58167 Document of The World Bank FOR OFFICIAL USE ONLY Connecting Afghanistan A Policy Note on Rural Access in Afghanistan June 2007 Sustainable Development Unit Afghanistan Country Management Unit South Asia Region ABBREVIATIONS AND ACRONYMS ANDS Afghanistan National Development Strategy ARTF Afghanistan Reconstruction Trust Fund CDC community development council CNTF Counter-Narcotics Trust Fund DFID Department for International Development (UK) EC European Commission IDA International Development Association IP implementation partner IRAP integrated rural access planning IRRIP Interim Rural Roads Investment Plan JPMU Joint Program Management Unit JSDF Japanese Social Development Fund LIWP Labor-intensive Public Works Program MIS/GIS Management Information System/Geographic Information System MPRNIP Master Plan for Road Network Improvement MRRD Ministry of Rural Rehabilitation and Development MPW Ministry of Public Works NGO non-governmental organization NDF National Development Framework NEEP National Emergency Employment Program NEEPRA National Emergency Employment Program for Rural Access NERAP National Emergency Rural Access Project NIB National Interministerial Board NRAP National Rural Access Program NRVA National Risk and Vulnerability Assessment NSP National Solidarity Program PDCU Program Development and Coordination Unit PIU Project Implementation Unit PRT provincial reconstruction teams PSIB programmatic support to institution building SY solar year SC Steering Committee UNDP United Nations Development Program UNOPS United Nations Office for Project Services USAID United States Agency for International Development CONTENTS Page ABBREVIATIONS AND ACRONYMS ............................................................................ i EXECUTIVE SUMMARY ................................................................................................. i I. CURRENT STATUS.................................................................................................. 1 From NEEP to NRAP ..................................................................................................... 1 Funding Sources.............................................................................................................. 1 Achievements.................................................................................................................. 2 Institutional Arrangements.............................................................................................. 4 Implementation Arrangements........................................................................................ 4 Access Planning and Prioritization under NRAP ........................................................... 6 Safety Net, Design Standards, and Technology.............................................................. 7 Current Policy Environment ........................................................................................... 7 II. KEY ISSUES AND CONSTRAINTS........................................................................ 8 III. RECOMMENDATIONS...................................................................................... 10 ANNEX 1: NRAP TIMELINE......................................................................................... 13 ANNEX 2: NRAP ORGANOGRAM............................................................................... 14 ANNEX 3: PROVINCIAL DISTRIBUTION .................................................................. 15 ANNEX 4: MAP............................................................................................................... 16 ANNEX 5: CAPACITY BUILDING ACTIVITIES ........................................................ 17 Tables Table 1: Financial Contributions as of December 2006 ..................................................... 2 Table 2: Provincial Distribution of NRAP.......................................................................... 3 Table 3: Achievements as of December 2006 .................................................................... 3 EXECUTIVE SUMMARY 1. The objectives of this Rural Access Policy Note are to (i) provide a first-level analysis of rural access issues in Afghanistan, (ii) identify key issues, achievements, and challenges in the provision of rural access, (iii) make recommendations for the Government's rural access policy, and (iv) stimulate the policy dialogue on provision of rural access between the Government of Afghanistan and stakeholders including civil society, NGOs, development partners, and private enterprise. 1 2. Nearly 80 percent of Afghanistan's people live in rural areas. Incidence of rural poverty is high as the country recovers from the long-term effects of the nearly thirty years of conflict and a recently ended six-year drought. The insurgency in some parts of the country continues to hamper development efforts. As lack of rural access is one of the sources of rural poverty, and holds back the replacement of the poppy economy with licit rural livelihoods, the World Bank has supported the provision of rural roads to improve rural accessibility. 3. The Transitional Government launched the National Emergency Employment Program (NEEP) in 2002 as one of four National Priority Programs designed to provide targeted social protection for vulnerable groups throughout the country. The goal was to reconstruct and develop the country's rural access infrastructure while offering short-term jobs that would provide a safety net for the rural poor. After a review of the initial experience, the activities under NEEP were reoriented in early 2005. Targeted interventions for the poor were by-and-large left to other ongoing programs (such as the National Solidarity Program), while NEEP evolved into the National Rural Access Program (NRAP). NRAP creates infrastructure that improves the access of rural people to economic and social opportunities, health services, and education facilities. Though NRAP makes some use of labor-based technologies, providing employment is not its main goal and the emphasis on social protection is more limited than at the outset of NEEP. 4. NEEP/NRAP has proved to be one of the Government's most successful national priority programs. Since 2002, approximately US$193 million have been invested in rural roads; about 9,000 km of rural roads have been rehabilitated throughout the country, and about 3,000 villages have been connected to town centers. 2 The program has provided nearly 13 million labor-days of employment. The rural road network is estimated at between 30,000 and 50,000 kilometers, and the rehabilitation works to date have restored about 25 percent of it to maintainable condition. This achievement is a fragile one, however, because about 7,000 of the kilometers restored were improved with relatively little technical input, and the longevity of the works depends heavily on how well they are maintained. Quite likely a substantial portion of the roads that were improved early on are already suffering from severe deterioration. The durability of works has improved, however, as the program focus has shifted from providing a social safety net to providing quality rural roads. 5. The attempt to pull different projects together under the single umbrella of NRAP was a visionary approach to coordinate donor funding and to ensure coordination between the two concerned sector ministries, the Ministry of Public Works (MPW) and the Ministry of Rural Rehabilitation and Development (MRRD). It was proposed to have work at the policy level 1 The Policy Note draws substantially on work carried out by John Howe during early 2007 as well as on the NEEPRA Final Review Report (2005) and the draft report, "NEEPRA and its Reorientation Towards Building Rural Access" (November 2006). 2 Source: PDCU, "NRAP Summary Achievement" from inception up to December 2006. i embodied by a Steering Committee, portfolio management entrusted to a Program Development and Coordination Unit, and implementation carried out by project implementation units in the two sector ministries with very substantial assistance by the Implementation Partner, UNOPS. 6. Unfortunately this institutional approach did not work as well as it could have. The planned Steering Committee met only twice after its creation, leaving the Program Development and Coordination Unit (PDCU) largely without guidance and supervision. Other permanent coordination mechanisms, such as the Consultative Group and Working Group for Agriculture and Rural Development, as well as the ones for Transport, which function under the umbrella of the Afghanistan National Development Strategy, apparently have not been used thus far to discuss rural access. Afghanistan's provincial reconstruction teams, whose interventions are security- and development-oriented, have so far not been involved in the donor coordination, and knowledge of their work is very limited. Results on the ground have nevertheless been impressive, especially in view of the security situation and the weak capacity in the ministries. 7. Fundamental challenges remain. First, a national priority program is being implemented without a policy framework. Second, there is no effective coordination between the providers and beneficiaries of access infrastructure. Local, district, and provincial authorities, as well as ministries that rely on rural roads (education, health, agriculture, etc.) and the provincial reconstruction teams, need to be more systematically involved in decision making, and this needs to be complemented by a knowledge base on rural access needs and the state of the road network. The prioritization of rural road investments is not sufficiently transparent or impact-oriented. Third, MRRD and MPW need to determine the most efficient and effective delivery mechanism for rural access, to distinguish their own core functions from functions that they could outsource, and to decide on the appropriate level of deconcentration (i.e. how many of their staff should be posted in the provinces, with what functions). Fourth, based on the technical standards the program is now using, the average cost of rural road rehabilitation in Afghanistan is more than US$30,000 per km, and with the currently available annual funding it would take about 100 years to upgrade the entire rural road network. An approach that is targeted to providing basic year- round access at the lowest acceptable level of service and using modern labor-based methods, whose costs per kilometer are lower, might be more appropriate. Fifth, investments have not been accompanied by a strategy and funding for maintenance. Sixth, so far no consideration has been given to gender considerations in rural access. 8. NRAP currently faces an acute funding shortfall and operational funds will be exhausted at the end of June 2007. The Ministry of Finance has agreed to provide bridge funding until a proposed IDA credit for the National Emergency Rural Access Project becomes effective. (Funding is for $112 million and effectiveness is expected by January 2008.) 9. Through the Afghan Compact the Government has committed itself to providing road connectivity to 40 percent of all villages (giving access to 19 million people in 38,000 villages) by the end of 2010, thereby increasing access to markets, employment, and social services. It is not clear on what basis these targets were set or if they are achievable. The Government has also committed to provide sustainable funding for road maintenance by the end of 2007. To honor these commitments and to ensure that accessibility is sustainable, the focus of Government policy needs to be sharpened, so as to make interventions more effective, and broadened, so as to provide a more comprehensive response to needs. It will take many years to bring about the necessary changes, so both short and medium-term perspectives are needed. ii 10. The recommendations of this Policy Note are directly derived from the key constraints. Five recommendations should receive top priority: a) define a National Rural Access Policy and then formulate a multi-year program for investment and maintenance of tertiary and secondary rural roads; b) set up a coordination mechanism for planning and implementation that brings together different levels of government, ministries, donors, and the provincial reconstruction teams; c) give priority to defining a strategy for rural road maintenance and its implementation; d) strengthen the bottom-up planning and prioritization process, introduce decision- making tools for road prioritization and investment programming, and make investment programming more transparent and better protected from the influence of pressure groups; and e) adopt an approach of providing basic year-round access at lower unit costs than the current approach, so as to benefit more people, and develop appropriate local technical specifications for this purpose. 11. Other recommendations should be addressed in the medium term: a) clarify the institutional responsibilities of MPW and MRRD and implement administrative reforms in these ministries, such as appropriate deconcentration and pay-and-grading; b) introduce modern labor-based techniques and make these the instrument of choice for rural road improvements, particularly for the tertiary road network; c) enhance the capacity of the public and the private sector through pragmatic and targeted interventions; and d) assess gender needs in access provision. 12. Many of these recommendations are cross-cutting; to plan and implement them would involve several ministries. Investments and recurrent expenditures in rural access are substantial and should be leveraged in the most effective manner. It is therefore recommended that the Ministry of Finance take the lead in bringing this Policy Note to the attention of Government, donors, and other actors. iii I. CURRENT STATUS 1. The National Rural Access Program (NRAP) that emerged from the National Emergency Employment Program (NEEP) is one of Afghanistan's most effective national priority programs. Given the country's acute shortage of rural roads and the extreme isolation of the rural population, the program has appropriately focused on rehabilitating rural roads. Until recently its management was largely ensured by an implementing partner, UNOPS, which provided the structure and systems for project execution; policy development, institutional strengthening, and capacity building were secondary objectives to implementation. Recently the ministries, especially the Ministry of Rural Rehabilitation and Development (MRRD), have become more involved. From NEEP to NRAP 2. The Transitional Government launched the National Emergency Employment Program (NEEP) in 2002 as one of four National Priority Programs designed to provide targeted social protection for vulnerable groups throughout the country. NEEP's rationale was to be "....the main instrument for the reconstruction and development of the country's rural access infrastructure, which also provides a safety net to the rural poor through the provision of short-term employment opportunities." 3 In the first year of implementation, the government spent US$28 million on NEEP, and the following year raised this allocation to about US$65 million, reflecting the importance of the program and the degree of government commitment to it. 3. When NEEP was designed, few data were available on poverty in Afghanistan, because the country had been in conflict for decades and its institutions were not equipped to collect the needed data or do research. In the course of implementing the program, changes in approach have been made in the light of experience and as new survey data have become available. 4. After a review of the initial experience, the activities under the NEEP umbrella were reoriented in early 2005. Targeted interventions for the poor were by-and-large left to other ongoing programs (such as the National Solidarity Program), while NEEP evolved into the National Rural Access Program (NRAP). NRAP emphasizes empowering rural people by creating infrastructure that improves their access to economic and social opportunities, health services, and education facilities. Though NRAP makes use of some labor-based technologies, providing employment is not its main goal and the emphasis on social protection mechanisms is more limited than at the outset of NEEP. Funding Sources 5. Since its establishment, NEEP/NRAP has been implemented through four major projects assisted by the World Bank (see Annex 1 for a timeline): The Government's Emergency Community Empowerment and Public Works Project received Bank support through the Labor-Intensive Public Works Project (2002-04; IDA funding US$16.78 million), which provided employment in rural areas at a minimum wage, as a safety net, through the sustainable rehabilitation of irrigation capacity and provincial and district- level roads. 3 "Transitional Islamic State of Afghanistan Strategic Framework and Work Plan (SY 1384)", prepared by the NEEP Joint Program Management Unit. 1 The National Emergency Employment Program-1 (2002-ongoing, with Afghanistan Reconstruction Trust Fund (ARTF) funding of US$51.9 million) was launched to provide bridge funding until the National Emergency Employment Program for Rural Access (NEEPRA) was prepared (2003-07; IDA funding US$39.2 million). Both of these projects were initially designed to provide targeted social protection and strengthen key rural access infrastructure, so as to improve the livelihoods of the rural poor through: (i) emergency short-term employment on labor-based subprojects; and (ii) technical assistance for program implementation. The National Emergency Employment Program for Reintegration and Alternative Livelihood (2004-08; Japan Social Development Fund Grant, US$19.6 million, administered by the World Bank) supports the reintegration of ex-combatants into society and also contributes to the government's Alternative Livelihood Program in poppy producing areas. It seeks to create opportunities for employment and development of vocational training and business management skills for the beneficiaries. Government also provided funds received from the Programmatic Support Institution Building (PSIB) project. These contributions have been implemented in accordance with the project documents formulated for NEEPRA. 6. In addition, the Ministry of Rural Rehabilitation and Development itself implements activities with funding from the European Commission and the Counter-Narcotics Trust Fund (CNTF), and, for Helmand Province, from the UK Department for International Development. Table 1 gives summary information by donor. Table 1: Financial Contributions as of December 2006 Donor Amount (US$ million) Implementing Partner ARTF-NEEP 1 51.9 UNOPS EC 10.5 MRRD JSDF 19.6 UNOPS PSIB 15 UNOPS WB-LIWP 16.8 UNOPS WB-NEEPRA 39.2 UNOPS UNDP 3.1 UNOPS CNTFa 5.3 UNOPS DFID (for Helmand) 15.4 MRRD Source: NRAP Monthly Progress Report, December 2006. a MRRD is not involved in counter-narcotics operations but receives funding for road rehabilitation from CNTF. 7. A difficulty for NRAP has been that funding has been made available with relatively short time horizons, not allowing for multi-year programming of works and consultant support. Delays in funding releases have also hampered implementation. 8. NRAP currently faces an acute funding shortfall and operational funds will be exhausted at the end of June 2007. The Ministry of Finance has agreed to provide bridge funding until the proposed IDA credit for NERAP becomes effective. Achievements 9. NRAP has disbursed funds in all 34 provinces (Annex 3) and Government has used the program to demonstrate its commitment and to provide tangible development activities throughout the country. Table 2 summarizes provincial disbursements, completed road rehabilitation, and disbursements/population as of December 2006. The substantial variations 2 in the road distance rehabilitated per province partly reflect the slow progress made in some provinces because of security constraints. 10. Road maintenance efforts are on a pilot basis only. This is of growing concern to the Government and donors because the sustainability of rural roads is severely comprised if routine maintenance is not carried out as soon as rehabilitation works are finished. So far, only 8 km are under maintenance in Badakhshan, 2.5 km in Kabul, 3.9 km in Kapisa, and 10 km in Kunduz. Though funding has been made available from the ARTF contribution, maintenance apparently is not yet considered a real priority. Table 2: Provincial Distribution of NRAP Region Disbursements Population KM of road Disbursements (US$) improvements per capita (US$) Badakhshan 9,292,005 805,000 341 11.5 Central Highlands 9,221,580 1,356,100 692 6.8 East 6,909,998 2,147,600 771 3.2 Kabul 16,413,216 4,976,000 1,456 3.3 North 8,420,664 3,165,800 1,137 2.7 North-East 8,519,318 2,422,800 1,079 3.5 South 8,341,700 2,465,500 804 3.4 South-East 11,877,204 2,364,000 1,460 5.0 West 4,895,024 2,393,600 170 2.0 Source: NRAP Monthly Progress Report, December 2006. Note: Central Highlands: Bamyan, Wardak, Daykundi; East: Nangarhar, Laghman, Kunar, Nuristan; Kabul: Kabul, Parwan, Logar, Kapisa, Panjshir; North: Balkh, Samangan, Saripul, Faryab, Jowzjan; North East: Kunduz, Baghlan, Takhar; South: Kandahar, Nimroz, Zabul, Urozgan, Helmand; South East: Paktiya, Paktika, Khost, Ghazni; West: Herat, Ghor, Farah, Badghis. 11. Under LIWP and NEEP the program was more diverse than it now is. In addition to financing the rehabilitation of rural roads, funding was provided for 748 irrigation schemes, 965 wells, 89 water reservoirs and water supply networks, 95 schools, 15 clinics, 135 shelters, 41 other buildings, and 29 plant nurseries. In addition to road rehabilitation, NEEPRA financed the rehabilitation of eight provincial airfields and the construction of the MRRD headquarters in Kabul. The bulk of funding, however, has been for roads and structures. Table 3 gives details of achievements. Table 3: Achievements as of December 2006 Activity Total Km of road rehabilitated since inception 9,002 Km of road rehabilitated since March 2005 1,386 Km of ongoing road construction 852 Labor days generated since inception 12,948,409 Labor days generated since March 2005 2,870,766 Running meters of structures completed since inception 39,963 Running meters of structures completed since March 2005 28,368 Number of bridges under construction 21 Running meters of bridges under construction 749 Number of bridges built since March 2005 16 Running meters of bridges built since March 2005 296 Number of bridges completed since inception 101 Running meters of bridges completed since inception 922 Number of districts targeted 292 Number of provinces targeted 34 Contracts issued since the start of the program 1,982 Source: NRAP Monthly Progress Report, December 2006. 3 12. Since 2002, NEEP/NRAP has generated nearly 13 million labor days (Table 3). Though initially intended to provide a safety net, in practice the NEEP activities were not well targeted to the poor: nearly four out of five poor households had no workers employed in the program, and nearly one in five of the workers who were employed came from relatively wealthy households. NRVA 2003 data established, after NEEP had been designed, that 78 percent and 76 percent of the poorest households were not participating in cash-for-work or food-for-work programs respectively. The same data also showed that 17 percent of the participants in the cash-for-work programs and 16 percent in the food-for-work programs came from the richest 20 percent of households. Institutional Arrangements 13. In 2003 it was envisaged that the Ministry of Rural Rehabilitation and Development would be the lead ministry for NEEP, in partnership with the Ministry of Public Works and the Ministry of Irrigation, Water Resources, and Environment. To provide an effective framework for coordination and decision making, it was proposed to set up an Interministerial Board to be chaired by MRRD and with representation from all the relevant ministries. 4 Instead, a Steering Committee was set up, comprising only MRRD and MPW (Annex 2 provides an organogram). This represented a major step back from the holistic vision of bringing together the providers and clients of rural access. Even the Steering Committee met only twice since its creation and never fulfilled the roles assigned to it. 14. Rural access issues do not seem to have featured prominently on the agendas of the Afghanistan National Development Strategy working groups for Agriculture and Rural Development and Transport. But these groups could be a useful forum to present and debate questions on rural access and to seek coordination across ministries. Rural access considerations should also be included in the various initiatives of Government such as Agriculture, Roads, Trade, and Irrigation or Rural Enterprise Development. Implementation Arrangements 15. The actual evolution of program management units did not follow the path at first envisaged. At the design stage of NEEPRA, the Joint Program Management Unit was intended to be responsible for the daily management of the program and to report to the proposed Interministerial Board. Once established, the JPMU received technical assistance from the International Labor Organization (ILO), and a senior Program Advisor was attached to the Unit. 16. Both MRRD and MPW set up their own project implementation units (PIUs) to be in charge of implementing sub-projects, leaving the JPMU to coordinate their work. Each PIU was set up to function independently. UNOPS provided support to manage contracts, procurement, and finances, as well as the units' management information systems 5 , and the PIUs were assisted by chief technical advisors. The establishment of separate PIUs recognized the different functions and community relations responsibilities of each ministry. For example, MRRD has a much closer involvement with local communities through the types of roads it has been responsible for, and the method of contracting with local communities for Level 1 roads. 6 Therefore this ministry needed a widespread, constant presence at the provincial level. MPW's PIU had fewer sub-projects to manage, though these 4 Ministries of Public Works; Irrigation, Water Resources and Environment; Agriculture and Animal Health; Urban Housing and Development; Martyrs and Disabled; Women's Affairs; Labor and Social Affairs; Returnees and Repatriation; Public Health; Education; Frontiers and Tribal Affairs; Planning; Reconstruction; and Finance. 5 As noted earlier, implementation assistance for both MRRD and MPW has been provided by an implementation partner, UNOPS. For MRRD the IP was selected on a competitive basis, while MPW continued with UNOPS which was already supporting the ministry. 6 So-called Level 1 projects are implemented by community members and use labor-intensive technology. 4 were technically more demanding (higher-level roads and structures), and therefore it did not need as much of a permanent presence on the ground. 17. Following a recommendation by the NEEP Review of 2005, the JPMU was renamed the Program Development and Coordination Unit (PDCU) and took responsibility for creating and maintaining MIS/GIS data, undertaking evaluation and research, and identifying and addressing training needs within the Government as well as the private sector. 7 The PDCU has served as a focal point for reporting and relations with donors. But because the Steering Committee did not function as intended, the PDCU was left without policy guidance. Proposals that it put forward, such as the Strategic Framework and Work Plan SY 1384, were not endorsed by the corporate level. 18. Program implementation has been very successful overall, and the results on the ground attest to that. The role of the implementing partner, UNOPS, in providing a stable organizational structure and reliable fiduciary systems must not be underestimated. The MTR report recommended "the gradual phasing out of implementing partners (over 3-5 years) with the assumption of core public sector functions such as MIS, logistics, procurement, planning, policy, standards, financial management, and human resource/human and institutional development by the ministries..., it is important to point out that there is a need to ensure that IPs transfer the necessary capacity to their Government counterparts. Technical capacity transfer is obvious but IPs need to transfer management capacity particularly to the higher echelons of management to ensure the smooth delivery of a program as complex as NEEP." 8 19. It is difficult to quantify the transfer of capacity that has taken place because no baseline of staff competencies was established at the start of the program. 9 Capacity building has taken place across all job categories and has focused on training on the job. Annex 5 lists the capacity building activities that have been undertaken. The technical assistance products prepared include: policy, strategy, and concept papers; procedures and guidelines for JPMU/PDCU office management; procedures and guidelines for JPMU programming and resource allocation; project proposals, contracts, and agreements; job descriptions, terms of reference, and roles and responsibilities of the Steering Committee, PDCU, and PIUs; budget staff/IP performance; progress reporting formats and progress reports; communication and PR training workshops, and presentations; M&E strategies, systems, procedures, and indicators; on the job training in applying labor-based rural roads engineering; NEEP/NRAP and NEEP- JPMU reviews; planning; guidelines on social inclusion; contracting procedures and formats; MIS systems, procedures, and formats; and technical standards. 20. In the fiduciary areas of procurement and financial management, UNOPS has put systems in place but retained primary responsibility. Staff training in these areas has been very limited and a recent assessment of MRRD's project management capacity concluded that it was weak. 21. NRAP has established a detailed management information system that contains data on works, contractors, contracts, etc. Improvements are still necessary to ensure data coherence between reporting periods, minimize changes in definitions, and restrict access for data entry. The change in program orientation from social safety net operations to providing rural access using more of an engineering approach makes comparisons difficult. 7 That Review also recommended that the PDCU should cover all national programs that could benefit from integration of delivery mechanisms, not just NEEP. Had this recommendation been adopted, it could have gone a long way towards achieving efficiencies and better coordination of the very substantial donor assistance that targets the rural areas. 8 "National Emergency Employment Programme: Final Review Report." March 2005. 9 However, a training needs assessment was carried out for specific functions: functional analysis/job descriptions, mapping exercise, competency profiles, and staff assessments. 5 22. Data analysis and use for management purposes are still lacking, and so far no systematic monitoring has been done for evaluation purposes. When NEEPRA was prepared, lack of data and analytical capacity made it impossible to put in place a baseline study, and therefore it has been impossible to assess the development outcome of the program. Trust funding has just been received to carry out a retrospective evaluation of NEEPRA and also to develop a methodology for a baseline study for a proposed IDA-funded project, the National Emergency Rural Access Project (NERAP). Access Planning and Prioritization under NRAP 23. Access planning under NEEP/NRAP was ostensibly based on the following strategic principles: (i) investment in road network connectivity should follow the center­periphery principle, which means that initially more resources should be invested to develop and connect the higher-order road network before roads of a lower order are developed; (ii) notwithstanding this principle, there should be a broad balancing in the amount of investment between the secondary (MPW) and tertiary (MRRD) road systems; and (iii) investment allocations ought to reflect the relative vulnerability of different provinces. 24. Initially MRRD used a Provincial Accessibility Allocation Model to distribute the available budget among provinces based on provincial population and a composite vulnerability index as measured by relative levels of year-round access to basic services. 25. Starting in 2003, however, MRRD helped individual provinces to identify their priority infrastructure investments. Each district listed its needs, which were not limited to roads. The top two of these priorities were then voted on at a provincial workshop, with district representatives, so as to keep overall investment within a nominal allocation of US$500,000 per province. 10 Technical surveys and refinements of each project were undertaken by MRRD. 26. How the overall access planning approach has worked in practice is difficult to discern. MRRD is working on, but has yet to complete, an inventory of rural roads and an Interim Rural Roads Investment Plan (IRRIP). Certainly MRRD has invested substantial staff time in preparing the IRRIP and the database and maps that have been produced have greatly increased the knowledge of the rural road network. MPW's complementary Master Plan for Road Network Improvement Project exists, but its current status is uncertain. Planning coordination between the two ministries appears weak. 27. The actual choice of sub-projects can vary significantly from the plan submitted by NRAP. The selection is the responsibility of the government through each of the ministries, and plans can be changed outside the processes used by NRAP. Work plans may be modified for a variety of reasons, including pressure from interest groups and deteriorating security. Security is deteriorating in the southern and central regions, where both MRRD and UNOPS have had casualties. Several staff have received threats and warnings. In addition, criminal activities against staff and offices are increasing. 28. The map for the Central Highlands Region (Annex 4) illustrates that sub-projects there are widely scattered and so far have not contributed much to establishing connectivity between the strategic (regional, national, provincial) and local (district, community) road networks. Having scattered worksites increases the cost of delivery and probably has a negative impact on quality. It is not uncommon for site engineers to travel a couple of days to reach a construction site. The cost of transporting material and machinery can be quite high and working conditions for the contractors are difficult. 10 In practice this was modified by the province's population and could range from US$400,000 ­ 600,000. 6 Safety Net, Design Standards, and Technology 29. As noted earlier, the initial stages of the rural access program put more emphasis on job creation than on quality infrastructure. LIWP improved about 4,200 km ­ without structures such as culverts or bridges ­ at a cost of about US$2,700/km. Under LIWP, NEEP 1, and NEEPRA, communities were contracted to implement sub-projects (so-called Level 1 contracts) and about US$17.6 million have been spent to improve 6,500 km in this way. 30. The reorientation of the program in 2005, to give primacy to providing rural access, brought with it a shift to an engineering approach. Under LIWP the average cost per km for MPW's sub-projects was US$9,300; under NEEP-1 it increased to US$10,055, and under NEEPRA it has reached US$30,000. There are no agreed-upon standards for Afghan rural roads, and existing technical documents have clearly been imported from other countries. The current approach and the standards that have been adopted seem appropriate for major highways. The same cannot be said for smaller rural roads. Strictly applying such a design approach to the expansion of the rural road network would be very costly: at the program's current average cost of US$30,000/km for rural road rehabilitation, and based on current annual funding levels, it would take more than 100 years to connect all villages to the road network. Communities are contracted to carry out improvement works and contractors use a mix of equipment and labor. But labor-based techniques are not systematically used and not enough training has been provided to ensure that labor is used efficiently. Whilst MRRD had a credible track record in labor-intensive works in the 1970s, it has little experience of the modern labor-based techniques that have since evolved internationally. International technical assistance was provided but there has been no sustained effort or political buy-in in support of these methods. Current Policy Environment 31. The 2005 mid-term review of NEEPRA concluded that: NEEP has been driven by "outreach" aims and has been implemented in an overall policy vacuum. Whether it delivered rural roads and related infrastructure, or irrigation and conservation, NEEP has been driven primarily by the desire to attain maximum geographical coverage while injecting cash to address emergency needs in rural areas--income as well as non-income transfers. There has been no explicit rural development strategy within which this national program was ensconced. More poignantly, while its bulkiest investment was in rural roads--it was not underpinned by a cohesive rural transport or an overall transport policy. The Implementation Partners and the sector ministries did attempt to rationalize rural access investments from a development perspective using standard allocation parameters and consultative mechanisms. The absence of a policy framework, however, was laid bare by, amongst other things, the absence of planning for the future sustainability of these investments. 11 32. This situation still persists, and NRAP continues to be implemented without a policy framework. In March 2005, the JPMU prepared a "Strategic Framework and Work Plan" including targets from SY1384 to SY1388. It is not clear whether this document has been applied. 11 "National Emergency Employment Programme: Final Review Report." March 2005. 7 II. KEY ISSUES AND CONSTRAINTS 33. This chapter discusses the crucial issues for rural access in Afghanistan. 34. A national priority program is being implemented without a policy framework: NRAP has successfully delivered about 9,000 km of rural roads across the country ­ a remarkable achievement under extremely difficult circumstances. But the lack of a rural access policy has left unanswered a number of key questions on institutional arrangements, planning and programming, and tradeoffs between quality and quantity in the provision of rural roads. Donor funding has been used in an ad hoc manner and no multi-year, realistic rural access program has been developed and presented to the donor community. 35. Need for more effective coordination between the providers of access infrastructure and the beneficiaries, and for more transparent decision making: Currently there is no effective coordination mechanism between MPW and MRRD and the ministries that need rural roads, such as Education, Health, Irrigation, Agriculture, and Counter-Narcotics. Further, no effective coordination takes place even on a program level among the National Solidarity Program, the National Area-Based Development Project, or the Horticulture and Livestock Project. Important players in Afghanistan are the provincial reconstruction teams, whose interventions are security- and development-oriented, but other agencies appear to have very little exchange with these teams. MRRD has been holding consultative workshops at the district and provincial levels, and the authorities at these levels are involved in the identification of needs but not the decision making. These consultations have taken place without reference to budget constraints and have therefore produced a long list of needs that cannot realistically be met. As noted in Chapter 1, investments are scattered over a large geographic area with little apparent concern for connectivity. Frequent changes in the work program suggest that not only security considerations but pressures from interest groups may be affecting the choice of subprojects. 36. Ministries' core functions, network responsibilities, and implementation arrangements need to be defined: The implementation of NEEP and NRAP relied on the services of an implementation partner because of the need to deliver tangible benefits quickly despite the capacity constraints in the ministries. Moving forward, the ministries need to determine the best delivery mechanisms for rural access improvement and management, and to distinguish the core functions that are necessarily within their domain from those functions that can be permanently outsourced to the private sector. The ministries also need to continue the public administration reform process, such as by introducing pay-and-grading, and decide on the appropriate level of deconcentration of services. Though the private sector is still emerging, its capabilities might be able to grow faster than ministries can be reformed. 37. Capacity constraints in public and private sector: As in other national programs, the severe capacity constraints in the public sector have so far been mitigated by the presence of an international partner, in this case UNOPS. This has been an effective solution up to now, but Government deplores the associated cost and the loss of sovereignty. UNOPS has put in place a reasonably dependable fiduciary system and has paid attractive wages. The reputation and strength of MRRD seems largely due to successful national programs, especially the National Solidarity Program and NRAP, but the ministry itself still needs to carry out a comprehensive institutional analysis and to define its vision. The salaries paid by the national programs are not affordable in the long run and the tight labor market makes it unrealistic to adjust salaries downward. MPW has not effectively started any institutional reforms and has been a passive partner in NRAP. The private sector has re-emerged since 2002 and the massive inflow of donor resources, coupled with the large exodus of qualified professionals during the prolonged war, is stretching the availability of contractors. There are no engineering firms that could be used for survey and design, or supervision of works. 8 Afghanistan lacks technical colleges and universities offering quality education and therefore there might be an acute shortage of qualified staff for years to come. 38. Managing the tradeoff between coverage and quality: As the program has moved from being essentially a safety net instrument to providing rural roads of a relatively high standard, the unit costs of works have risen substantially, though so too has their durability. As noted, resources are scarce, and even at the level of the secondary road network the situation is critical: apparently 108 of 396 districts do not have access to their provincial capitals. Given the extreme isolation of Afghan villages, a shift in program standards might be warranted, towards simply providing basic year-round access at the lowest acceptable level. Thus the program would first tackle obvious obstacles to all-season movement, and then, if demand exceeds expectations, carriageways could be widened and stronger surfaces installed that are appropriate to actual traffic volumes. 39. Funding for rural road maintenance: Spending on rural access in Afghanistan has been substantial, at US$193 million over the series of programs since 2002. But the investment is in danger of being lost within a couple of years if an effective maintenance strategy is not put in place very soon, at least for core networks in each district. Very limited funding has been made available in recent years, and maintenance efforts are small and piecemeal. The focus remains on rehabilitation to reach maximum coverage and there may be some resistance to returning to the same areas with funding for maintenance. Since the Government is unlikely to be able to finance the maintenance of rural roads, alternative solutions should be found, such as co-funding by road users. The latter solution is not currently feasible, however, given the high technical standards now in use for road rehabilitation. Roads restored to these standards require expensive and mechanized maintenance, which is beyond the reach of the users. 40. Potential for using modern labor-based techniques: International experience shows unequivocally that there is considerable potential for establishing modern labor-based methods in Afghanistan 12 , and significant benefits can be expected to result. A labor-based approach would provide large numbers of legal employment jobs in road construction and maintenance even though it could not offer high enough wages to compete with the poppy industry. A labor-based program cannot be introduced as a "quick fix" but needs systematic preparation, pilot testing, and training of ministry's staff and contractors. 41. Gender considerations in rural access provision: NRVA surveys show that access is significantly less for women and girls in all aspects than for men or boys. These differences are most pronounced at the household level with respect to education, health, and other services. These differences should be analyzed and community-supported ways identified to address these issues. 12 Potentially, labor-based operations can employ five times more labor, inject US$3,000 ­ 5,000 per km into the local economy on major rural road rehabilitation, generate income multipliers in the range 1.5 ­ 2.8 due to spending during construction, and be 25-30 percent cheaper than comparable capital-intensive methods. See Willem Keddeman, "Of Nets and Assets: Effects and Impacts of Employment-intensive Programs - A Review of ILO Experience," Socio-Economic Technical Papers No. 1. International Labor Office, Geneva, January 1998. Successful labor-based programs have been run in countries such as Botswana with a population density of just 1 person per square km ­ a fraction of that prevailing in Afghanistan. 9 III. RECOMMENDATIONS 42. Afghanistan is probably one of the most access-deprived countries in the world. Lack of roads holds back social and economic development, the replacement of the poppy economy with licit rural livelihoods, and the fight against the insurgency. The key constraints identified above clearly show that the rural access agenda needs to be tackled with the involvement of stakeholders from Government, the private sector, and the provincial reconstruction teams (PRTs). It also seems necessary to identify recommendations for the short and medium term. 43. The following five recommendations should be undertaken without delay: Define a National Rural Access Policy and a multi-year investment and maintenance program. This Policy Note can, it is hoped, serve as a basis for the Ministry of Finance (MoF) to initiate a dialogue with ministries, donors, the provincial reconstruction teams, and the private sector. It is recommended that MoF call for a workshop that will identify the rural access agenda and define the key elements of the policy. (Box 1.) Based on the policy decided upon, Government and development partners should formulate multi-year programs for investment in tertiary and secondary rural roads. Set up coordination mechanisms. Under the leadership of MoF an effective coordination mechanism should be established for the different levels of government, ministries, donors, and the PRTs. This mechanism is needed for planning and implementation purposes. An interministerial steering committee should be set up to review rural access program orientation and implementation. Give priority to defining a strategy for rural road maintenance. The recently improved 9,000 km of roads are in danger of falling into disrepair if no concerted effort is made to maintain them. Innovative approaches are needed to leverage the available funding. It is recommended that the proposed NERAP include financing for the elaboration of the maintenance strategy and a component for maintenance works that would be channeled through Afghanistan's community development councils. Introduce and apply transparent mechanisms for road prioritization and programming. The current prioritization process and resulting investment program are not sufficiently transparent and are open to frequent changes. To build a more robust decision making process, a synthesis of MPW's road master plan and MRRD's ongoing Interim Rural Roads Investment Plan should be (i) completed and integrated so as to provide a single national assessment; and (ii) complemented by an (equity) analysis that will show which provinces and districts are the most disadvantaged. Regular consultations should be held with the community development councils, district development assemblies, provincial development councils, client ministries, and PRTs. In these consultations the budget constraint needs to be highlighted so as not to raise expectations that cannot be fulfilled. Decision- making tools such as Integrated Rural Access Planning, the Road Economic Decision Model, and the Highway Design Model should be introduced to enable staff to provide their senior managers with robust arguments for investment decisions ­ that can also be used to fend off pressures for ad hoc adjustments of these decisions. 10 Adopt an approach of providing basic year-round access. The program is now taking a classic engineering approach to road provision, at unit costs that will inevitably limit its coverage. Instead, it is recommended that priority be given to providing reliable, all-season access at the lowest acceptable level of service; considerations for travelers' speed and comfort should be incorporated only at a later stage. Adopting this changed approach would require the ministries to develop appropriate technical standards and apply these specifications systematically. Box 1: Core Principles of the NRAP Strategy The core principles of NRAP, as defined in JPMU's "Strategic Framework and Work Plan" issued in 2005, might provide a useful platform to develop an Afghan rural access policy. General Accountability and transparency principles Effectiveness and competitiveness Coherence Uniformity Sustainability Efficiency Institutional Capacity development within government development Funding through a Government Trust Fund (ARTF) principles Government-led coordination Government execution Reporting through the government The government's primary role and function as a manager and not an implementer Private sector-led development with the Government in a facilitating role Social Resource allocations proportional to geographic (provincial) distribution of the rural targeting population and adjusted with a composite provincial vulnerability index principles Resource allocation takes already resourced projects into account Equality in opportunity for the targeted population Adherence to international labor standards and conventions Setting wage rates at the level of unskilled agricultural labor in the agricultural slack season Timing, location, and size of activities designed to avoid competition for unskilled labor Specific targeting through the selection of appropriate labor-based activities Specific targeting through piece work and task work implementation modalities Rural access Quality and durability development Functionality principles Available capacities for planning, implementation, and supervision Competitive bidding in contracting works to private contractors or NGOs Multi-year planning and resource allocation mechanisms linking physical planning and available resources Planning, prioritization, and selection of infrastructure schemes based on rational processes Infrastructure works' life-cycle costs to be considered, including maintenance costs Prioritization of still maintainable infrastructure including economic criteria and using objectively verifiable information. Labor productivity as basis for costing infrastructure works and labor payment Source: NEEP JPMU, "Strategic Framework and Work Plan SY 1384." 44. Other recommendations will require a longer-term approach, integrated with ongoing reforms. They are: Clarify institutional responsibilities. The responsibilities of MPW and MRRD should be clarified with the help of an improved road classification. Responsibility for maintenance needs to be revisited and decided on a 11 pragmatic basis. The maintenance of tertiary roads should be managed at the lowest level, i.e. by community development councils, with technical advice from the National Solidarity Program's facilitating partners and later from MRRD. Institutional analyses should distinguish the core functions of ministries from those services that can be outsourced on a long-term basis. The roles in building and maintaining rural roads of sub-national government (community development council clusters, district development assemblies, provincial development councils) and of de-concentrated ministerial services need to be assessed. Introduce modern labor-based techniques. Afghanistan's rural economy could benefit substantially from the introduction of such techniques for road rehabilitation and maintenance. Modern labor-based interventions should become the recommended technique of rural road improvement, particularly for the tertiary road network. A training program should be developed for ministry staff and contractors, drawing on the experience of successful labor- based programs in Afghanistan during the 1970s and outside the country. Enhance the capacity of public and private sectors. Capacity building efforts have not been very successful in Afghanistan even though nearly US$1 billion has been spent on assistance to the country since 2002. It has been recognized that it is next-to impossible to build capacity in dysfunctional ministries, and that new approaches are necessary. Reforms in MRRD and MPW need to be continued, institutional analyses carried out, and pay-and- grading reforms started. The ministries need to foster a system of corporate governance. The services of UNOPS have enabled the delivery of a successful national program but the substantial technical assistance is considered unsustainable by Government. During the new program phase, targeted capacity building should take place so that the support from technical assistance can be drawn down gradually. Fiduciary systems need to be shifted to the ministries without compromising their effectiveness and reliability. A comprehensive training program should be set up for contractors and the emergence of private engineering firms should be encouraged. There is a need to provide credit facilities to contractors so that they can expand, and this should be seen in the wider context of developing banking services and encouraging the private sector. Assess gender needs in access provision. While employment opportunities for women in the rural roads sector are extremely limited in Afghanistan, the access needs of women and girls need to be better understood and taken into consideration in planning. It is recommended that a study of gender-sensitive access needs be carried out to establish the baseline. 12 ANNEX 1: NRAP TIMELINE Overview of WB-led Funding for Rural Access LIWP (IDA) LIWP (IDA) US$16.69 M US$16.69 M Effective 6/11/02 Closed 12/31/04 [Labor- NEEP-1 intensive works (ARTF) NEEP-1 (ARTF) NEEP-1 (ARTF) NEEP-1 Component of US$25.4 M US$16.62 M US$36.82 M (ARTF) Emergency Effective Reduced 6/4/03 Increase 11/28/05 US$50.62 M Community 3/14/03 Increase US$16 Empowerment And Public [National Works Project] Emergency Employment NEEPRA (IDA) Proposed NERAP Project] US$39.2 M (IDA) US$120 M Effective 7/22/03 [Afghanistan Closing 9/30/07 NEEP DDR/RLS [National Emergency (JSDF) Employment US$19.6 M Project for Effective 8/28/04 Closing 3/30/2008) Rural Access] [Disarmament, Demobilization and Reintegration/Reintegration and Alternative Livelihoods] [Japanese Social Development Fund] 2002/3 2003/4 2004/5 2005/6 2006/7 13 ANNEX 2: NRAP ORGANOGRAM NRAP Steering Committee (MRRD, MPW) Program Development and Coordination Unit (PDCU) MRRD - PIU UNOPS MPW - PIU Regional PIUs Kabul, Southeast, East, Northeast, Badakshan, North, West, South, Central Highlands 14 ANNEX 3: PROVINCIAL DISTRIBUTION Province US$ Labor days Road Population US$ per disbursement generated Rehabilitation population (km) Badakhshan 9,292,005 716,024 341 805,000 11.5 Badghis 1,046,762 198,056 49 420,000 2.5 Baghlan 3,692,549 506,147 437 762,000 4.8 Balkh 2,204,068 174,716 215 1,073,000 2.1 Bamyan 6,437,719 909,592 332 379,200 17.0 Daikundi 345,169 391,000 0.9 Farah 639,619 34,244 19 428,800 1.5 Faryab 1,871,952 208,055 171 840,400 2.2 Ghazni 2,445,323 164,637 259 1,040,000 2.4 Ghor 2,438,692 136,941 290 585,900 4.2 Helmand 4,194,733 229,754 275 782,100 5.4 Herat 3,208,643 446,240 102 1,544,800 2.1 Jawzjan 1,410,444 293,139 286 452,000 3.1 Kabul 6,931,878 692,499 432 3,071,600 2.3 Kandahar 2,071,500 312,697 341 990,100 2.1 Kapisa 1,626,286 226,393 155 374,500 4.3 Khost 2,261,198 341,240 397 487,400 4.6 Kunar 297,806 65,239 146 381,900 0.8 Kunduz 3,418,419 685,710 484 833,300 4.1 Laghman 2,482,777 331,941 216 378,100 6.6 Logar 2,070,131 218,896 310 332,400 6.2 Nangarhar 2,889,153 336,407 320 1,261,900 2.3 Nimroz 234,829 41,328 30 138,500 1.7 Nuristan 1,240,262 171,460 89 125,700 9.9 Paktika 1,761,337 364,068 396 369,100 4.8 Paktya 5,409,346 669,701 408 467,500 11.6 Panjsheer 803,344 24,088 14 130,400 6.2 Parwan 2,427,615 148,945 220 560,800 4.3 Samangan 2,047,325 281,656 289 327,700 6.2 Sari Pul 886,875 163,626 176 472,700 1.9 Takhar 1,408,350 193,959 158 827,500 1.7 Uruzgan 1,060,827 63,019 61 297,200 3.6 Wardak 2,553,962 206,235 325 506,300 5.0 Zabul 779,811 77,028 97 257,600 3.0 Total 83,890,709 9,633,680 7,840 22,096,400 3.8 15 ANNEX 4: MAP 16 ANNEX 5: CAPACITY BUILDING ACTIVITIES · MIS: Training to staff of NEEP/NRAP, MRRD and MPW on data entry, data verification and validation, database development, database administration; · IT: Training of JPMU/PDCU and PIU staff on basic computer skills; · Management: On-the-job training on various management and programming aspects; · Social Inclusion: Training and orientation of national consultants and MRRD staff regarding the implementation of social inclusion methodologies, approaches and procedures in the program; · PR and Communication: On-the-job training to government civil servant counterparts and MRRD staff on basic aspects of public relations and communication (including practical training courses on subject matters like photography and writing articles for newspapers); In 2007 the Public Relations Office has conducted workshops in three locations teaching a range of PR and communications skills designed to promote the program and identify human interest aspects of NRAP. These programs involve NRAP, PRRD, MPW, RAL, and local staff. To date more than 60 staff have attended. · Planning: Training delivered to MRRD and NEEP/NRAP civil servants and national advisors regarding the application of planning procedures and processes that had been developed by the program; · Reporting: On-the-job training to the JPMU/PDCU and PIU national advisors on reporting; · Monitoring and Evaluation: On-the-job training to PDCU national advisors on the application of NRAP's monitoring and evaluation tools and systems; · Financial management and Administration: On-the-job training to national consultants of the JPMU in financial and administrative management; · Labor-Based Appropriate Technologies (LBAT): On-the-job training and formal training to engineering staff and advisors at national and regional level on the application of LBAT, including the selection of pavement options, the use of appropriate technologies and equipment and the geometric design of rural roads. In addition, on-the- job training was provided to local contractors. · Training of Trainers: In conjunction with the Engineers Workshop in January 2006, conduct training in basic training techniques for 46 engineers. · IRRIP Workshops: MRRD PIU have completed a provincial-based series of workshops covering the requirements to prepare proper road inventories. This included the re-design and modification of systems and procedures, and the conduct of workshops in each province. · Road Construction Engineering Training: Training was provided by MPW to 32 engineers and 85 site technicians based on internationally recognized standards for rural roads construction. Internationally recognized engineering standards, particularly appropriate for application at rural roads in developing countries, have been used throughout the project cycle; · Executive Training: Provision of executive level training through international forums, conferences and by working directly with international technical assistance staff of the NEEPRA-MRRD implementing partner (UNOPS) through which a solid implementation support capacity was established; · Internships for Engineering Students: MRRD established an internship program for 25 final-year engineering students of the Kabul Polytechnic University in 2006. MPW provided 45 students (21 in 2005 and 24 in 2006) from the 4th class of Civil Engineering field of Faculty of Engineering, Kabul University with Practical Training opportunities. 17 They received extensive survey, design, and site supervision knowledge during the Practical Training Semester of 4.5 months. · Survey Equipment: Over 50 percent of engineers have received training in the use of Total Work Station. · Surveyor Training ­ Total Station ­ Over 50 percent of engineers (approximately 60 staff) have now received training on the Total Station surveying system. · Regional Meetings: To improve communications, to devolve training activities and to achieve economies of scale, Regional offices instituted regionally-based meetings combined with training workshops appropriate to their needs. This resulted in an estimated 240 training days. 18