IMPLEMENTATION COMPLETION AND RESULTS REPORT (TF-91120) ON A GRANT IN THE AMOUNT OF US$57 MILLION) TO THE ISLAMIC REPUBLIC OF AFGHANISTAN FOR A

Transcription

1 Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Document of The World Bank Report No: ICR IMPLEMENTATION COMPLETION AND RESULTS REPORT (TF-91120) ON A GRANT IN THE AMOUNT OF US$57 MILLION) TO THE ISLAMIC REPUBLIC OF AFGHANISTAN FOR A KABUL, AYBAK, MAZAR-E-SHARIF POWER PROJECT Sustainable Development Department Afghanistan South Asia Region September 28, 2013

2 CURRENCY EQUIVALENTS (Exchange Rate Effective September 28, 2013) Currency Unit = Afghani Af 1,000 = US$18.03 US$ 1.00 = Afghani US$ 1.00 = Special Drawing Rights AFGHAN FISCAL YEAR March 21 March 20 (until December 2011) March 21 December 20 (in 2012) December 21 December 20 (from December 2012) ABBREVIATIONS AND ACRONYMS ADB AEIC APSDP ARTF DABM DABS DO EIRP EPRP ESSF ESS GoA GWh IANDS ICR IDA ISDS ISN KAMPP KfW kv LC LV MACA MEW MOF MoU MV Asian Development Bank Afghanistan Energy Information Center Afghanistan Power System Development Project Afghanistan Reconstruction Trust Fund Da Afghanistan Breshna Moassesa Da Afghanistan Breshna Sherkat Development Objective Emergency Infrastructure Rehabilitation Project Emergency Power Rehabilitation Project Environmental and Social Safeguards Framework Environmental and Social Safeguards Government of Afghanistan Gigawatt hour Interim Afghanistan National Development Strategy Implementation Completion Report International Development Association Integrated Safeguards Data Sheet Interim Strategy Note Kabul, Aybak, Mazar-e-Sharif Power Project Kreditanstalt für Wiederaufbau Kilovolt Letter of Credit Low Voltage Mine Action Center for Afghanistan Ministry of Energy and Water Ministry of Finance Memorandum of Understanding Medium Voltage

3 MVA NEPS NTS PCB PDO PMF PISU QALP SDU TA USAID Megavolt Amperes North East Power System Northern Transmission System Polychlorinated Biphenyl Project Development Objective Project Management Firm Project Implementation Support Unit Quality Assessment of the Lending Program Special Disbursement Unit Technical Annex United States Agency for International Development Vice President: Philippe Le Houerou Country Director: Robert J. Saum Sector Manager: Julia Bucknall Project Team Leader: Richard Spencer ICR Team Leader: Richard Spencer

4 AFGHANISTAN Kabul Aybak and Mazar-e-Sharif Power Project CONTENTS Data Sheet A. Basic Information B. Key Dates C. Ratings Summary D. Sector and Theme Codes E. Bank Staff F. Results Framework Analysis G. Ratings of Project Performance in ISRs H. Restructuring I. Disbursement Graph 1. Project Context, Development Objectives and Design Key Factors Affecting Implementation and Outcomes Assessment of Outcomes Assessment of Risk to Development Outcome Assessment of Bank and Borrower Performance Lessons Learned Comments on Issues Raised by Borrower/Implementing Agencies/Partners Annex 1. Project Costs and Financing Annex 2. Outputs by Component Annex 3. Economic and Financial Analysis Annex 4. Bank Lending and Implementation Support/Supervision Processes Annex 5. Beneficiary Survey Results Annex 6. Stakeholder Workshop Report and Results Annex 7. Summary of Borrower's ICR and/or Comments on Draft ICR Annex 8. Comments of Cofinanciers and Other Partners/Stakeholders Annex 9. List of Supporting Documents Map: Project Area... 47

5 A. Basic Information Country: Afghanistan Project Name: ARTF - Kabul- Aybak/Mazar-e-Sharif Power Project Project ID: P L/C/TF Number(s): TF ICR Date: 09/28/2013 ICR Type: Core ICR Lending Instrument: SIL Grantee: Original Total Commitment: Revised Amount: USD 57.00M Environmental Category: B Implementing Agencies: Ministry of Energy and Water Cofinanciers and Other External Partners: GOVERNMENT OF AFGHANISTAN USD 57.00M Disbursed Amount: USD 49.90M B. Key Dates Process Date Process Original Date Revised / Actual Date(s) Concept Review: 08/22/2007 Effectiveness: 11/20/2007 Appraisal: Restructuring(s): 09/19/ /14/2012 Approval: 10/04/2007 Mid-term Review: Closing: 12/31/ /31/2013 C. Ratings Summary C.1 Performance Rating by ICR Outcomes: Risk to Development Outcome: Bank Performance: Grantee Performance: Unsatisfactory High Unsatisfactory Unsatisfactory C.2 Detailed Ratings of Bank and Borrower Performance (by ICR) Bank Ratings Borrower Ratings Quality at Entry: Unsatisfactory Government: Unsatisfactory Quality of Supervision: Moderately Implementing Moderately Unsatisfactory Agency/Agencies: Unsatisfactory Overall Bank Performance: Unsatisfactory Overall Borrower Performance: Unsatisfactory i

6 C.3 Quality at Entry and Implementation Performance Indicators Implementation QAG Assessments Indicators Performance (if any) Potential Problem Project Yes at any time (Yes/No): Problem Project at any time (Yes/No): DO rating before Closing/Inactive status: Yes Unsatisfactory Quality at Entry (QEA): Quality of Supervision (QSA): None None Rating D. Sector and Theme Codes Original Actual Sector Code (as % of total Bank financing) Central government administration 8 Power 92 Transmission and Distribution of Electricity 100 Theme Code (as % of total Bank financing) Conflict prevention and post-conflict reconstruction Infrastructure services for private sector development E. Bank Staff Positions At ICR At Approval Vice President: Philippe H. Le Houerou Praful Patel Country Director: Robert J. Saum Alastair J. McKechnie Sector Manager: Julia Bucknall Salman Zaheer Project Team Leader: Richard Jeremy Spencer Sunil Kumar Khosla ICR Team Leader: ICR Primary Author: Richard Jeremy Spencer Gilles Marie Veuillot F. Results Framework Analysis Project Development Objectives (from Project Appraisal Document) The development objective of the project is to help provide reliable and quality power to the consumers in the target areas of the cities of Kabul, Aybak, and Mazar-e-Sharif. Revised Project Development Objectives (as approved by original approving authority) NA ii

7 (a) PDO Indicator(s) Indicator Baseline Value Original Target Values (from approval documents) Formally Revised Target Values Actual Value Achieved at Completion or Target Years Indicator 1 : Increase in power supply to project areas in Kabul. Value 427 GWh (in Afghan SY 482 GWh (in quantitative or 1386) Afghan SY 1389) Qualitative) 1,219 GWh Date achieved 06/18/ /30/ /31/2010 Comments (incl. % 253% of the baseline target achieved by originally expected completion date. achievement) Indicator 2 : Providing grid power supply access to the people of Aybak (% of population). 25% of Aybak population 25% of Aybak 25% of the Value connected to local grid population population now quantitative or fuelled by diesel connected to connected to the Qualitative) generators. substation. substation. Date achieved 06/18/ /30/ /31/2013 Comments (incl. % achievement) Indicator 3 : Value quantitative or Qualitative) All customers previously connected to local grid who received power from diesel generator now receive power from substation. Increase in power supply to customers in Mazar-e-Sharif GWh (in Afghan SY 1386) 113 GWh 197 GWh Date achieved 06/18/ /31/ /31/2010 Comments Grid-supplied electricity by 2010 was already 197 GWh in Mazar-e-Sharif; (incl. % 174% of the target had been reached. achievement) (b) Intermediate Outcome Indicator(s) Indicator Baseline Value Original Target Values (from approval documents) Formally Revised Target Values Actual Value Achieved at Completion or Target Years Strengthen the low voltage (LV) distribution network in parts of Kabul supplied Indicator 1 : by Junction 2 as measured by number of transformers installed and commissioned. Value (quantitative or Qualitative) None. Works completed. Works completed. Date achieved 06/18/ /30/ /31/2011 Comments (incl. % achievement) 100% achieved with 11 months delay. Installed and commissioned 135 transformers. iii

8 Strengthen the low voltage (LV) distribution network in parts of Mazar-e-Sharif Indicator 2 : City, and measured by the number of 20/0.4 kv transformers installed and commissioned. Value (quantitative or Qualitative) None. Works completed. Works completed. Date achieved 06/18/ /30/ /31/2011 Comments (incl. % achievement) Indicator 3 : Value (quantitative or Qualitative) 100% achieved with 11 months delay. Installed and commissioned 62 transformers; delivered 161 for future installation by DABS. Construct a 220/20kV substation and associated 20kV lines at Aybak to provide quality, reliable and clean grid power to the residents of Aybak town (indicator being completion of Aybak substation). No substation. 220/20kV Aybak substation completed. 220/20kV Aybak substation completed. Date achieved 06/18/ /30/ /08/2012 Comments (incl. % 100% achieved with 18 months delay. achievement) G. Ratings of Project Performance in ISRs No. Date ISR Archived DO IP Actual Disbursements (USD millions) 1 06/19/2008 Satisfactory Satisfactory /02/2008 Satisfactory Satisfactory /23/2009 Satisfactory Satisfactory /13/2009 Satisfactory Satisfactory /27/2010 Satisfactory Satisfactory /11/2010 Satisfactory Moderately Satisfactory /27/2011 Moderately Satisfactory Unsatisfactory /01/2012 Moderately Satisfactory Unsatisfactory /20/2012 Satisfactory Moderately Satisfactory /30/2012 Moderately Satisfactory Moderately Satisfactory /25/2013 Unsatisfactory Unsatisfactory H. Restructuring (if any) Restructuring Date(s) Board Approved PDO Change ISR Ratings at Restructuring DO IP Amount Disbursed at Restructuring in USD millions Reason for Restructuring & Key Changes Made 09/19/2011 MS U Extension of closing date iv

9 Restructuring Date(s) Board Approved PDO Change ISR Ratings at Restructuring DO IP Amount Disbursed at Restructuring in USD millions Reason for Restructuring & Key Changes Made 06/14/2012 MS U Extension of closing date I. Disbursement Profile v

10

11 1. Project Context, Development Objectives and Design 1.1 Context at Appraisal With the establishment of the Islamic Republic of Afghanistan and presidential elections in 2004, Afghanistan emerged from two decades of instability. Parliamentary elections followed in 2005 and a new Government was established in Significant progress was noted in a number of sectors over this period and strong economic growth was recorded. The Bank s Interim Strategy Note (ISN) of 2006 recognized that security remained extremely volatile and that the country was confronted to daunting governance and capacity challenges all of which called for continued, massive support to secure Afghanistan s stabilization and recovery. The long lasting conflict had a severe impact on the power sector, both directly from dilapidated generation, transmission and distribution infrastructure and indirectly from serious management issues at Da Afghanistan Breshna Moassesa (DABM) including lack of appropriate governance structure; overstaffing and under-qualification of staff; poor customer management; billing and collection; and below cost-recovery tariffs. 1 As a result, access to power was estimated at 6%, one of the lowest rates in the world with only 234,000 people connected to the public grid, of whom 30% were in Kabul with other provinces having even less access, and rural areas being virtually un-served. The World Bank provided limited support to the sector in 2002 under a multi-sectoral grant, the Emergency Infrastructure Rehabilitation Project (EIRP). In 2004, drawing on a series of studies 2, the Bank approved a more comprehensive US$105 million Emergency Power Rehabilitation Project (EPRP, IDA Credit 3933) which provided financing for rehabilitation of the 100MW Naghlu hydropower plant and transmission line to Kabul, and extensive rehabilitation of the Kabul distribution system. In 2005, the Bank also started processing a second Emergency Power Rehabilitation Project (EPRP II). It was to support one of the government s identified priorities for the sector: completing the North East Power System interconnector for the transmission of low-cost power from Uzbekistan to Kabul and points along the route. Delays in implementation led to a Highly Unsatisfactory rating for EPRP, and EPRP II was subsequently dropped. In 2006, EPRP underwent a major restructuring in which support to rehabilitation of distribution networks in Kabul was significantly reduced and funds were reallocated for the rehabilitation the NEPS Interconnector. 1 Until September 2009, the entity responsible for power supply, Da Afghanistan Breshna Moassesa (DABM), was a department of the Ministry of Energy and Water (MEW). With World Bank and other donor support, DABM was converted into Da Afghanistan Breshna Sherkat (DABS), focused on developing into a fully commercial power utility outside the MEW while remaining under state ownership. In this ICR, for clarity we refer to DABS as the national electric utility, the reader should note however that pre-september 2009 the term DABS refers to its predecessor, DABM. 2 Including a review of short term critical investment options; a road map, outlining measures needed in the midterm and an electricity sector policy setting forth strategic sector reform issues. 1

12 1.2 Original Project Development Objectives (PDO) and Key Indicators (as approved) The PDO, as stated in the Grant Agreement, reads: The objective of the Project is to help provide reliable and quality power to the consumers in the target areas of the cities of Kabul, Aybak, and Mazar-e-Sharif. The following indicators were retained: PDO: no high level Project outcome indicator provided. Outcome 1 Increase the power supply in project areas in Kabul: total grid electricity supplied to Kabul increased to 482 GWh by March Intermediate Result indicator 1: strengthen the low voltage (LV) distribution network in parts of the capital city of Kabul supplied by Junction Station 2, as measured by the number of transformers installed and commissioned. Outcome 2 Providing grid power supply access to the people of Aybak: twenty five percent of population of Aybak with access to grid-based power. Intermediate Result indicator 2: construct a 220/20 kilovolt (kv) substation and associated 20 kv line to Aybak to provide quality, reliable and clean power to the people of Aybak town with the indicator being the completion of the Aybak substation. Outcome 3 -Increase in power supply to customers in Mazar e-sharif : total grid electricity supplied to Mazar e-sharif increased to 113 GWh by March ; Intermediate Result indicator 3: strengthen the low voltage (LV) distribution network in parts of the Mazar-e-Sharif City, as measured by the number of 20/0.4 kv transformers installed and commissioned. 1.3 Revised PDO (as approved by original approving authority) and Key Indicators, and reasons/justification Neither the PDO nor outcome indicators were modified. 1.4 Main Beneficiaries The underlying rationale for supporting the power sector was that improving access to power would bring both a relief to households and contribute to private sector development, and eventually to economic growth. After early donor support focused on restoring generation and transmission, improving distribution emerged as a priority to ensure that the new supply of power could reach consumers. The Project was primarily aimed at the people of Kabul, Aybak and Mazar-e-Sharif. 1.5 Original Components The Project consisted of the following four components: (i) Part 1 - Distribution System Rehabilitation for Kabul: through the rehabilitation and extension of the distribution system to connect medium voltage (MV) trunk lines and cables to customers in Kabul; and reorganizing 3 This is after the expected Closing Date at the time of Project approval. 2

13 (ii) (iii) (iv) the existing distribution system to optimize loading to reduce losses and provide satisfactory power supply voltage to customers in Kabul. Part 2 - Establishment of a 220 kilovolt (kv) Substation at Aybak and Interconnection with Distribution System: through the establishment of a new 220/20kV, 16 Megavolt Amperes (MVA) substation at Aybak and 20kV interconnection with the existing Aybak distribution system. Part 3 - Power System Rehabilitation for Mazar-e-Sharif: through the rehabilitation of the distribution system in Mazar-e-Sharif by upgrading the existing MV distribution network from 6kV to 20kV; connecting the distribution network to the 220/20kV substation; rehabilitating and replacing the low voltage distribution network; and upgrading the Mazar-e-Sharif substation capacity to meet the requirements of the distribution network. Part 4 - Institutional Capacity Building: through the provision of capacity building and implementation support to MEW and DABM including for: (a) operation and maintenance of the distribution systems; (b) implementation of infrastructure metering and accounting of energy supplied in Kabul, Aybak and Mazar-e-Sharif, (c) establishment, operation and hand-over to DABM of customer care centers; (d) communication of Project plans to consumers; (e) Project management support; and (f) other capacity building needs identified during implementation. 1.6 Revised Components No revision of Project activities was recorded. 1.7 Other significant changes The Closing date for the Project was extended three times: (i) by 21 months (to September 30, 2011) in November ; (ii) by 9 months (to June 30, 2012) in September 2011; and (iii) by 9 months (to March 31, 2013) in June Key Factors Affecting Implementation and Outcomes 2.1 Project Preparation, Design and Quality at Entry Choice of activities: Analytical work had identified the need to increase electricity access and quality of supply on the NEPS, focusing on urban centers to support growth and to avoid possible conflict with underserved populations. Donor support had been identified for projects in a number of cities along the route of the NEPS interconnector. More funding was needed in some of them to complete connection to the grid, in particular in Aybak, whose power supply relied exclusively on a diesel plant at the time, and Mazar-e- Sharif, the largest urban center supplied by the NEPS after Kabul. Further financing was also needed in Kabul, which, in addition to the new power from the NEPS interconnector, 4 A three year implementation period was expected but only two years agreed at project approval, pending extension of the parent Afghanistan Reconstruction Trust Fund. 3

14 was to benefit from an additional 100 MW of thermal generation financed by USAID. Preparation documents indicate that the decision to support the rehabilitation and expansion of distribution networks in Kabul, Aybak and Mazar-e-Sharif was made on the basis of extensive discussions with GoA and donors, to match ongoing investments in each of those cities. Analysis underlined the importance of moving the sector towards financial sustainability, including through a move to cost recovery tariffs. In keeping with the notion that structural reform was a long run agenda and that a first step should be to corporatize and improve commercial practices of the utility, it was decided to provide support complementary to the commercialization of DABM, including through the provision of metering equipment and development of customer care centers. Selection of physical investments was thus based on a sound analysis and a good understanding of government priorities. Capacity building: the Project design focused almost exclusively on the rehabilitation of infrastructure. Capacity building was only addressed through limited support to project implementation under Component 4 and by suggesting that supply and installation contracts should include a provision for operations and maintenance. The lack of capacity had, however, been recognized to be a challenge in 2006 and had been a contributory cause to the initial failure of EPRP. Balancing the support to infrastructure rehabilitation with support to capacity building seems to have been overlooked. Implementation arrangements were adopted wholesale from EPRP, following its restructuring in late Procurement strategy relied, again, on a small number of design, supply and install contracts, for speed and to minimize implementation complexity. Disbursements followed existing practices, with a high degree of centralization in the Treasury Department of the Ministry of Finance (MoF). The Memorandum of Understanding (MoU) signed between MEW and MoF in 2006 to address coordination problems under EPRP was amended to make it applicable to the Project. The duties of the external Project Management Firm (PMF) to assist in day-today implementation, contract management and supervision were extended to cover the Project. There appears to have been some attempt to define the respective roles of MEW and DABM, although coordination between them was not discussed. With the restructuring of EPRP so recently completed, implementation design benefited from the experience it had provided, while also offering some economies of scale. Risks and mitigation measures: risks were identified covering substance and process, with two, deterioration of security or reversal in political process and weak implementation capacity rated high and substantial respectively. Several other risks were rated moderate or low. Project design was adjusted to take into account the main risks by: building enough time and flexibility into the Project to address security risks; and by following infrastructure designs standardized by previous studies, using experienced consultants and establishing counterpart core teams for subcomponents. In the light of experience with EPRP and other Bank-financed projects in Afghanistan, the risk assessment may have been over-optimistic, especially with respect to the implementation period. Quality assessment: no QER was carried out for KAMPP. A Quality Assessment of the Lending Program review of the Project was carried out in April Project appraisal is not documented. 4

15 2.2 Implementation Three main supply and installation contracts were planned under KAMPP, namely for: (i) the rehabilitation of the Kabul distribution network; (ii) the rehabilitation of the Mazar-e- Sharif distribution network; and (iii) the construction of a new 220/20 kv substation at Aybak and augmentation of 220/20 kv substation at Mazar-e-Sharif. All three were concluded within six months from Project effectiveness. The main factors affecting project implementation, of varying degrees of predictability, have been reported as follows. Overall country context: Afghanistan remained a difficult environment in which to operate and the following are reported to have affected the Project: o Security: security concerns have affected implementation throughout including, for example, the need for demining in Aybak, persistently volatile security in Aybak and Mazar-e-Sharif, and attacks upon Indian nationals in 2008 in Kabul which resulted in Indian contractors recalling employees. There has been a marked deterioration in security in the project areas, especially in Aybak between 2008 and 2013, rendering it impractical to conduct site visits. o Resistance to change and vested interest: in a country where rebuilding an operating administration remains a challenge, implementing some activities at the local level sometimes proved difficult, for example implementation by local branches of DABM. Similarly, resistance to change was reported due to opposing local interest as well as interference by individuals and other government departments which are reported to have included demands for financial compensation to allow work to continue. 5 o Transportation and climate: poor roads and careless handling resulted in damage to some equipment. Winter conditions in Afghanistan make construction possible only during part of the year and harsher than normal conditions are reported over the life of the Project, resulting in delays in works. Institutional: Despite adapting project design to take into account the experience with EPRP, institutional factors still affected project implementation, including: o Poor government coordination: including: (i) lengthy approval of LCs; (ii) lengthy approval of customs clearance which resulted in delays in bringing equipment on site and sometimes in the destruction of goods; (iii) budgeting issue and availability of funds including a haitus in funding at the start of each fiscal year; (iv) difficulties in securing and retaining rights of way; (v) poor coordination with municipalities on their plans and insufficient consultation with local stakeholders and (vi) poor coordination between DABS and MEW after DABS became autonomous from MEW. This became particularly acute as contractors reached the key milestones of completion, commissioning and handover. The contractor was bound to MEW, but MEW was informally 5 Although no formal complaint is reported in this respect. 5

16 bound to accept assets only with DABS s agreement which was often not forthcoming. o Weak implementation capacity: underlying a number of those problems is the overarching capacity deficit in post-conflict Afghanistan. Country systems for disbursements and auditing are centralized in MoF to make up for the lack of sufficient qualified personnel, but this creates bottlenecks and limited incentives to develop capacity within MEW. The use of a PMF could not eliminate all the challenges. Under the PMF each problem was solved on an ad-hoc basis which did not resolve the underlying problems. The performance of the PMF deteriorated significantly from about 2010, and its contract was allowed to lapse which led to a six-month hiatus during which project progress was negligible, until a new PMF was hired. Technical issues: KAMPP has been confronted with technical problems, affecting the quality of goods and installation: o Goods: problems are reported with respect to the quality or appropriateness of goods. In particular availability, quality, and installation of poles used in distribution networks hampered the progress of work by all contractors. While this was addressed through increased supervision by the authorities, shortages of poles meeting appropriate standards were also later reported. Transformers supplied under the Kabul distribution rehabilitation component experienced early failures which were initially attributed to overloading but which later were found to be the result of defective manufacture. The meters procured under the fourth component were found to be unfit for purpose, due both to their dimensions and to their high level of sophistication. o Contractors: quality issues are also reported with some contractors including the civil works to construct a sub-station in Aybak. Although technical specifications had to be met for recruitment, there is general agreement that the overall Afghan environment remains unattractive for most qualified bidders, resulting in less qualified contractors and suppliers bidding. Examples are reported of contractors without adequate experience or showing poor procurement planning. Low technical capacity by MEW and DABS may also have contributed to this situation. o Design change and time overruns: A change to the scope of work under the Mazar-e-Sharif distribution component is reported when the authorities asked that the MV and LV system be installed underground, thus significantly increasing costs while also requiring a change to the materials used. Funds under the Project were not sufficient to cover this change and GoA had in part to commit to provide the additional goods needed from separate sources, and DABS was assigned to do the work which was also a source of delay. o Operations and maintenance: starting in 2012, supervision documents report issues in connection with the operation and maintenance of rehabilitated infrastructure, caused in part by the protracted handover from MEW to DABS, raising serious concerns over the sustainability of those assets, in particular the distribution system installed in Kabul. 6

17 2.3 Monitoring and Evaluation (M&E) Design, Implementation and Utilization Choice of indicators: no high level outcome indicator was stated for the Project. One outcome indicator was defined for each of the three components covering infrastructure rehabilitation, but not for the institutional capacity building component. The outcome indicators appear to have three shortcomings. First, they were not stated clearly or consistently across the three components although they each had the same broad objective of providing or increasing grid power supply; for Kabul and Mazar-e-Sharif they were stated in terms of an increase in the quantity of electricity supplied, while for Aybak in terms of the proportion of Aybak s population with access to grid-based electricity. Second, the indicator measured an increase in the aggregate supply of electricity while the PDO was to provide reliable and quality power. The causal link between an increase in the amount of power and the quality and reliability of that power is by no means straightforward. Better distribution is not the only factor affecting quality and reliability. Third, other activities in the sector were being financed from other sources in Kabul and Mazar-e-Sharif. While acknowledged, they were not taken into consideration, thus raising attribution issues. Quality of baseline and monitoring data appears to be questionable; the data source is unknown and may have underestimated the start-of-project baseline. It may also have conflated generation, which has been systematically captured since 2006, with consumption which has not, thus ignoring losses. For example the end of project indicator for Kabul was the supply of 482 GWh by March 2010, yet according to published data in Afghanistan Energy Information Center (AEIC), generation in Kabul was 649 GWh in In Aybak, the goal was to reach 25% of the population with grid power as opposed to diesel. In 2007, DABS records indicate that there were 4,646 residential customers in Aybak, equivalent to about 23-46,000 people. Yet the population of Aybak city, according to the Central Statistics Office, was just 26,400 in 2010 implying that the substantial majority of households in Aybak were connected. In any event, all those connected to the Aybak distribution system, whether within the city proper or in the surrounding area, would benefit from the change once the substation was complete, thus making any headcount of population or connections irrelevant. Use of monitoring data is documented throughout the project lifetime, although sometimes infrequently, especially in 2010 and 2011 when there was no resident TTL. The first data for the Kabul and Mazar-e-Sharif components were provided by the end of 2009 and were used to monitor project progress, although a little inconsistently. The Aybak indicator was not useful for monitoring progress in the substation, although completion of civil works, electrical installation and energization were used as proxies. 2.4 Safeguard and Fiduciary Compliance Safeguards The Project was rated Category B and not expected to raise significant environmental concerns. No Integrated Safeguards Data Sheet (ISDS) was prepared. The Environmental and Social Safeguards Framework (ESSF) prepared for EPRP was used without change. The ESSF was based on the following principles: (i) recognition that detailed designs of activities may not be known at appraisal; and (ii) requirement that all proposed 7

18 components be screened to ensure that environmental and social risks can be adequately addressed. Two potential safeguards issues were identified: presence of polychlorinated biphenyls (PCBs) in transformer oil and land acquisition for the substation in Aybak. While the ESSF provided for arrangements for land acquisition, it was agreed that PCB testing would be conducted and appropriate handling procedures subsequently prepared if necessary. Efforts to strengthen capacity were pursued during the early implementation stage through the PMF. Streamlined Environmental and Social Safeguards (ESS) to be used by contractors were also prepared. A safeguards focal point was selected and regular monitoring is reported with the support of PMF. Significant training was also provided to improve the government s and contractors capacity to implement safeguards, with materials translated in local languages. During the implementation stage, safeguards issues arose as follows: Mine clearance, which was handled by Mine Action Center for Afghanistan (MACA) as per the ESSF; Rights of way, which was handled in compliance with the provisions of the ESSF and the ESS, including consultation with local community along the proposed routes at preparation and implementation stages. A complaint handling committee was also established at each Project site. No land acquisition was necessary under the project; PCB testing: testing took place but did not reveal the presence of PCBs. Safeguards supervision is regularly documented through 2009, but less information was made available from 2010 through 2013 and supervision aide memoires did not include a specific section on safeguards. This may have been because few safeguard issues were at stake under the Project by this stage but changes in TTL with an interim period in 2010, and departure of the PMF in 2011, which was only replaced in 2012, may also have affected safeguards supervision. Financial Management Some financial management concerns were reported, including lack of internal control mechanisms and late submission of the external audit for SY1391 ( ). Auditing procedures are handled at the central level through the Control and Audit Office and so the delay was not considered a Project-specific shortcoming. There was some delay in responding to the auditor s recommendations following the SY1391 audit, but these were resolved before final project closing. The Project was consistently rated in the satisfactory range. Procurement Very few contracts were to be procured under KAMPP and they all were awarded promptly after effectiveness. The project did not experience procurement problems although institutional and capacity constraints affected contract management such as in connection with LCs and customs clearance. 8

19 2.5 Post-completion Operation/Next Phase The sub-station in Aybak was energized and in operation before Project close but some civil works and commissioning were not completed. Concerns also exist over the sustainability of the infrastructure rehabilitated under the two other components, due to a mixture of poor operation and poor quality of the equipment installed. Contractual remedies including liquidated damages, cashing of performance bonds and extended warranty periods are being employed to ensure both the completion of the Aybak substation and the correction of faults in the Kabul and Mazar-e-Sharif distribution systems. These are complex tasks and the PMF whose contract is now funded under a separate project (Afghanistan Power System Development Project, TF , APSDP) has been tasked to assist the Government on this front. A new power project is under preparation, expected for delivery in FY15 that would aim at developing DABS s capacity for planning, operation and maintenance of electricity distribution infrastructure. It is part of a longer term plan to shift investment activities to DABS and seeks to address the need to mitigate the risk of dilapidation of infrastructure. 3. Assessment of Outcomes 3.1 Relevance of Objectives, Design and Implementation Afghanistan remains in the bottom 10 percent globally in electricity consumption per capita (~ 100 kwh per year) and only percent of its population is connected to the grid, giving it one of the lowest rates of electrification in the world. The Afghanistan National Development Strategy (ANDS) notes that Energy is [Afghanistan s] top economic development priority, and [its] economy s hope, singling out electricity as the motor that powers [Afghanistan s] growth. The National Energy Supply Program (NESP), a national priority program developed in the context of setting out a strategy of support for Afghanistan post the 2014 transition, further demonstrates the priority both country and donors place on support for the power sector. The Bank s FY12-14 Interim Strategy Note also recognizes the need to continue to address the energy needs of businesses and people, and to support cross-border trading of energy. In light of the rapidly evolving demand for power in Kabul, and the importance of Mazar-e-Sharif as a commercial and industrial center, and Aybak as a provincial city, the choice was sound. Hence the objective was and remains relevant. Project design was simple and informed by experience with EPRP, in particular taking account of country conditions. Implementation started well and demonstrates the solid design, but the Project soon became bogged down as a result of deteriorating security, weak coordination in government, poor contractor performance, including of the PMF and limited oversight by MEW and the Bank. This suggests weaknesses in the implementation support plan, and coincided with both the absence of a dedicated TTL in country and relatively low budget spent on supervision. 3.2 Achievement of Project Development Objectives Assessing the achievement of the PDO is difficult for several reasons. First, no high level outcome indicator was defined. Second, uncertainties about the quality of the baseline and, to a lesser extent, the end-of-project data for Kabul and the relevance of the 9

20 indicator for Aybak cloud evaluation. In Mazar-e-Sharif, accounting for 40 percent of project cost, it can be said with reasonable certainty that the component indicator was achieved. In Kabul, accounting for a further 36 percent of cost, the increase in power supply between from 649 GWh in 2007 to 1,219 GWh in 2010 is a larger absolute increase than can have been envisaged when the indicator of 482 GWh for March 2010 was set during preparation. Hence it might also be argued that the Kabul objective for quantity of power was reached. Third, additional quantities of power do not necessarily demonstrate better quality or reliability. There is some limited evidence to suggest that better quality electricity is now being more reliably supplied. Evidence drawn from bidding documents and reports at the time suggest that in 2007 power was rationed in Kabul with households receiving 4 hours supply every other day. With the completion of the NEPS interconnector, most of Kabul now receives power 24 hours a day 6. Attribution is an issue as many donors contributed to the completion of the NEPS interconnector, but the Project may claim some credit, since the rehabilitation of the LV distribution system was needed to complete the last mile to consumers. Another proxy for improved quality is the peak load that can be served per household, since it implies that there will be fewer voltage excursions while also making power available at the time needed by the consumer. Empirical evidence from 2006 to 2011 shows a steady increase in both the number of households served and the peak load served in Kabul after 2008 when the NEPS interconnector was completed, as shown in Figure 1. This suggests a likely improvement in the quality of supply because brownouts and other quality problems are normally symptoms of overloading, which has been significantly reduced. Figure 1: Peak Load and Households Served in Kabul 400, Number of Households and Peak Load (kw) 350, , , , , ,000 50, Residential Customers (Left axis) Peak Load (Left axis) kw Per Household (Right Axis) Peak Load per Household (kw) 6 Some areas of greater Kabul have yet to be connected to the grid. 10

21 In Mazar-e-Sharif, there is some limited evidence that suggests that the upgrading of the MV distribution system from 6kV to 20kV has improved the quality of power delivered to consumers, as shown in Figure 2. The evidence is limited to a survey of five houses at evening peak and is thus not statistically robust. Figure 2: Delivered Voltage to Households in Mazar Served at 6kV and 20kV D eli ve re d V ol ta Proper 220 V Line The Aybak indicator must be disregarded, but the qualitative change from diesel to grid supply has been achieved with the energization of the substation, and with this an approximately 15-fold increase in the amount of power available for distribution. Again, quantity is not a direct indicator of quality but it the same arguments adduced for Kabul and Mazar will apply in principle. Hence there is some evidence that the PDO has been achieved, but it is circumstantial and uncertain. 3.3 Efficiency For economic evaluation purposes, the project is divided into three subprojects: the Kabul distribution rehabilitation; the Aybak substation; and the Mazar substation extension and network rehabilitation. Between them, the three subprojects accounted for 95 percent of project costs. For evaluation of the Project, only the Aybak and Mazar subprojects are considered. The project financing for the Kabul subproject was for two contracts, for LV systems and meters, with a total value of $22.9 million, or about 30 percent of total costs, with the balance financed by EPRP. To avoid risk of double counting benefits, the results from the Kabul subproject have been excluded from the evaluation of the Project, which means the efficiency assessment is based on about 56 percent of total project financing. Table 1 gives details of the economic benefits from each of the three subprojects and further details of the analysis of the two considered in the Project are in 11

22 Annex 3. The analysis for the Kabul transmission, substation rehabilitation project is included in the ICR for EPRP. 7 Table 1: Economic Costs and Benefits of KAMPP Subprojects Sub-Project Cost (million $) Estimated NPV ( million $) IRR (%) Kabul Transmission, Substation and Distribution Rehabilitation 8 Aybak Substation % Mazar-e-Sharif Substation and Distribution Rehabilitation % Based on this analysis, the Aybak subproject was economically efficient, but the Mazare-Sharif subpoject was not. Combining the net present value of these two subprojects gives a negative net present value for the project. [Some element of economic efficiency will have been lost as a result of the delays in completion of the subprojects by about 39 months]. 3.4 Justification of Overall Outcome Rating Rating: Unsatisfactory The project was relevant and well designed. While the outcome indicator for one of the three subprojects has been met, it is hard to tell whether they have been for the other two, and in any case are of questionable value in demonstrating the achievement of the objective. There is circumstantial evidence sufficient to show that the quality and reliability of supply has been improved in Kabul but not in either Mazar-e-Sharif or Aybak. Efficiency was poor for the two subprojects entirely financed by the project. There are thus major shortcomings in the achievement of the objective and in its efficiency. 3.5 Overarching Themes, Other Outcomes and Impacts (a) Poverty Impacts, Gender Aspects, and Social Development No specific poverty or social impact study has been carried out. Per capita electricity consumption and economic growth are 90% correlated suggesting access to electricity is not only critical for improving livelihoods but also allow the private sector to grow, implying the project would eventually have a poverty reduction impact. (b) Institutional Change/Strengthening The project did support some institutional change to support the transition of operation of the power sector by MEW to DABS, but it was limited in extent. The project may have heightened awareness of the need for improved capacity in both MEW and DABS. 7 Report No. ICR435 8 Included for reference only 12

23 (c) Other Unintended Outcomes and Impacts (positive or negative) None. 3.6 Summary of Findings of Beneficiary Survey and/or Stakeholder Workshops Not applicable as this is a core ICR. 4. Assessment of Risk to Development Outcome Rating: High Development outcomes have been limited, and they are at risk from three factors. First, poor quality of some of the goods and installation work may curtail its useful life. Second, overloading and misuse, the result of poor training, weak technical capacity and inadequate operational and maintenance may result in damage to the assets created. Third, country factors such as insecurity, political instability and weak governance also represent risks that could directly affect the achievements of the Project. 5. Assessment of Bank and Borrower Performance 5.1 Bank Performance (a) Bank Performance in Ensuring Quality at Entry Rating: Unsatisfactory Activities chosen for financing by the Project were in line with available analysis and country priorities. They were discussed with GoA and underwent a systematic internal review process. The QALP assessment carried out in 2010 rated design to be satisfactory, taking into account the simple design and choice of implementing arrangements. There was no QER and appraisal was not formally documented. The Project documents seem to suggest that it was a spin off from the larger and more well-established EPRP, aimed at filling gaps it had not been able to finance following its restructuring. Judging from the budget expenditure during preparation (Annex 4) the Project seems to have been prepared on a shoestring. It is possible that supervision and restructuring of EPRP may have served many of the purposes of project preparation, but without documentation this cannot be assessed. The division of the LV and the MV contracts for the Kabul distribution rehabilitation across two projects was a significant failing and provides the strongest evidence that the project was never thought of as a discrete entity separate from EPRP. The risk presented by weak capacity and institutional rivalry between MEW and DABS, were not mitigated, and if any capacity was built it was at MEW which is not the body which will be responsible for future investment or operations and maintenance. The balance of support between infrastructure rehabilitation and capacity building, in particular contract management and supervision, should have been debated as should the means by which operations and maintenance were to be carried out. Monitoring and 13

24 evaluation design seems to have been weak with no arrangement for collection of data put in place. (b) Quality of Supervision Rating: Moderately Unsatisfactory Supervision is well documented throughout the life of the Project. There is a sense of disconnect between the realities of KAMPP and the willingness to address its specific problems. Quality, contract management and government coordination problems continued to arise and as soon as one was solved, another seemed to appear, necessiting several extensions. Despite this, and the questions on the Project s sustainability, satisfactory ratings for achievement of the PDO extended through Both the 2010 and 2011 requests for an extension of the Closing Date displayed confidence that the PDO remained achievable. This appears to show a want of candor in supervision. There seems to have been insufficient attention to rigorous monitoring of project progress, and may have been caused by a mixture of: security issues that made visits difficult; overreliance on the PMF, the quality of whose work started to deteriorate in about 2010 and the interruption of whose mandate further increased delay and left quality issues unaddressed; and a global handling of all three power projects in the portfolio in which the Project represented just a fraction of the many issues that affected all of them. Some of the values for indicators were not completed or updated over several supervision periods and using one of the many restructurings to review the results framework and update it was a missed opportunity. The closing of the Project left a number of technical issues unresolved, in particular defects at the Aybak substation and the Kabul distribution network. Legal remedies including liquidated damages, performance bonds and extended warranty periods may be available to address faults and MEW is applying these as each contract is closed, with the support of the PMF, the services of which continue, financed by another project. Bank supervision showed a number of significant shortcomings that could have been avoided and which, if addressed, may have been able to compensate for some of the weaknesses in design. (c) Justification of Rating for Overall Bank Performance Rating: Unsatisfactory As one dimension of Bank performance is rated Unsatisfactory and one Moderately Unsatisfactory, overall rating is Unsatisfactory. 5.2 Borrower Performance (a) Government Performance Rating: Unsatisfactory Continuing coordination issues among Government stakeholders, first experienced with EPRP, between ministries and with DABS, from central level to local level, from fiduciary to technical issues, was one of the main impediments to implementation. To a 14

25 large extent, this is a result of Afghanistan s post-conflict situation and an illustration that rebuilding an operative, capable public administration remains one of the country s greatest challenges. The decision to rely on country systems for disbursement and financial management, while justified in terms of capacity building, may also have contributed to the problems encountered. It should be underlined that this rating is not geared towards one institution in particular, as the various authorities involved in the Project, in particular MEW and MoF are said to have always shown, when solicited, responsiveness and willingness to sort problems out. Seeking speedy, coordinated action from various public stakeholders required an almost day-to-day involvement from those involved in Project implementation. The project took twice as long as originally planned, and as such was a major shortcoming warranting an Unsatisfactory rating. (b) Implementing Agency or Agencies Performance Rating: Moderately Unsatisfactory The lack of capacity at MEW significantly impacted the Project and timely implementation continued to be an issue throughout. Weak contract management was probably the single most important constraint, and the inability to address technical issues in contracts, especially relating to equipment quality, for example poles and distribution transformers, caused serious delays. Some positive aspects were reported over the life of the Project, in the fields of financial management and procurement coordination between MEW and DABS remains problematic. Several of these shortcomings are significant and indicate a Moderately Unsatisfactory rating. (c) Justification of Rating for Overall Borrower Performance Rating: Unsatisfactory As one dimension of Borrower performance is rated Unsatisfactory and one Moderately Unsatisfactory, overall rating is Unsatisfactory. 6. Lessons Learned Based on this evaluation, five lessons present themselves: Even when conceived as a follow up to an existing operation, any new project must be properly assessed as such It is hard to avoid the impression that the project was seen as a means of adding funds to EPRP for rehabilitation investments, based on needs identified, feasibility and implementation arrangements prepared earlier. Yet a rigorous appraisal of the project on its own merits might have revealed its design shortcomings. By this association, the lessons from EPRP, set out in the ICR also apply to the Project. 9 Capacity is often the main issue and is as important as the infrastructure itself As is common for infrastructure projects in post-conflict environments, priority was given to physical rehabilitation. While this may have been justified in terms of country and donor dialogue, it was also recognized at the time that this effort was 9 Ibid. 15

26 somewhat marginal given the immense needs of the country. Infrastructure has been rehabilitated but at risk of falling into the build-neglect-rebuild cycle due to lack of capacity for operation and maintenance. The proposal to provide for an obligation of operation and maintenance of the infrastructure rehabilitated under the contracts did not prove effective because contractors demobilized and enforcement of legal remedies to complete works or correct defects now needs to be sought. Power projects call for a precise definition of indicators, in particular when the sector is also supported by other donors Indicators chosen for two of the three rehabilitation components under KAMPP related to the quantity of power supplied and proved of little use to measure the quality and reliability goal of the project. Data adequacy was also a problem, and an early restructuring might have been an effective way to allow the approach to monitoring to be rethought. In countries with a history of conflict and poor capacity, consideration should be given to relative or incremental indicators, for example the number of households connected and metered, and the PDO framed around these simpler and more easily measured metrics. On site supervision is necessary It is unclear the extent to which infrequent supervision may have contributed to the quality issues encountered but more site visits could at least have allowed for a swifter, more informed reaction by the Bank. Security conditions in Afghanistan make field visits difficult and sometimes impossible to arrange. Some flexibility in design is essential, including the early identification of alternative activities to finance. Similarly, the inability to travel on site should be considered at inception, including by exploring alternative options through modern technology. In the end, though, there is no substitute for seeing for oneself. Corporatization and commercialization of utilities is a multifaceted reform and has significant institutional implications Poor coordination between MEW and DABS is a serious challenge to the smooth development and operation of the sector. It is, therefore, important that those issues be considered together at project design, especially if it is known that further sector development is planned during the project s lifetime. At the time the Project was prepared, the corporatization of DABS had not yet been completed and thus MEW continued to own both the project and the assets created by the project as well as have authority over DABS. Many of the coordination problems encountered towards the end of the Project could have been avoided had responsibilities for project implementation been considered sooner. 7. Comments on Issues Raised by Grantee/Implementing Agencies/Donors (a) Grantee/Implementing agencies MEW noted that the main challenge it faced during project implementation was the evolution of its roles and those of DABS. In the early part of the project DABM was a department within MEW. Upon corporatization, DABS became autonomous and hence MEW s authority was reduced but its implementation responsibilities were not. MEW recognizes that its capacity is an area in need of continuous improvement, and was disappointed to note that its performance had been rated Unsatisfactory, though it agrees 16

27 with the reasons given for this rating. MEW also considers that the project was impacted by the lack of an in-country TTL from 2010 to Country specific factors are also noted as affecting the ability to implement the project satisfactorily, most notably the post-conflict situation, the physical situation and the geographical location. It also points to the weaknesses of contractors affecting the timeliness of implementation. (b) Cofinanciers/Donors Not applicable. (c) Other partners and stakeholders Consultations were carried out with ADB, KfW and USAID as part of the preparation of this ICR. All donors confirmed the challenges of operating in the Afghan environment on the same issues that KAMPP was confronted with: volatile security; coordination among institutions (including MEW and DABS); weak planning, implementation and project management capacity (the use of PIUs offering only a limited solution); and quality of contractors. ADB, which has the largest power portfolio, confirmed facing significant delays as a result. Overall, there is recognition that power remains an issue of primary importance for the development of the country with efforts needed on generation, transmission and distribution. While Kabul is now relatively well served compared with the rest of the country, in part thanks to the combined efforts over the past decade (including the project), there is also a sense that continued efforts will be required in the Kabul subproject, in particular for distribution. The three following challenge emerge as priorities: Training: need to focus on capacity building in operation, maintenance and supervision with more efforts needed on design (availability of translated materials, training with clear deliverables; proper monitoring and follow-up); Institutional reform: need for clarification of mandates, in particular between MEW and DABS as a normal evolution should be for the power utility to be in charge of managing infrastructure investments and operation (which, in turn, may mean channeling external assistance through DABS); and Coordination among donors: whereas dialogue and exchange of information are said to have been good, more coordination is needed. This applies in particular to the planning of assistance (ensure coherence of interventions) and institutional arrangements (use of DABS, role of MEW and of other institutions, e.g. the Environmental Agency). The preparation of the new Power Sector Master Plan is considered by all as a positive step for improved coordination. 17

28 Annex 1. Project Costs and Financing (a) Project Cost by Component (in USD Million equivalent) Components Appraisal Estimate (USD millions) Actual/Latest Estimate (USD millions) Percentage of Appraisal Distribution System Rehabilitation for Kabul Establishment of a 220kV Substation at Aybak Power System Rehabilitation for Mazar-e-Sharif Institutional Capacity Building Total Baseline Cost Physical and Price Contingencies Total Project Costs Total Financing Required (b) Financing Source of Funds Type of Cofinancing Appraisal Estimate (USD millions) Actual/Latest Estimate Percentage of (USD Appraisal millions) 10 Afghanistan Reconstruction Trust Fund IDA (from existing Cr. 3933) Joint As of September 28,

29 Annex 2. Outputs by Component 1) Outputs based on Outcome Indicators Outcome Indicator(s) Achievements by close of Project Outcome 1 Increase the power supplytotal grid electricity supplied to Kabul Target met by 253%, with 1,219 GWh in project areas in Kabul. increased to 482 GWh by March attained by Outcome 2 Providing grid powertwenty five percent of population of Target met - substation completed; all supply access to the people of Aybak. Aybak with access to grid-based power consumers on Aybak distribution system now connected to the NEPS interconnector Outcome 3 - Increase in power supplytotal grid electricity supplied to Target met by 174%, with 197 GWh to customers in Mazar e-sharif. Mazar e Sharif increased to 113 attained in GWh. 2) Outputs per Project component Component Achievements by the Close of the Project Part 1: Distribution System Rehabilitation for Kabul The work has been substantially completed with completion and operational acceptance certificates issued in Since then, however, a series of transformer and other equipment failures have occurred. Those issues are being addressed by corrective actions by MEW or DABS and through legal remedies/actions against the contractor. Part 2 : Establishment of a 220 kilovolt (kv) Substation at Aybak and Interconnection with Distribution System The Aybak substation is reported to have been commissioned and energized by the contractor without involvement of MEW and there have been several equipment failures. In addition, several essential civil works including retaining walls, drainage and internal access roads have not been completed which put at risk the stability of the substation. Those issues will need to be addressed by corrective actions by MEW or DABS and through legal remedies/actions against the contractor. Part 3 - Power System The Mazar substation work was substantially completed in 2011 and is in operation. Rehabilitation for Mazar-e-Sharif Major problems were later reported and MEW has directed that the commissioning be redone. Distribution networks have also been rehabilitated; however, additional materials (made necessary by a design change) still need to be handed over by MEW to DABS. In addition, cracks appeared on some poles and 519 of them may need to be replaced which, again, is a warranty issue to be dealt with by the contractor. Part 4 - Institutional Capacity Building All the single phase and three phase meters have been shipped and received by the contractors. Also, meter reading devices, computer base stations and testing equipment have all been handed over to DABS in Kabul and Mazar-e-Sharif. Necessary training has been conducted in Kabul and Mazar-e-Sharif for the DABS staff. 19

30 Annex 3. Economic and Financial Analysis For economic evaluation purposes, the project is divided into three subprojects: the Kabul distribution rehabilitation; the Aybak substation; and the Mazar substation extension and network rehabilitation. Between them, the three subprojects accounted for 95 percent of project costs. For evaluation of the Project, only the Aybak and Mazar subprojects are considered. The financing for the Kabul subproject was for two contracts, for LV systems and meters, with a total value of $22.9 million, or about 30 percent of total costs, with the balance financed by EPRP. To avoid the risk of double counting benefits, the results from the Kabul subproject have been excluded from the evaluation of the Project, which means the efficiency assessment is based on about 56 percent of total project financing. Table 3.1 gives details of the economic benefits from each of the three subprojects and further details of the analysis of the two considered in the Project are in Annex 3. The analysis for the Kabul transmission, substation rehabilitation project is included in the ICR for EPRP. 11 Table 3.1: Economic Costs and Benefits of KAMPP Subprojects Sub-Project Kabul Transmission, Substation and Distribution Rehabilitation Cost (million $) Estimated NPV ( million $) IRR (%) Aybak Substation % Mazar-e-Sharif Substation and Distribution Rehabilitation % * The Aybak and Mazar-e-Sharif substation work was contracted together so it is not possible to precisely match actual disbursements with Aybak or Mazar-e-Sharif substation specific line item costs. Aybak Substation The benefits of Aybak Substation sub-project are estimated based on the avoided costs of the next-least-cost alternative to achieve the subproject objective. The objective of the subproject was to provide grid-connected power to the residents of Aybak. Before the project, Aybak received power from a local distribution system connected to a 1,100 kva diesel generator that connected the generator to approximately 3,500 households. Thus the counterfactual is simple to continue to run the existing diesel generator. The following assumptions are relied upon to conduct an economic benefit analysis. (a) The existing diesel generator has a heat rate of 14,000 Btu/kWh, an efficiency of 24%, which amounts to diesel fuel consumption of 375 liters per MWh; (b) The cost of diesel is 60 Afs per liter ($1.09 at the current exchange rate of 55 Afs/USD). This is the current cost and is assumed to hold constant in real terms; (c) The resulting fuel cost of generation is 40 cents per kwh. The capacity factor of the plant is 80%. It is high because the average capacity per household is just 250 Watts, so the load factor (average/peak) is quite high. 11 Ibid 20

31 (d) The cost of grid power is 6 cents per kwh. This is the cost of the imported power from Uzbekistan delivered by the substation. Although this does not include the full cost of delivering the power, it is approximately the weighted average residential tariff, so it is the price of electricity if not the full cost. Based on these assumptions, the results of the economic benefit of the project is computed to have a net present value $6.7 million at a discount rate of 12%, and an internal rate of return of 25%. More details are presented in Table 3.2. Table 3.2: Economic Benefits of Aybak Substation Item Without Project With Project Cost of Project ($) 8,146,428 Available Capacity (MW) Households Connected 3,500 3,500 Household Peak Load Served (KW) Price of Power (Cents/kWh) 40 6 Annual Generation (MWh) 6,167 6,167 Annual Cost of Generation ($) 2,466, ,033 Net Present Value, $ (@12%) 6,710,226 IRR 25% This is the most conservative analysis possible. This analysis is simple as it compares the cost of the project to the avoided cost of the least-cost alternative which was continuing to run the existing diesel generator. The project however not only replaced the need for this 100 kva (about 0.88 MW) of diesel generation; it provided a substation and transformer with a capacity of 12.8 MW, which can deliver nearly 15 times the power of the diesel. Given the existing distribution system, this allowed the average peak capacity per household to increase from 0.25 kw to 3.66 kw, which is more than twice the average capacity per customer in Kabul of 1.5 kw. There was no formal benefit analysis done at appraisal. Mazar-e-Sharif Substation and Distribution Rehabilitation 12 The primary quantifiable impact of the project was increasing the capacity of the Mazar substation from 32 MVA (26.6 MW) to 82 MVA (65.6 MW), an increase of over 250%. This allowed DABS to increase the number of customers by nearly 50% from 2008 to 2010 and still nearly double the peak capacity available to each customer from 0.56 kw 12 The total cost for the sub-project was $23.7 million. Since the Mazar sub-project included one complete contract (MEW/S-502) plus only components of two others (MEW/S-503 and MEW/S-508) that also included the substation in Aybak and meters for Kabul, it is not possible to precisely match actual disbursements with Mazar s specific line item costs. However, for MEW/S-503, the Mazar portion of the contract value is taken to be effectively the same as the actual project cost. For MEW/S-508, the actual disbursed amount ($2.8 million) is pro-rated by the share of the meters that were supplied to Mazar (10,000 of 50,000, or 20%). 21

32 to 0.98 kw over the same period. The impact of the project on DABS customers in Mazar-e-Sharif can be seen in Table 3.3 Table 3.3: DABS Customers in Mazar-e-Sharif Change Type Number Percentage Residential 45,531 66,779 21,248 47% Commercial 3,131 4,901 1,770 57% Total 48,662 71,680 23,018 47% Given the significant unmet demand, this should have allowed a corresponding increase in consumption. However, because the project didn t rehabilitate the low voltage (LV) distribution system, install any meters, or connect any additional customers, the actual project benefits were far more modest. The benefits of the project are estimated based on the avoided costs of the next least cost alternative to achieve the project s objective. The objective of the project was to improve the quality and reliability of grid-connected power in Mazar-e-Sharif. Before the project, the substation was insufficient to meet local demand and the distribution system was unreliable. 13 Mazar-e-Sharif substation received power from a 110 kv line from Uzbekistan and had two 16 MVA transformers. The project with enabled an increase in consumption that the expanded substation and partially-rehabilitated distribution system allowed. From 2008, before the project started, to 2010, when the project was substantially complete, consumption in Mazar increased from 177 GWh to 197 GWh, an 11% increase over two years. Crediting the project with this benefit, and assuming that it continued each year over a 20 year payback period, the estimate of the project s economic benefit, in this case the negative NPV are not marginal as summarized in Table As stated in the MEW/S-502 contract, the condition of the Mazar distribution system was described as follows: The reliability of the electricity supply is at a very low level due to load shedding as well as due to the frequent failures of the distribution network. The existing distribution system is also constrained by inadequate capacity of the network components such as power lines, cables, transformers etc. This has caused excessive voltage drops in most parts of the network causing severe difficulties to the consumers. It has also prevented new customers from being added to the network who are waiting for supply connection over a long period of time. The present condition of the network also poses a serious safety hazard to the general public, as well as to the operational (DABM) staff due to exposed or unprotected live network components, inadequate clearances etc. 22

33 Table 3.4: Summary of Benefits of Mazar Substation and Distribution Sub-Project Without Project With Project Cost of Project ($) 23,664,328 Available Capacity (MW) Households Connected 45,531 66,779 Household Peak Load Served (KW) Price of Power (Cents/KWh) 6 6 Annual Consumption (MWh) 177, ,283 Theoretical Capacity (MWh) 177, ,224 Annual Cost of Generation ($) 106,534,480 11,836,980 Net Present Value, $ (@12%) (13,115,886) IRR 0.15% The reasons for the negative NPV and marginally positive IRR can be attributed to the following: (a) The distribution system rehabilitation represented over 85% ($20.5 million) of the subproject cost, but since much of the LV work was left unfinished, the project failed to deliver the benefits of the increased substation capacity to customers. (b) Meter installation and actual customer connection were not part of the project scope. Therefore, the benefits of additional customers and increased access to electricity remain largely unrealized by the project. (c) The project includes the cost of meters, although no meters were actually installed as part of the project. The meters were delivered, but they did not fit in the meter boxes supplied. For the relatively minor marginal cost of properly completing the LV system, installing meters, and making customer connections, this project could have offered the people of Mazar-e-Sharif a much more significant benefit for the investment. 23

34 Annex 4. Grant Preparation and Implementation Support/Supervision Processes (a) Task Team members Names Title Unit Lending/Grant Preparation Nagaraju Duthaluri Lead Procurement Specialist ECSO2 Minerva S. Espinosa-Apurada Program Assistant SASDO Deepal Fernando Senior Procurement Specialist ECSO2 Abdul Wali Ibrahimi Operations Officer SASDU Asila Warkdak Jamal Consultant SASDE Sunil Kumar Khosla Lead Energy Specialist ECSEG Asha Narayan Sr Financial Management Specia SARFM Kenneth O. Okpara Sr Financial Management Specia SARFM Asta Olesen Senior Social Development Spec SASDS Mohammad Arif Rasuli Senior Environmental Specialis SASDI Responsibility/ Specialty Supervision/ICR Toufiq Ahmed Procurement Specialist SARPS Mohammad Anis Energy Specialist SASDE Abdul Mohammad Durani Social Development Specialist SASDS Nagaraju Duthaluri Lead Procurement Specialist ECSO2 Minerva S. Espinosa-Apurada Program Assistant SASDO Deepal Fernando Senior Procurement Specialist ECSO2 Julia M. Fraser Sector Manager EASTS Mariam Haidary Program Assistant SASHD Ahmad Rafi Hotofat Team Assistant SASDO Abdul Wali Ibrahimi Operations Officer SASDU Asila Wardak Jamal Consultant SASDI Sunil Kumar Khosla Lead Energy Specialist ECSEG Arun Kumar Kolsur Senior Procurement Specialist SARPS Asha Narayan Sr Financial Management Specia SARFM Mohammad Yasin Noori Social Development Specialist SASDS Wahida Obaidy Team Assistant SASDO Kenneth O. Okpara Sr Financial Management Specia SARFM Asta Olesen Senior Social Development Spec SASDS Abdul Hameed Quraishi Operations Officer SASDE Mohammad Arif Rasuli Senior Environmental Specialis SASDI Mohammed Edreess Sahak Program Assistant LCC3C Richard Spencer Country Sector Coordinator SASDE Gilles Veuillot Consultant SASDE Ameet Morjaria Energy Economist - YP SASDE 24

35 (b) Staff Time and Cost Stage of Project Cycle Lending Staff Time and Cost (Bank Budget Only) No. of staff weeks USD Thousands (including travel and consultant costs) FY Total: Supervision/ICR FY FY FY FY FY FY FY Total: To August 31,

36 Annex 5. Beneficiary Survey Results No beneficiary survey was carried out. Annex 6. Stakeholder Workshop Report and Results No stakeholder workshop was carried out. 26

37 Annex 7. Summary of Grantee's ICR and/or Comments on Draft ICR The Borrower s ICR is reproduced in full here. I. Introduction This report was prepared by the Ministry of Energy and Water of the Islamic Republic of Afghanistan (MEW), the implementing agency for the Kabul Aybak Mazar-e-Sharif Power Project (KAMPP). The KAMPP was composed of the following contracts: The KAMPP development credit agreement was signed on December 23, It was originally a two-year project with a closing date of December 31, 2009, although this date was required by the scheduled expiry of the ARTF grant, and the project was actually envisioned to last at least three-years. As it was, the project was extended just prior to its expiry three times by a total of 39 months. In November 2009, it was extended 21 months to September 30, 2011; in September 2011, it was extended nine months to June 30, 2012; and in June 2012, it was extended by a final nine months to March 31, With the four-month grace period for project close out, it would have ended July 31, 2013, but the grace period itself was extended by two months, so the revised closing date is September 30, The KAMPP was funded from an Afghanistan Reconstruction Trust Fund (ARTF) grant. The ARTF grant (TF AF) was for $57 million. Since the project is not closed as of the writing of this report, the final disbursement amount is not yet known. However, it is expected that final disbursements will essentially equal the total allocated funding. II. Status of KAMPP Physical Components Contract No. Description Contractor MEW 300/3 Kabul Distribution Rehabilitation Project Phase I Angelique MEW/S-502 Objectives of the Project Mazar-e-Sharif Distribution Rehabiliation Works JV ATSL/AEPC MEW/S-503 Mazar-e-Sharif and Aybak Substations IRCON MEW/S-508 Procurement of Single and Three Phase Electronic Energy Meters Angelique The World Bank s project development objective (PDO) stated in the grant agreement for KAMPP was the following: The objective of the Project is to help the Recipient provide reliable and quality power to consumers in the cities of Kabul, Aybak, and Mazar-e- Sharif. There was no formal performance indicator stated for measuring the achievement of this objective. However, in addition to the PDO, the project had a number of intermediate outcomes, indicators, and intermediate result indicators that are summarized in the following table: 27

38 KAMPP Objective, Outcomes, and Indicators Outcome Indicator Intermediate Result Indicator PDO: To help provide reliable and None Provided quality power to the consumers in the target areas of the cities of Kabul, Aybak, and Mazar-e-Sharif. Intermediate Outcome 1: Increase in power supply to project areas in Kabul Intermediate Outcome 2: Providing grid power supply access to the people of Aybak Intermediate Outcome 3: Increase in power supply to customers in Mazar-e-Sharif Total grid electricity supplied to Kabul increased to 482 GWh by the end of SY 1388 (March 2010) 25% of population of Aybak with access to grid-based power The following were the components of the project: Strengthen the low voltage (LV) distribution network in parts of the capital city of Kabul supplied by Junction Station 2, as measured by number of transformers installed and commissioned. Construct a 220/20 kv substation and associated 20 kv line to Aybak to provide quality, reliable, and clean power to the people of Aybak town (indicator being the completion of the Aybak substation). Total grid electricity supplied to Mazar-e-Sharif increased to 113 GWh by the end of SY 1389 (March 2011) Strengthen the low voltage (LV) distribution network in parts of Mazar-e-Sharif City, and measured by the number of 20/0.4 kv transformers installed and commissioned. 1. Distribution System Rehabilitation for Kabul: Rehabilitation and extension of the distribution system to connect medium voltage (MV) trunk lines and cables to customers in Kabul; and reorganizing the existing distribution system to optimize loading to reduce losses and provide satisfactory power supply voltage to customers in Kabul. 2. Establishment of a 220 kilovolt (kv) Substation at Aybak and Interconnection with Distribution System: Establishment of a new 220/20kV, 16 Megavolts Amperes (MVA) substation at Aybak and 20kV interconnection with the existing Aybak distribution system, currently fed through diesel generated power to switch to the grid based supply. 3. Power System Rehabilitation for Mazar-e-Sharif: Rehabilitation of the distribution system in Mazar-e-Sharif by upgrading the existing MV distribution network from 6kV to 20kV; connecting the distribution network to the 220/20kV substation; rehabilitating and/or replacing the 28

39 low voltage distribution network; and augmenting transmission facilities to meet the requirements of the distribution network. 4. Institutional Capacity Building: Provision of capacity building and implementation support to MEW and DABM including for: (a) operation and maintenance of the distribution systems; (b) implementation of infrastructure metering and accounting of energy supplied in Kabul, Aybak and Mazar- e- Sharif, (c) establishment, operation and hand-over to DABM of customer care center(s); (d) communication of Project plans to consumers; (e) Project management support; and (f) other capacity building needs identified during implementation. Note, however, that the institutional capacity building of Part 4 was effectively never implemented. There were also no performance objectives associated with this component. III. Assessment of Outcome Achievement of the Objectives KAMPP technically achieved its objectives, although the objectives were generally illdefined and measured with poor indicators. The PDO was simply to help provide reliable power in the target areas of the target cities, which of course any project could hardly fail to do. The Intermediate Outcomes were a bit more meaningful: Intermediate Outcome 1: Increase in power supply to project areas in Kabul Intermediate Outcome 2: Providing grid power supply access to the people of Aybak Intermediate Outcome 3: Increase in power supply to customers in Mazar-e-Sharif KAMPP did technically achieve the performance indicators of the Intermediate Outcomes. The indicator for Outcome 1 (Kabul) was for grid electricity supply to increase to 482 GWh by the end of SY 1388 (March 2010). However, in fact Kabul grid consumption was already 649 GWh in 2007, before the project even started. There was a significant increase in electricity consumption in Kabul after the project started, with consumption reaching 978 GWh by the end of 2009, although this was well before the project work in Kabul was actually completed, so the achievement of the indicator was entirely unrelated to the project. (In fact, the improvements realized in Kabul were almost entirely attributable to projects under EPRP, since the Kabul component of KAMPP was limited to rehabilitating the distribution system connected to a single junction station.) The indicator for Outcome 2 (Aybak) was for 25% of the population of Aybak to have access to grid-based power. While again, the project technically achieved this indicator, it is in fact an indicator that is unknowable and unaffected by the project. Calculating the percentage of the population connected to the grid requires three data points: (1) the 29

40 number of customers, (2) the average household size, and (3) the total population. Unfortunately, none of this information is accurate or even consistent. According to DABS, there were 7,492 residential customers in Aybak in With average household size conservatively estimated at 6 10 people, this corresponds with 45,000 75,000 people. However, according to the Central Statistics Office (CSO), the population of Aybak City, the urban center connected to the grid, was just 26,400 in Since the total population of Aybak (including the rural areas of the district) was about 100,000, the only way that the 25% number makes any sense is if it refers to the urban population as share of the total population of Aybak Province. In fact, if this is the case, the indicator was technically achieved before the project began, and since the project did not include any expansion of the distribution network, the project could not affect the indicator in any event. The indicator for Outcome 3 (Mazar) was that total grid electricity supplied to Mazar-e- Sharif should increase to 113 GWh by the end of SY 1389 (March 2011). This indicator was valid and legitimately achieved by the project. Reliable data for 2011 isn t available, but by the end of 2010, consumption in Mazar had increased to 197 GWh, 174% of the goal, and this can reasonably be attributed to the KAMPP project, specifically the expansion of the Mazar Substation. Also, while there were significant problems with the distribution system portion of the work in Mazar, upgrading parts of the system from 6 kv to 20 kv did at least anecdotally improve the quality of the delivered power. The improvement in delivered voltage is seen in the following graph: 250 Mazar Peak Load Power Quality Improvement Proper 220 V Line Voltage Delivered Voltage (V) Before (Previous 6 kv Network) A er (New 20 kv Network) Source: SMEC. Average voltage from five houses at evening peak load in March and May of The outcomes, indicators, and achievements are summarized in the following table: 30

41 KAMPP Achievement of Performance Indicators Outcome PDO: To help provide reliable and quality power to the consumers in the target areas of the cities of Kabul, Aybak, and Mazar-e- Sharif. Intermediate Outcome 1: Increase in power supply to project areas in Kabul Intermediate Outcome 2: Providing grid power supply access to the people of Aybak Intermediate Outcome 3: Increase in power supply to customers in Mazar-e-Sharif Indicator None Provided Total grid electricity supplied to Kabul increased to 482 GWh by the end of SY 1388 (March 2010) 25% of population of Aybak with access to gridbased power Total grid electricity supplied to Mazar-e- Sharif increased to 113 GWh by the end of SY 1389 (March 2011) Intermediate Result Indicator Achievements Assessment Strengthen the low voltage (LV) distribution network in parts of the capital city of Kabul supplied by Junction Station 2, as measured by number of transformers installed and commissioned. Construct a 220/20 kv substation and associated 20 kv line to Aybak to provide quality, reliable, and clean power to the people of Aybak town (indicator being the completion of the Aybak substation). Strengthen the low voltage (LV) distribution network in parts of Mazar-e- Sharif City, and measured by the number of 20/0.4 kv transformers installed and commissioned. Consumption was 978 GWh in 2009, 202% of the baseline. Consumption was 1,219 GWh in 2010, 253% of the baseline. Installed and commissioned 135 transformers. Aybak substation completed. All customers previously connected to local grid who received power from diesel generator now receive power from substation. Grid-supplied electricity was already 197 GWh in Mazar in 2010, 174% of the goal. Installed and commissioned 62 transformers; delivered 161 for future installation by DABS. Significantly improved power quality. Significantly Exceeded Goal (Not attributable to project) Achieved Goal (Not due to project, and indicator irrelevant) Exceeded Goal 31

42 Implementation Status of the Physical Components The status of the works of the component contracts is summarized in the following table: Status of KAMPP Physical Components Contract No. Description Contractor Consultant* Status MEW 300/3 Kabul Distribution Rehabilitation Project Phase I Angelique GFA Completed 1 MEW/S-502 Mazar-e-Sharif Distribution Rehabiliation JV Works ATSL/AEPC GFA Completed 2 MEW/S-503 Mazar-e-Sharif and Aybak Substations IRCON GFA Completed 3 MEW/S-508 Procurement of Single and Three Phase Electronic Energy Meters Angelique GFA Completed 4 *SMEC was original PMF for all contracts. GFA assumed the PMF role on indicated projects in January 2012 under contract MEW 257b-SSS implemented under the Afghanistan Power System Development Project (APSDP). (1) Assets in service and operational acceptance issued September 30, 2011, but a significant number (4,345) corrective action reports still outstanding, primarily concerning meter boxes. (2) Assets in service and operational acceptance issued July 31, However, 795 defective 9-meters are being replaced. (3) Assets in service and completion certificate issued December 8, Operational acceptance not issued since substations were commissioned without a witness from client or PMF. (4) Equipment delivered. Completion certificate issued June 20, There were, however, significant delays in the implementation of all project components, including delays in executing the contracts and performing the work. For executing contracts, the average period between MEW receiving bids to signing a contract was 125 days. The fastest turn around was 90 days and the worst was 190 days. The average period between contract signing and contract effectiveness was 95 days. These delays were caused by the multiple steps required to make the contract effective, including securing bank guarantees, navigating the internal processes between MEW and MoF, making the advance payment, issuing the letter of credit (LC), and finally getting the special commitment (SC) letter from the World Bank. However, these delays were significantly shorter than those previously experienced under EPRP. The times required to sign and make effective each contract are illustrated below: 32

43 KAMPP Contract Signing and Effec veness MEW 300/3 MEW/S 502 MEW/S 503 MEW/S Days Bid Received to Contract Signing Contract Signing to Effec veness All of the KAMPP contracts experienced enormous implementation delays. The most on-time project (in absolute terms), the meter supply contract, still took 320 days longer than expected, which was actually the worst delay in relative terms (178% of the original schedule). The worst-delayed project was the Mazar-e-Sharif distribution rehabilitation, which took 850 days (155%) longer than originally planned. The reasons for the delays included revisions of scope, delays in equipment, complications in installation, security and logistics issues, and generally poor performance of the contractor. With all of this, the average time of actual completion was over 2.5 times (153%) the original contractual time for completion. The original and actual completion times for each of the KAMPP contracts are illustrated below: KAMPP Implementa on Delays MEW 300/3 MEW/S 502 MEW/S 503 MEW/S ,000 1,200 1,400 1,600 Days Original Time for Comple on Addi onal Time Required All of the relevant contractual and performance dates are seen in the following table: 33