance used by RDA is sand sealing, a surface treatment with an effective life of about two years. In the early 1990s, RDA completed roughly 1,700 km of sand sealing per year. During the 1996 to 1998 period, RDA provided about 2,500 km of sand sealing each year, although the annual figure varies. This represents about 23% of the network each year. With its effective life of about two years, this effort is sufficient to mninimally maintain about 46% (23% x 2 years) of the total A and B roads network or, alternatively, to maintain the network with resealing, on average, every 4.3 years. Detailed information on the location of roads receiving sand sealing compared to their traffic volumes is not available. The Borrower's contribution to the ICR (Annex B) includes a summary of their Operational Plan for the immediate future. Details on the conditions of the road network are shown in Annex C. 38. There is further evidence of GOSL attention to the importance of maintenance in the highway subsector, through its increasing transfer of funds for maintenance of Provincial Council roads. This is a relatively new level of government, established in 1989. Since then the Provincial Councils are responsible for 15,300 km of Class C, D, and E roads. Through the Ministry of Transport and Highways (MTH), GOSL was transferring approximately Rs 90 million per year to the Provincial Councils prior to 1996. Since then the transfers have increased substantially: Rs 800 million in 1996 and 1997 and Rs 1,000 million in 1998 and 1999. The budget is Rs 1,500 million in 2000. The transfer mechanism is also evolving. As the Councils have developed administrative experience, they have taken on more direct responsibilities. Beginning in 2000, the Treasury will make the transfers directly to the Provincial Councils. There is no data available yet  14 to determine the effectiveness of these increased expenditures. Performance of the Provincial authorities in maintaining the secondary road network under their responsibility is critical to ensure sustainability of the road network. 39. MTH will continue to support to Provincial Councils through setting standards, regulating, training, technical assistance, and preparing District Plans (25 in total, with a 10 year planning horizon). A Road Sector Master Plan will be available in 2000 to guide priorities for the country as a whole. The design manuals and standards prepared through this project are being disseminated to all Provincial Councils and local municipalities to supply road and bridge standards. 40. RDA has improved its ability to undertake and manage construction and maintenance contracts with the formation of PMMU. The local contracting industry has increased its experience and skill in performing maintenance work. Both of these developments will contribute to better sustainability of the road sector. However long term sustainability will also depend on having an adequate stream of on-going construction contracting, to provide the newly created industry with a steady supply of work. 41. A project covenant not complied with was the implementation of recommendations from the Roads User Charges Study that taxes on diesel fuel be increased (raising the retail price, in 1992 figures, from Rs 11.55/1 to Rs 14.10/1, to be equivalent to petrol (gasoline). This represents an increase of 22%) and that annual registration fees for heavy trucks be increased (times three, to Rs 12,000). These recommendations have been under consideration by the GOSL since the report was completed in 1993. Because of their macroeconomic implications, the study recommendations were referred to MOF/Planning, but implementation of the recommendations has not taken place, and the matter still rests with MOF. During supervision, the lack of compliance with the project covenants was pointed out. F - IDA Performance 42. Project Identification. Project identification was satisfactory. This project was a continuation of earlier road development projects supported by IDA. The first road project was prepared by IDA staff, based on the country's list of road priority projects. Identification for the Second Roads Project was more detailed. It involved a preparation process whereby less than half of the detailed selection work was done by local consultants and the balance by international consultants. In the Third Roads Project about two-thirds of the identification effort was done by local consultants with only one-third being done by international firms. Realignment of road sections, or junctions, was not considered in project design. This was because (a) both IDA and Borrower sides felt that requirements for land acquisition would significantly delay implementation, and (b) delays in implementation would postpone - and thus lower - the benefits expected from the project. IDA's policy and guidelines on resettlement planning now require that a detailed social assessment of road rehabilitation and construction projects be conducted during project preparation. More attention to realignment and road junctions should be given during design of future road projects. 43. The selection of road segments for analysis and incorporation into the project was based on: (a) risk of physical failure, (b) completing gaps remaining in the priority road network from earlier investment programs, especially around Colombo, (c) expected areas of urban development around Colombo, and (d) longer distance access in the fast growing region south of Colombo. This explains the apparently "patchy" selection of roads around Colombo and the different set of inter-city roads selected to the south.  15 44. Preparation Assistance. Preparation assistance was satisfactory. Part of the effort to develop capacity in the highway sector has been to encourage an understanding of the value of maintaining existing roads rather than simply building new ones. This includes not diverting maintenance funds into reconstruction or new roads spending. 45. After the first and second roads projects, more attention was paid to the sector environment and the implementation details in this project. During project preparation an effort was made. to make better use of local consultants, backed up by international consultants, to establish the details of the investments required. In this project considerable effort was placed on clearly defining and describing the types of investment including road segments or bridges to be upgraded and presenting the analysis in detail in the appraisal report. Performance indicators were not included to monitor and evaluate the project during implementation. 46. Appraisal. The appraisal work was satisfactory. The Staff Appraisal Report (SAR) clearly documents the full details of the investment program and its analysis. At appraisal the final engineering designs were complete for about 200 km of roads and six bridges, nearly half of the project roads, which allowed a fast start on implementation. 47. The technical review of the project was detailed and robust. Cost estimates and economic analysis were completed for 25 road segments and 26 bridges. Traffic volume data provided by RDA were selectively cross-checked using field surveys conducted by Moratuwa University. The economic evaluation included a switch value analysis demonstrating that a cost increase of 418% or a benefits reduction of 81 % would be needed to drive the project to a zero net present value. 48. The SAR shows an understanding of the limits of the Borrower's procedures and provides remedies for this by setting up PMMU, the appointment of supervision consultants, the requirements for monitoring and an annual action plan showing adequate resources committed from GOSL. The possibility for delays was built into the original timetable by following the standard disbursement profile for this sector. 49. Supervision. With a total of 17 supervision missions over the lengthy 8-year project implementation period - or slightly over 2 missions each year - IDA supervision and progress reporting were satisfactory, and the performance ratings were appropriate. Performance indicators, however, were not retrofitted during project implementation. The slow progress in contract execution and cost control were the main concerns of IDA's supervision missions. The contract execution delays resulted mainly from the unsatisfactory performance of one of the two expatriate contractors and from the inexperience of some of the local contractors, which was expected given the nature of this component. The individual contract and component costs and the aggregated project cost and related disbursement percentage were closely monitored during supervision with PMMU. 50. There was considerable disappointment with the performance of an international contractor on Contract WB3/3 (Galle - Matara), and an inordinate amount of IDA supervision time was spent on this contract. The contractor failed to fulfill his contractual obligations and effectively abandoned the works, having completed only 12 km out of 88 km. The contractor had been prequalified by RDA as fully compliant. The bid came in much lower in price than that given in the Engineer's estimate and was the lowest bid received. The Tender Board, because of this low bid price, then recommended against awarding the contract to this contractor. However, IDA insisted on following the Procurement Guidelines (i.e., must accept the lowest responsive bid) and on having the contract awarded, albeit with an increased Performance Guarantee and a Detailed Implementation Program to be closely monitored. Once awarded, it became difficult for RDA to remedy the situation despite almost continuous urging by IDA supervision missions for  16 appropriate actions as allowed under the conditions of contract. The Borrower perceives IDA to be non-responsive on procurement decisions. 51. IDA's prompt and flexible responses to proposals dealing with the many problems experienced with the local contractors and to other project related matters conceming increased costs and revisions to disbursement percentages were a commendable feature of IDA's supervision performance (paragraphs 27, 60). 52. Apart from monitoring and reporting on the physical and financial progress of the project and ensuring that the suitable quality controls were in place, the supervision missions concentrated on the following important project matters: * institutional development of RDA's project management and administration, * project cost controls, * development of local contracting industry, and * studies on Road User Charges, Domestic Contracting Industry and Road Materials, Design and Construction Standards. 53. The extension of the credit closing date from June 30, 1998 to December 31, 1998 was undertaken at the request of GOSL and in the expectation that with this extension road contact WB3/3 and bridge contract WB3/29 could be completed. Despite the extension neither of these two contracts were completed. G - Borrower Performance 54. Project Identification and Preparation. Overall, the Borrower's project identification and preparation activities were satisfactory. The Borrower faced some difficulties in project identification and preparation because basic planning information was lacking, e.g., out-of-date or non-existent topographic surveys and pavement conditions data for some road segments. Another difficulty in project preparation was the specifications required by RDA for bridge design. RDA uses designs which are both material and labor intensive, and which result in slow construction. Requiring the use of these designs raised expected costs and reduced the amount of work planned. Adopting the Bridge Design Manual, prepared as part of this project, should improve future project preparation. 55. Implementation. The Borrower's performance in implementing the project was generally satisfactory. Following the system set up under the IDA-supported Second Roads Project, RDA managed the project through a Contract Construction Division. The Division was headed by a Director with Deputy Directors responsible for road and bridge contracts on a district basis. The Division was effectively supported by PMLMU. 56. Lengthy GOSL bid document approval and tender evaluation procedures (sometime taking up to 8 - 12 months) were the main causes of delays in project implementation. These procedures were discussed by IDA missions at all GOSL levels involved, and although the financial thresholds were raised to allow clearances by various departments and ministries, the procedures are still inordinately lengthy. This problem needs to be resolved by senior IDA staff with high level GOSL officials. 57. Expatriate consultants in joint venture with a local consultant assisted RDA in reviewing and revising designs done largely by local consultants and in the supervision of road works. Their performance on supervision was less than satisfactory on occasion and RDA had to request the removal and replacement of the engineer's representative because of his failure to apply strictly the specifications on drainage in one instance, and a seeming reluctance to recommend appropriate  17 contractual action against the poorly performing expatriate contractor. The replacement engineer's representative performed satisfactorily. 58. Local consultants were involved in the design of road works and performed satisfactorily. However, there was the problem of lack of responsibility between the design reviewer and the original designers in cases where errors of center-line location and resulting construction quantities occurred. 59. The bridge structure designs were done by RDA's Bridge Division. The designs were archaic and not cost effective. The Bridge Design Manual will help RDA upgrade and modernize its bridge designs. RDA has adopted the new Bridge Design Manual standards and is using them in other donor agency development projects as well. 60. In recognition of RDA's initially limited capacity to manage projects the project schedule allowed for a gradual start up with only one contract being tendered at a time. Much higher than expected cost estimates, which came in with the initial rounds of bids, delayed awarding the contracts while cost reduction approaches were examined, including alternative designs and split contracts. This need for redesign along with the limited capacity of local design contractors slowed the design work. The need for contract document re-drafting and re-approval also delayed letting contracts. 61. Despite the difficulties involved in managing inexperienced contractors and a reluctance to take appropriate contractual action against poorly performing contractors, RDA implemented the project satisfactorily and is to be commended for their flexible solutions in dealing with contractors' cash flow, equipment and materials problems (paragraph 27). 62. Project Covenants. Overall the project covenants compliance is rated satisfactory. The project covenants were complied with except for adopting the recommendations of the Road User Charges Study. As described in the sustainability section above, the study recommended a slight increase in the retail price of diesel fuel and a substantial increase in the annual license fee for heavy trucks. These recommendations have not been adopted. The recommendations were made to more carefully match the contribution of the road system users to the maintenance costs they create. GOSL's lack of acceptance of the recomnmendations, which has left in place inappropriate price signals, is a significant barrier to efficient evolution of the transport system. This issue is critical for sustainability of maintenance, indicating that future involvement in the sector should pay close attention to implementation of the study' s recommendation. 63. Operation. The Borrower's actions have been satisfactory. PMMU is an on-going operational aspect of this project. Setting it up was a condition of the Credit (Schedule 4-1). PMMU was operational with an expatriate Advisor by the time of the first supervision mission (March 1991). Despite a change in the Advisor after one year due to health reasons, and a period of staff turnover (April 1995 to May 1997) when no senior RDA staff were commnitted to PMMU, RDA has since supported it with adequate staffing, resources and attention. Its work has been expanded to manage and administer all foreign-aided development projects in the highway subsector. In its present form, PMMU is sustainable. Future work in this area would be to build on the initial success and to improve the supervision ability for more difficult contracts. H - Assessment of Outcome 64. Overall Outcome. The overall outcome of the project was satisfactory. This rating is based on the satisfactory, if partial, achievement of the physical objectives, the improved institutional capacity of RDA, and the development of the local contracting industry. The project would have been highly satisfactory if better project management for road contract WB3/3 had  18 resulted in problems being addressed in time to have an alternate contractor complete the proposed works. 65. Economic Evaluation. Economic analysis was undertaken to compare estimated ERR (from the SAR) with achieved ERR (based on actual results to date) for the project's roads segments. The SAR showed an overall ERR of 33% for the project, with a range for specific road links from 22% to over 80%. The revised analysis prepared by the Borrower shows an overall ERR of 52%, with rates of return ranging from 21% to 125% for specific road links. Overall traffic volumes on project road segments, at opening, were 14% higher than estimated in the SAR. Details by road link and assumptions used are provided in Annex C. I - Future Operation 66. In support of IDA's Transport Sector Strategy, future IDA support should address institutional change and policy reform of RDA: (a) from a public oriented to a mixed model agency, (b) to improve the efficiency of management of strategic needs, and (c) to improve human resource development. Institutional strengthening and RDA staff training should be continued, with emphasis on project management, environmental and social assessment, and planning. 67. GOSL needs to substantially increase the budget for roads to implement RDA' s ambitious priority investment plan for the next decade. This plan will require substantial resources for development and maintenance. At present, District Plans (including district road master plans) are under development. For any plan to be implemented it will be necessary to (i) increase the efficiency of expenditure, (ii) identify new sources of financing, (iii) attract private investment to complement public sector investment, and (iv) increase attention to sustainable and effective maintenance. 68. With the devolution of 15,300 km of roads to the Provincial level and the gradual emergence of road agencies for these subnational levels of government, there is an opportunity for RDA and external funding agencies to assist the newer agencies in developing their maintenance capability, including identification of sustainable maintenance funding sources. 69. Future investments in the highway subsector of Sri Lanka must balance maintenance with upgrading of strategic roads to provide additional road capacity. Past IDA projects have been concerned mainly with the physical rehabilitation of road pavements only. No realignment improvements or new alignments have been built, and no substantial increase in road capacity has been provided. Resettlement and environmental safeguards should be better incorporated into project design. 70. More cost-effective maintenance practices need to be adopted in view of the huge backlog in periodic maintenance. The lower cost single and double bituminous surface treatments must be considered and the more costly Asphaltic Concrete (AC) surfacings only used when additional pavement structural strength is needed. This would also help the local contractors since it is less equipment intensive. 71. The number of traffic accidents, already very high on some major roads, will increase dramatically. Road safety measures are already urgently needed. An effort should be developed with RDA/GOSL to support a comprehensive, staged, development plan for the expansion of road network capacity and the ancillary works needed to ensure a cost effective and safe road transport system.  19 72. IDA should assist RDA/GOSL in implementing a bridge replacement and rehabilitation program. This should be based on RDA's bridge condition inventory, which shows that a very high proportion of the existing bridge structures need to be either replaced or rehabilitated. J - Key Lessons Learned 73. The key lessons learned from the project are as follows: * Institutional reform is needed to enable RDA to move from a public oriented agency to a mixed agency and should: (a) improve efficiency through outsourcing of maintenance to contractors, (b) incorporate information technology in road management, (c) create an owner/provider model, (d) develop environmental and social safeguards in road development and maintenance, and (e) streamline management and decision-making. External financing agencies will play a major role in supporting RDA in the reform process as part of road sector lending. The availability of large external aided support, with limited conditionality, for new road development is a factor to take into consideration. Any future road sector lending should be preceded by a dialogue on reform with GOSL, RDA, and other major donors (paragraphs 28, 30, 56, 66). * GOSL should empower the Tender Committee and a revise the clearance procedures for the approval of bid documents to ensure timely contract awards. This could minimize project delays and improve decision-making (paragraphs 30, 56, 60). * During project design greater attention should be placed on: (a) planning and prioritization of investments, (b) holistic design to address safety issues, environment, and social safeguards, (c) monitoring and evaluation systems including performance indicators, and (d) sustainability, including funding and mechanisms for road safety. Appropriate design and standards, including attention to resettlement, environment, and alignment considerations, are critical to achieve efficient and effective results in construction and maintenance (paragraphs 42, 60, 69, 70, 71). * Support for local contractors needs to be improved, especially when fostering development of a new set of private sector companies. The lack of adequate working capital resulting from inexperience in financial management delayed several contractors (paragraphs 27, 31). Consideration should be given to a more comprehensive development program for the private construction sector as a whole, including more than simply updating government standards. There is also a need to have an adequate stream of on-going construction contracting, to provide the newly created industry with a steady supply of work (paragraph 40). * Decentralization of responsibility for the road network - if not well orchestrated - is likely to have an adverse effect on the sustainability of the strategic road network and the smooth functioning of road transport (paragraphs 38, 39, 68). * To ensure sustainable financing of road maintenance, it is critical to implement reforms in the road user charging system (paragraphs 22, 41, 62). * Bridge opening delays have a major effect on users by creating prolonged traffic bottlenecks at key locations. The economic analysis done at appraisal should explicitly consider the costs of delayed bridge construction in its risk assessment. The appraisal should also consider the potential for bridge construction delays in its treatment of risk mitigation and should include alternative designs (paragraph 14).  IMPLEMENTATION COMPLETION REPORT SRI LANKA THIRD ROADS PROJECT CREDIT 2183 - CE PART II - STATISTICAL ANNEX  21 TABLE 1: SUMMARY OF ASSESSMENTS A. Achievement of Objectives Substantial Partial Negligible Not applicable Macro Policies O O O / Sector Policies O / E