Nepal - Road Maintenance and Rehabilitation Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No, 19952 IMPLEMENTATION COMPLETION REPORT (Core ICR) ONA CREDIT IN THE AMOUNT OF SDR 36.6 MILLION (US$50.5 MILLION EQUIVALENT) TO THE GOVERNMENT OF NEPAL FOR A ROAD MAINTENANCE AND REHABILITATION PROJECT PROJECT ID: P010475 L/C NUMBER: 25780 - NEP DECEMBER 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.  CURRENCY EQUIVALENTS Currency Unit = Nepalese Rupees (NRs.) US$1.00 = NRs.45.47 (Appraisal May, 1993) US$1.00 = NRs.49.26 (FY 94) US$1.00 = NRs.49.94 (FY 95) US$1.00 = NRs.55.22 (FY 96) US$1.00 = NRs.57.03 (FY 97) US$1.00 = NRs.61.95 (FY 98) US$1.00 = NRs.67.95 (FY 99, until December, '99) GOVERNMENT FISCAL YEAR Mid-July to Mid-July WEIGHTS AND MEASURES Metric British/US Equivalent 1 kilometer (km) 0.62 miles (mi) ABBREVIATIONS AND ACRONYMS ADB Asian Development Bank BUMP Bridge Upgrading & Maintenance Program DDC District Development Committee DFID Department for International Development DOLIDAR Department of Local Infrastructure Development & Agriculture Roads DOR Department of Roads DRTU District Roads Technical Unit DTMP District Transport Master Plan EFYP Eighth Five Year Plan EIRR Economic Internal Rate of Return ERROM Eastern Region Road Maintenance Program FAR Financial Administration Regulation FIA Field Implementation Advisors FY Fiscal Year GEU Geo-Environment Unit HMGN His Majesty's Government of Nepal HMIS Highway Management Information System ICB International Competitive Bidding ICR Implementation Completion Report ID Institutional Development IDA International Development Association IRI International Roughness Index LRCC Local Road Coordination Committee  FOR OFFICIAL USE ONLY ABBREVIATIONS AND ACRONYMS (cont.) MLD Ministry of Local Development MOF Ministry of Finance MRCU Maintenance & Rehabilitation Coordination Unit NCB National Competitive Bidding NPC National Planning Conmission ODA Oversees Development Agency PAP Policy Action Program PIP Priority Investment Plan PLRP Pilot Labor-based District Road Maintenance & Rehabilitation Project QAG Quality Assurance Group RMDP Road Maintenance & Development Project RMP Road Maintenance Project RMRP Road Maintenance & Rehabilitation Project RPA Regional Procurement Advisor RSSDU Road Sector Skill Development Unit SAR Staff Appraisal Report SCAEF Society of Consulting & Architectural Engineering Firms SDC Swiss Development Agency for Cooperation SMD Strengthened Maintenance Division SRIP Second Road Improvement Project TESU Traffic Engineering & Safety Unit TRIP Third Road Improvement Project UNDP United Nations Development Program USD US Dollar VDC Village Development Committee Vice President Mieko Nishimizu Country Director Hans M. Rothenbuihler Sector Director Frannie Leautier Team Leader Guang Zhe Chen Task Team Leader Amer Zafar Durrani 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.  TABLE OF CONTENTS Page 1. PROJECT DATA ...................................................................................1 2. PRINCIPAL PERFORMANCE RATINGS ...................................................................................1 3. ASSESSMENT OF DEVELOPMENT OBJECTIVE AND DESIGN, AND OF QUALITY AT ENTRY .. 2 4. ACHIEVEMENT OF DEVELOPMENT OBJECTIVES AND OUTPUTS ................. .................................6 5. MAJOR FACTORS AFFECTING IMPLEMENTATION AND OUTCOME ............................................ 13 6. SUSTAINABILITY .................................................................................. 16 7. BANK AND BORROWER PERFORMANCE .................................................................................. 17 8. LESSONS LEARNED ................................................................................... 20 9. PARTNER COMMENTS BY DFID (ODA) .................................. ................................................ 21 10. ADDITIONAL INFORMATION .................................................................................. 22 iv  Project ID: P0 10475 Project Name: Road Maintenance & Rehabilitation Team Leader: Amer Zafar Durrani TL Unit. SASIN ICR Type: Core ICR Report Date: December 21, 1999 1. Project Data Name: ROAD MAINTENANCE & L/C Number 25780 REHABILITATION Country/Department. NEPAL Region South Asia Sector/subsector: Infrastructure/Transport KEY DATES Original Revised/Actual PCD January 6, 1992 January 6, 1992 Appraisal May 15, 1993 May 15, 1993 Approval March 15, 1994 March 15, 1994 Effectiveness April, 1994 September 12, 1994 MTR August 1, 1996 March 23, 1997 Closing June 30, 1999 June 30, 1999 Borrower HIS MAJESTY'S GOVERNMENT OF NEPAL Implementing Agency DEPARTMENT OF ROADS (DOR) Other Partners ODA, SDC, UNDP STAFF Current At Appraisal Vice President Mieko Nishimizu D. Joseph Wood Country Director Hans M. Rothenbuihler Ann 0. Hamilton Sector Director Frannie A. Leautier Marie Garcia-Zamor Task Leader of ICR Amer Zafar Durrani ICR Primary Author Amer Zafar Durrani Task Leader of Appraisal Juan Gaviria 2. Principal Performance Ratings Outcome: Satisfactory Sustainability: Likely Institutional Development Impact: Satisfactory Bank Performance: Satisfactory Borrower Performance: Unsatisfactory QAG (if available) ICR Quality at Entry: n/a Satisfactory Project at Risk at Any Time: Yes (Dec 1996 through Nov 1997) 1  3. Assessment of Development Objective and Design, and of Quality at Entry 3.1. Original' Objective: The primary institutional development objective of the Project was to address the development of road maintenance and rehabilitation capabilities and to support maintenance and rehabilitation activities, by: (a) insuring adequate funding of priority maintenance activities by establishing funding and budgeting arrangements which would channel a much greater, timely flow of funds to support maintenance within sectoral allocation ceilings based on a rolling 3-year priority investment program to be revised annually and reviewed with IDA; (b) strengthening sectoral planning by introducing network-based planning, initially to prioritize maintenance and thereafter to support the development of a Priority Investment Plan (PIP) and review process for all sector investments; (c) improving public sector implementation capacity to expand the coverage and enhance the quality of maintenance with a program of institutional strengthening and human resource development based primarily on training and technical assistance inputs (establish regional maintenance capabilities, coordinate and develop a comprehensive training program for DOR's staff) and increasing incentives to DOR staff with increased funding levels for road maintenance, enhanced quality control, and introduction of performance targeting (d) supporting development of private sector capabilities to do maintenance on contract by supporting on-the-job training, improving procurement of maintenance contracts, gradually developing a critical mass of yearly or multi-yearly maintenance contracts, streamlining contract supervision and insuring timely and regular funding and payments; and (e) testing ways to develop DDCs' road maintenance capabilities for local access roads (i.e., district roads) through a pilot project. 3.1.1 The direct economic impact of the project was to reduce vehicle operating costs and the delays associated with a deteriorated road system and generating much-needed rural employment through adoption of labor-based technologies, involving appropriate and efficient combination of existing capital or labor intensive techniques. 3.1.2 The project objectives were clear, and * was based on extensive consultations with stakeholders - specifically the Policy Action Programme (PAP) Workshop2 which emphasized the need for better planning, budgeting, and funding of road maintenance, and improving the road maintenance delivery mechanism/institutions, * supported IDA's country assistance strategy and sector strategy - shifting focus of sectoral allocations to maintenance, completing essential new construction, improving farm-to- market roads in Terai, and accessibility in the hills, Staff Appraisal Report, March 15, 1994 2 The Poticy Action Programme (PAP) Workshop held in Kathmandu during February 23-25, 1993. 2  * assisted HMGN's Eighth Five Year Plan (EFYP), 1993-97 for the sector, that was moderately consistent with IDA's strategy in the sector, and targeted maintaining the existing highways and feeder roads, establishing a revolving fund, and improving rural access, while also supporting new road construction. 3.1.3 Having said that the project objectives were responsive to HMGN/DORs circumstances - which included an absence of cohesive sectoral planning, neglect of road maintenance, a patchy project implementation track-record especially in the past four IDA projects, extra- ordinary administrative inefficiencies in the entire HMGN bureaucracy, governance issues, and highly active donors - the project objectives were ambitious. 3.1.4 There was some disconnect between the Bank strategy and the 'politically motivated' development priorities of HMGN. Therefore, even though EFYP did mention sustainability of investments, it did so with a distinctly 'new construction' flavor. This disconnect could be attributed to the divergence between the 'development' and the 'political' priorities in Nepal. The project objectives (a) and (b) targeted this issue, and the project included such requirements as the development of three year rolling plans, and the preparation and use of the Priority Investment Plan (PIP). 3.1.5 The project was complex. There were in excess of forty (40) sub-projects3 (works/projects, goods, and services) scattered in four of the five regions in Nepal. There were four co-financiers (which in itself was a wonderful achievement, and should not be derided) along with the government with, at times, varying objectives. The DOR's institutional improvements were elaborate, with fifteen or more areas of intervention in parallel. Post-facto interviews with project stakeholders suggest that the because of its complexity, the project was demanding for HMGN/DOR(and MLD) - the very fact that the project undertook such detailed and comprehensive institutional strengthening measures is indicative of the level of starting institutional capability prevalent at that time. 3.1.6 In summary, the project objectives were clear and responsive to the borrower circumstances; but at the same time they were ambitious. The project design was complex, and demanding for the Borrower. 3.2 Original Components: The project components designed to achieve the objectives were as follows. (a) Road Rehabilitation and Overlays: Supporting rehabilitation and overlays of 320 km of selected highways and 133 km of feeder roads of the strategic road network (30% of total Project cost) - comprising nine sub-projects. (b) Resealing and Re-gravelling of Strategic Network: Supporting a backlog periodic maintenance program of roads and bridges of the strategic road network in the Central, Western, Mid-Western and Far-Western regions including the resealing or gravelling, and minor bridge repairs, to 609 km of route (37% of total Project cost) - comprising twenty four (24) sub-projects. 3 Each sub-project procurement requiring anywhere from three to twelve months based on IDA's procurement guidelines. 3  (c) Pilot Labor-based District Road Maintenance and Rehabilitation: Supporting rehabilitation and maintenance of 147 km in 4 selected districts and promotion of appropriate rural transport technology to encourage community participation in road maintenance and related activities (6% of the total Project cost). (d) Institutional Development and Training: Supporting institutional strengthening of MOWT, DOR and four DDCs, and includes components financed in parallel by IDA, ODA and SDC (16% of total Project cost). The IDA financed components were: * Field Implementation Advisors (FIA) who provided substantial support for periodic maintenance activities, including quality control and strengthening the capacity of domestic contractors and consultants; * Continued support to the Mechanics and Operators Training Center: involved extensive training and improvement to the training center; * Consultant services for - - special studies on road construction and maintenance materials - report completed and now widely distributed; - assistance for the preparation of the Priority Investment Plan (PIP) - reports and strategy forms basis of most externally aided programs and partly adopted by HMGN/National Planning Commission of Nepal, 4 follow-up of the road contractor industry study - workshop held and working group subsequently established; ' conduct of an operational audit of maintenance and rehabilitation operations - completed; 4 development of a management plan for Bardiya National Park to address adverse environmental impacts of completed road construction under Cr.1515-NEP - not undertaken, due to overlap with U NDP activity, although strong mitigation actions established on the highway; and X preparation study for follow-up to pilot labor-based district road component. * Specialist assistance for maintenance of the Karnali Bridge - subsequently not required as inspections undertaken by DOR staff; * Implementation of the sector environmental assessment - completed; and * Vehicles and Equipment - procured and used on project activities. 3.2.1 The UNDP would support the District Road Services Development Unit the Ministry of Local Development (MLD) and labor-based training school at Butwal - completed: district transport master plans (DTMP) developed and contractors trained. 3.2.2 The SDC and DFID (then ODA) financed components were: * Establishment of a Road Sector Skills Manager position in the Road Sector Skills Development Unit (RSSDU) - done: skill assessment undertaken * Establish and organize on-the-job management training for DOR's staff and a small training/study exercise * Contractor training in cyclic maintenance operations - done: strengthened works quality and capacity of DOR and contractors * Bridge management - system established and data collected 4  * Geo-environmental/bio-engineering unit - Unit very effective in disseminating improved practices for sustainable slope stability protection * Plant management (Heavy Equipment Divisions/Mechanical Training Center) -major progress achieved in charging for external use of DOR equipment, leading to more efficient utilization and repairs, and increased private purchase of equipment * Road Safety Unit (TESU) - substantial progress in introduction of road safety analysis and mitigation measures, including two pilot implementations under the rehabilitation component * Strengthened Maintenance Divisions (SMD) - substantial improvements in effectiveness of routine maintenance management, including introduction of one-person contractors ("lengthmen") responsible for short road segments (e) Routine Maintenance: Supporting a gradually developed comprehensive routine maintenance program by contract (11% of total Project cost). 3.2.3 The project components' design was well linked to achieving the objectives set forth. The project's Institutional Development (ID) components were comprehensive and quite extensive: they took into consideration the project implementation requirements as well as long term DOR development needs. The works components were designed to leverage the ID interventions, such as the requirement for developing an annual rehabilitation and maintenance work program, with detailed prioritization, economic feasibility, and environmental assessments - hence building in a mechanism for utilizing products from all the ID interventions during project implementation. The project design broadly drew on relevant past lessons from prior projects in Nepal's road sector with a few exceptions. 3.2.4 The project components' design was ambitious with regard to DOR. The implementation schedule was unrealistic in relation to the implementation capacity within DOR/MLD/HMGN and not enough recognition was given to the overall HMGN system. Past experience had showed that shortage of funding had not been 'the' problem in the road sector in the previous ten years rather, it has been the lack of ability to coordinate, administer, and implement the funding where the problems have arisen - basically administrative and financial management capacity. Consider the following facts. At the time of project appraisal, and effectiveness, there were thirteen (13) donors active in the road sector and over fifty (50) projects in progress, or under negotiation. A review as part of the recently approved RMDP (Road Maintenance and Development Project) recognized that the demonstrated weaknesses in the financial management capacity of the DOR4. Given this, although the project start-up 'hiccups' associated with such an environment were recognized in the SAR, the time required for overcoming these hiccups was not factored into the project components' design. 3.2.5 There seemed to have been some divergence in views, at the time of project preparation, regarding DOR's capability to improve delivery of road maintenance and rehabilitation activities, as well as new road construction. Some views, based on lessons learnt from 4"..the level of coordination between various component managers was not adequate for purpose of efficient operations. There was only one accountant ofjunior level responsible for coordination and monitoring of financial transactions. In the first few years the project did not maintain basic financial control records .. frequent staff changes affected operations." Note by the Financial Management Specialist in the World Bank Office in Kathmandu. 5  previous road sector projects in Nepal and conclusions from the 1993 Policy Action Programme Workshop on Road Maintenance, indicated that the institutional capacity of DOR5, being a part of the broader HMG's civil service set-up, can not be improved in isolation, as has been demonstrated over the past 20 years. Hence it had been proposed that road maintenance operations should be planned outside the normal civil service - "to develop a non-governmental roads agency to provide a service to the government"6. 3.2.6 The project preparation team recognized the institutional weakness of DOR and designed a comprehensive program for strengthening its capacity. The involvement of four key donors in the road sector in the project through co-financing was a major step in improving donor coordination and consistency in policy dialogue which, it was hoped, would lead to substantial pressure on DOR to improve. However, the implementation of such comprehensive program itself turned out to be too demanding for the HMGN system (DOR/MLD/MOF/HMGN). 3.3 Quality at Entry: The quality at entry assessment at the ICR stage was based on the seven key dimensions: project concept, objectives and approach; technical, economic and financial analyses; environmental analysis; social and stakeholder analysis; institutional capacity analysis; readiness for implementation; and, risk assessment and sustainability. The project's quality at entry is rated satisfactory7 as: the project concept and objectives were clear and responsive to demand; the technical and economic analysis was satisfactory (most of the problems during implementation arose due to time elapsed between design and actual implementation); substantial stakeholder analysis; and a satisfactory risk assessment. The few shortcomings included: questionable implementation readiness; a financial management analysis (which was sufficient by the standards prevailing within the Bank at the time, but not by current Bank standards); absence of a defined exit strategy and unclear linkage to alternative project design; complexity; and some oversight on ability of the HMGN system to implement the project. 4. Achievement of Development Objectives and Outputs 4.1 Outcome/Achievement of Objectives and Output by Components: The project was able to improve road maintenance capabilities (as indicated in the original objectives) and supported maintenance and rehabilitation activities, while delivering a significant part of the envisioned economic impact of the project by targeting and improving significant sections of the strategic road network with high traffic flows. The project through, its development of the PIP as a base document, which was accepted by all donors and DOR (albeit less popular with HMGN), and its support in development of basic maintenance management systems and techniques, was able to effectively raise the profile of road maintenance needs and brought about a significant shift in DOR/donor strategy on roads in Nepal. This resulted in an overall improvement in the condition of the strategic road network. The overall project outcome is An agency employing 300 plus graduate engineers-traditionally DOR employs over 10 percent of the engineers in Nepal. 6Road Maintenance: Policy Action Programme Workshop and Background Papers - Workshop Proceedings, February 1993, Department of Roads, Ministry of Works and Transport, Nepal. 7 Based on a four point scale used by the Bank's Quality Assurance Group (QAG), i.e., Highly Satisfactory/Best Practice, Satisfactory, Marginally Satisfactory, and Unsatisfactory, the ICR team would have rated the quality at entry as marginally satisfactory. 6  satisfactory (see Annex 5) even though the wider HMGN system continued to be plagued by inefficiencies. 4.1.1 In 1992 over half (52%) of the main road network8 was recorded as being in poor condition, with a further 36 percent in fair condition only. By 1999, the proportion of the Strategic Highway Network9 in good and fair condition had increased to 88 percent of the total. Despite the reduction in scope, the project was able to implement works, under the rehabilitation and periodic maintenance components, on more than 30% of the network. A comparison of Strategic Highway Network condition between 1992 and 1999 is shown below: Network Condition Good Fair Poor Before the Project (1992) 11% 36% 52% After the Project (1999) 31% 57% 12% 4.1.2 While these improvements to the overall condition of the road network cannot be attributed solely (or directly) to the RMRP, they are indicative of the change in condition and emphasis towards maintenance that was achieved over the project period. 4.1.3 The considerable improvement of the Strategic Network condition over the past seven years is due primarily to the substantial programmes of 'back-log' maintenance undertaken over this period - including the Rehabilitation and Periodic Maintenance activities under the RMRP. Other major initiatives during the same period include the ADB Second and Third Road Improvement Project (SRIP and TRIP), the ODA-funded Eastern Region Road Maintenance Programme (ERROM) and the SDC-funded Arniko Highway Improvement Project. 4.1.4 The total length of road completed under the Rehabilitation Component was reduced by 50 percent, from the original 453 km to 228 km. The revised project list as actually implemented is presented in Annex 3, which includes also the actual expenditures, periods of construction, and roughness and traffic data. The overall costs of the component have increased from Rs 860 million in the SAR to an actual expenditure of Rs 1,184 million, which represents an increase in cost per km from Rs 1.9 million (USD 38,000 @ SAR exchange rate of Rs 501USD at SAR) to Rs 5.2 million (USD 76,500 @ ICR exchange rate of Rs 68/USD). 4.1.5 This increase in cost is attributed to: change in road conditions due to delay in actual implementation coupled with DOR ignoring maintenance of roads selected for the project at appraisal; substantial rain damages; inclusion of substantial geo-technical and traffic safety works; and changes in extent of work on some contracts (such as increased lengths and variations that caused quantities to exceed beyond the normal allowance under the contract; resulting in claims from contractors). Specific examples of these additional works [in keeping with the overall SAR objectives], that were financed under the project but were not originally envisaged, include completion of flood damaged bridges on the Prithivi Highway, traffic safety 8 Defined as the 'arterial' network of 2200 kIn: approximately equivalent to the Strategic Highways. "Expenditures in the Roads Sector" quoted in SAR, February 1994. 9 Black-topped sections only; total length 2007 km. The equivalent figures for the full Strategic Network of 3260 km of Highways and Feeder Roads are: Good 31%; Fair 57%; and Poor 12%. 7  improvements on Butwal-Tansen and Thankot-Naubise (additionally the length of improvement to both of these projects was increased), and flood damage repairs and off-road works at a number of sites, including Mugling-Narayanghat. The increased costs reflect essential additional works that were identified and defined during the detailed project design and implementation phases. Such increases are typical on road works in Nepal, given the fragile nature of the terrain and the potential for monsoon damage when improvement works are delayed. Similarly, increases are reported on other donor supported road projects in Nepal. 4.1.6 A total of 504 km of road were improved under the Periodic Maintenance Component, compared with an initial proposal for 609 km. The costs of the works increased from an estimated Rs 571 million to Rs 915 million, representing a doubling on a per km basis from Rs 0.9 million to Rs 1.8 million. The increases were due primarily to the delays in implementation and the resultant deterioration in road condition in the intervening period, plus increases in the scope of works including the adoption of bitumen seal, in place of gravel, on certain Feeder Roads in the Terai. Of the original 24 sub-projects, seven have been deleted from the programme and five have been combined to produce 12 sub-projects as shown in Annex 3. 4.1.7 The projects actually improved under the PLRP closely followed the original proposal: all 10 roads originally identified were improved at a total civil works cost of Rs 148.5 million, compared with the initial estimate of Rs 109 million. The length of road improved was 138.7 km compared with the proposal for 147 km. The proposed and actual works completed are summarized in Annex 3. 4.1.8 The project satisfactorily supported road maintenance and rehabilitation activities. Although some physical targets were curtailed, these were generally the more marginal components, with low traffic volumes, resulting in an improved overall project performance. The net effect of the project activities on the network, was complimentary to other donor agency interventions, resulting in an overall improvement in the network condition during this period, as indicated above. 4.1.9 The project was able to improve the road maintenance capabilities, a positive change in sectoral policies indicates that the project's contributions in this regard were satisfactory. DOR/HMGN commissioned the Priority Investment Plan (PIP) for the period of 1997-2006, which was completed in 1997. The PIP assessed resource constraints, funding requirements, sustainability of investment projects in Nepal, and environmental impact and policy considerations for road construction and maintenance. Some aspects of the PIP recommendations have been adopted by HMGN/DOR, but further progress is required in the acceptance of an overall balanced strategy towards maintenance and network development. While the political domain of HMG may not have accepted the PIP wholly, DOR and donor community (who fund 60%, or more, of the road sector budget) are essentially using PIP as a guide for planning. There has been relatively more emphasis on maintenance and rehabilitation, coupled with some increase in allocations for routine, recurrent, and periodic maintenance. Department of Local Infrastructure Development and Agriculture Roads (DOLIDAR) was set up within MLD to assist DDCs road maintenance and development. Overall the PIP has had a positive impact in re-ordering priorities with DOR towards maintenance and limited low-cost network expansion. 8  4.1.10 To ensure adequate funding of priority maintenance activities, the project's achievements were satisfactory; it helped achieve the following. Sixty percent of the road sector budget in HMGN's Ninth Five Year Plan has been directed towards road maintenance and rehabilitationl. In the fiscal year 1998/99, the budget allocation for recurrent and periodic maintenance (excluding donor-funded road projects) is about Rs. 420 million (USD 6.2 million), which is substantially higher than the Rs 100-200 million (USD 2-4 million) allocated in FY-1993/94, but short of the HM[GN agreement under RMRP to Rs 470 million in FY- 1997/98, and barely half of the estimated requirement of Rs 1,000 million per year. A amended Road Fund Act is being introduced and will assist in securing a regular and reliable source of funding for road maintenance. Efforts are now underway to establish a joint public- private sector Roads Board to oversee collection and expenditures of the road user funds for the maintenance of both the strategic and local road networks. The Nepal Roads Board Act has just been approved by the Cabinet, and is being forwarded to the Parliament for consideration. 4.1.11 As a result of the efforts of the Maintenance and Rehabilitation Coordination Unit (MRCU)11, DoR has systemized its data collection activities and now maintains a Highway Management Information System (HMIS) within the Planning Branch. This contains, on an annual basis, information on road condition (physical parameters, surface distress, roughness, etc) for all links in the strategic network (3260 km) and traffic count data for selected sites throughout the country. A road condition survey has been undertaken by consultants annually since 1995/96 with a detailed report produced in May each year. Since 1997/98 the data are available in electronic form. Traffic count data have been collected on a systematic basis since 1994/95 using automatic traffic counters with loops installed in the roadway. Other manual counts have also been undertaken and all available data compiled into the database. 4.1.12 The establishment of the HMIS, together with Strengthened Maintenance Divisions, the Geo-environmental Unit, and the Traffic Safety Unit (see below), have been major achievements in the capability building of DOR in recent years under RMRP, and are likely to be sustained. 4.1.13 The pilot labor-based district roads rehabilitation component (PLRP) has shown encouraging results in development of labor-based road construction techniques among loca' contractors, and institutional and planning capability among the district technical units. District Transport Master Plans (DTMPs) were developed, using a participatory approach, under the project, for the four districts, through DDC, VDCs, and DRTUs (district roads technical units). 4.1.14 Similarly, the Strengthened Maintenance Divisions (SMDs - SDC funding) have improved the capacity to plan (yearly plans of operation are being produced), program, and execute maintenance at the division level. The SMD approach has been shown to work in pTinciple, and DOR is gradually taking ownership of this strategy/approach. 4.1.15 The various interventions, under the project umbrella, to improve public sector implementation capacity through institutional development and human resource development 10 On a cautionary note, more than fifty percent of the road sector budget is financed through external donors, this figure could be reflecting donors targeting road maintenance and rehabilitation. " Set up in DOR in September 1991, as a result of a tripartite agreement between HMGN, British Govemment, and Swiss Government. RMRP did not directly fund the MRCU. 9  satisfactorily supported project implementation (immediate project needs) and made some impact in developing DOR's implementation and management capacity. No proper base line data for DOR as an institution is available, to really quantify the impacts of the RMRP - hence subjective reviews of literature, observations and interviews are used in this ICR assessment. Overall, the project implementation assistance was satisfactorily utilized. Field implementation advisors for DOR and maintenance activities were partially effective in achieving objectives. Technical training and assistance to contractors was partly effective. 4.1.16 The Institutional Development (ID) assistance was successfully utilized, but its 'take- up' within the DOR has been slow. Substantial progress has been made in improving the local capacity for environmental assessment, geo-technical engineering, and bio-engineering. The Geo-Environmental Unit has generated interest among the DOR engineers, and is an example in the region. The unit is successfully educating the DOR engineers about the relevance and importance of addressing environmental aspects of road sector activities early on and through out the project cycle. It has also produced tools to this end, including simple and practical Environmental Assessment Guidelines on environmental processes, mitigation measures, and sensitive areas. 4.1.17 The Traffic Engineering and Safety Unit (TESU) has also prepared and implemented some effective low cost road safety measures after carrying out road safety audits on parts of the network, and is collecting some accident data on the network. The Bridge Management unit did carry out an overall assessment of the bridges and structures on the network, but has not been able to carry out any periodic surveys, nor another overall survey, which is now required. The Road Sector Skills Development Unit (RSSDU) assisted DOR in better tapping training opportunities provided by various donor agencies, but was unable to contribute significantly in other areas targeted by the human resource development policy (improved incentives, staff development plans, etc). 4.1.18 In the context of the Institutional Development, the links made between physical rehabilitation (IDA-funded), human resource development (DFID and SDC), policy fornulation (DFID/SDC-MRCU) and field testing (DFID/EROM and SDC/SMD), began to show the potential of a coordinated donor approach, as opposed to a project-based approach used previously. 4.1 .19 Overall the ID components met with varying degrees of success. It was observed that while the ID approach was comprehensive and cohesive in design, many individual ID components remained project-based during implementation and did not benefit from coordination with other project elements. In general the 'system-based' components performed better that the 'people-based' components. 4.1.20 In supporting development of private sector capabilities, RMRP was partly successful. The project provided a major opportunity to the local consulting and contracting industry in strengthening their technical ability and financial standing. It was the first time that the private industry was able to carry out substantial work in the road sector for a substantial period. However, generally due to delays in procuring and processing, much of the work was undertaken (especially in the case of the periodic maintenance contracts) at great speed during the final months before the close of the project. Thus the project was only partly able to create a stable and sustainable flow of work in order to strengthen the capacity and capability of the 10  local contracting and consulting industry. This also resulted in, at times, poor perception of the private industry (and public agencies as well) in the eyes of the road users/general public. Regional maintenance capabilities were improved, but improvement in procurement of maintenance contracts, gradual development of a critical mass of yearly, or multi-yearly contracts, streamlining contract supervision, and ensuring timely and regular funding and payments was not sustainably achieved during the project. 4.1.21 The project was partly successful in testing ways to develop DDC's road maintenance capabilities. Pilot districts formed District Roads Technical Units (DRTUs) - DRTU staff underwent training - and prepared the District Transport Master Plans (DTMPs) which have been approved by the District Councils. Over 30 small contractors have successfully undergone training. Local Road Coordination Committees (LRCCs) were formed for the model and trial roads who were involved in PLRP and coached on various aspect of road works. Various approaches tested were documented by District Road Services Development Unit-Helvetas (DRSDU). The drawback in this component was the limited involvement of DDCs in project implementation. The component was managed by a unit under MLD at the central level, inconsistent with the decentralized approach envisaged for the component. A follow-up Rural Infrastructure Project by IDA has already been prepared and the issue of DDC responsibility and accountability in project implementation is being addressed during implementation. 4.2 Economic Rate of Return: The SAR presented (in Annex 11) an economic evaluation of the main civil works components, indicating an overall EIRR of 21.4 percent. The results12 of the analysis for each main components are summarized below. No evaluation was attempted for the Routine Maintenance Component nor for any of the training or institutional strengthening activities. Summary of Economic Evaluation Results from SAR Length Fin'cial Cost NPV EIRR l ~~~~~~~~~(km) (R m) *(Rs m) Rehabilitation of Strategic Highways 320 650.58 553.55 19.4% Rehabilitation of Feeder Roads 133 209.65 4.57 10.4% Periodic Maintenance 609 570.68 826.06 27.3% Pilot Labour Based Maintenance 147 26.72 29.77 36.0% Overall 1209 1,457.63 1,413.95 21.4% 4.2.1 A simplified spreadsheet approach has been adopted to undertake a revised economic evaluation of the rehabilitation and periodic maintenance components, taking account of the actual costs incurred, timing of implementation and updated traffic and roughness data. Individual cost and benefit streams have been developed for each of the seven rehabilitation and 12 periodic maintenance projects. 4.2.2 Wherever possible comparable assumptions have been made to those in the earlier SAR work. However the approach is somewhat different and has been applied uniformly to all projects as actually completed. The SAR economic analysis presented a detailed analysis of only two (heavily trafficked) sections of the strategic highway improvements, with only limited 12 Source: SAR, February 1994, Annex 11, Page 1. 11  analysis for the others: the justification for the periodic maintenance projects was based on a matrix approach. This economic re-evaluation applies a series of assumptions regarding road condition and deterioration to assess individual projects on a comparable basis: the validity of the assumptions is demonstrated in a series of sensitivity tests. 4.2.3 Overall, the results of the re-evaluation illustrate that the -project produced acceptable returns - despite the increased costs and reduced lengths of route improved. The reasons for this are the relatively modest costs of the periodic maintenance activities, the actual higher- than-forecast traffic volumes at completion and, most significantly, the increasingly poor condition of the roads in the without-project case (which was exacerbated by the fact that DOR did not maintain road sections included in RMRP from appraisal till the contractors started work on those sections). This latter effect was influenced by the delays in implementation and the consequent additional monsoon damage. Summary of Economic Re-Evaluation for ICR Length Actual Cost NPV EiRR (km) (Rs m) (Rs m) Rehabilitation of Strategic Highways 215.0 1,077.8 2,800.4 35.6% Rehabilitation of Feeder Roads 12.3 106.0 6.7 12.9% Periodic Maintenance 504.0 914.6 1,474.3 27.6% Pilot Labor Based Maintenance 138.7 148.6 264.5 30.3% Overall Total 870.0 2,247.0 4,526.3 31.4% 4.2.4 The significant improvement in the results of the rehabilitation of Strategic Highways is attributable to two heavily trafficked components (Naubise-Thankot and Mugling- Narayanghat). From an examination of the individual projects, it is evident that the greatest returns are from those roads firstly with the higher traffic volumes and secondly with high initial roughness. Rates of return for rehabilitation work exceed 20 percent for those projects where traffic volumes exceed 800 vpd - or slightly lower volumes if the initial roughness is high (e.g., over 12 IRI). However, even with high initial roughness, projects with low traffic (e.g., Naubise-Bhainse) do not justify expensive rehabilitation work. 4.2.5 For periodic maintenance, returns of over 20 percent were obtained where traffic levels exceeded 300 vpd: with lower traffic volumes the returns were marginal - even in cases with high initial roughness. With higher traffic volumes, the works were justified even in situations with relatively low roughness. 4.2.6 The Pilot Labor-Based Rehabilitation projects appear - on the basis of savings in transport costs alone - to generate adequate returns, due to the modest cost of the works, but only for roads with over 120 vehicles per day. Of the ten roads improved, two were in the hill and the remainder in the Terai. 4.3 Institutional Development Impact: The overall Institutional Development (ID) impact is considered satisfactory, whilst individual ID interventions had varied levels of success as described above. However, within the overall context of Nepal before the project, and the DOR in particular, substantial progress has been achieved over the duration of the RMRP. Importantly, these developments have been achieved within the overall structure of the DOR and not solely on an individual project basis as in previous projects. DOR's budget has 12  increased by 70% in the past five years Statistics shows that the annual expenditure as a ratio of annual budget allocation has increased from 50% in 1992/93 to 87% in 1997/98, indicating a gradual improvement in effective budget utilization. 4.3.1 Notable successes were achieved in the coordination of actions between a number of the donors active in the roads sector and the establishment of technical units within the DOR, which are being retained on a continuing basis. Overall the levels of data collection, analysis and storage (particularly within HMIS and MRCU) have shown significant improvements, and the Geo-Environmental and Road Safety Units are now established within the DOR structure. 4.3.2 As described earlier, RMRP addressed the institutional strengthening needs in two parts: the programs required for the immediate needs of the project and those for the longer-term development needs of the department. The activities for the immediate needs of the project achieved generally satisfactory outcomes. In terms of the long term developmental goals of DOR, the future of the various units established in DOR under RMP is important. It is encouraging to note that these units are being incorporated on a permanent basis. 5. Major Factors Affecting Implementation and Outcome 5.1.1 Factors outside the control of Government or Implementing Agency: Overall, the project design was ambitious and complex and the implementation schedule not entirely realistic. Insufficient allowance was made for preliminary activities and the learning process: the 'start-up' time was not factored in - in short, "too much, too fast" in an environment with low recipient capacity. This led to a protracted procurement process for civil works and consultants which has resulted in substantial delays in project implementation. 5.1.2 The overall capability of the private sector consultants and contractors was limited, more so at the start of the project, and the use of international 'sleeping partners' in Joint Ventures to bid for ICB contracts was detrimental to timely implementation. In certain instances, the international partner was introduced solely to allow for qualification under ICB, and took no substantive part in the actual works. The performance of both private domestic and para-statal contractors was also a major cause of implementation delays. Similarly the performance of ICB contractors was not up to the standard expected. The delays between project appraisal, project negotiations/approval, signing, and effectiveness resulted in a change of priorities and the nature of planned interventions. At times, the initial designs and specifications prepared by project consultants were not suited to site conditions resulting in the need for major variations and claims during construction. 5.1.3 Project design did not give appropriate weight to externalities - operating solely within DOR has limited impacts, many of the institutional constraints were elsewhere in HMGN, and whilst any individual project cannot solve these problems, each project must recognize the environment within which it has to be implemented. It was unrealistic to add to an already complex project a series of objectives related to civil service reform: this element was dropped at appraisal. 5.2.1 Factors subject to Government Control: The various tiers of government involved in the project were highly politicized: democracy in Nepal is but a decade old, with frequent changes in government, and the previous system's credo is still entrenched in all spheres of 13  life. At the time of project design, democracy had been restored for only two years. All levels of government administration are highly politicized resulting in institutional instability, frequent staff turnover, adversely affecting continuity, which leads in turn to poor performance of any programmed or project-based intervention. At the same time, it is important to recognize that while the advent of democracy - and subsequent government changes and instability - may have had a direct impact on continuity of personnel directly, or indirectly, related to the project, the systems - such as financial regulations, procurement procedures, centralized decision making - were the same before and after the advent of democracy. 5.2.2 HMGN was unable to follow planned policies included in Eighth Plan (FY-1993/97): for instance, the introduction to the Ninth Plan document indicates that of the planned target of 1778 km, 1092 km were actually built, but additional 1771 km of unplanned roads were built. The recurrent allocations for road maintenance have remained low in comparison with demand. Although maintenance allocations increased by a factor of three between 1992/93 and 1996/97, the share of maintenance allocation has not increased by similar factors, and actual expenditures were only around 75% of the budgeted amount. 5.2.3 There was an obvious lack of cross-sectoral coordination during policy formulation. Examples include introduction of new taxes affecting road projects without knowledge of what can be claimed back by contractors under subsequent legislation clauses, inflexibility in customs regulations, etc. 5.2.4 Financial releases (HMGN/MOF) against allocations were delayed considerably - resulting in an inability to program works effectively or to fully utilize available funds. Additionally, DOR was not able to utilize the retroactive financing facility which could have helped in overcoming many of the start-up delays. 5.2.5 Overly centralized decision making - within both DOR and HMGN - resulted in project processing decisions being delayed. 5.2.6 Some other key factors that were outside the purview of DOR alone, affecting project implementation and outcome, and identified at the project's mid-term review workshop, were: equipment import delays; custom clearances; visa processing; transfer of equipment from one site to another; disposal of low value/scrap equipment by contractors or government; controls on procurement of explosives, bitumen, cement, steel; problems in obtaining bank guarantees and foreign exchange; delay in approval of procurement master lists by MOF; lack of training, and continuity of financial officers related with project financial matters, and delays caused by them; problems with Ministry of General Administration in project staffing; timely resolution of land and forest issues; and conflict between various project procurement guidelines. 5.3.1 Factors subject to Implementing Agency Control: Within DOR, slow decision making, lack of complete awareness about modern contracting procedures and documentation, some apathy towards private sector, "too much" and at other times "too little" oversight, generally weak management at the project and division level, poor perfornance of DOR officers seconded to consultants, were some of the major factors that delayed project implementation. 5.3.1 The project's financial management system did not receive the in depth attention at appraisal that more recent projects have received following the introduction of Bank's OP/BP 14  10.02. Although the financial systems put in place serviced most of the projects basic needs, the lack of ability to coordinate and administer funding where problems arose is indicative of weak administrative and financial management capacity. A better financial management system would have improved project management in these respects, tightened financial controls and improved the monitoring of the project's physical and financial progress 5.4. Costs and financing: Cost changes in physical components varied from substantial in some contracts to marginal in others. The cost increases primarily resulted either directly or indirectly from the delays in project implementation: these delays led to increases in the scope of works required, resulting from additional monsoon damage, a lack of routine maintenance in the intervening periods, and off-road remedial works. As indicated earlier, additional road safety measures were included in two key projects. These cost changes are discussed in detail, in context of the economic re-evaluation, in Annex 3, while the summary of the costs comparison is shown in the following table. Components: SAR SAR Estimates by ICR Actual Totals Financier (IDA & related Expenditures (all in US$ Millions) * HMG funds ONLY) IDA HMG Total IDA HMG Total Rehabilitation 24.265 Civil Works 19.594 2.177 21.771 18.769 3.003 21.772 Emrg'cy/Prithivi Br 3.218 3.763 6.981 Consultants/Supervn 2.370 0.125 2.495 3.567 3.567 Sub-Total 21.964 2.302 24.266 25.554 6.766 32.320 Periodic Maintenance 30.270 Civil Works & Eq'mnt 15.519 1.518 17.037 12.242 3.262 15.504 Consultants/Supervn 3.673 0.192 3.865 3.092 3.092 Sub-Total 19.193 1.709 20.902 15.334 3.262 18.596 PLRP 4.785 Civil Works & Eq'mnt 3.426 0.170 3.596 3.333 0.222 3.555 Consultants/Supervn 0.162 0.059 0.221 0.292 - 0.969 Sub-Total 3.587 0.230 3.817 3.625 0.222 4.524 Instit Devel & Training 12.109 4.986 3.731 Routine Maintenance 8.885 0.023 0.0 Project Prep!n Facility 0.750 0.750 0.591 Total 81.065 50.502 48.835 * includes contingencies 5.4.1 It should be noted that the Rehabilitation Component was financed by IDA & HMG alone but that other components were jointly financed also by ODA, SDC and ITNDP. With the exception of the first column, the above figures relate to IDA funding (and associated HMG funding) only. The total project cost estimate (including contingencies) was US$ 81.1 million, of which the IDA funded element was US$ 50.5 million. 5.4.2 Actual IDA expenditures on the two main components (Rehabilitation and Periodic Maintenance) were US$ 40.9 million, compared with an initial estimate (including contingencies) of US$ 41.2 million. 15  5.4.3 For the Rehabilitation Component, it can be seen that the civil works expenditure (by IDA) was marginally increased due to the inclusion of the emergency works and Prithivi Highway Bridges - and that the HMG contribution was significantly increased. IDA expenditure on the Periodic Component was correspondingly reduced, but HMG expenditure was again increased. 5.4.4 For the PLRP Component, the actual expenditures are similar to the original estimate and include the UNDP contribution. 6. Sustainability 6.1.1 Rational for Sustainability Rating: The overall probability of maintaining the achievements of this project is likely, given that, as discussed earlier, the Priority Investment Plan is being used by DOR and other donors to push for better sectoral planning, with a focus on road maintenance and rehabilitation. The roads improved under the project are being subsequently taken up by the Strengthened Maintenance Divisions of the DOR and the DOR is proposing to sustain the successful ID initiatives under the RMRP. The Road Fund Act is in the process of receiving Parliamentary approval and the follow-up interventions by IDA to assist the DOR program are already underway - the Road Maintenance and Development Project and the Rural Infrastructure Project. 6.1.2 The physical assets created under this project are likely to be maintained through the Strengthened Maintenance Divisions, which automatically take up all maintainable road assets generated in their division. SDC (through its on-going SMD program) and IDA (through its follow-up project) will be closely involved with the further development of these decentralized routine maintenance capabilities. However, it is important that: HMGN does not default on its routine maintenance releases and increases allocations by 15% or more every year; HMGN does not cut the routine maintenance budget after the introduction of the road fund; DOR provides the required level of administrative and financial autonomy to the divisions for efficient operations; and divisions are able to retain good staff - i.e. they do not move to the central DOR office in Kathmandu. Another important aspect is the resolution of pending RMRP contractors' payments and claims that DOR has been unable to process in time to claim disbursement from IDA. 6.1.3 It is too early to comment on the sustainability of the long term ID interventions under RMRP, but generally sustainability can be rated as being likely in some cases and unlikely in others. As in the case of the physical achievement, a majority of these interventions are being supported further through the follow up IDA and other donor projects, such as the TESU, RSSDU, GEU (with social assessment capabilities being added), bridge management (through BUMP/DFID), SMD (through SDC), etc. DOR is also planning to assimilate the MRCU into its formal structure. It is however unclear how (or if) these units will maintain incentives to retain the trained professionals in the absence of donor support. 6.1.4 Continuing donor support will ensure the sustainability of key ID interventions in the short to medium terrn, although it will be necessary for DOR to assume responsibility in the longer term - as is currently proposed in the case of MRCU, SMDs and HMIS. 16  6.1.5 While continued donor support is not a valid argument for likely sustainability, in Nepal's case, where more than 60% of the roads sector budget is donor funded, withdrawal of the donor funding in the medium term [at least] is not likely [barring a force-majeure event]. 6.2 Transition Arrangements to Regular Operation: It is proposed that DFID will continue to support TESU and GEU, at least in part, under the Rural Access Program, and the Bridge Management Unit under BUMP. The resealing programs will be carried out partly with assistance from ADB s planned Fourth Road Improvement Project and the yearly periodic maintenance component being undertaken in three regions under RMDP, and partly from HMGN's own resources. 7. Bank and Borrower Performance Bank Performance: 7.1.1 Lending: The Bank's performance during lending was satisfactory. The project was fairly complex and ambitious. Although setting a high standard for the sector is not in itself a problem, DOR's institutional capacity to handle such a project was not adequately appraised. The policy, planning and institutional development dimensions of the road sector were however discussed at length during preparation, and as such received a substantial share of resources. 7.1.2 At the time of appraisal, under IDA s lead other key donors (DFID, SDC and UNDP) joined an umbrella institutional strengthening program to support some components of the institutional strengthening strategy previously prepared under an ADB technical assistance program. IDA agreed during preparation with DOR on the specific strategies for each unit in DOR. to a much greater detail than usual, including MRCU, SMD and GEU. These actions have changed the approach within DOR in an important (and hopefully sustainable) manner. 7.1.3 The human resource development and training was needed at an early stage, and should have received more attention during project preparation. However, human resource development has progressed reasonably well from previous projects into this one. The ICRs for the previous two projects noted that training was achieved but institutions lagged behind; this project has improved institutional capabilities substantially, but incentives and accountability are still problems. 7.1.4 Institutional development was a key focus during project preparation. It was agreed that training of DOR staff would be prepared in key areas and in a sustainable manner, and the project therefore supported the RSSDU, which probably lacked sufficient momentum and support from the donors during implementation. Lessons from past projects and consultations were selectively used in project design. 7.1.5 The project experience at the time of appraisal indicated that past projects had focused too much on physical activities favoring construction: this was reflected in project design which included a substantial shift to sustainable maintenance. Many past projects had suffered from delays in decision making and procurement, and during preparation an effort was made to broaden the understanding and procurement capacity in the DOR. In hindsight, the political 17  instability was much greater than anybody could have anticipated in a new democracy (only since 1991). 7.1.6 Another lesson brought in from general Bank experience was that much more emphasis was required on addressing the sustainability of investments and proper funding and management of maintenance - this was a major theme during preparation. Finally, even though at the time of appraisal the Bank did not have the benefit of the two ICRs (Cr. 1515 or Cr. 1922), the same team in charge of appraisal was supervising these operations. 7.1.7 As part of project preparation, IDA developed the foundation for major achievements of the project including: a quantum leap in road condition, more sustainable funding for maintenance, start-up of policy reform, attention to environmental safeguards, and institutional strengthening. These factors have contributed to the major changes that have taken place in the thinking and actions within the DOR. 7.1.8 A further important contribution of the Bank during the lending period was to set in motion a fundamental shift in the sector from a traditional public works agency dedicated to attending the politician needs (construction) to one where attention was given for the first time to maintenance. Moreover, it was a daunting task to get 4 donors together in an environment where previously the multiplicity of donors had been a major bottleneck. 7.1.9 During preparation IDA sought and achieved a very broad participation of DOR's middle management, anticipating the many changes of staff as a result of the political environment. The preparation team was successful in raising the ownership and stakes of the project. This provided a platform on which project management at the end had a much greater continuity and chance of succeeding in the implementation of the project. 7.2.1 Supervision: The Bank's performance during supervision was satisfactory - this has been assessed against: the focus on development impact; supervision of fiduciary aspects; appropriateness of inputs and processes; and supervision reporting. Given that there was no clear exit strategy identified during the lending phase, Bank staff showed timeliness, flexibility and foresight in solving the various problems that surfaced during project processing and in guiding DOR/HMGN. The Bank fielded on an average three missions staffed with appropriate skills each year. The Kathmandu-based Bank staff were available at all times to assist DORJIHMGN. The Bank staff used workshops and consultations very creatively to broaden the awareness of the problems being faced during the project implementation, which helped solve the problems quickly. The quality of supervision reporting was satisfactory. 7.2.2 However, a lot of the pending variations were not resolved until late in the credit period, when DOR realized they might lose the funding, and as a result some of the works completed may have quality problems associated with forced rapid execution of works. The design on the district roads under PLRP were probably too high for the traffic used in those roads. The complaint from the DDCs of their lack of involvement in the PLRP was not timely addressed. Due to the complexity of the project, with many sub-projects/components, it may not have been possible to address all problems in a timely manner. 7.2.3 Due to the delays in implementation, too much attention was paid to the physical components rather than institutional strengthening objectives. Specifically, insufficient 18  attention was given to the areas of procurement speed, DOR supervision, government obstacles to efficient works, and district involvement. The amount of supervision in the last 18-24 months was probably insufficient given all the activity, and contributed to the reduced effectiveness of donor coordination. 7.3 Overall Bank Performance: The overall Bank performance was satisfactory. Borrower Performance: 7.4 Preparation: Overall HMGN/DOR performance during project preparation was satisfactory. Most of the delays in starting negotiations and project effectiveness were attributable to the wider HNGN system which was strained by political turmoil at the time. 7.5.1 Government Implementation Performance: HMGN performance during project implementation was unsatisfactory. Sector policies were abrogated (as shown by the Ninth Plan's assessment of the Eight Plan); commitment to project varied (as also indicated during Mid Term Review Workshop in March 1997); there were delays and lack of coordination among government departments in operational decisions; and, there was a lack of commitment to adopting the PIP. 7.5.2 HMGN failed to adopt a 3-year rolling plan for road expenditures, which would have ensured continuity and certainty in the budgeting process and would have enabled a more consistent work program to be developed. Additional problems associated with HMGN include: (i) numerous delays and obstructions relating to customs, explosives, imports; and (ii) a very inefficient annual pattern of late release of regular non-project funds to DOR, resulting in poor planning and contracting practices. 7.6 Implementing Agency: DOR's performance was satisfactory. Considering the complexity of the project and DOR's capacity in general, they were instrumental in the project achieving an overall satisfactory outcome. Even though project compliance was poor and decision making was delayed at times, DOR's efforts towards the later part of the project helped deliver the desired improvements in the network. DOR keenly received and acted on donor and consultative advice, and used technical assistance relatively effectively. 7.7 Overall Borrower Performance: Not withstanding the project's satisfactory outcome, the overall Borrower performance is considered unsatisfactory, based on the problems associated with HMGN/MOWT/MOF relating to the late release of funds, lack of commitment to the strategic planning objectives, and continuing political interference. The project's satisfactory outcome was a direct result of the efforts put forth by the DOR managers and the coordinated interventions of the project's co-financiers in particular, and the broader donor community in general. The unsatisfactory performance rating of the Borrower as a whole is based on the numerous wider HMGN systematic issues that have been highlighted earlier in this ICR, and has been necessitated, partly, to raise the awareness of these issues during Bank's further dialogue with HMGN. 19  8. Lessons Learned 8.1 Notable success was achieved in the coordination of the activities of the multiple donors involved in the project, despite differing individual guidelines and procedures. 8.2 Human resource and institutional development reforms in sector departments (such as DOR) should be designed in view of the broader HMGN civil service rules and regulations. 8.3 While designing improvements in service delivery - such as better road maintenance - alternate delivery mechanisms, to the existing public set-up, should not be entirely discarded. A Roads Board is only now feasible - after 5 years of the RMRP - to a great extent because the RMRP was successful in contributing to the improvement of average quality of the strategic highways in Nepal. 8.4 The divergence in objectives amongst the public and private sectoT - civic obligations vs. paid service provision - should be clearly recognized and the private sector should not be out- rightly expected to absorb costs incurred due to poor performance of public institutions; concurrently, the private sector should be held strictly accountable to its performance guarantees. 8.5 The time provided for engineering investigations and design should be increased, and consultants should improve the efficiency of their design and supervision. They should provide rational professional liability for all designed works. Road works specifications need to be revised, utilizing the results of the Materials Study completed as part of RMRP. 8.6 Strengthened Maintenance Divisions have improved the capacity to plan, program and execute maintenance at the division level. Divisions need financial autonomy and timely budget releases - and clearer accountability for results - in order to effectively take on a stronger decentralized role in managing minor works and contracting out routine maintenance. 8.7 Periodic updating of initiatives like the PIP should be encouraged, and more effort is needed to ensure their adoption of these policy planning tools by NPC/HMGN. 8.8 Privatization/commercialization of government equipment and mechanical training programs should be accelerated. At the same time it should be recognized that in Nepal's road sector - where landslides and other natural disasters are frequent - certain force account works will be necessary as public service obligations: traffic levels are currently not high enough to make emergency interventions commercially sustainable. 8.9 Nepal is a land locked country with very few land links to neighboring countries and a weak economy with a small workload for numerous contractors to keep machinery engaged all year around. There is thus a continuing role for the commercialization of DOR's equipment rental facility as well as encouraging contractors to procure their own (when subsidized equipment was no longer available). 8.10 A careful scrutiny of the track record of contractors during pre-qualification and post- qualification is required. In cases where the Borrower has documented evidence of firms' non- performance or poor performance, the Bank has accepted the position taken by Borrowers to 20  disqualify firms based on that past performance. There is an urgent need to develop a database for comparative rating of consultants and contractors in the road construction industry, which can be subsequently updated regularly (this should be done by a third party, under contract). 8.11 Multiple procurement guidelines amongst the many donors and agencies in Nepal can create problems in project processing: there is an immediate need for a review of current procurement guidelines and development of the national procurement guidelines for goods, works and services for Nepal which are acceptable to all concerned. 8.12 Strengthening of basic institutional capacities for effective project management need to be prepared and implemented early in the life of the project. Procurement capabilities need to be bolstered from project design, with initial contracts based on completed designs, strict qualification criteria and awards planned before project effectiveness. 8.13 Labor-based projects with community-participation, which was introduced for the rehabilitation and maintenance of district roads, have shown good results in the development of institutional and planning capability at the district level. Districts have begun to develop their capacity for road planning and labor-based contracting of works through the preparation of DTMPs and forming DRTUs (aided by the new DOLIDAR). As a lesson learned to support the decentralized government system, DDCs should be allowed to actually prepare, procure, implement, and manage the work; whereas under the RMRP, project preparation and implementation was centralized and came under the responsibility of MLD. 8.14 Project impacts are generally more effective if they are prepared by stakeholders with beneficiary participation and where communities gain from their involvement in road rehabilitation and development. This process of decentralization to DDCs should be closely monitored, to ensure rapid interventions relevant to its success. 9. Partner Comments by DFID (ODA) 9.1 The Road Maintenance Project (RMP) was part of an umbrella Road Maintenance and Rehabilitation Project. The main objective of the RMRP was to ensure that Nepal's strategic road network would be maintained according to planned procedures. The RMP component supported the development of DOR capacity to plan and manage road maintenance by providing technical and managerial training to DOR and the private sector. 9.2 The RMRP umbrella programme supported by World Bank, SDC and DFID represents the first attempt to apply a coherent approach to support for HMGN road sector activities. Taking the previously neglected road maintenance activities as its focus it has been successful in rehabilitating sections of main road to a maintainable standard, and has assisted in developing capability within the Roads Sector in Nepal to undertake maintenance. One of the most interesting aspects for DFID is the links made between physical rehabilitation (World Bank), human resource development (DFID and SDC), policy formulation (DFID/SDC- MRCU) and field testing (DFID/EROM and SDC/SMD) to create this coherent approach. Possibly one of the more important lessons learnt has been how much more effective potentially a common donor approach can be than the type of projectised approaches applied previously. 21  9.3 Of course not all the components within RMRP were equally successful. As to be expected implementation of the capacity building components lagged behind the engineering components and ultimately met with varying degrees of success as evidenced by review of the DFID funded RMP components. On reflection RMP was perhaps overly ambitious in trying to introduce five different aspects of road sector skill at the same time. Perhaps a lesson for the future is to consider a phased approach over a longer time period for the institutional components based on an institution's capacity and willingness to take on board new initiatives. 9.4 Whilst acknowledging the fact that the different components of RMP have met with varying degrees of success, it is considered that the components identified were appropriate to improve Road Sector capacity to plan and manage road maintenance, and furthermore that these were complementary and have contributed to achievement of the overall RMRP objective. In cases where the components have had limited impact their failure can be attributed to skills transfer training being implemented prior to policy formulation and take up. 9.5 The provision of training prior to policy development is a recurrent problem on capacity building projects such as RMP as Donors/ Government often become anxious to demonstrate activity on the ground. In the case of RMP this tendency should have been mitigated by the fact that the project formed part of the umbrella programme. Unfortunately as the RMRP programme progressed the approach became fragmented, lacked overall co-ordination and the components comprising the whole began operating in a projectised fashion. This resulted in the RMP components trying to resolve issues within their narrow spectrum rather than taking a broader outlook. This observation is not intended as any criticism of a common approach but rather it is intended to highlight the need for common review, common monitoring and co- ordination to ensure that fragmentation does not occur. 9.6 It is too early to comment on the future sustainability of the capacity built by RMP. The programme ended with no clear exit strategy defined for each of the units established under the programme to implement the components. The challenge remains for DOR to identify which units should remain and which should be integrated into general DOR maintenance activities. To assist with this each component has developed procedures which could quite easily be integrated in to DOR management practices. Clearly a lesson for the future is to ensure that exit strategies are devised at least a year or two before programme end. 9.7 This report has been drafted by the World Bank in consultation with DFID and accurately reflects the partners' joint views concerning programme impact. 10. Additional Information 10.1 There was a strong feeling in the Nepal consulting industry community that the current method of procurement based on quality and cost is not yielding the desired objectives and the regulations governing consultants in Nepal are not efficient/effective. The bottom line is that they argue that consultant selection always boils down to the least cost proposal amongst the qualified consultants. 10.2 The following contribution is provided by Mr. Badan Lal Nyachhyon and forwarded by Mr. Subama B. Joshi, General Secretary of Society of Architectural and Engineering Firms, 22  Nepal. The SCAEF's comments received by the ICR team, are included here, with the intent to invite a dialogue between RPA and SCAEF. 10.3 "... Regulation governing Consulting Services: The existing regulations in Nepal governing the Consulting Industry in Nepal are - Financial Administration Regulation (FAR) - Company Act - Industrial Occupation Act - Income Tax Act 10.4. The basic flaws of these regulations in relation to the Consulting Industry are: * The pre-requisite criteria for professional development of the industry is not considered * FAR relies on non-professional ways of selection of consultants as low bid among technically responsive consultants. * Direct award of jobs to government subsidiary companies without considering technical capability. * The concept of working directors-professionals is not considered as against other business houses where professional liability is non-existent. * Tax system is not based on reality and fairness. Tax hassles are very much characteristics. * The government committedfacilities of tax rebate are practically not utilized due to cumbersome process and negative attitude of the government. 10.5 SCAEF has recommended that the Financial Regulation shall be reviewed in relation to the problems and issues mentioned above. A draft of such regulation is submitted to the government. SCAEF has also recommended that a Charter by law shall be sought with formulation of Consulting Services Act. 10.6 Pros and Cons of Evaluation of Priced Proposals: The World Bank's new approach of evaluation of the proposals for selection of the consultants include weightage to the financial proposal with score ranging from 70-85% for technical part and 30-15%forfinancial part. A case study of various combination of the weightage is illustrated in Figure 1. 10. 7 Figure 1 has three different distinctive parts. The left part where the technical score is 60% and over the effect offinancial difference is less and a the merit of technical excellence is predominant. The right part of the Figure indicates that when the technical score is less than 53%. The financial component becomes a decisive factor in wining the job. The middle part of the Figure may be called afield of Gambling when the decisive combination would be a guess work and wouldfall in the hands of the evaluator tofavorfor. 10.8 The clear message from this exercise is that one should quote less if technical merit is not quite clear and ultimately the job will be awarded based on the power of lower bid price irrespective of the technical merits. A big question arises before us. Whether we want to promote Gambling in the Profession and in the Project Life Cycle or we want to promote sound professional practice for better economy. The choice is in hands of the evaluators. It is quite clear that the promotion of the Gambling Scenario can not be the purpose of the Project 23  and the objectives of the Client. In this case, it is very much require that the negative effect of the priced proposal shall be eliminated by firmly establishing the criteria of evaluation of the financial proposal and maintaining transparency in evaluation by consideration of inclusion of external experts in evaluation team or provision of independent auditing prior to the contract award" Evaluation of Priced Proposals (for a set of given data) 753 '' |Tcnia 70 -_ __ __ ___ ___p * Zn _ 0 m 65_____ __ ,) 60 >. 55 -Financial Meri TeIchnical Merit 'Zon Zone 50 Gambling Zone Zn 4.0 3.6 3.2 2.8 2.4 2.0 1.8 3.2 2.1 3.2 2.0 2.0 2.8 2.4 1.8 1.4 1.0 Financial Proposal CurPnrce/MinPre +--FS85 15 -4 FS80 20 FS90 10 - FS95 5 FSIOO 0 +FS70 30 24  Annex I Key Performance Indicators Outcome/Impact Indicators Indicator Projected in SAR/PAD* Actual/Latest Estimate (a) HMG/N incorporates Priority Yes Done Investment Plan (PIP) recommendations for maintenance funding in National Planning (b) Priority Investment Plan completed Yes Endorsement in progress: 9th plan includes most and formally endorsed by HMG/N roads as recommended by PIP, with some additional and included in 9th plan. roads; Proposed 9th plan expenditure is 27020 Resources allocated to network on million compared with PIP recommendation 29631 basis of PIP recommendations million. over the full study period (10 years) (c) Length of roads with Average Yes Currently 32% of core roads in good condition and Daily Traffic greater than 250 58% in fair condition, which appears to be well vehicles per day in good/fair ahead of target. Length is being confirmed. A condition (Surface Distress Index substantial proportion of the core network is less than 3.0) progressively currently under construction. increased from 1,270 km in 1996 to 2,460 km at the end of 1999 (d) Annual value of contracts awarded Yes Increase from 95/96 to 96/97 was 24 % (from Rs. in road sub-sector increasing by 2969 million to Rs. 3182 million). 15% per annum (e) Designed serviceability of district Yes Initial works only recently completed; condition road maintained good; to be further monitored in coming missions. (f) Selected road maintained in Yes good/fair condition with average Surface Distress Index less than 3.0  Annex 2 Project Costs and Financing Project Costs by Components (in US$ million equivalent) Project Component Appraisal Estimates Actual/Latest Estimate Percentage of Appraisal Totatc without with Conktiligelcy HMG funids Total (IDA Conthiugency diCtributed IDA Coinpoaenit related to IDA & assoc IDA HMG Total* IDA funds Total * Canthigency distributed ~~~Coiitpoiieait 11MG funds) Road Rehabilitation 20.7 24.26 21.96 2.30 24.27 25.55 6.77 32.32 116.3 133.2 Periodic Maintenance 25.3 30.27 19.19 1.71 20.90 15.33 3.26 18.60 79.9 89.0 Pilot Labor-Based District Road Rehabilitation 4.2 4.78 3.59 0.23 3.82 3.63 0.22 3.85 101.1 100.8 Project Preparation and Implementation Support 0.74 0.75 0.75 0.75 0.59 0.59 78.8 78.8 Technical Assistance Institutional Development Technical Assistance 10.8 12.11 4.99 4.99 3.73 3.73 74.8 74.8 Routine Maintenance 7.4 8.89 0.02 0.02 0.00 0.00 0.0 Contingencies 11.9 Total 81.0 81.06 50.50 = 54.74 48.84 = 59.09 96.7 107.9 * Components with IDA involvement ONLY  Annex2b Project Costs and Financing Project Costs by Procurement Arrangements (In US$ million equivalent) Expenditure Procurement Method Categories Appraisal Estimate Actual/Latest Estimate ICB NCB OTHER NBF* Total ICB NCB OTHER NBF* Total 1. Works 26.98 9.87 2.43 12.42 51.70 25.47 9.31 1.90 12.38 49.06 2. Goods 4.52 - 0.80 0.91 6.23 4.25 - 0.78 5.03 3. Services - 13.06 7.54 20.60 - 11.43 7.39 18.82 4. Misc. = = = = = Total 31.50 9.87 16.29 20.87 78.53 29.72 9.31 14.11 19.77 72.91  Annex 2c I'roject Financing by Component (in US$ million equivalent) Component Appraisal Estimate Actual/Latest Estimate Percentage of Appraisal (%) Bank Governimlenit Cofinancier Bank Governemnt Cofinanicier Bank Government Co-financier 1. Road Reiabilitatioi and 21,863.600 2,301.800 24,190.000 2,572.000 110.64 111.74 Overlays 106 111.74 2. Resealing and Gravelling of 19,192.600 4,870.300 6,058.700 17,282.000 4,385.000 5,331.650 90.05 90.04 88.00 Strategic Net Road Rehabiloitation 3,587.400 292.200 967.500 2,895.000 236.000 967.500 80.70 80.77 100.00 Roa RehabiitationalDvlpe n 4. Instit.ional Developme.itaid 5,008.900 631.600 6,970.800 3,870.(X)) 630.00() 6,482.840 77.26 99.75 93.00 I rainitig 5. Routine Maintenlanlce 6,365.400 2,104.400 6,238.090 2,082.310 98.00 98.95 6. Project Preparationi Facility 750.000 591.0(0 78.80 50,402.500 14,461.300 16,101.400 48,828.000 14,061.090 14,864.300 96.88 97.23 92.32  Annex 3 Economic Re-Evaluation of Rehabilitation & Periodic Maintenance Components A. Introduction 1. The total project costs for the Nepal Road Maintenance & Rehabilitation Project (RMRP), as presented Staff Appraisal Report (SAR) of February 1994, were estimated at US$ 81million. The Road Rehabilitation and Periodic Maintenance (Resealing & Gravelling) Components represented 30% and 37% respectively of this total, as indicated below. Project Costs (US$ million) % of Project Component L Base . Local Foreign Total Costs Road Rehabilitation & Overlays 9.631 11.101 20.732 30 Resealing & Gravelling of Strategic Network 9.518 15.768 25.286 37 Pilot Labour Based District Road Rehab 2.121 2.113 4.234 6 Institutional Development & Training 2.549 8.205 10.754 16 Routine Maintenance 6.227 1.168 7.396 11 Project Preparation Facility 0.453 0.291 0.743 X Total Baseline Costs 30.500 38.646 69.146 100 Physical Contingencies 2.361 3.243 5.603 8 Price Contingencies 2.377 3.939 6.316 9 TOTAL PROJECT COSTS 35.238 45.827 81.065 117 Source: SAR, February 1994, Annex 8, Page 1 2. The SAR presented (in Annex 11) an economic evaluation of the main civil works components, indicating an overall EIRR of 21.4%. The results of the analysis for each main components (in Rupees) are summarized below. No evaluation was attempted for the Routine Maintenance Component nor for any of the training or institutional strengthening activities. Fin'cial Cost Econ Cost NPV EIRR (Rs m) (Rs m) (Rs m) Rehabilitation of Strategic Highways 650.58 585.52 553.55 19.4% Rehabilitation of Feeder Roads 209.65 188.64 4.57 10.4% Periodic Maintenance 570.68 513.61 826.06 27.3% Pilot Labour Based Maintenance 26.72 23.31 29.77 36.0% Overall 1,457.63 1,311.08 1,413.95 21.4% Source: SAR, February 1994, Annex 11, Page I B. Project Components as Originally Defined 3. At the time of project inception, the Rehabilitation Component contained six sections of Strategic Highway and three sections of Feeder Road. The characteristics of each and the anticipated economic returns are summarized in the following table. The evaluation of the Feeder Roads was based on extremely limited information.  Rehabilitation Component: Summary of Results of SAR Economic Evaluation Length IRI AADT Finc'lCost Costper Econ Cost NPV NPV per [ EIRR _ km (1993) 1993) (Rs m) km (Rs m) (Rs m) E (s0% km Highways: Thankot Naubise 12.171 12.6 1,435. 90.02' 7.40 81.02 385.60 31.68, 42.3% Naubise Hetauda 106.4 12.0 82 97.77 0.92 87.99 2.42 0.02 10.5% Mugling Narayanghat 36.24 5.1 1,291 63.04, 1.74 56.74] 40.87 1.13 16.2% Marshyandi Khairenitar 44.131 7.0 378 171.190 3.88 154.07 37.80 0.86[ 12.7% Butwal Tansen 36.85i 10.8 280 142.07' 3.86 127.86 52.20 1.42 14.8% Tansen Syanga 84.421 9.0 136| 86.49| 1.02 77.84 14.40 0.17 12.2% Sub-Total Highways: 320.21 650.58 2.03 585.52 533.29 1.67 19.1% Feeder Roads: i Birgunij Kalaiya 15 122.10 8.14 109.89 8.04 0.54 11.1% Tansen Ridi-Thamgas 68 55.28 0.81 49.75 0.631 0.01: 10.4% Nepalgunj Bagauda 50 32.27 0.65 29.05 1.44 0.03, 11.1% Sub-Total Feeder Roads: 133 209.65 1.58 188.69 10.1 1 0.08o 11.0% Total Rehabilitation: 453.21 860.23 1.90 774.21| 543.40 1.20 Source: SAR, February 1994, Annex 11, Tables 1-4 & Final Feasibility Report, ND Lea, May 1993 4. For Highways, an overall EIRR of 19.1% was demonstrated. It can be seen that improvement costs vary significantly from over Rs 7 million per km for Thankot-Naubise, including substantial reconstruction and slope stabilization, to less than Rs 1 million per km for Naubise-Hetauda, which involved only relatively minor works. The highest rate of return was achieved for the Thankot-Naubise section due to the substantial traffic volumes and high level of initial roughness. 5. Only limited information was available in regard to the Feeder Roads and it was agreed that further evaluations would be undertaken prior to implementation. The Birgunj-Kalaiya project includes the replacement of three bridges. [It is noted that the EIRRs quoted for the individual projects differ from those in the overall project assessment.] 6. The benefits were calculated using the HDM3 model calibrated and adapted for use in Nepal. The benefit streams over an assumed 20-year project life include: reduced vehicle operating costs due to improved road surface; reduction in future major rehabilitation costs through regular re-surfacing; reduced emergency maintenance resulting from improved drainage and off-road works; and road user savings due to reduced road closures. An annual traffic growth rate of 6% was assumed. 7. The without-project scenario assumed minimal maintenance (patching and pot-holing only) and a continuing deterioration to the road surface, resulting in ever-increasing vehicle operating costs. This approach (which, it can be argued, produces an unrealistic situation with an excessively deteriorated road surface) acts as a proxy for the inevitable major costly reconstruction that would otherwise have to be implemented. 8. For the Periodic Maintenance Component, an initial screening was conducted using a matrix approach with each potential section classified according to its traffic volume (AADT), roughness (IRI) and pavement condition (Surface Distress Index, SDI). The NPV per km was calculated using HDM3 for each cell in the matrix, so that each section could be prioritized and those with a negative NPV eliminated. 2  9. About 40 sections of the network passed the initial screening and subsequently 24 sections were selected in the Central, Western, Mid-Western and Far-Western Regions. The costs were estimated based on unit rates for the extent and type of improvement works envisaged. The results of the initial economic evaluation are presented in the following table. Periodic Maintenance Component: Summary of SAR Economic Evaluation Project: Lengt IRI I AADT Finc'l Cost Cost per Econ Cost NPV | NPV per EIRR h km (1993) | (1993) (Rs m) km (Rs m) (Rs m) 10% km Bhittamod Dhalkebar 43 5.3 00 49.04 1.14 44.14 18.65 0.43 13.6%/ Hetauda Lothar 41 4.71 1000 44.38 1.08 39.94 287.56 7.01 38.0% Lothar Narayanghat 33 4.7 1000 34.84 1.06 31.36 233.32 7 07 38.50/ Nawalpur Malangwa BT 9 15.9 70 4.76 0.53 4.28 9.88 1.10 17.2% Nawalpur Malangwa GR 16 15.9 70 18.90 1.18 17.01 Kakani Trisuli 63 12.0 70 56.41 0.90 50.77 10.28 0.16 13.0% Chand'pur GaurBT 34 9.1 96 17.51 0.52 15.76& 58.70 1.73 39.8% Chand'pur Gaur GR 8 15.0 96 9.63 1.20 8.671 * Godavari Puichoki 6 13.8 50 3.12 0.52 2.81' 3.32 0.55 26.9/o Syangja Kubinde 21 5.5 210 12.49 0.59 11.24; 0.91 0.03 10.8% Kubinde Pokhara 15 5.5 210 8.92 0.59 8.03 * Tansen Bartun/Hartok 4 13.5 100 2.07 0.52 1.86 5.77 1.44 37.6% Bhairahawa Lumbini 20 8.6 290 16.57 0.83 14.91 46.90 2.35 36.80/0 Khrishnagar Chanuta 20 6.8 250 16.80 0.84 15.12 13.64 0.68 23.8% Lumbini Tailihawa | 22 12.0 50 25.74 1.17 23.17 48.90 2.22 33.1% Kohalpur Chisapani 13 16.7 80 8.39 0.651 7.55 13.23 1.02 36.5% Chisapani BiTendranagar 1 78 16.7 80 93.46 1.20 84.11 1.98 0.03 10.6% Nepalgunj Gulariya BT 9 16.0 80 7.39 0.82 6.65 0.34 0.04 10.2% Nepalgunj Gulariya GR 26, 16.0 80 32.60 1.25 29.34 Kohalpur Jamunaha 221 5.7 150 10.34 0.47 9.31 1.54 0.07 13.6