Document of The World Bank Report No:ICR168 IMPLEMENTATION COMPLETION AND RESULTS REPORT ( IBRD-43390 ) ON A LOAN IN THE AMOUNT OF US$45 MILLION TO THE LESOTHO HIGHLANDS DEVELOPMENT AUTHORITY FOR LESOTHO HIGHLANDS WATER PROJECT - PHASE1B June 14, 2007 Water & Urban 1 Country Department 1 Africa Region CURRENCY EQUIVALENTS ( Exchange Rate Effective ) Currency Unit = Loti (plural: Maloti (M)) 1.00 = US$ US$ 1.00 = M7.05 (December 31, 2006) Fiscal Year GoL, LHDA, TCTA, DWAF: April 1 - March 31 ABBREVIATIONS AND ACRONYMS CALC Community Area Liaison Committee DBSA Development Bank of Southern Africa DRIFT Downstream Response to Imposed Flow Transformations DWAF Department of Water Affairs and Forestry ­ South Africa EAP Environmental Action Plan EIA Environmental Impact Assessment EIB European Investment Bank ERR Economic Rate of Return ESSG Environment and Social Services Group GDP Gross Domestic Product GoL Government of Lesotho IBRD International Bank for Reconstruction and Development ICM Integrated Catchment Management ICR Implementation Completion Report IDA International Development Association ISR Implementation Status and Results LEC Lesotho Electricity Corporation LFCD Lesotho Fund for Community Development LHDA Lesotho Highlands Development Authority LHWC Lesotho Highlands Water Commission LHWEP Lesotho Highlands Water Engineering Project LHWP Lesotho Highlands Water Project PAD Project Appraisal Document PDO Project Development Objective PoE Panel of Experts QAG Quality Assurance Group QSA Quality of Supervision Assessment RSA Republic of South Africa SACU Southern African Customs Union TCTA Trans Caledon Tunnel Authority VIP Latrine Ventilated Improved Pit Latrine Vice President: Obiageli K. Ezekwesili Country Director: Ritva S. Reinikka Sector Manager: Jaime M. Biderman Project Team Leader: Andrew Macoun Lesotho Highlands Water Project Phase 1B ICR CONTENTS 1. PROJECT CONTEXT, DEVELOPMENT OBJECTIVES AND DESIGN........................................ 1 1.1 CONTEXT AT APPRAISAL ...................................................................................................................... 1 1.2 ORIGINAL PROJECT DEVELOPMENT OBJECTIVES (PDO) AND KEY INDICATORS................................... 3 1.3 REVISED PDO AND KEY INDICATORS AND REASONS/JUSTIFICATION.................................................... 3 1.4 MAIN BENEFICIARIES, ORIGINAL AND REVISED .................................................................................... 3 1.5 ORIGINAL COMPONENTS....................................................................................................................... 4 1.6 REVISED COMPONENTS......................................................................................................................... 5 1.7 OTHER SIGNIFICANT CHANGES.............................................................................................................. 5 2. KEY FACTORS AFFECTING IMPLEMENTATION AND OUTCOMES ...................................... 6 2.1 PROJECT PREPARATION, DESIGN AND QUALITY AT ENTRY .................................................................. 6 2.2 IMPLEMENTATION................................................................................................................................. 8 2.3 MONITORING AND EVALUATION (M&E) DESIGN, IMPLEMENTATION AND UTILIZATION ................... 13 2.4 SAFEGUARD AND FIDUCIARY COMPLIANCE........................................................................................ 13 2.5 POST-COMPLETION OPERATION/NEXT PHASE .................................................................................... 16 3. ASSESSMENT OF OUTCOMES......................................................................................................... 18 3.1 RELEVANCE OF OBJECTIVES, DESIGN AND IMPLEMENTATION............................................................ 18 3.2 ACHIEVEMENT OF PROJECT DEVELOPMENT OBJECTIVES ................................................................... 18 3.3 EFFICIENCY ........................................................................................................................................ 19 3.4 JUSTIFICATION OF OVERALL OUTCOME RATING ................................................................................ 20 3.5 OVERARCHING THEMES, OTHER OUTCOMES AND IMPACTS ............................................................... 21 3.6 SUMMARY OF FINDINGS OF BENEFICIARY SURVEY AND/OR STAKEHOLDER WORKSHOPS.................. 22 4. ASSESSMENT OF RISK TO DEVELOPMENT OUTCOME.......................................................... 23 5. ASSESSMENT OF BANK AND BORROWER PERFORMANCE.................................................. 23 5.1 BANK.................................................................................................................................................. 23 5.2 BORROWER......................................................................................................................................... 24 6. LESSONS LEARNED............................................................................................................................ 25 7. COMMENTS ON ISSUES RAISED BY BORROWER/IMPLEMENTING AGENCIES/PARTNERS........................................................................................................................... 29 ANNEX 1. RESULTS FRAMEWORK ANALYSIS............................................................................... 30 ANNEX 2. PROJECT COSTS AND FINANCING................................................................................. 38 ANNEX 3. OUTPUTS BY COMPONENT .............................................................................................. 39 ANNEX 4. ECONOMIC AND FINANCIAL ANALYSIS...................................................................... 46 ANNEX 5. BANK LENDING & IMPLEMENTATION SUPPORT/SUPERVISION PROCESSES. 50 ANNEX 6. STAKEHOLDER WORKSHOP REPORT AND RESULTS............................................. 53 ANNEX 7. SUMMARY OF BORROWER'S ICR AND/OR COMMENTS ON DRAFT ICR ........... 58 ANNEX 8. COMMENTS OF COFINANCIERS AND OTHER PARTNERS/STAKEHOLDERS.... 70 ANNEX 9. LIST OF SUPPORTING DOCUMENTS ............................................................................. 78 A. Basic Information Hiland Water 1B Country: Lesotho Project Name: Project Project ID: P001409 L/C/TF Number(s): IBRD-43390 ICR Date: 06/18/2007 ICR Type: Intensive Learning ICR GOVERNMENT OF Lending Instrument: SIL Borrower: LESOTHO Original Total USD 45.0M Disbursed Amount: USD 21.7M Commitment: Environmental Category: A Implementing Agencies: Lesotho Highlands Development Authority Cofinanciers and Other External Partners: European Investment Bank (EIB) Development Bank of Southern Africa B. Key Dates Process Date Process Original Date Revised / Actual Date(s) Concept Review: 01/13/1993 Effectiveness: 06/09/1999 06/09/1999 Appraisal: 06/23/1997 Restructuring(s): Approval: 06/04/1998 Mid-term Review: 11/13/2000 Closing: 12/31/2004 12/31/2006 C. Ratings Summary C.1 Performance Rating by ICR Outcomes: Satisfactory Risk to Development Outcome: Low or Negligible Bank Performance: Satisfactory Borrower Performance: Moderately Satisfactory C.2 Detailed Ratings of Bank and Borrower Performance (by ICR) Bank Ratings Borrower Ratings Quality at Entry: Satisfactory Government: Moderately Satisfactory Quality of Supervision: Highly Satisfactory Implementing Agency/Agencies: Moderately Satisfactory Overall Bank Overall Borrower Performance: Satisfactory Performance: Moderately Satisfactory i C.3 Quality at Entry and Implementation Performance Indicators Implementation QAG Assessments Performance Indicators (if any) Rating Potential Problem Project No Quality at Entry None at any time (Yes/No): (QEA): Problem Project at any Quality of Yes Satisfactory time (Yes/No): Supervision (QSA): DO rating before Satisfactory Closing/Inactive status: D. Sector and Theme Codes Original Actual Sector Code (as % of total Bank financing) Water supply 100 100 Theme Code (Primary/Secondary) Access to urban services and housing Secondary Secondary Export development and competitiveness Primary Primary Infrastructure services for private sector development Secondary Secondary Rural services and infrastructure Secondary Secondary Water resource management Primary Primary E. Bank Staff Positions At ICR At Approval Vice President: Obiageli Katryn Ezekwesili Callisto E. Madavo Country Director: Ritva S. Reinikka Pamela Cox Sector Manager: Jaime M. Biderman Jeffrey S. Racki Project Team Leader: Andrew Macoun John A. Roome ICR Team Leader: Leonard John Abrams ICR Primary Author: Marcus J. Wishart F. Results Framework Analysis Project Development Objectives (from Project Appraisal Document) The main objectives of the project are: (a) to put in place the physical and managerial capacity for Lesotho to transform its principal natural resource of abundance - water - into export revenues that can be applied to poverty reduction and economic stability while; (i) protecting the environment and mitigating any adverse social and environmental impacts, and (ii) maximizing the local development spin-offs of the ii project in Lesotho; and (b) assisting South Africa in developing its lowest cost alternatives for supply of water to the Gauteng Region. Revised Project Development Objectives (as approved by original approving authority) (a) PDO Indicator(s) Original Target Formally Actual Value Indicator Baseline Value Values (from Revised Achieved at approval Target Completion or documents) Values Target Years Indicator 1 : Cumulative combined Phase 1A and 1B royalties (in 1995 Rand) of M375m by 12/2000, M650m by 12/2004 and M1150m by 12/2007. Value quantitative or M362 million M650 million M1,918 million Qualitative) Date achieved 11/30/1998 12/31/2004 12/31/2006 Comments (incl. % Upon closing of the loan royalties were 167% of December 2007 expectations. achievement) Indicator 2 : Managerial and institutional capacity in place to manage the project. Value Project cost estimates at Achievement of Project cost quantitative or appraisal = US$884.4 target completion estimates at loan Qualitative) million. dates within closing = US$628.9 budget estimates. million. Date achieved 11/30/1998 12/31/2004 12/31/2006 Comments Project costs in December 2006 were 30% below appraisal estimates, and (incl. % expected to be 5% below appraisal estimates at completion in 2044. Main achievement) project works approximately 30% below appraisal estimates and largely on time. Indicator 3 : Protection of the Environment and mitigation of any adverse social and environmental impacts. All disturbed sites All disturbed sites successfully restored; # rehabilitated; 8 fatalities on site; fatalities Value household significantly quantitative or Nil standards of living improved on Ph1A; Qualitative) maintained; household compensation standards of living paid; downstream maintained; impacts mitigated. compensation paid; downstream impacts mitigated. Date achieved 11/30/1998 12/31/2005 12/31/2006 Comments (incl. % Resettlement and development aspects were guided by comprehensive policy and achievement) implementation framework detailed in the text and documents on file. iii Indicator 4 : Local development spin-offs of the project in Lesotho. 3,000 jobs (16,500 person-years 9,000 jobs (20,000 Value employment) with person-years quantitative or Nil contracts employment) with Qualitative) generating M275m contracts generating (M460 million at M623 million (Dec Dec 2003 prices). 2003 prices). Date achieved 11/30/1998 12/31/2004 12/31/2006 Comments Local labor earnings estimated at M138 million, with sub-contracting (incl. % opportunities contributing an additional M300 million. 2545 VIP toilets to 126 achievement) Highland villages which had no piped water or sanitation at appraisal. (b) Intermediate Outcome Indicator(s) Original Target Formally Actual Value Indicator Baseline Value Values (from Achieved at approval Revised Completion or documents) Target Values Target Years Indicator 1 : Delivery of 2.2 m3/s of water from Matsoku Diversion Value (quantitative Nil 2.2 m3/s 2.2 m3/s or Qualitative) Date achieved 11/30/1998 03/31/2001 03/31/2001 Comments (incl. % 100% achieved. achievement) Indicator 2 : Delivery of 9.6 m3/s from Mohale to Katse Value (quantitative Nil 9.6 m3/s 9.6 m3/s or Qualitative) Date achieved 11/30/1998 01/31/2003 01/31/2003 Comments (incl. % 100% achieved. achievement) G. Ratings of Project Performance in ISRs Actual No. Date ISR Archived DO IP Disbursements (USD millions) 1 06/18/1998 Satisfactory Satisfactory 0.00 2 12/14/1998 Satisfactory Satisfactory 0.00 3 05/26/1999 Satisfactory Satisfactory 0.00 4 12/13/1999 Satisfactory Satisfactory 1.21 5 06/22/2000 Satisfactory Satisfactory 5.98 6 12/27/2000 Satisfactory Satisfactory 11.11 iv 7 06/21/2001 Satisfactory Satisfactory 13.46 8 12/10/2001 Satisfactory Satisfactory 14.31 9 04/29/2002 Satisfactory Satisfactory 15.58 10 05/23/2002 Satisfactory Satisfactory 15.85 11 12/10/2002 Satisfactory Satisfactory 17.39 12 05/28/2003 Satisfactory Satisfactory 18.96 13 07/11/2003 Satisfactory Satisfactory 18.96 14 12/01/2003 Satisfactory Satisfactory 20.36 15 03/03/2004 Satisfactory Satisfactory 20.36 16 05/27/2004 Satisfactory Satisfactory 20.36 17 11/24/2004 Satisfactory Satisfactory 21.60 18 04/02/2005 Satisfactory Moderately Satisfactory 21.60 19 09/21/2005 Satisfactory Moderately Satisfactory 21.60 20 02/03/2006 Satisfactory Moderately Unsatisfactory 21.66 21 06/26/2006 Satisfactory Moderately Unsatisfactory 21.66 22 12/12/2006 Satisfactory Moderately Satisfactory 21.66 H. Restructuring (if any) Not Applicable I. Disbursement Profile v 1. Project Context, Development Objectives and Design 1.1 Context at Appraisal 1. The Lesotho Highlands Water Project (LHWP)1 is designed to transfer water from the water-rich highlands of Lesotho to the dry Gauteng region of South Africa through a se- ries of dams, transfer tunnels and associated infrastructure, and to supply electricity to Lesotho. At appraisal in 1997, the Bank loan for Phase 1A of the LHWP was nearing clo- sure, South Africa needed to develop its next-least-cost water source for Gauteng, its in- dustrial and commercial heartland, and Lesotho, which at the time depended heavily on migrant labor remittances, needed to secure alternative sources of income to stabilize its economy. Although the World Bank had largely withdrawn from financing large hydrau- lic infrastructure pending the outcome of the World Commission on Dams it continued its support to the LHWP with a loan for Phase 1B of US$45 million. 2. Lesotho is a small, poor country of Box 1: Lesotho Highlands Water Project Phase 1 2.2 million people landlocked by South Phase 1A: Katse Dam on the Malibamats'o River Africa, upon which its economy is `Muela Dam on the Nqoe River largely dependent. At appraisal about Delivery tunnel to South Africa `Muela hydropower station 40% of GNP was derived from remit- Phase 1B: Mohale Dam on the Senqunyane River tances, with about 40% of the male la- Matsoku Weir on the Matsoku River bor force employed in the South African Delivery tunnel from Mohale to Katse mines. Economic growth was driven by Phase 1A of the LHWP, the service sector, and a small but rapidly expanding textile export industry. Despite significant challenges, Leso- tho had demonstrated its commitment and capacity to partner with the far larger, devel- oped economy of South Africa with the Lesotho Highlands Development Authority (LHDA) having successfully implemented the LHWP Phase 1A. The 1996 Country As- sistance Strategy (CAS) for Lesotho had four broad objectives: (i) Fostering Labor Inten- sive Growth; (ii) Investing in Human Resources; (iii) Maximizing the Poverty Reducing Impact of LHWP; and, (iv) Improving Institutional Capacity. 3. South Africa, emerging from the end of apartheid, was facing future water shortages, particularly in Gauteng, which is responsible for almost 60% of the national GDP and holds 42% of the urban population. By the mid-1980s water demand in Gauteng had out- stripped available supply from the Vaal catchment in which it is situated. The Country Assistance Note (1997) for South Africa identified four priority areas where the Bank could be of assistance: (i) Growth and Stability; (ii) Poverty and Inequality; (iii) Capacity Building; and (iv) Regional Issues. Given South Africa's acute water shortages and the opportunities for the "win-win" joint management and development of water resources by South Africa and neighboring states, regional issues were specifically focused on the joint development of infrastructure, especially water. 4. Corruption: The ICR for Phase 1A of the LHWP (1999) rated the project outcome as Satisfactory. Notwithstanding this, Phase 1A has subsequently been marred by revela- tions of corruption. The Government of Lesotho (GoL) has been widely acclaimed for its stance and action in this regard. It has secured six convictions to date, including a former 1The LHWP Treaty, signed in 1986 by the Republic of South Africa and Kingdom of Lesotho, outlines responsibilities until 2044 for the first of four phases. Phase 2 feasibility studies are currently underway. - 1 - chief executive of the LHDA, and is currently pursuing cases against a former chief dele- gate and a serving delegate to the Lesotho Highlands Water Commission (LHWC). The Bank cooperated in the investigations and has debarred one individual and two compa- nies, Acres International (in July 2004) and Lahmeyer International GmbH (in November 2006). Both companies are ineligible to be awarded Bank-financed contracts for a period of three and seven years, respectively. Another two individuals and three European firms, not involved in Bank-financed contracts, were also convicted in Lesotho. Fines and set- tlements now amount to M71.8 million (over US$10 million). Although not related to Phase 1B, to which this ICR pertains, these high profile cases have been addressed during the implementation of Phase 1B. 5. To support the GoL in meeting the costs of prosecution, the Bank offered supplemen- tary financing for an equivalent amount of eligible expenditures under any IDA Credit in Lesotho's World Bank portfolio in order to free up Government resources for the cost of the prosecutions. This offer was never acted upon, nor was the potential financial support offered by the EU and UK. The Bank did not give an undertaking to make direct pay- ments towards the costs of prosecution as it does not have a mechanism for providing such support. The Bank also provided Technical Assistance during the discovery phase of the corruption and advised the GoL on how much it could cost to prosecute the cases. 6. The experience of the LHWP and the pro-active role of the GoL have been used widely as a case study of good governance, highlighting how relatively small, developing states can successfully prosecute some of the largest multinational firms in the world. De- spite the corrupt practices identified in Phase 1A of the project, Phase 1B came in under initial cost estimates and there is no evidence to indicate that the quality of the project was compromised. 7. Sector Issues Expected to be Addressed by the Project: The Project focused on the water export sector in Lesotho and on the development of the least-cost supply of bulk water to Gauteng in South Africa. The project did not explicitly attempt to address broader water sector issues in either Lesotho or South Africa, although this was expected to be done in parallel. 8. Rationale for Bank Assistance: The World Bank had a long history of involvement in the LHWP, having: i) acted as the executing agency for the UNDP-financed consult- ants who supervised the LHWP feasibility studies in 1983; ii) provided an IDA Credit of SDR 8 million (US$9.8 million equivalent: Credit 1747-LSO) in 1986 for the Lesotho Highlands Water Engineering Project (LHWEP), which assisted with preparation of the LHWP; iii) made an advance of US$750,000 under the Bank's Project Preparation Facil- ity (PPF P356-LSO 1988) for Phase 1A; and, iv) supported Phase 1A through a US$110 million loan (Loan 3393-LSO) approved on July 23, 1991. This loan closed on March 31, 1999. 9. The Bank's financial role in Phase 1B of the LHWP was very minor (~4%), limited to engineering design and supervision of the main works (US$17.8 million), institutional support to LHDA, including the social and environmental Panel of Experts (PoE), the engineering PoE and the Disputes Review Board (US$26.7 million), and training - 2 - (US$0.5 million). However, both governments sought and valued the Bank's involvement since this (i) provided comfort to other lenders - in particular European Investment Bank (EIB) and the Export Credit Agencies, (ii) allowed the Bank to play a facilitating role be- tween South Africa and Lesotho and to provide monitoring and evaluation for the imple- menting agencies, and (iii) provided careful attention to sensitive issues, through compli- ance with the Bank's environmental and social safeguard policies. 1.2 Original Project Development Objectives (PDO) and Key Indicators 10. The project's main objectives articulated in the Project Appraisal Document (PAD) were: (a) to put in place the physical and managerial capacity for Lesotho to transform its principal natural resource of abundance - water - into export revenues that can be applied to poverty reduction and economic stability while; (i) protecting the environment and mitigating any adverse social and environmental impacts, and (ii) maximizing the local development spin-offs of the project in Lesotho; and (b) assisting South Africa in devel- oping its lowest-cost alternatives for supply of water to the Gauteng Region. 11. Key PDO indicators included: (i) Cumulative combined Phase 1A and 1B royalties (in 1995 South African Rand which has parity with Lesotho Maloti) of M375m by 12/2000, M650m by 12/2004 and M1150m by 12/2007. (ii) Managerial and Institutional capacity in place to manage the project by 12/31/2004. (iii) Protection of the environment and mitigation of any adverse social and envi- ronmental impacts (all disturbed sites restored; # fatalities on site; household standards of living maintained; compensation paid; downstream impacts miti- gated) by 12/31/2005. (iv) Local development spin-offs of the project in Lesotho: 3,000 jobs for Basotho (16,500 person-years employment) generating M275m by 12/2003. 12. The project objective defined in the Loan Agreement and the Treaty does not include explicit reference to broader environmental, social or development objectives. The Loan Agreement simply states, under Schedule 3: Description of the Project, that the "objec- tive of the project is to establish the physical and managerial capacity for Lesotho to earn export revenues by transforming its principal natural resources ­ water ­ into an export commodity". This objective reflects the Purpose of the Project as defined under Article 4 of the Treaty - to "enhance the use of the water of the Senqu/Orange River by storing, regulating, diverting and controlling the flow of the Senqu/Orange River and its affluents [sic] in order to effect the delivery of specified quantities of water to the Designated Out- let Point in the Republic of South Africa and by utilizing such delivery system to gener- ate hydro-electric power in the Kingdom of Lesotho." 1.3 Revised PDO and Key Indicators and reasons/justification 13. The objectives of the Project were not revised. 1.4 Main Beneficiaries, original and revised 14. The main beneficiaries identified in the PAD and captured in the PDO include: - 3 - a) The whole population of Lesotho: The population of Lesotho as a whole, and the poor in particular, were expected to benefit from revenues derived from royalties in terms of the Treaty. These were, in part, to be deposited into a dedicated Lesotho Highlands Water Revenue Fund to support community development projects in Lesotho. Once Phases 1A and 1B of the project were operating at full capacity it was expected to generate about M130-140 million per annum (1995 prices, US$36-38 million, equivalent to 4% of 1995 GDP and 10% of 1995 Government Revenues). b) Industrial and urban water users in the Gauteng region of South Africa: The Project was the least-cost solution to the provision of water to Gauteng. In terms of the project Treaty, approximately 44% of the cost savings between LHWP and the next-lowest-cost alternative was estimated to accrue to water users in Gauteng. The unit cost of incre- mental water from Phase 1B is about 2/3 that of the next-lowest-cost alternative. For Phase 1A and 1B combined this means savings of about US$30 million per annum. c) Project workers, Basotho contractors and suppliers to the project and LHDA employ- ees: During the construction phase of the project, further benefits were expected to accrue to Lesotho nationals, including about 3000 project workers (M200 million), Basotho con- tractors and suppliers to the project (M140 million), and LHDA employees (M250 mil- lion). These benefits, together with other LHWP related benefits, were estimated to amount to about 6.5% of GDP in 1998 and 3.9% of GDP in 2002. d) Local Highland communities: While suffering compensated losses, the local Highland communities were expected to benefit from new roads providing better access and from health services and other rural development activities including agriculture, range man- agement, forestry, fisheries and tourism. These were to be achieved through a compensa- tion approach that, under the provisions of the Treaty, would leave no project affected person worse off, and a development approach that, under the Bank's resettlement policy, would make those affected by the project beneficiaries of it. 1.5 Original Components 15. The project consisted of four components that were clearly linked to the PDO and key indicators. The description of the project detailed under Schedule 2 of the Legal Agree- ment differs, however, from that in the PAD. While most of the differences are in presen- tation, the Legal Agreement specifies seven components, of which Studies and Institu- tional Support for LHDA have been incorporated under component 4 below. The Legal Agreement included a specific component aimed at "strengthening the Lesotho Highlands Water Revenue Fund's capacity to manage efficiently Royalty Payments". The PAD does not include this component but details this as one of a number of institutional and imple- mentation arrangements. The four components described in the PAD were: Component 1: Main Project Works (US$531 million; 60%) · Mohale Dam: Construction of a 145m high concrete-face rock-fill dam at Mohale. · Mohale Tunnel: Construction of a 32km-long, 4.1-4.6m diameter concrete-lined tun- nel from Mohale to Katse, capable of delivering 9.6m3/s. · Matsoku Diversion: Construction of a 15m-high concrete weir at Matsoku and a 5.6km-long, 4.5m-diameter tunnel capable of delivering 2.2m3/s to Katse. - 4 - · Mountain Road and Mohale Access Road: Construction of associated advance infra- structure including upgrading of 59km and the new construction of 21km of mountain road. · Infrastructure: Construction of the Maseru bypass, Likalaneng Weir, and construc- tion camps/workers village, power supply and telecommunication facilities. · Construction Mitigation Program: Mitigation of environmental impacts of construc- tion: water quality management, pollution abatement, dust control and monitoring, noise control & monitoring, site rehabilitation, community impact avoidance, public safety, access management, site landscaping. Component 2: Engineering Supervision (US$105 million; 12%) · Engineering services for the tender design and the detailed design. · Construction Supervision services for the main project works. Component 3: Environmental and Social Action Plan (US$140 million; 16%) · Implementation of a full Environmental Action Plan including: i) Resettlement of affected communities; ii) Compensation for project-related losses; iii) Income restoration and rural development; iv) Natural environment and heritage; v) Public health program; and, vi) Program monitoring and evaluation. Component 4: Administration and Capacity Building (US$108 million; 12%) · Institutional Support for LHDA through short- and long-term engineering technical assistance. · Finance and administration (individual service contracts and technical assistance). · Environment and social issues (including the services of both an Engineering and an Environment/Social Panel of Experts and a Disputes Review Board). · Operations and maintenance (technical assistance). · Training and studies. · Operating and administrative costs for LHDA. 1.6 Revised Components The components of the Project were not revised. 1.7 Other significant changes 16. There were no formal changes in the project's original design, scope, scale or imple- mentation arrangements. The project closing date was extended twice, by one year on both occasions, from its original closing date of December 31, 2004 to December 31, 2006. This was primarily to address residual environmental, social and institutional is- sues. The original funding allocations for Phase 1B of the LHWP included a US$45 mil- lion IBRD loan and US$8 million from the Phase 1A IBRD loan. Of the original loan amount, US$9 million was cancelled on December 18, 2001 at the request of LHDA, fol- lowing authorization from the LHWC, due to less-than-anticipated technical assistance requirements. A request for the cancellation of a further US$6 million was submitted to the Bank in October, 2003, and another for US$7 million on January, 2005. At the clos- ing date of December 31, 2006, US$21.66 million (48.13%) of the original loan amount - 5 - had been disbursed. Cancellation of loan amounts, due to lower than anticipated techni- cal assistance, did not have an impact on the Bank's leverage, the effectiveness of Bank supervision missions or relationships with the Borrower. 2. Key Factors Affecting Implementation and Outcomes 2.1 Project Preparation, Design and Quality at Entry 17. The project was designed with clearly defined objectives linked to specific, measur- able indicators as a joint undertaking of the governments of the GoL and the RSA. Simi- larly, the Treaty, covering Phases 1A and B, clearly spells out the respective responsibili- ties of Lesotho and the RSA2. While there was no formal assessment of quality at entry, the project drew on experience from Phase 1A, with 19 specific lessons identified during preparation and reflected in the design of Phase 1B. 18. Economic Issues: At the time of preparation South Africa had begun to implement water demand management measures, the results of which suggested that Phase 1B could potentially be delayed for up to seven years. However, these measures alone would not have obviated the eventual need for Phase 1B; and the potential insecurity in water sup- ply was considered an unacceptable risk by the South African government. Significant sunk investments and an economic analysis at appraisal confirmed that any delay in im- plementing Phase 1B would have had significant cost implications (contractor re- mobilization, institutional administration and maintenance, SACU compensation, fore- gone hydropower benefits, the costs of continuing to run the alternative Tugela/Vaal pump storage system). The combination of these factors outweighed the potential cost savings of any delay. 19. Engineering Components: The project description under the Treaty included only the infrastructure components necessary to achieve the project's water transfer objectives. Drawing on outputs from the LHWEP and lessons from Phase 1A, there was an enhanced focus on local contractor participation through specific provisions in the bidding docu- ments for civil works and the design and tunneling between Mohale and Katse reservoirs was refined. Overall, the engineering component was well designed and provided a solid platform for implementation. 20. Social and Environmental Components: The environmental and social benchmarks for the development of hydraulic infrastructure have evolved significantly since ap- praisal. Indeed, the project was designed during a time of considerable debate on the fu- ture of large dams. While the Treaty was general on standards for the social and envi- ronmental components, requiring only "reasonable measures", the project benefited from the preceding LHWEP that financed detailed studies on the social and environmental as- pects of the LHWP area that resulted in the adoption by the GoL of a comprehensive en- vironmental action plan. Phase 1B incorporated many of the emerging lessons and, 2The RSA assumes responsibility for all costs relating to water transfer, with Lesotho having direct cost responsibility for the hydropower component and 50% cofinancing of any ancillary developments beyond compensation and income restoration. The Treaty provides Lesotho with 56% of the net difference between the cost of the LHWP and the next-least-cost alternative to provide water to Gauteng. The Bank loan for Phase 1B was made to the LHDA, with guarantees from both the GoL and the RSA. LHDA is responsible for project implementation within the borders of Lesotho and the Trans Caledon Tunnel Authority (TCTA) for project works in South Africa. Cost recovery is through the sale of all bulk water received by the South African Department of Water Affairs and Forestry (DWAF) and bulk electricity sales to the Lesotho Electricity Corporation (LEC). - 6 - guided by the Bank's safeguard policies, adopted what were considered at the time to be stringent measures that conformed with the findings of the 1994 Bank-wide Review of Resettlement. A comprehensive compensation package was designed which included in- dividual compensation for fixed assets, including dwellings, gardens, trees, fields, kraals and graveyards. Compensation for loss of communal grazing lands, access to common property, brushwood fuel, useful grasses, medicinal plants and wild vegetables was ad- dressed through community-based compensation and development programs in agricul- ture, tourism, and small business support. The design of Phase 1B included improved consultation procedures. Combined Area Liaison Committees (CALCs), facilitated by Community Liaison Assistants, were established and proved to be an effective mecha- nism for facilitating community consultations. An annual stakeholder conference was es- tablished to provide a forum for enabling stakeholders to discuss progress and perform- ance with LHDA. 21. The design of Phase 1B included ensuring completion of a full Environmental Impact Assessment (EIA) before preparing the Environmental Action Plan (EAP), preparing im- plementation plans and performance indicators prior to project commencement, and strengthening monitoring. Budget allocations for environmental and social components under the EAP were significantly increased, from US$67 million in Phase 1A, represent- ing around 5% of capital costs, to US$115.6 million or around 15% of capital costs in Phase 1B. 22. Financial and Institutional Arrangements: The financing structure included a large number of co-financiers, including multi-lateral agencies (14%), South African Capi- tal/Money Markets (73%), export credit-backed commercial loans (7%), and government contributions (2%). The legitimacy of the LHWP Treaty was questioned, given that it was signed in 1986 between the apartheid regime in South Africa and a military govern- ment in Lesotho. However, at the time of appraisal for Phase 1B, the new, post-apartheid government in South Africa and the democratically elected government in Lesotho re- affirmed their commitment to the project. A revision of the Treaty, Protocol VI, outlined a revised governance model for the LHWP. With this foundation and the experience of Phase 1A, the institutional arrangements for Phase 1B were well established. 23. Higher-level institutional risks identified at appraisal included the GoL's ability and willingness to assume responsibility for assets and activities as Phase 1B drew to conclu- sion, along with delays in the delivery and quality of outputs resulting from the re- organization of LHDA. Institutional relationships between LHDA, its Board and the Joint Permanent Technical Committee (JPTC: predecessor of the LHWC) for Phase 1A were acknowledged as being cumbersome and slow, resulting from micro-management, cum- bersome procedures and a lack of accountability. South Africa favored a full bi-national authority, with a joint Board allowing the placement of South African nationals in line positions. However, this proposal was not acceptable to Lesotho, and the alternative structure agreed under Protocol VI proposed that (i) the LHWC retain ultimate account- ability for the project, but shift to more of a policy-formulation and monitoring role, (ii) LHDA's Board assume a greater executive role, but its members be appointed on the ba- sis of merit and skills, by the LHWC, based on a set of proposals from GoL (who may at - 7 - their discretion nominate South African residents in addition to Lesotho residents) and be subject to performance contracts. 24. Inspection Panel Requests: Two Inspection Panel requests were filed during prepa- ration of Phase 1B. On April 23, 1998, a group of residents from Alexandra township in Gauteng claimed that Phase 1B would reduce resources available to improve water sup- ply for low-income communities. The Panel concluded that the concerns were more likely the legacy of apartheid policies than a consequence of the Bank not complying with its own policies and procedures, and the complaint was dismissed without further investigation. The second Inspection Panel request was filed on April 14, 1999 by a group of nine mining companies registered in Lesotho claiming that mineral lease rights on land in the Katse Dam area were unlawfully expropriated by the GoL without com- pensation. A judgment of the High Court of Lesotho ruled that the disputed mining lease was null and void. The Panel concluded that there was no direct link between any actions and/or omissions of the Bank and the harm claimed by the mining companies and did not recommend further investigation. 2.2 Implementation 25. The ratings of project performance reflected in Implementation Status and Results reports (ISRs) provided a clear and consistent indication of the project's performance that followed a logical sequence over the implementation period. Overall Implementation Per- formance was rated as satisfactory throughout the original implementation period. This was downgraded to moderately unsatisfactory during the two extension periods due to slow progress in implementing the agreed Action Plan covering unresolved social and environmental issues. Significant gains were subsequently made and the project imple- mentation performance was rated as moderately satisfactory at closing. There was no re- structuring of the project, reflecting the detailed preparation and sound engineering de- sign founded on the LHWEP and lessons learned from Phase 1A. Phase 1B was identi- fied in May 2000 as "At Risk", due primarily to slow disbursement, but was removed later the same year. The mid-term review (November, 2000) re-affirmed satisfactory pro- gress toward objectives, although identified the need to develop a strategy to accelerate implementation and enhance performance of the EAP. 26. At the time of the closing of the World Bank loan at the end of December 2006, the overall project (including the elements not financed by the World Bank) had cost US$624.3 million, 70.6% of the appraisal estimate which was US$884.4 million. However, the costs of the project did not end at the closure of the World Bank loan - final cost es- timates, extrapolated to 2044 and taking into account long-term compensation payments and programs, are expected to be 95.4% of appraisal estimates. 27. Economic Issues: Implementation of water demand measures in Gauteng and lower- than-expected population growth rates, resulting in lower-than-projected demand for wa- ter in South Africa, led to a reduction in the economic rate of return (ERR). However, the ex-post rate of return remained within acceptable limits (11.5%) and the project remains the lowest-cost alternative for supply to South Africa. In 2002 a cost/benefit analysis compared the economic value of water released to meet the IFRs with project costs under various IFR scenarios. Releasing water to meet Instream Flow Requirements results in - 8 - reduced water available for transfer to South Africa and therefore reduced royalty reve- nues and impacting the cost/benefit balance. The financial analyses showed compensat- ing downstream communities to be a more cost-effective alternative, but concluded that moderate increases in the IFRs would not compromise the project's economic viability and would provide additional benefits to downstream communities. 28. Actual revenues generated through royalties exceeded initial estimates with cumula- tive royalties at the end of 2006 amounting to M1,918 million, representing an average of 4.8% of GDP between 1998 and 2006. Combined SACU and tax revenues also exceeded 2003 targets, with M845 million in SACU revenues and M435 million in tax revenues. By the end of 2006 these contributions amounted to M1,060 million in SACU revenues and M1,124 million in tax revenues. Note that the apparent contradiction between lower than expected ERR due to reduced demand in Gauteng and higher than estimated royalty revenues is because the ERR is based on long-term projected demand, whilst royalties received are as a direct consequence of actual current demand. 29. Revenues earned from the 'Muela Hydropower Plant increased to M63 million in 2005 but remain below expectations due to an unfavorable fixed price tariff structure es- tablished during Phase 1A. Despite significant efforts during the supervision of Phase 1B, these issues remain controversial and unresolved. 30. Engineering Components: All engineering was completed by mid-2004 and the Main Project Works and Engineering Supervision components were continually rated as highly satisfactory. Delivery was typically on time and under budget with a high quality of supervision of the works. Problems were experienced with the 4.5m-diameter, 32km- long Mohale/Katse Transfer tunnel due to unexpected adverse geological conditions, which required additional pumping and grouting activities, and frequent mechanical fail- ures. Intervention from the Disputes Review Board resulted in the Contractor's claim be- ing reduced from an original US$80 million to an agreed amount of US$30 million. Im- poundment of the 947 million m3 Mohale reservoir was delayed when agreed prerequi- sites, relating primarily to IFRs, were not fully met. Agreement was reached for a phased implementation, with triggers linked to the completion of specific actions, with final im- poundment in July, 2003. 31. Social Components: Community development, compensation and resettlement, natu- ral environment and heritage, and monitoring and evaluation were all integral to the EAP, complemented with a separate public health program. The physical relocation component of these programs has been largely satisfactory, with 99 households resettled on time and within budget during the Stage 1 resettlement. Stage 2 resettlement experienced delays and cost-overruns, with 216 households relocated, 44 more than appraisal estimates. Stage 3 resettlement will assist 103 households that can no longer sustain their existing livelihoods due to land lost to Mohale reservoir and which cannot be offset by realloca- tion of other available land. Infrastructure, including access and feeder roads, schools and clinics, were all successfully provided, although with some delays. The water and sanita- tion program has exceeded initial targets, having provided piped water systems, 2545 VIP toilets and other facilities to 126 villages throughout the Phase 1B project area which had no piped water or sanitation at appraisal. - 9 - 32. The compensation framework included individual household compensation comple- mented by communal compensation programs for the loss of arable land, communal as- sets and downstream resources resulting from changes in instream flows. As land could not be valued at market rates, the project agreed to pay the value of agricultural produc- tion over a period of 50 years, with provisions for early withdrawal of lump-sums for ap- proved business investments. Excluding income restoration and training programs, M83 million has been paid out in compensation for project-related losses, with payments to continue, where still owed, for a further 40 years. While the policy framework is sound there have been implementation difficulties. Individual compensation payments were largely made on time. Advanced payments of lump sums supported by approved business plans have been made in 237 cases to support specific investments, with 360 of the 425 affected households (85%) having chosen to move out of the Highlands to Maseru or the foothills. Those below a "minimum threshold" of annual income (M3,760 at 1997 prices) have income supplementation for 10 years. However, communal payments have proven problematic, because the institutional apparatus for payment and utilization of compensa- tion have proven extremely difficult to establish. In addition, the project has paid out M27 million as a first 10-year payment under the IFR policy for compensation of pre- dicted downstream losses of natural resources to 54 communities in proximal reaches. Implementation weaknesses identified include insufficient information, a lack of techni- cal support to communities resulting in saturated markets, accelerated drawdown of cash investments, and poor record keeping. The LHDA, in collaboration with the GoL, has taken steps to address these issues. 33. The LHDA successfully implemented a comprehensive stakeholder consultation process. While the Community Liaison Assistants proved too costly to maintain after the completion of construction, the CALCs continued to provide an effective mechanism for engaging project-affected communities. These were complemented by an annual stake- holder conference to assess performance and provide feedback to management, and inter- nal grievance procedures. These procedures were under-resourced and did not perform adequately, resulting in aggrieved parties taking their complaints to the Lesotho national Ombudsman. The Ombudsman issued two major reports on the LHWP. Each report re- viewed several hundred issues from those affected by the LHWP. A substantial number were either rejected by the Ombudsman or were non-actionable complaints about policy. All but four of the first report's items have been resolved, but 107 (27%) of those arising from the second report are still pending. A third report is in preparation. The role of the Ombudsman as an independent, third-party arbitrator is a positive sign for civil society and accountable governance. 34. Environmental Components: Significant achievements were made under the envi- ronmental measures implemented in Phase 1B, particularly on downstream issues. All programs implemented under contractor arrangements were successfully executed and all disturbed sites were successfully restored. The eight site fatalities exceeded the zero tar- get as did the disabling injury frequency rate. However, both these indicators improved upon Phase 1A implementation and are within industry norms for large infrastructure construction projects in southern Africa. The environmental and social PoEs played an important role in supporting LHDA with technical advice on implementation issues. Op- - 10 - erationalisation of three nature reserves has bought 7,574 hectares of cultural heritage and biodiversity under protection in a country with only 0.4% of its land area so protected. The Katse Botanical Garden has entered into a number of international cooperative ar- rangements and seed sharing programs, with successful propagation of the rare and en- dangered endemic spiral aloe (Aloe polyphylla). 35. To address downstream issues, the project undertook a detailed assessment of the en- vironmental flow requirements for river reaches downstream of the Mohale and Katse dams. The IFR studies were undertaken during implementation of Phase 1B, after the EIA and engineering designs had been completed. These used a scenario-based method- ology for predicting Downstream Responses to Imposed Flow Transformations (DRIFT), within the framework of a river classification system, to address biophysical and socio- economic issues. These studies, complemented by an economic analysis of the various scenarios, led to the acceptance of an adaptive management approach. This approach sig- nificantly advanced the scientific foundations of IFR and enabled the development of an IFR policy and procedures designed to provide for the sustainable maintenance of tar- geted river conditions. These included performance standards to guide the mitigation of impacts, assessment of socio-economic losses and compensation values. The process re- sulted in changes to the dam outlet valves and operating rules, and an increase in the minimum flows specified in the Treaty by 300 to 400%. Management systems allow for the systematic review and audit of the performance standards, with the first audit being completed in March, 2007. The audit found that implementation has been approximately 60% fully or partially compliant with the Policy and Procedures and has identified issues likely to affect the sustainability of the IFR. It must be noted that the IFR work described here is emerging science with many of the elements of the Policy and Procedure having been developed during and through the project. The audit concluded that the LHWP's approach to IFRs was at the forefront of global practice and, building on this, proposed 100 "improvement action recommendations", including preparation of a manual to im- prove operationalisation of IFR releases. LHDA has committed to addressing these rec- ommendations, along with the optimal release of deficits in the IFRs, as they endeavor to improve performance, defining a progressive approach to IFRs and global best practice. 36. Funds available for technical assistance to support management and training were un- der-utilized, with only M392 million of M600 million disbursed, despite the clear need for building capacity. The important Integrated Catchment Management (ICM) program was started in May, 2005. This encompasses many of the environmental monitoring pro- visions and programs aimed at providing a platform for integrated planning and sustain- able development through improved community management of natural resources and other development activities,. 37. One of the highly significant project-related impacts identified in the EIA was the impact of the project on the Maluti Minnow. An action plan was agreed in December 2003 based on a population inventory and genetic assessment and 1,711 fish were trans- located to four reaches without predatory species in an effort to establish protected popu- lations. Having made initial resource allocations to implement a program of protection, M9.15 million was re-allocated in 2003/04 to provide seed capital for establishing the Lesotho Biodiversity Trust (LBT). Mandated with developing sustainable programs to - 11 - ensure preservation of the Maluti Minnow, the LBT started operations in 2005 and is monitoring existing populations. Three of the four translocated populations appear to have successfully adapted to alternative sites. Despite these efforts the protection of the Maluti Minnow was not secured by the end of the World Bank loan period. A proposal to create a physical barrier to protect populations from the migration of trout into streams upstream of the dam, although planned, had not been implemented by the time of the clo- sure of the World Bank loan. 38. Institutional Arrangements: Many of the Phase 1A institutional issues identified during appraisal persisted during Phase 1B. Cumbersome relationships and confusion re- garding institutional roles and responsibilities among LHDA, its Board and the LHWC negatively affected project performance. During 2005 four members of the LHWC were appointed to the LHDA Board. Although this achieved the objective of making the exper- tise of the LHWC available to the Board, it was never formalized by an amendment to the LHDA Order or gazettal of the appointments. Such action raised a high risk of conflict of interest and of compromising the countervailing checks and balances otherwise provided by an independent LHWC and Board. The possibility of LHDA undertaking further water resources development projects for the GoL according to its mandate was effectively pre- cluded by changes to the Board's membership and accountability. 39. The process of internal controls and approvals retains a number of inefficiencies. Placing such a high-profile project, with national impact, under the Office of the Prime Minister or under the Council of Ministers rather than the sectoral ministry (Natural Re- sources) may have facilitated greater ownership among the GoL and broader develop- ment opportunities. At the end of construction the LHDA was restructured in accordance with its change of function from implementation to operation and maintenance of the LHWP. Many of the developmental functions and assets of the LHDA were transferred to the relevant GoL departments. In many instances these departments did not acquire in- cremental budget resources as part of the transfer process, resulting in a decline in quality and poor maintenance. Resource and capacity constraints of the GoL undermined many efforts to broaden the project's potential development impacts. 40. Implementation Supervision: The Bank played an important supervision role. De- spite the small financial contribution, limited largely to institutional support and training, the Bank's supervision covered all project activities and provided comfort to other lend- ers through internationally accepted benchmarks for implementation practices. The high profile of the project, combined with the need to respond to increased environmental and social requirements, resulted in an intensive supervision effort. The Bank dedicated sig- nificant resources to supervision and responded to changes in implementation status with increased resources. Supervision resources were applied effectively and the Bank pro- vided continuity through a committed supervision team with a set of core competencies that were consistently applied throughout the implementation period. Long-term in- volvement of the supervision team's core members fostered credibility with the imple- menting agency, the Borrower and the governments of both Lesotho and South Africa. It enabled the Bank team to develop detailed knowledge of the project and promote trans- fers of skills and improving corporate capability. The Bank's supervision team was seen as providing an important review mechanism as partners in the development process. - 12 - This effort complemented that of the independent PoEs that addressed engineering, envi- ronmental and social issues. The total Bank budget for supervising all aspects of the LHWP during implementation averaged US$266,000 annually. Despite the higher than average absolute costs for supervision, the relative cost of supervision within the context of the total costs associated with Phase 1B show that the Bank was able to achieve sig- nificant leverage with limited resources. An assessment in 2002 by the Bank's Quality Assurance Group rated the supervision effort as satisfactory. 2.3 Monitoring and Evaluation (M&E) Design, Implementation and Utilization 41. A detailed results framework was defined during appraisal. Although considered "far too all-embracing" (QAG 2002), the indicators proved relevant in monitoring progress during implementation and evaluating overall performance toward achieving project ob- jectives. A comprehensive set of realistic, measurable indicators were clearly linked to project components. Monitoring of economic and engineering impacts related to specific quantitative targets for cumulative royalties, completion dates and cost targets for civil works, and employment impacts were supported by specific studies to determine macro- and micro-economic impacts. Monitoring of environmental and social impacts associated with the construction was defined under the contractual provisions for the major civil works. However, the arrangements for monitoring and evaluating the broader develop- ment objectives relating to environmental and social issues and poverty alleviation were less satisfactory. Many environmental and social indicators were not defined at appraisal, despite detailed environmental baseline studies. Insufficient management focus, indeci- sive procurement and lack of capacity to supervise socio-economic and environmental studies resulted in discontinuities between the three socio-economic monitoring contracts to determine the impacts of resettlement. As a result, data were not collected regularly or reported in a timely or effective manner. These problems have complicated determina- tions of project impacts on Highland communities and the extent to which the Treaty provisions have been met. 42. An internal Environment and Social Services Group (ESSG) with GIS capacity was assigned responsibility in LHDA for monitoring and evaluation. The ESSG was reorgan- ized with a new monitoring and evaluation unit established in 2003 to consolidate M&E and transition toward the operation and maintenance phase. This unit provides a central mechanism for coordinating, collecting and analyzing data and reporting to management to inform decision making and resource allocation. An appropriate monitoring and evaluation framework is in place but lacks the rigor and focus required to ensure efficient linkages between operational aspects and management decisions. Given the detailed framework, Bank supervision focused on continuous monitoring of prioritized implemen- tation indicators. The LHDA is developing a Management Information System to im- prove management and record keeping and improve the provision of compensation enti- tlements. 2.4 Safeguard and Fiduciary Compliance 43. The 1998 template for the PAD did not include specific provisions for listing Safe- guard policies triggered. Based on the review of safeguards arrangements during supervi- sion, the project triggered: - 13 - (i) Environmental Assessment (OP/BP 4.01) (v) International Waterways (OP 7.50) (ii) Involuntary Resettlement (OP/BP 4.12), (vi) Financial Management (OP 10.02) (iii) Natural Habitats (OP 4.04) (vii) Procurement (OP 11.00) (iv) Safety of Dams (OP 4.37) (viii) Disbursement (OP 12.00). 44. The Bank supervised the project regularly and an ISR was consistently prepared for each mission, with due attention to safeguard and fiduciary compliance. The 22 ISRs ex- hibit a clear consistency in rating each of the safeguard policies, although changing re- porting requirements for the ISRs meant that not all categories were rated on all 22 occa- sions. While overall Implementation Performance was considered satisfactory throughout the original loan period, ratings of Safeguard Performance were only considered satisfac- tory in four of the 13 ISRs for which it was rated, mainly due to non-compliance with the IFR Policy, delays in formulating the approach to stage 3 resettlement and weak perform- ance of other social protection programs. All satisfactory ratings were during implemen- tation, and ratings declined during the two extension periods. One of the justifications for further extension to the original loan period was to address residual social, environmental and institutional issues. A failure to adequately address these issues during the extensions resulted in the overall performance ratings slipping to Moderately Satisfactory and Mod- erately Unsatisfactory during this period. OP/BP 10.02, Financial Management and OP/BP 12.00, Disbursement 45. At the closing date of December 31, 2006, US$21.66 million of the original loan amount had been disbursed. Three cancellation requests were made due to revised cost estimates and lower-than-anticipated uptake of technical assistance. Financial Manage- ment performance was considered satisfactory during implementation but was down- graded to Moderately Satisfactory during the extension periods due to unreconciled cost allocations, and not due to safeguard issues. Accounting records were maintained effi- ciently and in chronological sequence, facilitating the validation process. Audits were submitted on time annually and with an unqualified opinion. Early audits were not, how- ever, prepared in line with IBRD expectations. This was due to the lack of a separate opinion on the Special Account and SOE procedures, compliance with International Au- diting Standards and accounts prepared in accordance with professional pronouncement of the International Standards Committee. These were rectified during the mid-term re- view. Internal auditing, however, remained problematic and, despite continued question- ing by the Bank's supervision team, the Borrower failed to adequately address internal audit capacity deficiencies. Had they done so, subsequent problems associated with poor record keeping for compensation payments may have been identified and corrected ear- lier, and the very costly and under-performing water and sanitation program replaced sooner with an in-house team. Reconciliations were done in a timely manner and inde- pendently reviewed. OP/BP 11.00, Procurement 46. Procurement was rated as satisfactory in all six ISRs for which it was assessed. The Bank team provided assistance during supervision and through a series of procurement clinics. Procurement plans were clear, with defined cost estimates, and were reconciled and updated to reflect LHDA's technical assistance strategies and priorities. The need for - 14 - LHDA to obtain internal approvals prior to requesting the Bank's "no objection" delayed submission and extended the time required to comply with procurement procedures. In response, the Bank agreed to a parallel review process, with Bank clearances requested at the same time as internal approvals, with provision for re-submission in the event that any terms and costs differed from that originally submitted. Delays resulting from the transmittal of original documents were resolved through agreement to use electronic transfers and clearly defined lines of communication. OP/BP 4.01, Environmental Assessment 47. The project was assigned `Category A' under OP 4.01, Environmental Assessment. Implementation of the Environmental Component was rated as satisfactory throughout implementation, with the exception of ISR16 in 2004 due to increasing non-compliance with the IFR policy. In contrast, the Safeguard Performance relating to OP 4.01 was rated as satisfactory by 12 of the 19 ISRs for which it was rated. Performance toward the end of implementation and during the period of the loan extension was rated as Moderately Satisfactory, Moderately Unsatisfactory, or Unsatisfactory, reflecting a failure to ade- quately address unresolved environmental issues. Although it was originally restricted in scope, the EIA identified 140 impacts, including 29 downstream and 48 other impacts. Of these, 31 were of high significance and two of very high significance: the loss of life through HIV/AIDS and extinction of the Maluti Minnow. Failure to implement all of the measures which would have better protected the Maluti Minnow demonstrates a lack of commitment from the GoL and raises broader questions as to the long-term sustainability of many environmental measures. OP/BP 4.04, Natural Habitats 48. The Natural Habitats policy was recorded as satisfactory throughout implementation. However, many of the persistent issues addressed under the EA policy related to the im- plementation of IFRs, the objective of which is to ensure the protection of downstream habitats. The conversion of a flowing river into a reservoir of standing water through im- poundment changed the upstream habitat and the transfer of water out of the basin results in degradation of the downstream aquatic habitat. The approach to use OP 4.01 as the op- erational instrument with which to ensure the protection of instream flows and aquatic habitats reflects a pragmatic decision providing a simpler approach that was better under- stood by the implementing agency. OP/BP 4.12, Involuntary Resettlement 49. Resettlement and compensation issues were identified as a major challenge during preparation, being more complex, and qualitatively and substantively different from Phase 1A. While the physical relocation and individual compensation of those directly affected by the loss of physical assets has generally been well addressed, broader devel- opment issues associated with livelihood restoration programs and downstream impacts have been more complicated and less satisfactory. Production in the Highlands is fragile; families had survived by means of remittances from household members in the mines of the RSA, who had been largely repatriated in the period just before the project began. De- termining the impacts upon affected households has been undermined by inefficiencies in the design of monitoring and evaluation systems. Results show that those affected by the - 15 - project are in about the same financial condition as before, with compensation having re- placed own production. There is some evidence that after a period of transition the reset- tlers are beginning to invest in new livelihood strategies, with membership in agricultural programs growing and purchases or rental of croplands increasing. However, the sup- plementary development programs that were to have been implemented have suffered from a lack of commitment from the GoL and the transition is probably slower than it might have been. OP/BP 4.37, Safety of Dams 50. The Safety of Dams was rated as satisfactory throughout implementation and the ap- pointed PoE provided high caliber advice. The risks of siesmicity, which was problematic during Phase 1A, was better addressed in Phase 1B through the establishment of an emergency preparedness plan, improved downstream communications and the installa- tion of siesmicity instrumentation. All of these measures proved important when in early February 2006, heavy rains caused the Mohale reservoir to fill rapidly to a spilling condi- tion. This resulted in significant dam settlement, along with downstream and cross-valley movements of the crest, which resulted in a crack forming in the concrete facing and an increase in seepage from about 50 l/s to around 600 l/s at the peak. A second crack at greater depth took place around March 16, 2006, although there was no evidence of in- creased seepage. Inspections by both the PoE and the Bank's dam specialist shortly after the failure confirmed that it presented no risk to the integrity of the dam. A study was commissioned in September 2006 to determine the extent and nature of the damage, de- termine the potential for further movement, and to prepare a work plan for remedial measures. It confirmed that the crack did not present any dam safety concern or major financial implications, with economic impacts limited to the cost of the repair works. OP/BP 7.50, Projects on International Waterways 51. The Projects on International Waterways safeguard was consistently rated as satisfac- tory or above throughout implementation. The Senqu River lies wholly within Lesotho in the headwaters of the Orange River, which flows through South Africa to discharge into the Atlantic Ocean. The Orange River catchment also includes parts of Namibia and Bot- swana. The Orange River mouth was declared a Ramsar site in 1991 and, following the collapse of the saltmarsh component of the estuary in 1995, was placed on the Montreux Record. The rapid degradation was attributed to diamond mining activities and the cumu- lative impacts of dam construction in the Orange River catchment. Bi-national relations with the riparian states were formalized with the establishment of the Orange-Senqu River Commission (ORSECOM) on November 03, 2000. 2.5 Post-completion Operation/Next Phase 52. Phase 1B built upon the foundations of Phase 1A to consolidate the LHWP's reputa- tion as a world-class, cooperative infrastructure project in a poor and under-developed country. The Bank has played a pivotal role through more than two decades of involve- ment, especially in the introduction and implementation of environmental and social safeguards. Sustainability of the LHWP is dependent upon (i) successful completion of all project components; and (ii) adequate ongoing operations and management of both Phase 1A and 1B. - 16 - 53. Economic Issues: Continued economic growth and water demand in South Africa will ensure that the water transfer components remain financially sustainable. With reve- nues paid from a proportion of the Vaal River water-user tariffs, servicing of the loan is assured. Attention to financial management capacity within LHDA, in particular to O&M costs, will support this. Failure to address the current tariff structure and operational ar- rangements will continue to undermine the financial sustainability of the 'Muela power station. The GoL's Poverty Reduction Strategy, supported by a solid budget allocation framework, will ensure that project revenues continue to provide an important source of revenue in support of poverty alleviation and economic stability objectives. The transition of project-affected people from dependence on compensation payments toward higher incomes depends on active investment of compensation in additional income generation. 54. Engineering Component: The project infrastructure is well constructed, based on mature technology and supported by strong staff and technical capabilities. The failure in the concrete facing of the Mohale Dam is receiving attention. The infrastructure should achieve its design life and meet all safety requirements. 55. Environmental and Social Components: The achievements of the project in imple- menting the Resettlement Action Plan including an innovative approach to compensation and livelihood reestablishment have also been acknowledged as world class (see para- graphs 31 to 33). This is ongoing work, however, which extends beyond the closure of the World Bank loan. Several key components of the LHWP, encompassing central ele- ments of environmental and social sustainability, are incomplete. These include securing the future of the Maluti Minnow, commitment to the arrangements for and management of communal compensation, capacity building in enterprise management, and following through on the project's development program. In addition, the project's impact on HIV/AIDS remains uncertain, and planned prevention and support actions in affected communities have not yet been fully implemented. 56. The IFR policy, procedures and monitoring system designed for the LHWP are glob- ally precedent-setting and central to maintaining downstream environmental sustainabil- ity and protecting natural-resource-based livelihoods. Implementation in the absence of a strong water-resources or environmental regulatory framework is proving difficult (e.g., missed flood flows) and, based on current governance and systems, environmental sus- tainability may suffer over time. Specifically, downstream monitoring and evaluation ac- tivities are lagging and are of moderate quality. The national framework and internal ac- countability both for ensuring environmental flows and for reporting flow operations are unclear, with successful implementation at risk from inadequate systems, communication and cooperation at the operational level. 57. Institutional Arrangements: One of the primary objectives of the project was to put in place the physical and managerial capacity to transform water into export revenues. The institutional arrangements have proven successful in realizing this objective, and LHDA has in place the managerial capacity in terms of systems, staff and governance structures to operate the infrastructure. However, managerial and institutional sustainabil- - 17 - ity is likely to be a challenge in the medium term due to internal organizational and gov- ernance issues, particularly the relationship between the LHDA executive, the Board and the LHWC. This is particularly important as there is no regulatory framework to ensure compliance, accountability or transparency of operations. The project has put in place initial key operating parameters to realize multi-purpose objectives, but there does not appear to be strong capability or inclination to approach decision-making from a multi- ple-objectives framework. 58. Transition Arrangements and Recommended Bank Follow-up: The current transi- tion plan includes ongoing monitoring of compensation and resettlement by LHDA to the end of the program in 2044 and continued involvement of the social/environmental PoE to advise LHDA. However, these activities are unlikely to mitigate the sustainability risks identified above. A set of transition activities that address managerial systems and prac- tices is needed, such as adjusting the composition of the PoE, ongoing technical assis- tance for monitoring and internal governance, the introduction of an Environmental Man- agement System, and targeted skills development. The Bank-funded Water Sector Im- provement Project, which is part of the ongoing country program, is addressing the legal and regulatory framework for the sector and could be used to address these sustainability needs. Given the unique and progressive nature of the IFR work, there is value in con- tinuing to monitor implementation, particularly following-up on the recommendations of the IFR audit. 3. Assessment of Outcomes 3.1 Relevance of Objectives, Design and Implementation 59. The objectives of Phase 1B of the LHWP continue to be highly relevant in the context of Lesotho. Dependence on the regional economy, in particular South Africa, and fluctua- tions in the textile industry continue to make Lesotho susceptible to changes beyond its control, which increases the importance of translating its rich abundance of water into sustainable revenues. Continued development of least-cost sources of water remains highly relevant to the expanding South African economy. 3.2 Achievement of Project Development Objectives 60. Of the four explicit objectives under the PDO three can be considered to have been met or exceeded. Water export revenues have exceeded expectations and made signifi- cant contributions to the limited economic base in Lesotho, mitigating the economic im- pacts of a changing national and regional economy. The impact of royalty revenues on poverty is difficult to determine. Despite the failure of the LFCD, the government's solid budget allocation framework, commitment to social spending and the Poverty Reduction Strategy provide appropriate mechanisms for the application of royalties to poverty alle- viation. The project has also increased access and improved infrastructure along with wa- ter supply and sanitation in the Highlands, in addition to significant short-term employ- ment benefits for local communities (over 9,000 jobs generating nearly M400 million in earnings) and national companies (over M623 million). Water supplied to Gauteng (7,034 Ml per day) through the Vaal system has continued to support service expansion to previ- ously disadvantaged communities and economic growth in Gauteng, which contributes 10% of the GDP of the entire African continent. Direct environmental impacts associated - 18 - with the civil works were mitigated or successfully rehabilitated, and the project has re- sulted in the formation of a number of nature reserves, the Katse Botanical Gardens and the Lesotho Biodiversity Trust. While specific PDO indicators for environmental and so- cial impacts have been met, addressing resettlement within a development framework remains a challenge and ensuring long-term environmental sustainability is susceptible to shifting priorities, requiring a continued commitment and the maintenance of implemen- tation capabilities. Box 3: Poverty Alleviation and the LHWP Lesotho continues to be one of the world's poorest countries, with one third of the population living on less than US$1 per day. Since 1995 the Human Development Index has declined and Lesotho now ranks 28th lowest on the HDI. One of the primary objectives of the LHWP was to generate export revenues that could be applied to poverty reduction and economic stability. This could be achieved either through the government's normal function of public budget allocation or through a specifically targeted off-budget mechanism to channel all or part of the funds directly to poverty reduction projects and activities. A Development Fund was established in 1992 un- der Phase 1A to channel part of the project revenues to community-driven development projects in Lesotho. Inadequate criteria for project approval and poverty targeting, poor construction quality and inadequate op- erations and maintenance resulted in the Development Fund being replaced with the Lesotho Fund for Community Development (LFCD) in March, 1999. LHDA was replaced as manager by a Board, reporting to the Prime Minister, with technical support for project preparation and implementation. The LFCD, supported by a US$4.7 million Learning and Innovation Credit, was a prerequisite for World Bank support of Phase 1B. However, the LFCD did not perform adequately for a range of reasons and was rated as Highly Unsatisfac- tory. The route of off-budget poverty reduction mechanisms, in the case of Lesotho, was not successful. Through the formal budget framework, however, the LHWP has made significant contributions to economic stability and poverty alleviation through contributions to government revenues. Cumulative royalties from Phase 1B have contributed M1,918 million to the national economy, with Phase 1B contributing an average of 4.8% to the GDP. Public expenditures account for about 40-50% of GDP and the GoL has a sound budget allocation framework. The country's Poverty Reduction Strategy includes a number of activities to address poverty and nearly two-thirds of projects funded under the 2004/5 development budget were deemed to be directly pov- erty related. Social sectors account for roughly 40% of public expenditures, with the GoL providing free pri- mary education and expanding health care. The LHWP experience demonstrates that off-budget support, within the context of a solid allocation frame- work, may not be an appropriate mechanism for using project revenues to address poverty alleviation. 3.3 Efficiency 61. Economic Rate of Return: Phase 1B had a lower than expected ERR due mainly to much lower water demand growth than projected at appraisal: from 15.9% in 1998 to 11.5% in 2006. An increased allocation to environmental flows also reduced the project yield and hence the ERR. While a detailed discussion is presented in Annex 8, three im- portant points should be noted about the key driver, namely water demand in the Vaal River system. First, it was clearly not anticipated that water demand growth would de- cline so drastically shortly after implementation. Second, it is unclear whether the as- sumption of 1.5% p.a. growth in consumption for the coming decades is too conservative given actual recent growth rates. And lastly, the project helped bring about the reductions in water demand growth as it contributed significantly to the development of water de- mand management policies in South Africa. - 19 - Description Estimated at Estimated at Appraisal Completion Cost-Benefit Analysis: · Net Present Value (NPV) NPV=US$1.2bn at 10% NPV= US$230m at 10%* · Economic Rate of Return (ERR) ERR=15.9% (1998) ERR= 11.5% (2006) * The apparent large discrepancy between the NPV at appraisal and at completion is due mainly to the re- ductions in water demand that became apparent after appraisal. It is also, to a small extent, affected by the decision to increase IFRs. See Annex 4 for explanation. 62. Despite the changes in water demand growth projections, the ex post assessment sug- gests that Phase 1B carries an acceptable ex post rate of return. 63. As an efficiency analysis, the economic rate of return does not include the following indirect benefits and costs: value of capacity and institutional development, potential in- direct and induced economic benefits from supplier contracts and potential economic de- velopments around the reservoir. 64. Cost Effectiveness Analysis: A cost effectiveness analysis was undertaken at ap- praisal to compare Phase 1B of the LHWP with alternative supply augmentation schemes, specifically the Tugela/Vaal Transfer Scheme and the Orange Vaal Transfer Scheme. Phase 1B was found to be much less expensive than the two alternatives, with unit reference values being 30-45% lower. Since these alternative schemes have not been implemented there is no updated information available on current cost estimates. Since the project costs of Phase 1B have been lower than anticipated it is highly likely that the cost advantage over these alternative schemes was maintained or increased. Description Estimated at Estimated at Appraisal Completion Cost Effective- 20% cheaper than At least 20% cheaper than alternative options, as costs were ness Analysis alternative supply lower than at appraisal. options (Assumes alternative project costs remain unchanged). 3.4 Justification of Overall Outcome Rating Rating: Satisfactory 65. Phase 1B has achieved or exceeded the project's development objectives. The physi- cal capacity to transfer water is highly satisfactory, generating revenues in excess of ini- tial estimates. These have proved successful in strengthening government revenues in Le- sotho and providing water for Gauteng in support of continued economic growth and ser- vice delivery. Significant managerial and engineering capacity was developed within Le- sotho and proved itself capable of implementing large dam projects. The contribution of the project to the determination of environmental flows is of global significance, although long-term commitment and sustainability remain to be proven. The Resettlement Action Plan has largely been completed, successfully providing replacement housing and associ- ated services. Residual issues relating to Stage 3 resettlement, compensation payments and the income generation programs in support of sustainable livelihoods are scheduled to continue until 2044. Delays in the implementation of a number of important social and environmental activities (e.g., ICM, compensation payments, Maluti Minnow) contrib- uted to shortcomings in achieving the project's social and environmental development - 20 - objectives, re-affirming the challenges of addressing resettlement and downstream im- pacts with a development framework. 3.5 Overarching Themes, Other Outcomes and Impacts (a) Poverty Impacts, Gender Aspects, and Social Development 66. The LHWP has made significant contributions to local community development, with short-term labor opportunities under Phase 1B providing 9,000 jobs, equivalent to 20,000 person-years of employment. This is considerably higher than anticipated and it is esti- mated that more than M294 million has flowed into the local economy through associated wage earnings. Those severely affected by Phase 1B are considered better off, with an average monthly income around M729, of which M326 is received as cash compensation. Resettled households average around M674 a month, of which nearly two-thirds is com- pensation. Monthly income among communities upstream of the Mohale dam average around M248 a month, not accounting for own-production. The impact of these compen- sation payments is difficult to analyze. Agricultural production and household nutrition in areas affected by the LHWP have followed national trends of decline. While improve- ments in Highland infrastructure have reduced traveling times to the Mohale region (99km from Maseru) from more than a day on unsealed, hazardous tracks to less than two hours on a high-quality paved road, the project has not determined the ancillary benefits. The water supply and sanitation program has increased coverage to nearly 85% of all households in the project area from zero in 1993. 67. The spread of HIV as a result of the large construction workforce associated with the project and improved transport corridors was identified as one of only two highly signifi- cant potential impacts relating to Phase 1B at appraisal. Lesotho has the 4th highest infec- tion rate in the world, after Botswana, Swaziland and South Africa, with antenatal sur- veys indicating that HIV/AIDS is disproportionately concentrated in urban areas. In Le- sotho the most plausible explanation for the high level of HIV/AIDS is due to infection through migrant miners working in South Africa (GoL/UNDP, 2003). The direct impact of the project on HIV prevalence is difficult to determine given deficiencies in monitor- ing and evaluation systems. In response to the epidemic the project prepared an action plan, distributed over 200,000 condoms to clinics, project areas and within LHDA along with 2,400 awareness pamphlets annually, and implemented a number of peer training programs and workplace strategy. The loss of public health staff after the re-organization and the failure of other GoL agencies to take up these programs have undermined their long-term contribution. (b) Institutional Change/Strengthening 68. The managerial capacity put in place under the project has successfully allowed Leso- tho to meet the project objective of transforming its principle natural resource into export revenues. Significant engineering capacity was developed through both phases of the LHWP, but was largely disbanded through re-organization to prepare for the transition from construction to operation and maintenance. Restructuring reduced the overall staff complement from 480 to 239. The process was highly inefficient with the recruitment policy and procedures proving overly complicated and cumbersome. A number of key staff were also lost to more competitive positions in the RSA. Key positions remained - 21 - vacant at loan closure, including financial and accounting managers, internal audit and the chief executive, who continues to serve in an acting role on short-term appointment. During this period (2004 to 2006) there was little staff training. There remains a need to resolve the lines of responsibility and accountability within LHDA and to clarify relation- ships between LHDA, the Board and the LHWC. Furthermore, accountability for meeting social and environmental benchmarks needs to be clearly defined. LHDA has undertaken impressive initiatives to provide transparency and external accountability, including the annual stakeholder conference and posting of reports to the web, including a number of commissioned audit reports. (c) Other Unintended Outcomes and Impacts (positive or negative, if any) 69. The intervention of the Ombudsman in providing affected communities recourse to a third party was an unintended but outstanding success. This may have tangible benefits to the GoL's current process of decentralization. Another unintended outcome was the ex- emplary example provided by the GoL in response to Phase 1A corruption. The case study is now widely used as an example of a how political commitment, adequate re- sources, and persistence can enable a small country to successfully prosecute large inter- national firms. The Bank team was also able to facilitate an agreement between the GoL and RSA enabling emergency releases from the LHWP in the event of flow in the Cale- don River falling to levels that cannot support communities dependent on that source for their water supply, including Maseru, the capital of Lesotho. 3.6 Summary of Findings of Beneficiary Survey and/or Stakeholder Workshops 70. Two stakeholder workshops and further consultations were held in Lesotho and South Africa to gather information on project impacts, discuss the project experience and derive lessons learned as part of the ICR. A broad range of stakeholders were identified with 163 invited in Lesotho and 70 in South Africa. Both workshops followed a structured fa- cilitation approach using a scorecard as the instrument for stimulating discussion and so- liciting input from participants. High attendance (over 100 participants) in Lesotho re- flected the high profile of the project, with participants representing all sectors, from government, NGOs and community based organizations, the private sector and special- ists. Low attendance in South Africa (10 participants) highlighted the lower profile of the project, with participants representing mainly government and bulk water suppliers, al- though NGO and civil society representatives were also present. 71. Stakeholders ranked various aspects of the project, positively highlighting the role of the Bank along with the importance of environmental and social considerations. While improvements in environmental issues in Phase 1B relative to Phase 1A were acknowl- edged, certain constraints limited their effectiveness. These included risks and impacts that were not fully addressed and the need for earlier and more proactive incorporation of environmental and social considerations into the design phase. Incorporation of IFRs into the project should be done early to inform the engineering design and financial modeling. Sufficient capacity within the implementing agency was recognized as important for en- vironmental and social issues to be adequately addressed. An important component of this is to ensure sufficient capacity for monitoring and evaluation. Widening communica- tion and stakeholder involvement was needed, with enhanced collaboration between the - 22 - partner countries at an operational level and better mobilization of South African re- sources to enhance regional cooperation and complementarity of resources, especially for monitoring and collaborative research. Functionality of the project should be optimized to ensure that the dams serve multiple purposes. 4. Assessment of Risk to Development Outcome Rating: Negligible or low 72. The physical and managerial capacity is in place to ensure continued transfer of water to South Africa. Economic growth and increasing demand for water in South Africa will ensure that royalties continue to accrue to Lesotho. There is negligible risk to the physical infrastructure over the next 75 to 100 years, although there are moderate risks associated with existing institutional arrangements. Moderate risks associated with the sustainability of some social and environmental development programs require attention to ensure long-term monitoring and evaluation complemented by institutional commitment from management and government. 5. Assessment of Bank and Borrower Performance 5.1 Bank (a) Bank Performance in Ensuring Quality at Entry Rating: Satisfactory 73. The engineering aspects associated with Phase 1B were well designed, drawing on lessons from Phase 1A. Detailed analysis of the project's economic and financial aspects during preparation provided a solid framework, clearly demonstrating the benefits of di- versifying the economy in Lesotho and providing the least-cost solution to Gauteng's wa- ter supply problems. Specific design provisions based on lessons from Phase 1A helped to ensure local development spin-offs and to maximize employment opportunities. Measures to mitigate social and environmental impacts linked to construction included clear measurable indicators, although those for determining the maintenance of living standards, downstream impacts and broader development objectives were not as well de- fined resulting in moving performance indicators, accentuating problems of ownership and commitment to environmental and social aspects. (b) Quality of Supervision Rating: Highly Satisfactory 74. Bank supervision encompassed all aspects of the project, not only those under the IBRD loan, and was exemplary. The Bank fielded 22 missions during implementation over eight years, with missions taking place at least every six months. Missions were typically less than 10 days and comprised two to 12 members. The Bank maintained a core supervision team comprised of a range of competencies, including engineering, op- erations, management, water resources, environment and social specialists, had strong support from Bank management and worked effectively to maintain good relations with the Borrower. The Bank's Quality of Supervision Assessment (2002) concluded that su- pervision of all aspects of infrastructure contracting and development were excellent. While the QSA concluded that the key issue of IFRs should have been tackled earlier and more forcefully this was not realistic given the development of IFR technologies, to which the project contributed, and would have delayed implementation of other project - 23 - activities. The issue of inadequate attention to using and improving the M&E system identified in the QSA evaluation was rectified, with the detailed results framework pro- viding an important tool for monitoring progress (see Annex 1). In addition to informing Bank management on implementation issues, aide memoires were also distributed widely among NGOs and other stakeholders with the permission of LHDA. This proved an im- portant factor in ensuring accurate reporting and also served as an important, independent monitoring and evaluation tool. Areas where supervision could possibly have been im- proved were in tackling the key issue of IFRs earlier and more forcefully, and in using and improving the M&E system. Support through regional Bank budget allocations, as opposed to country budgets, ensured adequate resources, with annual supervision costs averaging US$266,000. (c) Justification of Rating for Overall Bank Performance Rating: Satisfactory 75. Despite the relatively small financial contribution to Phase 1B, the Bank played a sig- nificant role in preparation and implementation. Technical and financial support provided assurance to international contractors and financial institutions, whilst enabling the Bank to act as an objective facilitator between South Africa and Lesotho. Application and ad- herence to the Bank's safeguard policies provided international benchmarks that ensured high-quality standards. However, discrepancies between the provisions of the Treaty, the PAD and the Legal Agreement introduced ambiguity in relation to the scope of the pro- ject's objectives, undermining commitment from some key stakeholders. 5.2 Borrower (a) Government Performance Rating: Moderately Satisfactory 76. In successfully completing Phases 1A and 1B of the LHWP, the Government of Leso- tho has made a very significant achievement of world class standards. During prepara- tion the GoL worked closely with the Bank team. The GoL strengthened the enabling en- vironment for water resources development, having adopted a water resources manage- ment policy in 1999 and appointed a Commissioner of Water in 2002 to coordinate water sector policy and planning. The commitment to combating corruption has been exem- plary. However the performance of the GoL, and the LHWC as the designee for the pro- ject under the provisions of the Treaty, has been mixed with some shortcomings. The GoL exhibited limited commitment to addressing the project's broader social and envi- ronmental objectives and failed to capitalize on a number of significant opportunities. Examples are the failure to resolve important issues on a timely basis (such as the LFCD and authorization of key appointments), the shortcomings in effectively transferring as- sets from the LHDA to respective government departments, the failure to resolve 'Muela hydropower issues and bulk electricity tariffs, and the failure to use the capacity devel- oped within LHDA to address water resource development needs in other parts of Leso- tho as provided for in the LHDA Order. (b) Implementing Agency or Agencies Performance Implementing Agency: Lesotho Highlands Development Authority Rating: Moderately satisfactory - 24 - 77. The performance of LHDA has been largely satisfactory, with some short-comings in implementation. LHDA is highly committed to achieving the development objectives and has successfully implemented the engineering and construction phases of the LHWP such that Lesotho has received revenues in excess of original estimates and South Africa has an assured supply of water that augments bulk supply in six provinces. Phase 1B drew on lessons learned from Phase 1A, resulting in greater involvement of Lesotho interests in the project and increased local economic development. Many challenges such as chang- ing environmental and social benchmarks have been successfully addressed. Stakeholder consultations and public accountability have been encouraged. Long delays in the imple- mentation of a number of social and environmental programs and concerns around their long-term sustainability, along with continued uncertainty surrounding the future of the Maluti Minnow reflect short-comings in the overall performance. (c) Justification of Rating for Overall Borrower Performance Rating: Moderately satisfactory 78. The Borrowers' overall performance is rated as moderately satisfactory. Successful completion of infrastructure under Phase 1B represents a major achievement. Organiza- tional efficiencies in implementation have been undermined by governance arrangements, the protracted restructuring and by resource capacity constraints. The absence of strong government institutions able to deliver services to the Highlands undermined implemen- tation, with LHDA having to assume responsibilities beyond its mandated responsibilities or resources. Less-than-satisfactory implementation of environmental and social pro- grams detracts from the engineering achievements and overall project success of Phase 1B. 6. Lessons Learned 79. The project represents the culmination of significant Bank involvement in the Leso- tho Highlands Water Project over more than 20 years. The project has responded well to the changing context and focus on hydraulic infrastructure over this period and a number of positive lessons have been learned (see Annex 9). The lessons to be drawn from the project are of importance to subsequent water resources development in Lesotho, to the World Bank and its involvement in future such projects, and to the broader global debate on the construction of large hydraulic infrastructure and their impacts. The lessons learned are necessarily summarized in this report - consideration is being given to compi- lation of a specific publication detailing the lessons learned through the LHWP. 80. Best practice confirmed. Many of the experiences of this project confirm lessons learned over the past decade in large infrastructure projects around the world. These are lessons such as:- a) Environmental issues need to be integrated into the design of large water storage and transfer projects and considered within an integrated basin-wide management frame- work (beyond the specific basin or sub-basin in which the project is situated) that ex- plicitly considers the cumulatively impacts of all developments and not only the spe- cific project under consideration. b) Monitoring and evaluation of large, complex projects needs to ensure that data is col- lected consistently, reported in a timely manner and integrating into the decision - 25 - process. To ensure data consistency and compatibility, M&E contracts should be de- signed to cover the full period of implementation. LHWP Phase 1B had multiple M&E contracts which, compounded with procurement difficulties, resulted in M&E shortcomings. Effective data management systems improve integration, adaptation and decision process. c) Economic analyses integrating full cost (and benefit) accounting that addresses envi- ronmental, capital and operating costs along with the broader cost/benefits associated with avoided costs strengthen the project design and decision process. Robust rates of return are required ex-ante to accommodate attrition during implementation. Addi- tional economic benefits can be derived from ensuring multi-purpose designs, incor- porating hydropower, flood management, tourism, fisheries and additional environ- mental benefits. d) Independent opinion through the establishment of Panels of Experts provides an im- partial mechanism facilitating expert input to implementation, particularly important in joint international projects, and contributes to capacity building and the transfer of knowledge. Selection for such panels should consider the relative merits of interna- tional experience against regional knowledge, and ensure independence of the Princi- pals in order to avoid any potential conflicts of interest. e) Engineering lessons derived from the construction of Mohale Dam have already been applied to the design and construction of subsequent concrete faced rockfilled dams globally, including the 198m high Kárahnjúkar dam in Iceland. Lessons include con- crete slab and joint design, quarrying techniques, rockfill layering, construction se- quencing and materials testing, both in-situ and in the laboratory. f) Joint development of international water resources, as demonstrated by the LHWP, can leverage potential benefits in excess of individual approaches as well as contrib- uting to socio-political co-operation. With at least 34 of Africa's rivers shared by two or more countries, such lessons are particularly relevant to maximizing opportunities for poverty alleviation through economic growth and development of the continents water resources. g) Combating corruption requires Government commitment, political will and persis- tence. Efforts to address corruption in public procurement are more likely to be suc- cessful in an environment of good governance that includes an independent judiciary, adequate resources for prosecution, an empowered civil society and a free press. The Bank has strong preventative mechanisms, including procurement procedures and le- gal frameworks, and supports capacity-building to prevent and pursue corruption, but lacks judicial powers to gather evidence and prosecute. More support is needed at the project level to develop indicators of potential corruption, including resources to en- able Task Teams to pursue allegations or suspicions of corruption. 81. Social objectives need to be clearly articulated - broad objectives such as "no af- fected person should be left worse off" need to be distilled in detail for the specific cir- cumstances of the particular project. Resettlement, implemented within a development framework, is a complex and under-estimated process that requires dedicated resources and realistic timeframes. Replacement for the loss of physical assets and cash payments are but two aspects of compensation that should be provided in a timely fashion and complemented by long-term interventions aimed at income restoration and livelihood - 26 - protection. Ensuring effective communication mechanisms, through a diversity of activi- ties, including liaison committees, stakeholder conferences, public information centers and individual liaison officers, along with grievances procedures and an independent, third-party adjudicator are central to effective participation of affected people. Clearly defined policies and procedures are important for defining roles, responsibilities, and stakeholder entitlements, while defining a framework for facilitating independent evalua- tion. 82. Environmental Flow Requirements to maintain ecosystem functions should be de- fined within a national river classification framework that enables strategic, integrated catchment planning to evaluate the impacts and trade-offs of different scenarios. These should include social impacts. Determination of environmental flow requirements is a long, iterative process based on complex science that does not always provide definitive quantitative results and should be based on the precautionary principle. A clearly defined policy and procedures supported by operational manuals identifying institutional roles and responsibilities ensure independence and accountability. The adoption of Adaptive Management principles in the LHWP Phase 1B project is an important lesson. This is necessary because the impacts of large infrastructure projects may only emerge over con- siderable lengths of time, beyond the period of a World Bank loan, and because of the large range of influencing factors and the uncertainty of how the environment will react to these factors. Incorporating the requirements of proximal and distal reaches into the IFR and accommodating both social and environmental requirements creates a complex, long-term management process which was innovatively addressed in the project and which provides important lessons for other such projects. 83. Procurement processes related to large infrastructure projects can either provide im- portant development opportunities to build local and national capacity and income gen- eration, or act as barriers to local enterprise. Based on the experience of Phase 1A this issue was specifically addressed in Phase 1B with considerable success. Building capac- ity takes time and needs to be nurtured through specific initiatives. Requiring bids to in- clude local partners can be complemented with initiatives facilitating partnerships be- tween local and international firms. Provisions such as the size and application of bank guarantees and requirements for advanced payment should be reviewed with the objec- tive of opening procurement possibilities to local and national contractors. In addition to contributing to the project at hand, such approaches contribute to the overall developmen- tal impact of large infrastructure projects and can be regarded as capacity building in- struments. 84. Institutional arrangements for the implementation of large infrastructure projects need to be clearly defined. Such projects often involve the formation of well-resourced extra-governmental implementing agencies. This needs to be regarded as a developmen- tal opportunity for the government with the potential to strengthen capabilities in related government departments and build agencies which will be better able to replicate similar projects. This potential was largely unrealized in the case of LHWP Phases 1 A & B due to a range of factors including political influence. A related issue is the transfer of assets from the implementing agency - the LHDA - to the respective government departments. - 27 - Generally the departments were poorly prepared and under resourced which may under- mine the sustainability of the assets. Project design should include how asset transfer will be achieved and consider budget allocation frameworks and clearly defined institutional responsibilities. Provisions should be made to track post-project asset performance and ensure they maximize potential benefits. 85. Design and Implementation. Current World Bank operational policies and proce- dures for project preparation - the Lending stage - do not provide sufficient time or re- sources for the preparation of large, complex projects such as the LHWP. Elements such as detailed design, resettlement and compensation planning, engagement of stakeholders, full environmental impact assessment including IFR design, building of client capacity etc. are too complex, lengthy and expensive to fit within the Lending stage of the project cycle. There is no formal mechanism to accommodate this. As the Bank increasingly reengages in such large infrastructure projects, different, often ad hoc, mechanisms are being used to undertake the initial preparation phases including incorporating detailed preparation work as components of other projects. One problem with this ad hoc method is that the host project of which the preparation is a component is often not a water re- sources development project and the project team is generally not aware of the complexi- ties of large infrastructure works and their preparation. The lessons learned in LHWP Phases 1A & 1B and in similar projects are that intricate and detailed work needs to be done in sensitive areas such as working with affected communities and assessment of en- vironmental impacts at the front end of the projects in order to avoid problems later on, and yet this crucial stage of project preparation does not fit well within existing instru- ments and project cycles of the Bank. Another related important lesson from LHWP Phase 1B is that it benefited greatly from the lessons learned in Phase 1A which illus- trates the benefits of regarding such large projects in a phased approach. Phasing large projects enables the lessons from early phases to influence the design of latter stages and the incorporation of changes in the global context for example processes such as the World Commission on Dams. This is clearly evident from the experience of Phases 1A and 1B of the Lesotho Highlands Water Project. 86. The World Bank's role. The role of the World Bank in large infrastructure projects has been changing in the past several years as the proportion of finance provided by the Bank decreases compared to other sources. The role is changing from a primary role of financing agent to adding value in other areas such as improving the developmental value of projects through the application of Safeguards, improved project preparation, capacity building/institutional development and in providing comfort to other investors. Stake- holders and LHDA acknowledged the valuable role of the Bank, but wanted to see more skills transfer and longer involvement in monitoring and evaluation. In this role, the Bank provides comfort to other lenders/donors and enhances the broader development impact and economic, social and environmental sustainability of the project. This role is most effective when integrated with complementary projects or activities in the country (e.g., through country assistance strategies or sectoral programs). In addition, supervision costs, supported through regional budgets, need to ensure continuity of teams that encompass a wide range of competencies. Trust funds provide an important instrument to support - 28 - learning and development and significantly add to the value of core project activities, leveraging greater development impacts. 7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners Building on the assessment of Phase 1B of the LHWP undertaken by stakeholders during the ICR workshops and the individual interviews, a range of stakeholders were invited to comment on the ICR. These included the Borrower and Implementing Agencies, along with co-financiers and other national and international stakeholders. In addition to the comments received on the project performance during the workshops and interviews, de- tailed written comments on the ICR were received from the LHWC and LHDA, the De- velopment Bank of Southern Africa (DBSA), and a joint comment from the International Rivers Network (IRN) and the Environmental Defense (ED). The DBSA was a cofinan- cier and both NGOs have a long-standing interest in the LHWP. The Borrower's evalua- tion is provided in Annex 7 and their comments have been addressed in the revisions to the ICR. The comments from DBSA and IRN/ED are provided in Annex 8. - 29 - Annex 1. Results Framework Analysis Objective Key Performance Indicators Actual Value Achieved at Completion Target Years Comments A. CAS OBJECTIVE Contribution in Lesotho to: A.1. Reduction of poverty; and (a) Improvement in key poverty indicators for Lesotho Declining UNDP HDI index since 1995 and increasing pov- Exogenous factors ­ loss of remittances, impact of ­ specially those related to access to services. erty. Increased enrolment through free primary education HIV/AIDS, drought. (68% in 1998 to 85% in 2002) & improved WS&S coverage. (b) Social Fund operational June 1998, with: Legally declared March 1999 Community Development Support Project ICR (P058050: Operational July 2000 Report No: 27944-LSO) provides detailed assessment of LFCD performance: Highly Unsatisfactory (i) appropriate poverty targeting mechanisms; Projects mostly rural access roads, footbridges, water supply Mechanism appropriate but poorly implemented, Opera- systems and soil conservation. tion Manual not adhered to. (ii) about M75 m per annum disbursed and 2001/02: M40 m 2004/05: M30 m Cumulative expenditure between 1991 (LHRF) and 2002 2002/03: M30 m 2005/06: M15 m (LFCD) was M200m. Plans are to move the Fund to new 2003/04: M14 m 2006/07: M 5 m local government Community Councils established mid Annual average: M22.33m 2005 and still to be capacitated. (iii) measures of poverty alleviation (e.g. km of ac- 101 projects identified: 58 projects completed, 26 suspended 1,100km rural roads, 210 small earth-fill dams, 60 foot- cess roads, income to communities, access to wa- and 17 ongoing by December 2006. bridges and some forestry and soil conservation works by ter). 2002. Average monthly employment was 3,702 workers on full- time equivalent basis. A.2. Enhancement of economic (a) Contribution to GDP: stability in Lesotho 1998: M260m (6.5%) 2000: M 237 m (3.8%) Average contribution of Phase 1B to GDP 1998-2007 = 2000: M183m (3.9%) 2005: M 2,702 m (4.8%) 4.8%. (b) Contribution to GoL Revenue Cumulative Royalties to 2006: M1,918 million. 1998: M145m (7.4%) 2000: M 227 m (9%) Hydro-power revenues to 2006: M411 2000: M197m (9.3%) 2005: M 1,594 m (3.9%) Contribution towards assisting South Africa in: A.3. the regional development of (a) LHWP continues to be a model for joint water Phase 2 feasibility studies initiated. infrastructure on a win-win basis development. A.4. contributing towards poverty (a) Access to urban water supply by local urban com- 1994 to March 2006, 3,474,137 people in Gauteng (32% of April 2006: an estimated 107,420 people remained with- alleviation and empowerment munities increases. population) were provided with improved water supply. out access to any form of formal water infrastructure. through effective management of (b) Productivity rates for water in urban areas The Gauteng economy averaged 3.3% growth between 1995 water resources. (c) 95% of incremental water going to high value users & 2002, compared to national average of 2.7%, and in- including industry and domestic consumption. creased its contribution to the national GDP from 32.6% to 33.9%. (d) Incidence of water restrictions in Gauteng de- Data not available. creases. - 30 - Objective Key Performance Indicators Original Target Values Actual Value at Completion Target Years Comments B. DEVELOPMENT OBJECTIVES B.1. Capacity in place for (a) Physical Capacity: Lesotho to transform its prin- (i) cumulative combined Phase IA and IB 12/2000: M 375m 12/2000: M 497.3 m cipal natural resource of royalties (in 1995 Rand) of: abundance ­water- into ex- 12/2004: M 650m 12/2004: M 948.8 m port revenues: 12/2006: M1,150m 12/2006: M1,162.0 m (ii) cumulative increased SACU revenues M745 million with M845 SACU M1,124m tax revenue by 2005. from Phase 1A and Phase 1B M370m of tax reve- M435m tax revenue M1,060m SACU revenue by 2007. nues by 2003 (b) Managerial and Institutional Capacity in place to manage the project: (i) localisation of key long term positions 1997: 3 of 26 mana- 2006: 3 out of 13 managerial positions filled by expa- 1997 structure for Phase 1B implementation lasted by 2003, without any loss of performance gerial positions held triates (HR, Finance * Divisional Manager) till 2003. by expatriates by 2003/04 re-structuring for operation and mainte- 2003 nance complete 12/2006. (ii) measures of LHDA institutional per- formance: · Audits < 4 months of FY end 5 months (average) · WACC < 15% 14.24 % (2003/04) · Financial cost < 33% 51.83% (2003/04) · Contractor credit <5% with >90 days 3.77% (2003/04) · Industrial disputes days lost <1.8% Achieved. Industrial Officer engaged from 1998 to 2003 to facilitate industrial relations. · New governance structure in place LHDA Order amended 2000. New policies/procedures developed (by 2003). Governance model refined 2003. Restructuring disrupted progress. · New Board appointed December, 1998 April, 2003 04/2003-03/2006 mandate extended 1 yr to oversee restructuring. · JPTC members appointed December, 1998 June, 2002 Renamed Lesotho Highlands Water Commission (LHWC). · LHDA staff performance 90% satisfactory Not yet completed: sample indicates >90% satisfactory Performance review process approved Oct, 2005. 2 cycles completed Dec, 2006. 3rd cycle to be com- pleted by April, 2007. · Key measures from LHDA transfor- Targets not defined Business Planning system adopted in 2001. Another organizational restructuring (downsizing mation process Performance management system (2000). for operation and maintenance) started in 2004, New governance in place (2003). with the recruitment process extending to the end Groups formulated own charters, visions and missions of 2006, but governance system remained un- aligned to those of the LHDA. changed. - 31 - Objective Key Performance Indicators Original Target Values Actual Value at Completion Target Years Comments · Successful renegotiation of Treaty December, 1998 Treaty renegotiation 1998-99, with the outcome of Protocol VI also required amendment to the 1986 Protocol VI signed in June 1999. LHDA Order. B.2. Lowest cost alternative (a) ERR based on consumer surplus analysis 16 % 11.5% See detailed analysis: Annex 4 for supply of water to the (b) Unit reference value of water of Ml.20 m-3 Phase1B costs < forecast, cost advantage over alterna- Gauteng Region compared to Ml.72 m-3 for next best alternative tives stands or exceeded. (at 12% IRR) B.3. Any adverse social and As specified in monitoring and evaluation environmental impacts miti- plan. Details below. gated (a) Workforce and Worksite (i) All disturbed sites restored to contract No baseline required Site rehabilitation completed by December, 2003. Highly satisfactory Contractor performance. impacts mitigated standards within one year of decommissioning Feeder roads (75 km) rehabilitation completed Decem- and to full use within 5 years ber, 2004. (ii) Death rates on site Zero 8 deaths (1998 to 2003) Phase 1A: 27 deaths (1991 to 1997) (iii) Disabling Injury Incident Rate <1 per annum Dam: DIFR = 7.361 South African National Occupational Safety Asso- Tunnel: DIFR = 7.160 ciation (NOSA) DIFR target = 1.5. Matsoku: DIFR = 7.950 DWAF construction target is 5, achieving ~3.4. Project Average: = 7.500 (iv) Observance of OH/IH standards ISR Minor Disab % Data based on external audit every 6 months. Ma- Tunnel: 1.65 843 187 22.2 jor works completed Sept, 2003. Matsoku: 2.70 355 47 13.0 Dam: 1.27 325 74 22.0 Project Total: 1.93 (v) LHDA and contractor severe accidents < 0.1% per annum Severe accidents range 0.1% to 0.4%; Major works completed Sept. 2003. 0.2% in last year of construction. (vi) Workforce STDs rates < baseline Syphilis Test < 7.9% Workforce STDs rates c10% of affected community Monitoring not regularly undertaken. levels (which are c100 per 1,000 patients). (b) Household standards of % Income Restored long term and short term H/h income >M2,500 1998/99 = M7, 860 / annum Varies among project impacted households: living have been maintained, food shortage preva- food shortage prevalence = 44% Not affected 41.7% if not improved (i.e. compen- lence < 77% 2005/06 = M8, 748 / annum Moderately affected 48.6% sation completed, Income food shortage prevalence = 51.8% Severely affected 46.4% Restored (c) Natural Communities and Key indicators not decreasing within Monitoring only undertaken once and recom- populations of key species Phase 1B area: mended 3 year intervals. exist in sustainable densities: ·Natural vegetation (i) Index of Range Condition Baseline established, monitoring under ICM. ICM Contract 1044 started May 2005. (ii) Vegetation biomass Baseline established, monitoring under ICM. ICM Contract 1044 started May 2005. ·Avifuana (iii) Index of avian abundance Baseline established, monitoring under ICM. ICM Contract 1044 started May 2005. ·Spiral Aloe (iv) Individual counts per site Not carried out. Kate Botanical Garden successfully propagating (v) Seedling/adult ratio per site Not carried out. and sharing seeds through international exchange programs. - 32 - Objective Key Performance Indicators Original Target Values Actual Value at Completion Target Years Comments ·Maloti Minnow (vi) Density of fish by sub-catchment 2003 Dec: Policy & Action Plan Approved. LBT mandated with protection but no barriers in Translocation of 1711 fish made to: place and populations remain susceptible to preda- Maletsunyane River = 600 tory invasive species. Makhaleng River = 300 Captive breeding populations have not been estab- Jorodane River = 300 lished. Quthing River = 400 (d) Downstream impacts (i) To be defined after IFR study June, 1999 IFR Study completed October, 2000 mitigated. (ii) IFR Policy & Procedures adopted Adopted July, 2003. Policy & procedures included performance targets and considered pioneering. (iii) Mitigation & Compensation Program in 23 of 54 Local Legal Entities (LLEs) established in Comprehensive approach at forefront of IFR im- place. proximal reaches (1,2,3) received M27 million com- plementation. pensation in 2004 (10 year tranche). Remaining Distal reaches communities to be determined later reaches (7 & 8 ­ Mohale downstream) are still being subject to impact verification monitoring. organized. (iv) Biophysical Monitoring Started July 2005 under Contract 1237. Consultant leads LHDA in-house team. (v) SocioEconomic Monitoring Started June 2005 under Contract 1204. Consultant leads LHDA in-house team. (vi) River Condition Targets maintained Data collection does not facilitate adequate assessment. See LHDA IFR Audit report for detailed consid- erations. (vii) IFR Audit After every 5 years of Compliant = 59 (24%) Audit completed March 2007. Success reflected by policy Partially compliant = 89 (36%) management commitment to process, which is highly satisfactory to date, and to implementation Non-compliant = 84 (34%) of recommendations. Not applicable = 14 (6%) (viii) Policy Revised Every 5 years after Policy to be reviewed following audit. Procedures reviewed and amended. IFR (e) Quality of water harvested 15 standard water quality measures do not Monthly water sampling, testing and report production National standards and guidelines for Lesotho still and delivery maintained exceed guidelines at any time. within acceptable limits of Rand Water and DWAF. being developed by GoL. (f) Public health maintained Key Health Indicators (morbidity and mortal- Baseline not determined. or improved for (resettled & ity) for Workforce, Resettlers and Host Com- Contract 1204 study (Dec. 2006) to assess. host) munities at least stable versus baseline 4. Local development spin- offs maximized: (a) Economic spin off effects (i) Jobs for Basotho 3,000 jobs by 12/03 9,000 jobs or 20,000 person-years of employment. 14,500 person years on main contracts only. from construction (16,500 ppl yrs). (ii) Basotho contracts generating M460 million by M623 million Main contracts only. Expected labor earnings were December 2003. estimated at M275million by Dec 2003 in the PAD (1995). Escalating these figures to account for inflation = M460 million. (iii) Local labor from highland communities 30% by Sept, 2000 24% (M138m earnings) All unskilled labor sourced locally. - 33 - Objective Key Performance Indicators Original Target Values Actual Value at Completion Target Years Comments (iv) Sub-contracting supply opportunities for M150 million by M300 million. M148 million from subcontracts and M152 from Basotho December, 2003 consultancy. (v) Stage 2 housing to local contractors (in- 60% LHDA team established due to high local contractor Successfully done in-house for significantly lower cluding owner built). costs. cost. (b) Local road infrastructure (i) Access times/cost of transport to clin- 99km from Maseru to Maseru to Mohale takes ~2 hours on sealed road. Infrastructure benefits not quantified: appears to be provided with wider benefit ics/shops reduced. Mohale took ~1.5 dys Access improved 1997 with feeder roads in 2004. the most valued benefit to highland communities than affected populations (see Contract 1204) (ii) Cost of basket of imported goods falls in Not determined. Benefits not quantified. Improved infrastructure, lower traveling times local market. likely reduced transport costs and cost of goods locally. (iii) Community infrastructure used Not determined. Not directly assessed. See contract 1204 and workshop findings for rec- ognized value. (c) Ecological responsibility (i) Catchment Community Conservation September, 1999 Not yet established. Contract 1044: ICM started May 2005; CCCs to be and awareness increased Committee (CCCC) fully operational formed by the 3rd year. Despite delays appears to among local communities be progressing well. with the full involvement of (ii) GoL reps fully active in CCCC September, 1999 See above. As above. GoL in integrated catchment management (iii) Ecotourism generates 10% of locally 2003 Eco-tourism program not yet initiated. ~M20 million spent on studies. generated income (iv) Attainment specific index of environ- No index determined. Study outsourced 1998/99, implemented by Mohale Combined Phase 1A and 1B study. mental awareness Field Office Branch. (d) Soil Erosion limited in the (i) Bare ground index declining To be implemented under ICM. ICM Contract 1044 started May 2005. catchment area (ii) Rate of soil loss in cultivated catchments below 12 t/ha/yr To be implemented under ICM. ICM Contract 1044 started May 2005. (iii) Areas impacted by gullies under conserva- To be implemented under ICM. ICM Contract 1044 started May 2005. tion regime (iv) All cultivated lands in catchment provided To be implemented under ICM. ICM Contract 1044started May 2005. with erosion protection (e) Public health improved for (i) Public Health System established by LHDA September, 2003 Leribe Trauma Unit & Lejone: 2003 all villages in the catchment, successfully transferred to GoL (Phase 1B Katse and Mohale facilities: 2004 i.e. over and above the reset- activities accepted in budget). tled and host communities (ii) Action Plan Prepared September, 2000 MoU (Action Plan) prepared 2003 but only signed in Hand-over negotiations were protracted. 2005. Divestiture Policy only adopted in 2004. Facilities operated at much lower standards by GoL. (iii) Key Health Indicators (morbidity & mor- Not defined. To be determined under Contract 1204. tality) improving - 34 - Objective Key Performance Indicators Original Target Values Actual Value at Completion Target Years Comments (iv) Antenatal HIV rates fall from 16% 14% by 2000 There are no records of ANC HIV tests HIV general prevalence summarized as follows: 1993 Community Baseline = 0.5% 12% by 2003 Antenatal clinics don't routinely test for HIV 1996 Community Survey = 6.3% 1999 Contract 669 study: = 21.1% 2000/01 Contract 669 study: = 28.9% 2005/06 Contract1204 survey = 23% 2000 National HIV prevalence = 23.7% 2003 National HIV prevalence = 28.5% (v) STD patients HIV prevalence falls from 38% by Sept, 2000 2000/01 study (C669) found STDs clinic attendances HIV was most prevalent among 15-19 yr olds 40% 35% by Feb, 2003 high. Clinic treatment for STDs higher among affected (67%). 37.5% of wage earners at risk. groups. (vi) Diarrhea episodes/child/yr. fall where to 6 by Sept, 2000 Not baseline determined. Phase 1A WATSAN reconnaissance study WATSAN in place. to 2 by Sept, 2003 Contract 1204 to assess. (2003/04) reported 15.8% of under-fives had an episode of diarrhea in last two weeks in Katse (20.9%) Lejone (13.2%) and Matsoku (15.3%) (vii) Helminthic infection reduced 10% by 2000 Baseline not determined. Helminthic infestations are not common in the 25% by 2003 Highlands & subsequent surveys did not report helminthes (viii) Stunting falls from 47% - 35% by 2000 40.7 % by 2000 (MRC) Routinely collected growth data (2000) for under - 10% by 2003 59.2% by 2002 (C.699) fives found less than 20% stunting in 'Muela (0- 10.8%), Lejone (0-3.5%) and Mohale (0-10%). 36.2% by 2004 (C.1204) (ix) Severe PEM falls from 7% - 5% by 2000 5.1 % by 2000 (C.669) Severe PEM is not reported in all follow-up sur- - 3% by 2003 4.7% by 2004 (C1204) veys. (x) Antenatal clinic attendance increases to - 70% by 2000 Baseline not determined. Only 45% of women delivered their last babies 70% by 2000 and 85% by 2003 - 85% by 2003 under the supervision of a medically trained health worker. - 35 - Objective Key Performance Indicators Original Target Val- Actual Value Achieved at Comments [Inputs] ues Completion Target Years D. Project Components D.1. Construction of main (i) US$530m US$530m US$377.79m (M2,500.91m) Completed well below appraisal estimates even with some contracts over Project Works budget. (ii) Contractors for Mohale Dam (145m) M550m M535.25m (US$70.35m) Completed within revised time (year later), within contract sum and budget. (iii) Contractors for Mohale Tunnel (32km) M915 M1,118.70m (US$155.56m) Completed 8 months later, above contract sum and budget due to claims. (iv) Contractors for Matsoku Diversion (15m M240m M228.90m (US$32.92m) Completed 2 months later but within contract sum and budget. weir and 5.6km tunnel) (v) Contractors for Infrastructure M485m M617.90m (US$119.96m) Completed within time but above budget due to increased scope. D.2. Engineering Supervision: (i) US$105m US$105m US$76.45m (M515.06m) Completed below appraisal estimates despite additional costs = Consultants to do detailed US$17.93million. design and supervision for: (ii) Mohale Dam M160m M129.68m (US$17.04m) Completed within budget. (iii) Mohale Tunnel M150m M189.01m (US$25.53m) Completed above budget and estimates due to extended construction length and claims (disputes) processing. (iv) Matsoku Diversion M60m M52.52m (US$7.19m ) Completed within time and budget. (v) Infrastructure M40m M52.14m (US$7.0m ) Completed within time but above budget due to increased scope. D.3. Environmental and US$140m US$140m US$82.7m (M622.17m) Social Action Plan (a) Resettlement of affected (i) Stage 1 Resettlement M19.7m M11.26m Closed: Stages 1 and 2 above estimates due to increased numbers and communities in three phases, (ii) Stage 2 Resettlement M35.1m M53.69m additional community infrastructure costs. including required social and physical infrastructure and a (iii) Stage 3 Resettlement M10.3m 0 Not yet started: modalities still being worked out. community participation (iv) Access Roads M145.9 M222.55m Closed: 75km all-weather gravel road & 3 and 3 bridges constructed with programme. in-house team; 3 bridges were tendered out. (v) Community Infrastructure M24.2m M16.81m Closed. (vi) Community Participation M5.7m M2.77m Open. (b) Compensation for project (i) Procurement and delivery of grain M3.5m M3.73m Closed. related losses: (ii) Arable land compensation M8.7m M65.02m Closed. · Verify physical losses · Develop and maintain (iii) Communal assets M11.1m M11.10m Closed. compensation register (iv) Individual Property M1.0m M1.53m Open. · Develop mechanisms for (v) Temporary land acquisition M1.8m M1.84m Closed. delivery of all forms of compensations · Effect delivery (c) Income Restoration and (i) Agriculture M22.1m M15.26m Incorporated into ICM Contract 1044, started in May 2005. Training (inc. credit facilities) (ii) Range Management M4.8m M0.39m Incorporated into ICM Contract 1044, started in May 2005. and Rural Development- (iii) Forestry M4.8m - Incorporated into ICM Contract 1044, started in May 2005. - 36 - Objective Key Performance Indicators Original Target Val- Actual Value Achieved at Comments [Inputs] ues Completion Target Years including crop diversifica- (iv) Tourism M8.0m M22.07m Closed. tion, forestry, tourism, Fisher- (v) Training and Income Generation M11.0m M3.72m Open. ies and Women in Develop- ment (d) Monitoring and Evalua- (i) Objective Impact M1.3m M10.60m Open. tion (ii) Participatory M0.6m - Open. (iii) M&E Unit support M1.1m M2.39m Open. (e) Natural Environment and (i) Integrated Catchment Management M7.7 m M8.28m Started in May 2005. Heritage including (ii) Environmental Resources Conservation M20.7 m M19.69m Open. - watershed management - environmental conservation (iii) Environmental Resources Monitoring M16.4m M10.42m Open. - mitigation of and compensa- tion for downstream environ- mental impact, - water quality monitoring, - environmental awareness and a rare and endangered species protection program. (f) Public Health Program (i) Healthcare Services & Monitoring for M42.2 m 0 Costs included in construction contracts for: Construction Workforce (ii) Leribe Trauma Unit (LTU) functional M22.3 m M21.20m Closed. (iii) Healthcare services for local communities M2.2 m M2.50m Closed. in project area (iv) Capacity building for ministry of health M1.1 m M0.53m Closed. (v) Water and Sanitation for communities in M15.0 m M112.85m Open. project area (beyond affected communities) D.4. Institutional Support and (i) US$110 m US$110 m US$84.32m Capacity Building: (ii) Short and long term Technical Assistance: M626.83m ·Engineering M344m M156.29m Closed. ·Finance & Admin M384m M65.39m Closed. ·Environment & Social Issues M600m M392.02m Open. ·Operation & Maintenance M48m M13.13m Included in operation and administration costs below. (ii) Training 0 Included in operation and administration costs below. (iv) Studies as required: M3.54m ·Instream Flow Requirements M15m ·Organization & Manpower Study M3.44 (v) Operation & Administrative costs for LHDA M421.97m Open. - 37 - Annex 2. Project Costs and Financing (a) Project Cost by Component (in USD Million equivalent) Components Appraisal Estimate Latest Estimate (12/06) Percentage of (USD M) (USD M) Appraisal Main project works 531.00 377.80 71.15 Engineering supervision 105.00 78.50 74.76 Environmental and social action plan 140.00 88.30 63.07 Administration and capacity building 108.00 84.30 78.06 Total Baseline Cost 884.00 628.90 71.14 Financing Costs 248.00 Total Project Costs 1,132.00 (b) Financing Source of Funds Appraisal Estimate Latest Estimate (12/06) Percentage of (USD M) (USD M) Appraisal Borrower 26.00 8.41 32.35 Development Bank of Southern Africa 47.00 32.31 68.74 European Investment Bank 109.00 45.34 41.60 IBRD 45.00 21.66 48.13 IBRD Phase 1A Loan 3393 8.00 8.00 100.00 Export Credit Backed non- CMA Commercial Loans 60.00 35.26 58.77 Non-CMA Commercial Loans (ECA Co-payment) 11.00 8.65 78.64 RSA Capital / Money Mar- kets and Water Users 826.00 503.43 60.95 - 38 - Annex 3. Outputs by Component Objective Key Performance Indicators Original Target Val- Actual Value Achieved at Completion Comments ues Target Years C. PROJECT OUTPUTS C.1. Major Infrastructure (i) Delivery of water from Matsoku Diversion 2.2 m3 /s water by 2.2 m3 /s of water by May 2001 from Mat- (Mohale Dam, Mohale Tun- March, 2001 soku Diversion nel and Matsoku Diversion) (ii) Delivery of water from Mohale to Katse 9. 6 m3/s by January, 9. 6 m3/s delivered from Mohale to Katse Impounded November, 2002 but only reached Full Supply completed to allow water 2003 by February, 2006 Level in January, 2006. transfer on time and within cost (iii) Overall Cost of Phase 1B limited to US$1.2 billion US$628.90 million (December, 2006) 2010 estimate: US$651 (73.6% of appraisal cost) US$1.2bn. 71% of appraisal cost (US$884.4m) 2044 estimate: US$843 (95.3%) O&M over project life- time. (iv) Mohale Dam ready for Impounding October, 2001 November, 2002 1 year behind due to unexpected geological conditions. (v) Mohale Dam and Tunnel completed within October, 2002 February 2002 The TBM finished excavating the 14.6km tunnel February cost Dam: US$125m - Mohale Dam: US $70.35m 2002. Water delivery from Mohale to Katse on February, 2006. Tunnel: US$208m - Mohale Tunnel: US$155.56m (vi) 59km upgraded Access Roads completed December, 1997 December, 1997 Completed within budget. (vii) Mohale Camp transferred January, 1998 June, 1998 Completed within budget. (viii) 21km new Maseru Bypass complete July, 1998 July, 1998 Completed within budget. (ix) Emergency Preparedness Plan implemen- Bank approved 1997. tation (x) Reservoir Induced Seismicity mitigation June, 2000 RIS system installed and operational2001 plan (xi) Estimated time and cost to completion of each contract at key benchmark dates: · By MTR (9/30/00) - Dam 60% / M300m 57% / M304 m - Tunnel: 80% / M725m 97.86% / M 1094 m - Diversion: 95% / M230m ~100% / M 229 m · Contractors mobilised: - Dam: April, 1998 March, 1998 - Tunnel: January, 1998 February, 1998 - Diversion: January, 1998 February, 1998 C.2. Compensation imple- (i) Communities and LHDA prepared for April, 1998 Sept., 1997: Policy approved The Policy also covered Phase 1A area in April 1998. mented for those impacted by commencement of new compensation policy April, 1998: Policy operational in all the project (as set out in the with all mechanisms agreed & in place, in- project areas EAP) cluding register - 39 - Objective Key Performance Indicators Original Target Val- Actual Value Achieved at Completion Comments ues Target Years (ii) Households and communities compensated Annual compensation (for lost arable Delays in 2005 led to public demonstration. per EAP in right amount and on time land) paid regularly and on time. Others (individual and communal assets, lump- sums, etc) started in 2002 and experience delays. (iii) Less than 5% complaints remain unre- Complaints resolution took much longer 396 complaints received. 43 were non-actionable com- solved after 6 months, 25% more than 3 than intended. Ombudsman involved in plaints, 126 rejected, 87 resolved before report published, months (by 9/00) resolution. 107 (27%) complaints remain unresolved. (iv) Arable land loss < 900ha 1275 ha No more than determined by Contract 1012 (Resettlement and Development Study). Any further land-loss (above full supply level) has been due to other project works such as advance infrastructure and feeder roads. (v) Grazing land loss < 1700ha 1635 ha No more than determined by Contract 1012.. (vi) Review of compensation policy completed September, 1999 Started in 2003 and ongoing. Policy proved sound in general, with minor amendments made. (vii) Detail Final Resettlement plans by: - Stage 1: Dec, 1997 - Stage 1: March, 1998 - Completed before start of construction. - Stage 2: June, 1998 - Stage 2: June, 2004 - Resettlement priority given to villages closer to water - Stage 3: June, 2002 - Stage 3: Not yet started. level so as not to delay impoundment. - Stage 3 concept is being finalized. C.3. Affected Populations (i) Resettlement of 99 h/hold pre-construction - Stage 1: 99 h/hold - Stage 1: 99 households moved by March - Stage 1 completed on time. Satisfactorily Resettled per (3/98), 172 h/hold post inundation (1/01), up resettled by Mar, 1998 1998. the EAP including 1 latrine to 89 h/hold post inundation (12/03) - Stage 2: 216 h/hold - Stage 2 resettlement started with those affected by full per hh, 301 water/hh, a gar- by Jan, 2001 - Stage 2: 216 h/holds resettled between supply level. den, kraals and fencing as - Stage 3: up to 89 2001 and 2004. - Stage 3 concept is being finalized. required and payment of h/hold by Dec, 2003 - Stage 3: not yet implemented. disturbance allowances (ii) Houses replaced Sept, 2000: 340 All houses replaced (per m2) with new and Most preferred consolidating floor areas of traditional huts Sept, 2003: 450 mostly consolidated plans as per the above into modern houses, rondavel sometimes requested. resettlement schedule. (iii) Three multipurpose development centers September, 2003 Not developed. The idea was abandoned in favor of utilizing existing train- (clinics, workshops etc.) ing facilities. C.4. Host and resettled com- (i) 80 classrooms December, 2003 March, 2004: 7 foothill schools (20 new There were a further 2 replacement schools and a church. munities provided with ade- classrooms, 6 staff rooms, 4 kitchens) Numbers revised during implementation to address needs. quate infrastructure September, 2004: 7 highland schools (13 new classrooms, 5 staff rooms, 4 kitchens) & 2 schools renovated. (ii) 300 latrines at 15 schools December, 2000 March, 2004: 7 foothill schools (86 VIPs) Implementation started mid 2003 with LHDA construction Sept, 2004: 7 highland schools (16 VIPs) team due to high contractor costs. Total: 102 VIPs at 14 schools - 40 - Objective Key Performance Indicators Original Target Val- Actual Value Achieved at Completion Comments ues Target Years (iii) 4 bridges December, 2000 July 2004: 3 foot & 3 vehicular bridges Part of feeder roads and reservoir crossing system. 3 bridges design and construction outsourced. (iv) 100 kms roads September, 2003 July 2004: 75 kms all-weather road Other resettlement related accesses as well as over 100 km earth access for WATSAN project constructed by LHDA team. (v) Water supply and sanitation for September, 2000: 22 Water supply systems provided at destina- Latrines to hosts supplied only in the highlands through host/resettled villages at 30l/pp/day and 1 September, 2003: 48 tions of more than 3 resettled families. WATSAN project where applicable i.e. catchment areas. latrine per 1 h/h. Latrines provided only to those resettled as part of replacement housing. (vi) Water minders trained and committees September, 2000: 44 Done for all systems (2 per system). All systems had to comply with the Dept. of Rural Water established. September, 2003: 96 Supply standards (minders included) for long term opera- tion. (vii) Actions per detailed timetable in Reset- All actions implemented within the RDAP RDAP implementation is ongoing with varying levels of tlement & Development Action Plan. framework. success. C.5. Effective Participation (i) CLAs functioning 1998 ­ 2003 Disbanded due to high operational costs. mechanism operational and (ii) CALCs functioning 1998 ­ present CALCs continuing. sustainable. C.6. Income restoration and (i) Business plans developed September, 2000: 270 Dec 2000: 102 Of 244 plans, 202 approved and paid to-date as invest- development September, 2003: 365 Sept, 2003: 154 ments of: rental units in town (70), taxis (45), agricultural investments (36), milling equipment (8), shops (9) and Sept, 2004: 244 markets/ savings investment (34). (ii) Annual job equivalents Sept, 2000: 400 Figures never determined. About 160 jobs could be deduced from the above invest- Sept, 2003: 400 ments if running successfully. (iii) Households in trained courses September, 2000: 180 2000: 304 (cumulative) Training in: carpentry, catering, sewing, construction, September, 2003: 290 2003: 421 (625 cumulative) book-keeping, record keeping and cooperatives formation. 2006: 235 (860 cumulative) (iv) Established credit facility for households September, 2003: 50 Not implemented. The idea abandoned in the light of its failure in Phase 1A area. (v) Agricultural extension service in place September, 2001 Initiated in 2000 with services introduced Capacity constraints with GoL departments limited extent incrementally over time. of service provided. (vi) Agricultural input centers established September, 2001 Existing centers used; Mohale centre established 2002. C.7. Monitoring and Evalua- (i) Indicators agreed. June, 1998 1998/99: first defined (Contract 669) Problems in continuity resulted in indicators being refined, tion System in place and 2000/01: refined (C669) evolving to embrace livelihoods by 2004. Created prob- operational lems with data compatibility. 2004: more refinement (pilot testing) 2005: more refinement (Contract 1204). (ii) Database & questionnaire designed July, 1998 July, 1998 For 1998/99 Contract 669. but see above. Database not compatible and used by subsequent M&E contractors. - 41 - Objective Key Performance Indicators Original Target Val- Actual Value Achieved at Completion Comments ues Target Years (iii) Participatory criteria September, 1998 May, 1999: completed Through Contract 669. (iv) Annual Participatory evaluations under- from October, 1999 Annual stakeholder workshops undertaken Stakeholder evaluations of LHDA performance: taken since 1999 with evaluation of LHDA 2000 82% 2001 73% performance. 2002 80% 2003 69% 2004 87% 2005 72% 2006 86% (v) Surveys and workshops undertaken October, 1998 June, 1999: 1st evaluation completed Participatory approach was suspended in favor of ques- October, 1999 1999 & 2001 participatory workshops tionnaires. Piloted Sept 2004 and now underway (2005-06) through Contract 1204. October, 2000 held (vi) Catchment Community Conservation December, 1999 Not started. Under Contract 1044: ICM, started May, 2005. Committee established C.8. Natural Environment (i) Environmental Awareness program September, 2000 1998/99 study, implementation through Combined Phases 1A and 1B. Program launched and operational Mohale FOB (2000) Integrated Catchment Man- (ii) Recycling program established and trans- Sept, 2000: established 2000-2002: established refuse receptacles, agement established with full ferred to local communities and contractors Sept, 2003: sustainable cleaning campaigns, can-crusher plant but support of GoL and commu- not sustained. nities, and awareness and (iii) Biosphere planning complete with sites Dec, 1999: approved ICM Contract 1044, started May, 2005 Three nature reserves (7574 hectares) of cultural heritage recycling campaigns under- and conservation plans approved by 12/99 conservation plans will address over next 5 years. and biodiversity under protection. way with reserves established and handed over by 2005: handover 2005 C.9. Establishment of com- (i) Sites with sustainable Maloti Minnow 9 by October 2000 1711 fish relocated to 4 sites. Monitoring ongoing to determine sustainability prehensive Environmental populations 15 by 2003 No physical barriers constructed to protect populations. Resources Conservation Program (ii) Captive breeding program operational. Tried and discontinued. Translocation became the preferred option. (iii) Sustainable medicinal plant nurseries 2 by September, 2000 5 nurseries established under Contract 5 by 2003 1054 and handed over to communities. (iv) Commercial spiral aloe nurseries September, 2000: 1 No Spiral Aloe nursery established. Katse Botanical garden established international seed shar- September, 2003: 3 ing program. & spiral aloe propagation successfully im- plemented. (v) Land use imagery system operational and Sept, 1998: operational 2000 & 2002: Draft monitoring reports Staff shortage made the system un-sustainable. providing biannual reports Biannual reports C.10. Establishment, opera- (i) Annual statistics on vegetative cover, soil Starting Sept, 1999 Not started. Contract 1053 (Env. Monitoring) did not undertake, activi- tion and handover of a com- erosion, key terrestrial and aquatic fauna ties to be implemented under Contract 1044: ICM, started prehensive set of environ- population characteristics May, 2005. mental resource monitoring (ii) Annual independent construction site Done bi-annually during construction systems and procedures audits period. (iii) Tri-annual wetland reports Starting March, 1999 Not started. Under Contract 1044: ICM, started May, 2005. (iv) Monthly water quality reports on 12 key Monthly sampling and reporting on 12 parameters. variables. - 42 - Objective Key Performance Indicators Original Target Val- Actual Value Achieved at Completion Comments ues Target Years (v) Downstream impact monitoring system September, 2000 Biophysical established July 2005. established Socio economic ­ June 2005. C.11. Public Health Pro- (i) Workforce pre-employment health report June, 1998 June, 1998. Part of construction specification. gram. (ii) Worker Health care reports produced Quarterly Worker Health Care reports Workforce Clinical Services quarterly were prepared by the contractor ·Clinical services and facili- (iii) 2000 health pamphlets and 50,000 con- Pamphlets: 2000/yr pamphlets: 150-300/mth (2,400 annually) Total condoms distributed in all clinics and LHDA = ties provided for workforce doms distributed annually in 1B area. Condoms: 50,000/yr condoms: 800-1000/mth (to June 2003) 200,000 (June2000). Public Health interventions declined ·Occupational health moni- after restructuring with loss of 2 staff. toring system in place (iv) Patients treated at Leribe Trauma Unit 7,000 / annum Out-patients treated In-patients admitted ·Leribe Trauma Unit (LTU) 1992/93 ­ < 1,000 1993/94 ­ < 1,500 1992/93 ­ 55 1993/94 - 180 functional 1994/95 ­ > 2,000 1995/96 ­ > 4,500 1994/95 ­ 180 1995/96 - 240 1996/99 ­ < 6,000 1999/2000 ­ > 6500 1996/99 ­ 245 1999/2000 - 279 2000-2003 ­ No data 2000-2003 ­ No data (v) Leribe Trauma Unit cost per project patient Not defined 1999 Expenditure = M5.63 million Direct cost per project patient is difficult to calculate as Patients treated = 6,035 outpatients and 279respective costs were not disaggregated. Average costs can admissions. be estimated. In-patients incur more costs. Average cost per patient = M891.00 (vi) Handover of LTU. 2003 March 2003: hand over to MOHSW Standards declined due to GoL capacity constraints. C.12. Local Communities (i) Health centers complete or repaired Sept, 2000: 2 Hosp/Clinics Operation date The dates given indicate when the clinics originally started Health Care: 2003: 4 1. Katse 1992 March providing services, inclusive of those inherited from Phase (a) Comprehensive primary 1A (pre 1998 period). There were no other health centers 2. Lejone 1992 health care services provided or sub-centers in Phase 1B. 3. Mohale 1998 June (b) Health monitoring system in place 4. LTU 1992 (c) Primary health care ser- (ii) Antenatal mother attendance rate Sept, 2000: 70% No data available. vices strengthened Sept, 2003: 85% (d) Campaign undertaken per (iii) Universal primary immunization Sept, 2000: 85% Measles immunization: 98% LHDA assisted MOHSW personnel. village annually 2003: universal TB (BCG) immunization: 80 ­ 95% Polio immunization: 98% (iv) Indicators and monitoring procedures from Sept, 1998 Final monitoring (Contract 1044) under- LHDA encouraged to release data into public domain. published way. (v) Health status in ESSG database March, 1999 Contracts 669 and 1204 created GIS based Data in the GIS system, but operational capacity declined due project health status data base to staff shortage. C.13. Strengthen Ministry of Health and Social Welfare and CHAL: - 43 - Objective Key Performance Indicators Original Target Val- Actual Value Achieved at Completion Comments ues Target Years (a) Critical mass of personnel (i) 50% of required staff identified and trained 1999: identified All staff identified by MOHSW and 5 senior personnel trained: 1 from Queen 2 Hospital and 4 trained in public health sys- 2002: trained CHAL were trained by end of 2005. from MOHSW. tems research, monitoring & (ii) M&E programs in place September, 2002 Staff for LTU trained in clinical skills to Most staff lost upon transfer - due to lower GoL remunera- evaluation run LTU specialized services and the tion levels and as a result, service standards declined. other 4 clinics by MRI. (b) Critical mass of personnel (iii) 50% of required staff identified and Dec, 1998: identified All staff identified by MOHSW were trained in clinical skills and trained Dec, 2002: trained trained. specialized equipment and procedures (iv) Phased clinic handover March, 2003 Lejone clinic ­ Nov. 2002 Transfer policy was developed along the way. Asset Dives- LTU ­ Mar. 2003 titure Policy was only adopted in 2004, and long-term sustainability remains an issue. Katse & Mohale ­ Dec. 2004 (c) On the job training for (v) Facility heads identified and training pro- September, 1999. Staff development & training by the con- Staff or skills retention became a problem after take-over. management grams completed tractor. (vi) 1 staff/facility trained in HR and finance September, 2002. Staff development & training by the con- Staff or skills retention became a problem after take-over. tractor. (vii) Practical skills upgrade September, 2001. Staff development & training by the con- Staff or skills retention became a problem after take-over. tractor. (ix) Handover of clinics to GoL September, 2003. Lejone clinic ­ Nov. 2002 Transfer policy was developed along the way i.e. Asset LTU ­ Mar. 2003 Divestiture Policy was only adopted in 2004. Katse & Mohale ­ Dec. 2004 C.14 Water supply and sani- (i) VIP latrines (1/5 HH) in 15 extra vil- Sept, 2000: 30 December 2005: 126 villages Started with Contract 1055; then LHDA team took over tation to Mohale catchment lages/yr Sept, 2003: 75 VIP latrines: 2545 households mid-2004 and completed work in Dec. 2005. Figures not communities (over and above on target but far in excess of appraisal estimates. affected and host communi- (ii) Refuse disposal in 15 extra villages/yr Sept, 2000: 30 December 2005: 126 villages Not on target but in excess of appraisal estimates. ties) Sept, 2003: 75 Refuse disposal: 2545 households (iii) Water supply at 30/l/hh/day at WATSAN Sept, 2000: 30 December 2005: 126 villages Some systems serve more than one village. standards in 15 extra villages/yr Sept, 2003: 75 Water supply systems: 105 systems for Implementation started in 2003. 126 villages (iv) 2 water minders per water supply trained Minders trained for all 105 water supply systems C.15. Administrative and (i) Consultants appointed per procurement Largely achieved ­ see Annex 1 Section Capacity Building plan (on time). B1.b.ii (a) Technical Assistance (ii) Outputs as specified in individual terms of Done. Technology transfer to locals how- recruited and delivers accord- reference with respect to both work program ever, proved difficult to effect, monitor ing to specific ToRs completion and technology transfer and on the and verify. job training. - 44 - Objective Key Performance Indicators Original Target Val- Actual Value Achieved at Completion Comments ues Target Years (b) Training program de- (i) Training Policy in place September, 1998. No formal Training Policy developed. LHDA operated with interim policies throughout. signed based on full assess- ments and tied to annual reviews (c) Capacity in place for (i) Initial Training Plan and rolling 3 yr plans September, 1998 Limited training based on interim plan. Varied from on the job training to short- and some long- management of training pro- and detailed annual plans by Sept each year. term training (mostly for operation and maintenance). gram (d) Training program imple- (i) Implementation of annual plans and related Interim plan was implemented but the mented per plan budget expenditure (# of trainees completed, effectiveness was never assessed. cost per trainee). (e) IFR Study complete and (i) IFR study completed September, 1999 August, 2000 Pioneering approach developed with scope increased to mitigation plan developed develop scenario based approach for policy making. and agreed (ii) Agreement on IFR releases and/or mitiga- June, 2000 July 2003: Policy and Procedures ap- Policy and Procedures define release plan and mitigation tion plan proved and implemented. requirements. Biophysical monitoring and Socio-economic monitoring mid-2005. (f) O & M study completed (i) O&M Study completed for period beyond December, 2001 Start: 2003 The study went in phases, overlapping with implementa- 2003. Completion: 2004 tion that started January 2004. (ii) New Organization agreed October, 2002 New structure (phased): 2004. Approved in phases starting from the top. Performance was adversely affected by long transition period (2004 -2006). (g) LHDA Administration (i) Treaty and Order amended. December, 1998 1998-99: Treaty renegotiation. functions continue to improve June 1999 Protocol VI signed. LHDA Order amendment: 2001. (ii) Re-organization completed. December, 1998 Started January 2004 and finished June December 2006: only 13 out of 239 approved positions still 2006. unfilled. (iii) New Board appointed December, 1998 April 2003 1st mandate: April 2003 to March 2006, extended by a year to enable restructuring completion. - 45 - Annex 4. Economic and Financial Analysis Ex Post Rate of Return Calculation and Cost Effectiveness Analysis Introduction. Calculation of the ex post economic rate of return of LHWP-Phase 1B is made for the ICR. The most important changes that have happened between the ex ante and the ex post situation are changes in water demand in South Africa, the implementation of an IFR policy, as well as changes in the project costs. The impact of these factors will be discussed in detail below. The baseline comparison for the economic rate of return is 15.9% as was calculated in the Project Appraisal Document in 1998. Methodology and Assumptions. The rate of return assessment is undertaken from 2 perspec- tives: the project as a whole and RSA as the main beneficiary of the water. The difference be- tween the two perspectives is that in the latter perspective we do not consider the additional hy- dropower generation made possible through Phase 1B water and add the royalties as a cost. In both scenarios, the expenditures of Phase 1A are considered sunk, including the expenditures for the hydropower plant and the delivery tunnels to South Africa which are both used by Phase 1B but will not be considered a project cost. The basic framework for the analysis is a Consumer Surplus Analysis, i.e. to what extent water the project has increased the consumer surplus of wa- ter consumers in the Vaal River System. Details of the method can be found in the LHDA Docu- ment: The Economics of Phase 1B (LHDA 1998) or the LHDA Document: The Economic Costs and Benefits of Changing In-Stream Flow Requirements below Phase 1A structure (LHDA 2002). The assumptions and data sources are also described in detail in the latter document. The aim of the assessment here was to mimic the assumptions made in the PAD as closely as possible to get a direct comparison of the changes that occurred during implementation. Key Drivers of Changes in the Economic Rate of Return: Water Demand. Projections of wa- ter demand seriously affect the economic rate of return. At the same time, there is great uncer- tainty surrounding accurately projecting water demand. Phase 1B was beginning to be imple- mented at a time when water demand in South Africa was undergoing significant changes for three reasons. First, as a result of the 1994-95 water restrictions, water demand levels fell and never fully recovered to the old demand growth curve. Secondly, in conjunction with the planning for this project, DWAF began instituting water demand management policies to explicitly limit the growth of water demand. These policies included significant increases in water tariffs (which were also justified as necessary to pay for LHWP), a change in the water law that ended free unlimited access of riparian users to water, agreements to require irrigators to progressively pay for the water, and several campaigns to stop leakage and promote water conservation. LHWP was an important catalyst of promoting these changes so that water demand reductions achieved as a result of these initiatives are also partly a benefit and World Bank support to Phase 1B was spe- cifically linked to continuation of these water demand management activities. Third, the demo- graphic dynamics in the Vaal River System slowed down considerably due to fertility decline, reductions in in-migration, and increases in AIDS mortality. Just before the PAD was produced in 1998, it became apparent that water demand levels were not as high and were not rising as fast as had previously been assumed and that it would, in principle, be possible to postpone Phase 1B by up to 7 years. After a careful economic assessment of the benefits and costs of postponement (see paper on the Costs and Benefits of Postponing Phase 1B in the project files, World Bank, 1998), it became clear that it was more economical to proceed, given the state of project preparation. The assumptions of the rate of return calculations were however, already changed, and it was assumed that the demand levels in the Vaal River System were lower, implying that Phase 1B water would only be necessary by 2009 and that water demand was growing by only 3% a year. Further as- sessments of water demand led to even lower growth curves. Demand curves produced by DWAF - 46 - in 1998 implied water demand growth of only 1.7-2.2% and projections made by Rand in 2000 implied water demand growth of only 0.8% a year. The realism of these projections are difficult to assess especially given more recent developments. Two developments of particular note are that the projections of AIDS mortality need to be rethought given recent changes in access to ef- fective AIDS treatment, and that actual water demand in the Vaal River System (at least the por- tion attributable to Rand Water, the most important water user) has been rising by 2.2% in 2001, 4.6% in 2002, 6.3% in 2003, 2.2% in 2004, and 1.1% in 2005, thus questioning the much lower assumption of only 0.8% growth. Rand Water and DWAF are currently reassessing water demand in the Vaal River System but the outcome of this process will not be known until later in 2006. For this analysis, the baseline assessment makes the following assumptions regarding water de- mand: Phase 1B water would only have been needed by 2009 (in line with the assumption at ap- praisal which still seems reasonable) and Phase 2 water would only be needed by 2017. The im- plied growth of water demand of this scenario is 1.5% per year which appears reasonable given the discussion above. Key Drivers of Changes in the Economic Rate of Return: IFR. At appraisal, only minimal allowance had been made for IFRs which were limited to the mandates in the Treaty. As a result of a thorough assessment of IFRs and the associated policy decisions, the IFRs at Phase 1B were significantly increased and in addition, compensation was paid to downstream users. The impact on the project economics is two-fold. First, due to the higher releases, the resulting water transfers are lower, thus limiting these project benefits. We assume that the yield of the Phase 1 system is now 773MCM as a result of these higher IFRs. Secondly, the compensation paid to affected downstream communities has (slightly) increased project costs which are now included. Key Drivers of Changes in the Economic Rate of Return: Project Costs. As shown in Table 1, the assessment of costs of the project have varied considerably over time. From 1993 to 2000, each successive cost plan projected significant declines in costs, which was largely achieved by much lower than expected bids for construction costs, and lower estimated administration costs. Since May 2000, the costs appear to have increased somewhat and are about 20% higher in 2005 than in 2000 (in real terms). But the main reason of the higher reported costs are that in the latest estimate, O+M expenditures and administration expenditures over the life of the project are for the first time being included and spending until 2010 in all other categories is considered. But even if these costs over the life of the project are included, the project costs remained lower than initially considered (and much lower if thus initial assessments had included O+M and admin costs over the life of the project). Despite considerable efforts and expenditures to address social and environmental issues, they make up only about 13% of total project costs (and a mere 2% is spent on compensation). Conversely, two-thirds of total project costs went for engineering and construction expenditures. The expenditures on administration and O+M are also considerable, even for a project of this magnitude. Overall, the project is one of the few such large projects where costs at the end have remained below the assessment at the beginning and the project au- thorities should be very pleased with this outcome. This is even more apparent, when one exam- ines Table 2 which compares the current US$ cost projections at appraisal with the actual costs. Even if one makes an very generous allowance for administration and O+M costs for the life of the project, the project remains considerably cheaper than at appraisal. As it appears that O+M and Administration costs after the construction phase were not included in the PAD cost assess- ment, the second column of Table 2 is probably the most relevant comparison and there project costs have been over 30% lower than at appraisal. Clearly, the devaluation of the Loti vis-à-vis the dollar in the periods of high costs (around 1999-2002) has also contributed to this very fa- vourable picture in $ terms. As a result, changes in project costs have had very little impact on the rate of return calculation. - 47 - Rate of Return Calculation: Base Results. The results of the base analysis using actual costs, the current IFR regime, and assuming 1.5% water demand growth and a possible delay of 1B by 7 years, are shown in Table 3 alongside several sensitivity analyses. Due to much lower demand growth than assumed at appraisal, the rate of return has dropped from 15.9% to 11.5% (15.5% to 11.1% from RSA's perspective). While this poorer performance is lamentable, one should bear in mind three things. First, it was clearly not anticipated that water demand growth would be revised downward so drastically shortly after implementation. Second, it is unclear whether the assump- tion here of a 1.5% growth is not too conservative, given recent developments. And lastly, the project helped bring about the reductions in water demand growth as it contributed significantly to the development of water demand management policies in South Africa. So the project clearly was able to achieve its overarching goal which was to improve security of water availability in South Africa by increasing supply and by helping to promote policies to limit demand.3 Rate of Return Calculation: Sensitivity Analyses. Table 3 also shows a range of sensitivity analyses. The most important finding is that the project economics is heavily affected by water demand projections. If water demand growth was as high as 2.5% a year (which is not unreasona- bly given recent trends), then the rate of return would jump to 17.0%, while it would drop to 7.6% if water demand growth was 0.8% and water demand levels were so low that further augmenta- tion would not be needed until 2025. The project economics is also quite significantly affected by the assumed water demand elasticity.4 At the same time, the minor changes in costs and the sig- nificant change in the IFR regime has negligible impacts on the project economics. Conclusion. This ex post assessment suggests that despite drastic changes in water demand pro- jections (partially brought about by the project), Phase 1B carries an acceptable ex post rate of return. The project economics are largely driven by assessment of future changes in water de- mand in the Vaal River System. Cost Effectiveness Analysis The PAD also included a cost effectiveness analysis which compared Phase 1B with competing supply augmentation schemes, particularly the Tugela Vaal Transfer South (TVTS) and the Or- ange Vaal Tranfer Scheme (OVTS). LHWP 1B was found to be much cheaper than the other two, with unit reference values being 30-45% lower. Since these alternative schemes have not been implemented, there is no updated information on costs available. Since the project costs of Phase 1B have been lower than anticipated, it appears highly likely that the cost advantage of Phase 1B over these alternative schemes was maintained or even increased. References LHDA. 1998. The Economics of Phase 1B. Maseru: LHDA. LHDA. 2002. The Economic Costs and Benefits of Changing In-stream Flow Requirements be- low Phase 1 Structures. Maseru: LHDA: World Bank. 1998. The Costs and Benefits of Postponing Phase 1B. Washington DC: The World Bank. Table 1: Project Costs in Real Maloti (millions in 2004 prices) 3In this context it should be noted that the project was paid for by the water users in South Africa and the higher water tariffs that came as a result of the project were a key ingredient of the water demand manage- ment policies. 4The Economics of Phase 1B (LHDA 1998) used an elasticity of 0.4. Later assessments suggested that an elasticity of 0.3 might be more reasonable (see LHDA 2002) which have been used as a baseline here. - 48 - Total Construction EngineeringAdministrationEnvironment Of which O+M Social Compensation May 1993 6157.8 3958.1 573.2 959.7 666.8 218.9 0.0 May 1997 5711.4 4094.8 539.3 540.0 537.3 336.9 0.0 May 2000 5097.5 3235.5 601.2 570.6 648.7 150.9 34.3 Nov 2002 5179.2 3243.9 650.6 541.5 672.9 169.3 63.0 Nov 2005 5930.0 3267.2 664.9 901.3 769.2 145.3 327.4 Share 2005 55% 11% 15% 13% 2% 6% Notes: Up until the May00 Cost Plan, only costs until 2003 are considered. In 2002, costs until 2006 are included. In the 2005 cost plan, costs until 2010 are considered. Starting in May00, compensation costs over the life of the project are included. In the Nov 2005 row, allowance has been made for the first time for spending for O+M and Administration over the entire life of the project (2044). Table 2: Project Costs in Current US$ Appraisal Actual until Projected until Projected for 12/06 2010 project life (2044) Construction 436.0 380.4 380.4 380.4 Engineering 87.9 77.0 77.0 77.0 Environment/Social 115.6 82.7 96.8 96.8 Administration/O+M 87.5 84.3 97.5 259.5 Contingencies 157.4 Total 884.3 624.3 651.7 813.6 Note : Projected until 2010 includes compensation spending over life of project. Projected until 2044 includes a (very generous) allowance for administration and O+M expenditures over life of project. Table 3: Ex Post Rate of Return Calculation and Sensitivity Analyses Project Perspective RSA Per- spective PAD 1998 (Water Demand Growth 3% a year, Phase 1B 15.9% 15.5% needed 2009, & Phase 2 in 2014) (NPV M. 4500m.)5 Baseline Rate 2005 (Water Demand Growth 1.5%, Phase 11.5% 11.1% 1B needed 2009 & Phase 2 needed in 2017) (NPV M. 845m.)6 Sensitivity Analyses Water demand elasticity 0.4 (instead of 0.3) 10.1% Costs down 10% 11.9% IFR to Treaty Releases 11.7% Water Demand Growth 0.8% & Phase 2 needed 2017 10.7% Water Demand Growth 0.8% & Phase 2 needed 2025 7.6% Water Demand Growth 2.5% & Phase 2 needed 2013 17.0% NPV expressed in 1995 prices to make them comparable to the PAD. A real discount rate of 10% is assumed in these calculations. 5This is equivalent to $1.2billion in 1995 prices and exchange rates. 6This is equivalent to $230m in 1995 prices and exchange rates. - 49 - Annex 5. Bank Lending and Implementation Support/Supervision Processes (a) Task Team members Names Title Unit Responsibility/Specialty Lending Supervision / ICR Leonard John Abrams Sr Water Resources Mgmt. Spec. AFTU1 John E. Ambrose Consultant AFTU1 Dan R. Aronson Consultant AFTEG Devendra Bajgain Operations Officer AFTU1 Roxanne Hakim Sr Anthropologist AFTS2 Rafik Fatehali Hirji Sr Water Resources Spec. ENV Stephan Klasen Consultant AFTU1 Kisa Mfalila Consultant MNSRE Jonathan Nyamukapa Sr Financial Management Specialist AFTFM Peter Okwero Sr Health Spec. AFTH1 Peter Watson Consultant AFTU1 Marcus Wishart Young Professional AFTU1 Daryl Fields Sr Water Resources Spec. ETWWA (b) Staff Time and Cost Staff Time and Cost (Bank Budget Only) Stage of Project Cycle No. of staff weeks USD Thousands (including travel & consultant costs) Lending FY98 -70.96 FY99 0.00 FY00 -0.03 Total: -70.99 Supervision/ICR FY98 0.00 FY99 0.00 FY00 46 257.06 FY01 27 194.52 FY02 32 281.87 FY03 26 318.42 FY04 28 337.96 FY05 33 372.24 FY06 42 409.71 FY07 7 0.00 Total: 241 2171.78 - 50 - Mission Composition Stage of Project Cycle No. of Persons and Specialty Performance Rating (e.g. 2 Economists, 1 FMS, etc.) Month/Year Count Specialty Implementation Development Progress Objective Identification/Preparation October 4 ­ 14, 1993 4 Task Team Leader (1); Environmental Spc. (1); Operations Officer (1); Pro- [Ph1A supervision, 1B planning] curement (1) February 12 ­ March4, 1994 8 Task Team Leader (1); Operations Officer (1); Resettlement (1); Environ- [Ph1A supervision, 1B planning] mental Spc. (1); Water Resource Management (1); Procurement (1); Country Officer (1); Operations Advisor (1) October 4 ­ 21, 1994 5 Task Team Leader (1); Operations Officer (1); Resettlement (1); Social [Ph1A supervision, 1B planning] (ODA); Dam Engineer/Geologist (1) November 1 ­ 22, 1996 11 Task Team Leader (1); Power Engineering (1); Dam Engineer (1); Dam En- [Ph1A supervision, 1B pre- gineering/Geologist (1); Participation (1); Economist (1); Social (1); Envi- appraisal] ronmental Spc. (1); Water Resources/Economist (1); Natural Resource Mgmt/Water Resource Mgmt (1); Financial Analyst (1) Appraisal/Negotiation June 23 ­ July 11, 1997 8 Task Team Leader (1); Power Engineering (1); Participation (1); Economist [Ph1A sup., Ph1B appraise] (1); Social (1); Environmental Spc. (1); Legal (1); Management ­ Country Director (1) October 26 ­ November 14, 1997 3 Participation (1); Political Economist (1); Social (1) [post-appraisal mission] Supervision May 24 ­ June 3, 1998 2 Dam Engineer (1); Dam Engineering/Geologist (1) June 27 ­ July 2, 1998 2 Task Team Leader (1); Political Economist (1) November 15 ­ 25, 1998 4 Task Team Leader (1); Social Scientist (1); Institutional Spc. (1); Environ- S S mental Spc. (1) February 4 ­ 12, 1999 1 Natural Resource Mgmt/Water Resource Mgmt (1) April 28 - 05/14/1999 [completion 7 Task Team Leader (1); Social Scientist (1); Dam Specialist (1); Environment S S Ph1A, supervision 1B] Spc. (1); Economist (1); Institutional Spc. (1); Lawyer (1) November 3 ­ 9, 1999 1 Natural Resource Mgmt/Water Resource Mgmt (1) November 9 ­ 19, 1999 7 Task Team Leader (1); Pr.Water Specialist (1); Inst. Develpt. Spec./M&E S S (1); Pr. Social Scientist (1); Environment Spec. (1); Engineering Spec. (1); Economist (1) May 2 ­ 12, 2000 8 Task Team Leader (1); Engineering (1); Resettlement (1); Environment (1); S S Economist (1); Institutional Issue (1); Procurement (1); Water Resources/IFR (1) November 2 ­ 13, 2000 Task Team Leader (1); Engineering (1); Resettlement (1); Environment (1); S S Economist (1); Water Resources/IFR (1); Procurement (1); Financial (1); April 23 ­ May 4, 2001 7 Task Team Leader (1); Engineering (1); Social Specialist (1); Environmental S S - 51 - Spec. (1); Economist (1); Water Resource Spec. (1); Procurement Spec. (1) October 29 ­ November 6, 2001 7 Task Team Leader (1); Engineering (1); Resettlement (1); Rural Develop- S S ment (1); Environment (1); Economic And Financial (1); Procurement (1) March 5, 2002 1 Task Team Leader (1) S S April 22 ­ 30, 2002 8 Task Team Leader (1); Adviser-Consultant (1); Engineering-Consultant (1); S S Social Aspect (1); Resettlement (1); Economics-Consultant (1); Water Re- sources (1); Management ­ Dir Inf (1) September 17, 2002 8 Task Team Leader (1); Adviser-Consultant (1); Engineering-Consultant (1); S S Social (1); Water Resources (1); Environment-Consultant (1); Financial Management (1); Economics - Consultant (1) October 28 ­ November 1, 2002 3 Task Team Leader (1); Consultant (1); Economist (1) March 3 ­ 12, 2003 7 Task Team Leader (1); Environment (1); Social Development (1); Econom- S S ics (1); Engineering (1); Strategic Issues, IFR (1); Management ­Sector Manager (1) June 9 - 13, 2003 2 Task Team Leader (1); Strategic Issues (1) S S September 16 ­ 24, 2003 8 Task Team Leader (1); Environment (1); Social Development (2); Econom- S S ics (1); Engineering (1); Strategic Issues (1); IFR (1) December 1 - 5, 2003 4 Task Team Leader (1); Social Development (1); Strategic Issues (1). S S Vice President Africa (1). March 22 - 30, 2004 7 Task Team Leader (1); Strategic Issues (1); Social Development (1); Envi- S S ronmental (1); Dam Safety (1); Water Resources IFR (1); Financial Man- agement (1) August 31 ­ September 13, 2004 10 Task Team Leader (1); Strategic Issues (1); Social Development (2); Envi- S S ronmental (1); Dam Safety (1); Water Resources IFR (1); Financial Man- agement (1); Economics (1); Management ­ Dir Inf (1) January 17 ­ 21, 2005 4 Task Team Leader (1); Strategic Issues (1); Social Development (1); Public S S Health (1) April 4 ­ 12, 2005 8 Task Team Leader (1); Strategic Issues (1); Social Development (2); Envi- MS S ronment (1); Public Health (1); Financial Management (1); Management ­ Country Director (1) September 5 ­ 15, 2005 12 Task Team Leader (1); Strategic Issues (1); Social Development (2); Eco- MS S nomics (1); Environmental (1); Water Resources IFR (1); Public Health (1); Financial Management (1); M&E IFR (1); ICR (2) December 12 ­ 15, 2005 5 Task Team Leader (1); Strategic Issues (1); Social Development (1); Sector MUS S Mgr (1); Management ­ Country Director (1) May 15 ­ 23, 2006 4 Task Team Leader (1); Strategic Issues (1); Social Development (1); Finan- MUS S cial Management (1); October 16 ­ 26, 2006 9 Task Team Leader (1); Strategic Issues (1); Social Development (1); Eco- MS S nomics (1); Water Resources IFR (1); Public Health (1); ICR (3) January ­ February, 2007 5 Task Team Leader (1); ICR (4); ICR Mission & workshops - 52 - Annex 6. Stakeholder Workshop Report and Results The World Bank Group and Lesotho Highlands Development Authority hosted two workshops in February 2007 in Maseru, Lesotho, and Gauteng, South African (details below). The objectives of the workshops were to: i) share information regarding the design, development, implementation, operation and impact of Phase 1B of the LHWP; ii) review and discuss stakeholder viewpoints and concerns in these contexts; iii) describe and discuss the lessons that have been learned by stakeholder groups through the project lifecycle of LHWP Phase 1B, including lessons for finan- cial management; iv) review and discuss current vehicles for assessment, evaluation and reporting around LHWP Phase 1B; and v) propose measures to improve and sustain the above. The follow- ing provides a summary of the detailed "Report on Stakeholder Consultations" available through the project files. Table 6.1: Summary of the LHWP Phase 1B ICR Stakeholder Workshops Lesotho South Africa Venue Maseru Sun Hotel, Maseru Birchwood Hotel, Gauteng. Date 8 February 2007 13 February 2007 Participants invited 163 70 Attendees7 Approximately 100 10 Key stakeholder groups LHDA Department of Water Affairs and For- LHDA Board of Directors estry Lesotho Highlands Water Commission Water Research Commission Lesotho Government Randwater Directly affected communities Alexandra Development Forum Private sector Environmental Monitoring Group Civil society ­ NGO & CBO Wits University PoE, academia and consultants Specialists and consultants The workshops followed a structured facilitation approach using a scorecard (Table 6.2) as the instrument for soliciting input from participants. Table 6.2: Scorecard category definitions Category Definition 1. Project Were the objectives of Phase 1B appropriate and clearly defined? Were the components and Design organization realistic and adequate to achieve the objectives? Did the project design properly address the potential risks and project impacts? 2. Governance Was there sufficient Government commitment to achieving development objectives? Did the project governance structures provide an enabling environment to achieve the objectives and ensure timely resolution of implementation issues? 3. Environmental Were environmental issues adequately identified during project preparation and addressed during implementation? 4. Resettlement Were resettlement issues adequately addressed? Has the project provided sufficient opportuni- ties toward assisting those affected by the project? 5. Stakeholder How were relationships with stakeholders? and project affected communities ? Was there Involvement sufficient coordination and communication with stakeholders? leading to stakeholder issues being adequately addressed? Were stakeholders seen as project partners? 6. LHDA Was LHDA committed to achieving the development objectives, have sufficient resources, Performance staff and arrangements to ensure timely resolution of implementation issues? 7. World Bank Did the World Bank assist in ensuring quality of design and implementation and help achieve Performance the development outcomes? Were Bank policies and supervision arrangements adequate and timely? 8. Development To what extent has the project achieved its objectives, contributed to poverty alleviation and Impact economic stability. Have there been significant development impacts in the country as a result of the project 9. Overall Has the project achieved its stated development objectives? Performance 7Excluding World Bank Group and SRK representatives - 53 - Breakout groups evaluated the project by rating the nine categories in the scorecard, reflecting on key lessons learned and the strengths and weaknesses of World Bank's role. Group rapporteurs presented the groups' results in a plenary discussion. The format differed between Lesotho and South Africa due to differences in both the number and nature of stakeholders in attendance. While the morning session in South Africa comprised presentations by the World Bank Group and LHDA representatives8, the afternoon session focused on scorecard evaluation by individuals in attendance. At both workshops the ratings, lessons and institutional roles were discussed in plenary. Performance: The outcomes of the Lesotho workshop are summarized below (Figures 6.1 and 6.2). The following emerged as dominant patterns among stakeholders in Lesotho: · Project Design and World Bank Performance received the most favourable ratings by the groups, while Resettlement and Compensation, LHDA Performance and Environment received the least favourable ratings. The average ratings of the worst performing categories were gener- ally above that suggesting any moderate shortcoming, indicating that project performance was generally perceived to be on the more favourable side of the rating continuum. · While there was good agreement (to within one rating category) in the ratings between groups on most themes, in particular Governance and Environment, there was wide variation in the scores of Development Impact and Stakeholder Involvement. The variation in these categories resulted from the higher (less favourable) scores given by the Civil Society ­ NGOs and CBOs group, that felt Development Impact and Stakeholder Involvement had MAJOR and SEVERE shortcomings, respectively. · There were also consistent differences in the scores of some of the breakout/stakeholder groups, implying differences between the perceptions and understanding of the different breakout groups in project performance. While LHDA (both Regions and Central) view the project in the most favourable light, Civil Society ­ NGOs and CBOs, and the Private Sector groups perceive the pro- ject to have had MODERATE to SIGNIFICANT shortcomings. Average High Low 0 More favourable 1 1 1 1 1 1 1 1 1 1.88 2 2.13 2 er 2.33 2.38 2.50 2.57 2.63 2.63 scoega 3 3.00 3 verA 4 4 4 4 4 4 4 5 5 Less favourable 6 6 Project Design World Bank Overall Governance Development Stakeholder Environment LHDA Resettlement & Performance Performance Impact involvement Performance compensation Rating category Figure 9.1: Average, highest and lowest scores per rating category for the Lesotho workshop 8A presentation was also given by Dr Johan Rall on the Maluti Minnow (Appendix F) - 54 - More favourable 0 Average High Low 1 1 1 1 1 1 1.22 2 1.89 2 2.00 2 2 2 2 er 2.38 2.44 Sco 3 3.00 3 age 3.38 3.50 verA 4 4 4 4 4 5 Less 6 favourable 6 LHDA Regions LHDA Central Civil Society - Government LHWC & LHDA Specialists Private sector Civil Society - Communities Board NGOS & CBOs Breakaway/Stakeholder Group Figure 6.2: Average, highest and lowest scores9 per breakout or stakeholder group for Lesotho. In South Africa the following emerged as dominant patterns among stakeholders: · Resettlement and Compensation and World Bank Performance received the most favourable ratings by stakeholders, while LHDA Performance, Environment, Stakeholder Involvement and Governance received the least favourable ratings. Only three stakeholders provided a rating for Resettlement and Compensation, and so less certainty exists regarding this figure. · It is evident that there is some agreement between Lesotho and South African stakeholders re- garding perceptions of project performance: both groups identified LHDA Performance and Envi- ronment, and World Bank Performance as the poorer and better performing themes respectively. Divergence between Lesotho and South African stakeholder perceptions include Governance and Project Design, which were ranked less favourably and more favourably respectively by South African stakeholders. · Note that the average ratings of the worst performing themes are generally still above rating category `3' i.e. MODERATE shortcomings in performance, indicating that performance was generally perceived by stakeholders to be on the more favourable side of the rating continuum. Development Impact: There was general agreement by stakeholders that the project had a posi- tive development impact on South Africa and Lesotho. On average, ratings indicated only MI- NOR to MODERATE shortcomings. The project brought about significant economic and infra- structural improvement, although poverty levels in the country were still high. Stakeholders be- lieved that functionality of the project should be maximized to ensure that the dams are `multi- purpose' and that the involvement of small to medium enterprises and local service providers should be extended in any future phases. The positive impact related to economic development in South Africa that was not hampered by water shortages. 9Potential scores range from 6 (least favourable) to 1 (most favourable). Score definitions are given below: 1=NO shortcomings in achievement of objectives, efficiency and relevance 2=MINOR shortcomings in achievement of objectives, efficiency and relevance 3=MODERATE shortcomings in achievement of objectives, efficiency and relevance 4=SIGNIFICANT shortcomings in achievement of objectives, efficiency and relevance 5=MAJOR shortcomings in achievement of objectives, efficiency and relevance 6=SEVERE shortcomings in achievement of objectives, efficiency and relevance - 55 - Bank Performance: Stakeholders identified a number of areas where they felt the Bank per- formed well, specifically: · International credibility: The Bank provided international credibility to the project with its sound policies, professional approach and expertise. This helped in ensuring project-funding se- curity. · Supervision missions and advice: The Bank's professionally conducted supervision missions provided solid project monitoring and advice, and were critical in ensuring that important issues such as environmental and resettlement issues received attention, thus creating certainty and value for all participants. However, the Bank's expectations of LHDA sometimes exceeded LHDA's capacity and mandate and it was suggested that the Bank should have held stakeholders accountable only for the issues they were responsible for and which were under their control. · Support for good governance: The Bank supported good governance and dealt with corruption effectively, e.g. disbarment of fraudsters. · Skills transfer: LHDA Central and Regions both felt that the Bank was successful in its skills transfer. · Basotho interests: The Bank successfully advocated the Basotho interests in the project. Stakeholders also identified a number of areas where it was felt the Bank could have been more effective, specifically: · Wider consultation with stakeholders: Consultation with other stakeholder groups during su- pervision missions should have been conducted e.g. with NGOs, vulnerable groups such as the aged, youth and vulnerable households; · Local procurement: The Bank should have further encouraged local procurement of appropri- ately qualified service providers. Procurement procedures should be revised to prevent delays, and country-specific tendering policies and procedures should be formulated. · Local capacity building: The Bank should ensure that skills are transferred to locals, both from Bank staff and from the Panel of Experts. · Pressure on Lesotho Government ministries: It was felt that further government involvement in the project was required, and the Bank should have exerted pressure on government departments in specific areas, e.g. regarding the registration of Local Legal Entities and land allocation. · Environmental requirements: Closer supervision of the environmental impact assessment was required in addition to ensuring earlier incorporation of environmental requirements into the de- sign and implementation phases of the project. It was felt that LHDA had insufficient capacity to adequately deal with environmental and social issues and to implement the Environmental Action Plan (EAP). · Inflexibility and moving the goal posts: The Bank was perceived to be unrealistically inflexible and at times moved the goal posts. Stakeholders suggested that the Bank should be more recep- tive to alternate ideas and should be a better listener. · Use of leverage to influence performance: the Bank should use its influence and conditions of loans to improve the performance of project implementation agencies and achieving broader de- velopment objectives. This includes: - Exerting pressure on Government ministries: It was felt that there were legislative and gov- ernance shortcomings of the Government, and that the Bank could have done more to address these shortcomings. - Holding back finance more frequently in response to performance failures and breaching of loan agreement conditions. · Large numbers of review/appraisal entities can create an administrative burden. · A feedback loop is required between the Panel of Experts, the World Bank and the specialists to ensure a review and consensus on sensitive issues. · Further attention should be paid to ensuring that the institutional arrangements of a project are efficient and accountable early in the project. - 56 - Key Lessons Identified by Stakeholders: This section describes the key lessons for consideration in future infrastructure projects. It is not intended to provide an exhaustive list of all lessons identified by stakeholders at the Lesotho and South Africa work- shops, but rather aims to synthesize the key themes common to both sets of stakeholders. · Environmental and social considerations are critical to a project of this nature: Incorporating environmental and social considerations into all project phases is essential for project sustainabil- ity. The following points are notable: - There was an improvement in the consideration of environmental issues in Phase 1B rela- tive to Phase 1A; however, certain constraints limited the effectiveness of this. - Risks and impacts were not fully addressed (e.g. downstream impacts in South African eco- systems). This related to the need for earlier and more proactive incorporation of environ- mental and social considerations into the design phase. - Instream Flow Requirements (IFR) were incorporated into the project retrospectively and should be incorporated into the engineering design and financial modelling from project in- ception. There was a perception that the IFR specialist studies were too subjective and varia- tions in results between studies along with the substantial financial impact of the studies war- ranted a review of the methods used. - Sufficient capacity within the implementing agency is important for environmental and so- cial issues to be adequately addressed. An important component of this is to ensure sufficient capacity for monitoring and evaluation. - Large components of the implementation of environmental and social programmes lag by years. - There are major shortcomings in red data species protection, resulting in the potential ex- tinction of the Maluti Minnow. - An environmental impact assessment is an important development tool to guide project de- sign. · Widening communication and stakeholder involvement: Variations in scores shows that there are varying perceptions and levels of understanding of the project, which underlines the need for additional/more focussed communication with different stakeholder groups. Robust communica- tion systems and active participation by stakeholders throughout project cycle is required. This system should be based on the principles of transparency and accountability, particularly in rela- tion to communications with directly affected communities, and also free, prior and informed. An additional point to note is that the communication should be widened to include all relevant stakeholders, notably those from partner countries that are receptors of project impacts, in this case South African stakeholders. · Enhanced collaboration between the partner countries needs to take place at an operational level. There should have been more mobilization of South African resources to enhance regional cooperation and complementarity of resources, especially for monitoring and collaborative re- search. There should be future collaboration between LHDA and the Crocodile Marico Catch- ment Management Agency. · The functionality of such projects should be optimized to ensure that the dams serve multiple purposes: opportunities for clean electricity generation through hydropower should be exploited to improve access to energy and generate revenues for Lesotho. This has been incorporated into the Phase 2 planning, where consultation with local electricity suppliers will take place. In addi- tion, the project could have served to provide Lesotho with an alternate water supply. - 57 - Annex 7. Summary of Borrower's ICR Lesotho Highlands Development Authority P.O. 732, Maseru 100, Lesotho. Telephone: (+266) 22 311280 Fax: (+266) 22 310065 Email: lhwp@lhda.org.ls Our Ref.: SCS/M&E/39/07/MO June 13th, 2007 The World Bank , African Region (AFTU1) 1818 H Street NW Room J11-06 Washington D.C. Dear Mr. Macoun, LHDA'S COMMENTS ON THE ICR REPORT FOR LHWP PHASE 1B Lesotho Highlands Development Authority acknowledges receipt of the Implementation Completion Re- sults Report (ICRR) for Phase 1B of the Lesotho Highlands Water Project. The report records the perform- ance of the parties in delivering the objectives of the project now that the World Bank loan has come to a close. Through strong commitment, hard work and focus, the parties, in strategic partnership with various other stakeholders, managed to successfully deliver on three of the four Project Development Objectives identi- fied in the 1998 Project Appraisal Document. Most of the environmental and social impacts have and con- tinue to be addressed with the exception of a few cases where challenges remain. The rating that has been accorded to the LHDA's performance, as the implementing body is noted. It is our hope that the good lessons learned from the implementation of Phase IB will be taken into consideration in the design and implementation of any other similar projects. We wish to acknowledge the important role that the World Bank played both in financing but more impor- tantly in monitoring and supervision of the project. This latter support was instrumental to the successful implementation of this otherwise very ambitious and complex project. Our comments on the final draft of the ICRR are attached to this letter for your consideration and incorpo- ration into the final document. Yours Faithfully __________________ Masilo Phakoe Chief Executive a.i Copy: Board Chairman Lesotho Highlands Water Commission - 58 - 1.0 Introduction The Lesotho Highlands Water Project (LHWP) is primarily the water transfer scheme (from the Lesotho Highlands to the Republic of South Africa industrial heartland of Gauteng Province), with a system of dams and water transfer and delivery tunnels. The LHWP includes four phases, of which the implementation of Phase 1A and Phase 1B were committed under a Treaty between the two countries. Phase 1A was completed in 1997 with Phase 1B starting in 1998. Construction ended in 2003. Implementation of associated social and environmental mitigation measures are continuing. Institutional transition from construction to operation and maintenance started in 2003, for substantial completion in 2006. Implementation of the LHWP Phase 1B was partly fi- nanced by the World Bank, with a loan amounting to US$45 million. The original loan period (1998-2004) was twice extended by 12 months (to December 2006), mainly to see to the conclu- sion of the residual environmental, social and institutional related work. It is in the light of this loan closure that this Project Implementation Completion Report (ICR) has been produced. Based on information presented in the 1998 Project Appraisal Document (PAD), the aim is to assess the project's achievements and derive lessons learned to inform operation and maintenance activities and subsequent water resources development projects. 2.0 Assessment of Development Objective and Design, and of Quality at Entry Project Goals and Development Objectives: The main objectives for the LHWP were: (a) to put in place the physical and managerial capacity for Lesotho to transform its principal natural re- source of abundance water into export revenues that can be applied to poverty reduction and eco- nomic stability ­ while (i) protecting the environment and mitigating any adverse social and envi- ronmental impacts ­ and (ii) maximizing the local development spin-offs of the project in Leso- tho; and (b) to assist South Africa to develop its lowest cost alternative for supply of water to the Gauteng Region. These objectives remained unchanged throughout Phase 1A and 1B, implying sound conception. The development objectives were determined by the demand for water in South Africa and its availability in Lesotho that offered opportunities for joint development in support of the two countries. The validity of these objectives has been supported by the need for Phase 1B, benefits derived as discussed herein and the continued process of assessing the feasibility of subsequent phases of the LHWP. Project Components: The project included four components: A) Construction of Main Project Works comprising: i) Mohale dam, which is a 145 metres high concrete face rock-fill dam; ii) Mohale Tunnel, comprising a 32km long and 4.1-4.6m diameter concrete lined tunnel capable of delivering 9.6m3/s of water from Mohale to Katse; iii) Matsoku Diversion with a 15m high concrete weir, 5.6km long and 4.5m diameter tunnel capable of deliv- ering 2.2m3/s to Katse; iv) Associated Advance Infrastructure, comprising the upgrading of 59km and construction of a new 21km of mountain access road, as well as construction of the Maseru bypass, construction of Mohale Township, power supply and telecommunication facilities; v) mitigation of environmental impacts of construction - with all of them costing US$531 million or 60% of total costs. B) Engineering Design and Construction Supervision (US$105 million or 12% of total costs). C) Environmental and Social Action Plan (US$140 million or 16% of total costs) for impacts mitigation containing i) Resettlement of affected communities, ii) Compensation for project re- lated losses, iii) Income Restoration and Rural Development, iv) Natural Environment and Heri- tage; v) Public Health program, and vi) Program Monitoring and Evaluation. D) Institutional Support and Capacity Building (US$108 million or 12% of total costs) through technical assistance in Engineering, Finance and Administration, Environment and Social issues, - 59 - Operation and maintenance, training, services of Engineering and Environment/Social Panel of Experts, Disputes Review Board, as well as operating and administrative costs for LHDA. Total project costs came to US$884 before adding financing costs of $248 million, which brought the total to US$1,132 million. These components could be regarded as soundly conceived since, like the project goals and objectives above, they did not change throughout the implementation period and largely achieved their stated objectives. Quality at Entry: Preparation of Phase 1B in 1998 derived significant benefit from incorporating lessons from Phase 1A. At the socio-political level, both Lesotho and RSA were then under de- mocratic rule, and riparian issues with the concerned states of Namibia and Botswana had been successfully addressed. The four riparian states had also been developing the joint water re- sources management framework called The Orange-Senqu River Commission aimed at promot- ing and coordinating the watershed related activities (signed in 2000). This was within the broader framework of the revised SADC Protocol on Shared Watercourses aimed at promoting `sustainable and coordinated management, protection and utilization of shared watercourses' (also signed in 2000). Internally, Lesotho was finalizing its first Water Resources Management Policy (adopted in 1999) to better coordinate water sector activities through amongst others, es- tablishing the office of the Commissioner of Water (in 2002) to coordinate water sector policy and planning. Mechanisms for establishing the new governance system were being established, with the LHDA undergoing transformation and restructuring to enable Phase 1B implementation alongside the operation and maintenance of Phase 1A completed facilities. On the social and en- vironmental side, the Phase 1B Environmental Impact Assessment (EIA) had been carried out, and the associated management plan (Environmental Action Plan) had been adopted in late 1997. All these provided a quality setting for Phase 1B implementation with the outcome as detailed below. 3.0 Achievement of Objectives and Outputs 3.1 Outcome/Achievement of Objectives Capacity in place for Lesotho to transform its principal natural resource of abundant ­ water- into export revenues: at an overall level, this objective was achieved, with the completion of construc- tion works and the delivery of water to RSA, generation of hydropower in 'Muela, and the receipt of related project royalties. The PAD set the specific targets, whose achievements are detailed in Annex 1, but can be summarized as follows: the economic stabilization targets were met and sur- passed by 2003 (end of construction) as the project had then been contributing an average of 4.8% to the GDP as against the target of 3.9%. Cumulative royalties collection had come to M835.3 million against the target of M650 million, while the combined SACU and tax revenues collection of M1,280 million exceeded the target of M1,115 million. Operation of the Social Fund to target poverty reduction was less successful. The `improved' Le- sotho Fund for Community Development (LFCD) was established in July 2000 to succeed the `failed' Lesotho Highlands Revenue Fund (LHRF) of 1991 but has to-date, fared not better. Like its predecessor, it had the aim of reducing unemployment through provision of labour intensive rural infrastructure development projects (access roads, footbridges, water supply systems, soil conservation procedures, etc.). By 2005, the Fund had identified 101 projects with current status (Sept. 2006) as follows: 58 projects completed, 26 suspended and 17 ongoing with overall expen- diture of M114 million. By March 2002, M93 million had been spent, and average monthly em- ployment was 3,702 workers on full-time equivalent basis. Both LHRF and LFCD had, by 2002, completed 1100kms rural roads, 210 small earth-fill dams, 60 footbridges and some forestry and soil conservation works at an overall cost of M200 million. Funding had been declining as the implementation outcome was deemed unsatisfactory, and other fund utilization alternatives were being explored. Both Funds suffered from political interference that undermined project identifi- - 60 - cation mechanisms, which had ramifications on the quality of project design, execution and out- puts, as well as on the outcome effectiveness and sustainability. The current idea of moving the Fund to the newly established system of Community Councils (elected in 2005), looks no less risky, as the new entity is still to develop technical and other operational capacities. Though it could be argued that the Fund assisted in mitigating the poverty effects at the time when migrant remittances from the RSA were declining rapidly, the elusiveness of management and outputs quality, as well as sustainability have necessitated exploration of alternatives. Another achieved target was that of localising key long term positions in LHDA by 2003. In the 1997 approved structure, 5 out of 30 managerial positions were originally held by expatriates, and by end of 2003, that had declined to 2 out of the reduced 27 managerial positions. The year 2003 signaled the end of construction and the start of LHDA restructuring meant to `downsize' to the core business of operation and maintenance. By end of 2006, only 2 of the 13 new structure managerial positions were held by expatriates. The restructuring process had been long (2003- 2006) and massive (60% staff reduction), and this had naturally disrupted staff performance. Other institutional improvement targets were met but not on time. The new governance structure was achieved in 2002 as against the 1998 target date due to delays caused by the requirements to first renegotiate the 1986 LHWP Treaty, which happened in 1998/99, and whose outcome was Protocol VI signed in June 1999. This in turn required the amendment of the 1986 LHDA Order, which happened in 2000. Following this, the new project governance structure agreed by the two parties in 1999 gave rise to the LHWC (2002) with a new mandate of overall accountability for the project, while the new LHDA Board, which was answerable to the LHWC, came into effect in April 2003. In summary, the institutional related objectives were achieved but only for the con- struction period. The transition to operation and maintenance had the effect of undoing some of the gains e.g. loss of planning and construction related skills, capacity and performance. The original (PAD) indicators failed to clearly visualize the post construction phase requirements and define appropriate related and guiding targets. Lowest cost alternative water supply to Gauteng: The objective seeks demonstration that the Eco- nomic Rate of Return (ERR) would be 16% based on consumer surplus analysis, with the unit reference value of water being M1.20/m3 compared to M1.72/m3 for the next best alternative at 12% IRR. The ERR was recalculated and found to be 11.5%, far lower that the PAD's 15.9%. The calculation is based on the water demand projection of 1.5% as opposed to the PAD's 3%. The ERR has been found to be sensitive to water demand projections that were revised down- wards as a result of the DWAF's restrictive water management policy (increased tariffs, increased irrigation water charges, promoting water conservation and repairing leakages, etc.) that started in 1995. This policy reduced the demand significantly, exacerbated by declining fertility, increases in HIV/AIDS mortality, and the Gauteng area in-migration reduction. The project costs reduction gain was off-set by IFR increases. This ex post rate of return (11.5% ERR) is still deemed accept- able relative to other options; and current data is not available to compare alternative solutions identified during preparation. Nevertheless, the fact that Phase 2 feasibility studies are being un- dertaken, suggests that it remains the most cost effective solution (see Annexes 1, 4 and 5 for fur- ther details). Any adverse social and environmental impacts mitigated: the extent of objective achievement here varied from one programme to another. Annex 1, Section 4 provides detailed consideration of specific targets. In summary, the construction related components (resettlement housing, ac- cess roads, water supply and sanitation provision, community infrastructure, etc) had objectives achieved. This is mainly due to the fact that such jobs are technically straight-forward, i.e. re- sourced and run on contract lines, involve people less, and take fixed (shorter) durations. On the other hand components with longer duration (compensation, income restoration, public health, ICM, natural environment, monitoring, etc), realized less, due to the heavy reliance on people's - 61 - involvement, being amenable to bureaucratic, political and social controls or interferences, rely- ing more on internal (limited skilled) staff than outsourcing (contracting out), generally attracting less resources (commitment) than outsourced jobs, and hence carrying high sustainability risk. Local development spin-offs maximized: of all the project objectives, this one is the most broadly achieved, with targets surpassed in most areas (more details under components achievements - Section 3.2 and Annex 1). In summary, the economic spin-off target of 16,500 person-years of employment was surpassed with 20,000 person-years of employment reached during construc- tion, earning some M623 million against the target of M275 million. The local contracting, sub- contracting and supply services earned M300 million, which is double the target. Labour from the Highlands constituted 40%, earning some M249 million. While the indicators above can be con- sidered transient project benefits, there were significant permanent benefits that include the feeder roads system (completed in 2004) around the reservoir that were intended to compensate for im- peded access. This consisted of 75kms of all-weather gravel road, 4 vehicular and 3 foot bridges. Other local accesses related to the resettlement construction include a further 100kms of earth access pioneered for water supply and sanitation project to the remotest villages of Mohale catchment. The key performance indicators reflecting the local impact of this infrastructure on access times, cost of transport, and imported goods along with utilization were never directly as- sessed. However, given that travel times prior to the project took an average of 8 hours from up- per Mohale to Maseru, the improved infrastructure network has reduced travel time to less than 3 hours, improving accessibility to services, as well as reducing local prices of goods. Other indica- tors are covered under the component achievement below. 3.2 Outputs by Components Component A: Construction of Main Project Works: The physical targets for this component were all achieved, with some variances on costs and time. Mohale dam delayed by a year but was completed within budget. Mohale Tunnel was completed some 20 months later, with costs over- runs (22% more). These overruns were mainly caused by unforeseen geological conditions that saw frequent TBM breakdowns as well as encounter with subterranean water which in turn gen- erated several claim dispute settlements. Matsoku Diversion was completed 2 months later but within budget, while the Infrastructure was completed on time but above budget (27% more). The PAD's overall construction costs target of US$531 million was more than met when the overall costs came to US$378 million or 71%. The target of keeping the overall costs of Phase 1B below US$1.2 billion was achieved as the overall costs by December 2006 stood at US$629 million or 71% of the appraisal estimates. Component B: Engineering and Supervision: The PAD had estimated US$105 million for this component for design and construction supervision services for Mohale Dam, Mohale Tunnel, Matsoku Diversion and associated infrastructure. All targets were met with the overall services costs of US$79 million or 81% of the appraisal estimates. These costs include the supporting or miscellaneous engineering jobs such as geo-technical investigations, surveying and mapping, monitoring, Expert Panel and Dispute Review Boards, all totaling US$17.93 million. Component C: Environment and Social Action Plan: Valuable lessons were learnt from Phase 1A for enhanced social and environmental care, and US$140 million was allocated to this component during preparation. The costs cover programmes that vary in duration from a few years (starting in 1998) to 2003 for majority, with the rest running to 2006 while a few go still to 2010. Pro- grammes that run longer (e.g. 50 years project's lifetime), such as compensation and monitoring, will require extra resource allocation. In general, implementation of the EAP has been behind schedule and by the end of 2006, the overall expenditures were US$82.7 million, or 71.5% of the appraisal target. The main sub-components and achievements are given below. - 62 - Resettlement of affected communities in three phases: the physical resettlement aspect of stages 1 and 2 was successful, with the timely Stage 1 (pre-construction) resettlement of 99 households, and the Stage 2 (pre-impoundment), also successfully resettling the expanded 216 households between 2001 and 2004. The PAD's stage 2 target had been 172 households resettled by January 2001, but this was based on the impoundment target of November 2001 which was delayed by a year. This involuntary resettlement part (Stages 1 and 2) was deemed a success in that the displaced went to destinations of their choices (most chose foothills where relatives re- sided) where replacement housing (with choice of plans) was of equal or better value, with addi- tional amenities of sanitation, fencing and heating facilities. Relocation also involved moving displaced persons with their belongings, including graves relocation, compensation (replacement or cash) for kraals, stables, and other structures, as well as providing related communal assets compensation as seed capital for communal development ventures in their destination areas. Those resettled were further compensated with a disturbance allowance for the first 3 years. Those with annual income below the minimum threshold (M3,760 in 1997 prices) had their in- come topped up for 10 years after resettlement. However, the post resettlement livelihood maintenance compliance remain to be confirmed (through ongoing M&E studies) and there remains significant debate among stakeholders. Unfor- tunately 2 previous monitoring studies, proved too deficient to inform the debate (see M&E sec- tion below), and it is hoped that the current one (Contract 1204) would succeed in early 2007. The Stage 3 resettlement (post impoundment and voluntary upon losing over 50% arable land) has not yet been implemented but is still under consideration; with delays caused by a long drawn-out debate on merits of relocation versus `in-situ' development. Finality on the issue is still pending. Details of other related target achievements are outlined in Annex 1 and 2, suffice here to show that community infrastructure was successfully provided in the form of local access roads to the resettlement areas as well as water supply to the resettlement and host areas (completed by 2004 in the Foothills). In terms of schools infrastructure, success has been good in that all 7 Foothill schools were completed (20 new classrooms, 6 staff rooms, 4 kitchens and 86 VIP latrines) by March 2004 while for similar 7 Highlands schools (total of 13 class rooms, 4 kitchens, 5 staff- rooms/book stores, 16 VIPs in 2 schools, and renovations to 2 schools) the project was completed in September 2004. Participation of the project affected people, initiated through the Resettlement and Development Study (Contract 1012) in 1996-97 has been central to the resettlement process, structured around the Area Liaison Committees (ALCs) which were facilitated by Community Liaison Assistants (CLAs) that linked it with LHDA. The structure was eventually transferred from the Consultant to the LHDA in late 1997 (end of study period) from where it is still operat- ing. The CLAs were abolished in 2003, after construction completion, due to their operational costliness. Compensation for project related losses: the targets as set out during the preparation stage were largely met subject to the following variances: i) while the Compensation Policy was consulta- tively designed, adopted and implemented on time (from September 1997 for 1B and April 1998 for 1A retrofitting), the review was delayed to 2003-06 period, implying minimal design and op- erational shortcomings; ii) communally owned assets compensation disbursement was started 2 years later in 2002 following the adoption of cooperatives formation concept as a means of ac- cessing funds; iii) individually owned assets outside of arable land compensation also started in earnest in 2002. However, delays in disbursing these compensation items, including delays in attending to the resettlement housing construction defects, generated so many complaints that complainants sought the intervention of the Ombudsman who generally ruled in their favour, or- dering LHDA to redress the situation. LHDA is working intensively to address the Ombudsman's - 63 - recommendations by implementing the finding where possible and reviewing the finding where it challenges policy.. Compensation paid to-date totaled M83 million including communal assets, which is 319% of the allocated sum of M26 million. This variance can be explained by the fact that the estimate related only to the annual payments for construction period (1998 to 2003). Downstream compensation disbursed to-date totaled M27 million for 54 communities as the first 10 year tranche, with other communities still being organized. It was not budgeted for during PAD preparation because IFR studies were still underway (see Annex 2). In summary, while the Policy was regarded as gener- ally good, it was the application that was fraught with problems requiring solution. This solution would have to, amongst others, address the inadequacy of operational resources including skills, bureaucratic constraints, community and other stakeholder empowerment and involvement, community and public relations, acceleration of lump-sum payments to get people off annual compensation register, effective monitoring and evaluation, responsive complaints management system, etc. (see Lessons Learnt). Income Restoration and Training (including credit facilities) and Rural Development: this has been one of the least successful programmes. While targets were partially met, the scale of operation was too small to make significant impact. The programme was limited by the fact that: i) while agricultural programmes (maize seed and seed potato production, general maize produc- tion, poultry, vegetables and fruit trees), training and income generation are all under limited scale implementation, those of forestry and range management have to await the ICM that started mid-2005; and tourism promotion is still to start (under HERRIEP that started in Phase 1A); ii) the idea of a credit facility was abandoned, given the failures in Phase 1A; iii) only 154 business plans had been developed by 2003 (244 by 2006) with 202 paid or implemented; iv) records of annual jobs created have never been kept but they are unlikely to surpass the target considering that those 202 approved investments comprise: rental units in town (70), taxis (45), agricultural investments (36), milling equipment (8), shops (9) and markets/ savings investment (34); v) cu- mulative totals of those trained reached 625 by 2003 and 860 by the end of 2006. Training is now targeted to various individual and communal compensation funds investment ventures (income generation), while before, they comprised imparting skills in carpentry, catering, building, sew- ing, book-keeping and record keeping (more information in Annex 1). Monitoring and Evaluation: The targets of the 1998 first round of monitoring were met through the first socio economic and epidemiology monitoring study (Contract 669) of 1998/99 that com- bined Object Impact with Participatory M&E, that started in July 1998. In fact between 1998 and 2006 only 3 rounds of monitoring studies were carried out as against the annual target, caused largely by delays in the consultancy procurement process (scope of work approval, repeated bid- ding processes, and bids adjudication and contracting process). Baseline data were all defined in that first 1998 study, but the study outcome was deemed deficient as it failed to conclusively pro- vide an answer on the attainment of the standard of living maintenance (or otherwise), which has Treaty compliance implications. Improvements were made for the second round that started in August 2000. However, while the raw data was deemed satisfactory the results and analyses were not deemed satisfactory, lacking multi-variate analyses and the resultant standard of living index for compliance monitoring. The idea of securing a different consultant to re-analyse the data was discarded in favour of procuring an improved third round. Improving the scope of work through the Panel of Experts, deciding on the approach (in-house team versus outsourcing) and to procur- ing the consultant took 3 years, with the resultant Contract 1204 starting in July 2005. It is still underway as of December 2006, and it covers the downstream as well, with the results expected in the first half of 2007. It is obvious the current approach requires further improvements. The restructuring of 2003 brought all LHDA's monitoring under one roof (socio economic, health, - 64 - IFR, water quality, etc.) but with reduced staff complement and dearth of skills. Monitoring on regular basis and utilizing the results still remains a serious challenge for LHDA. Natural Environment and Heritage: Overall achievement of the objectives here is rather low (see Annex 1). Mitigation measures associated with construction, including annual independent construction site audits, were achieved under the construction supervision contracts. Protection of the natural environment through the ICM programme (watershed management, range, hot-spots or in-situ conservation, wetlands, soil and water conservation for both Phases 1A and 1B areas) only started in May 2005 (through Contract 1044) as opposed to the appraisal target of 1999. This was caused by delays in developing scope of work for each programme (2000-2002), in deciding to combine them under one umbrella contract (1044), as well as the 2 year long procurement process. Only one Biological Monitoring study was undertaken from 2001 to 2003. It showed satisfactory status relative to the baseline study of 1996/97, with small variances here and there in some species composition, which were attributed to habitat deterioration (overgrazing). Three reserves were established to ensure protection of biodiversity and cultural heritage. A follow up study was recommended, with monitoring at 3 year interval, but lack of skills in LHDA con- strained their carrying out. Preservation of Maloti Minnow was identified during preparation as a key concern. The objective was partially achieved when the study and implementation resulted with the trans-location of 1,711 Minnows to 4 new rivers and related monitoring is continuing. Captive breeding experi- ment showed no success and was abandoned. The proposal to construct two barriers to prevent upstream migration of minnow predators (trout from Katse reservoir) is under consideration. The Biodiversity Trust for developing sustainable programmes was established in 2003/04 to ensure minnow preservation with the seed capital of M9,146,011. The Trust started operations in 2005, and it is entrusted to monitor the situation closely and recommend appropriate action including barriers construction if warranted. The project has pioneered the development of environmental flows, with studies started in 1996 and completed in 2001. From there a comprehensive policy and set of procedures governing ap- plication were developed and approved in June 2003. Developed through participatory scenario approach, and informed by cost-benefit analysis framework developed to compensate for impacts associated with changes in flow ­ the policy paid out M27 million as a first 10-year tranche to 54 cooperative communities of proximal reaches 1,2 and 3. Other communities in reaches 7 and 8 (Mohale downstream) are in the process of receiving their part. Other communities in other reaches will be compensated later subject to impacts verification through monitoring. Down- stream monitoring of biophysical (Contract 1237) and socio economic (Contract 1204) started in June and July 2005 respectively and the information is used to inform further development revi- sion of the policy and procedures. IFR Policy will be reviewed every 5 years following the audit, but the Procedures would be amended from time to time. Water quality monitoring (12 key parameters) is carried out through monthly water sampling and testing (through RSA laboratories). The analysis when compared with South African standards consistently shows that the quality of water delivered is satisfactory. The development of formal quality standards specific to rivers in the projects is only starting. Public Health Program: Those health related programmes associated with the civil works were achieved while sustainability of most of the long term ones remain problematic. Targets concern- ing workforce health care, transfer of health facilities to GOL, and Water and Sanitation project, were all achieved. The Leribe Trauma Unit (LTU) functioned successfully under the project man- agement, treating patient ranging from 1,000 to 6,800 annually until its transfer to GOL in March 2003. Health services were successfully provided to the local communities until 2004, after which - 65 - staff depletion (due to downsizing and end of construction) caused service levels to decline. Other health facilities handed over to GOL were Lejone clinic in November 2002, and Katse and Mo- hale in December 2004. Preparations for hand-over had started 3 years earlier. Monitoring showed that GOL's Ministry of Health operated the facilities at their own normal standards which are understandably lower than the project's. Monitoring of health indicators formed part of socio economic monitoring (Contract 669 and 1204) progress as described earlier under monitoring and evaluation. The Project made significant contributions to improving water supply and sanitation services in the highland regions. Upon completion of work in the phase 1B area at the end of 2005, 105 water supply systems had been provided to 126 highland villages (some villages shared systems), 2,545 VIPs for same number of households, as well as 2,545 refuse disposals and 2,545 soak-aways. With regard to HIV/AIDS, LHDA emphasized protective measures in both project area and workplace. HIV/AIDS Personnel policy was formulated and approved and MOU was established with Lesotho Red Cross Society, Ministry oh Health and Social Welfare to ensure development and sustainable HIV/AIDS mitigation projects within the project area. As part of dynamic education awareness, Billboards and T-shirts bearing HIV/AIDS messages were used. Male and female condoms are annually distributed at workplace and public places within the pro- jects area. Component D: Institutional Support and Capacity Building for LHDA: the success of this component has been rather short term, relating mostly to support during the construction phase (1998-2003) for technical assistance and special studies. In financial terms, the cumulative ex- penditure to December 2006 stood at US$84.3 million or 77% of the appraisal estimate (US$110 million). However, the operation and maintenance costs run indefinitely, and are expected to reach about US$97.5 million (111.4%) by 2010. While construction targets were largely achieved (Annex 1), other indicators relating to governance and training show mixed success. The formal training policy and plan, scheduled for September 1999 was never prepared, with only interim measures in place until the advent of the 2003 organisational restructuring. Professional training that took place between 1998 and 2003 mainly involved working as counterparts to the expatriate specialists, attachment to consultancies, some short term and limited long term academic training (mainly geared to support operation and maintenance). Training (except for engineers) was re- stricted to SACU area, resulting in the the Bank's training funds not being utilized. 3.3 Project Net Present Value/Economic rate of return and Financial rate of return These were calculated by the Bank at appraisal, and have been so recalculated (by the Bank) as appended in Annex 4 and 5. As mentioned in Section 3.2 above, the ERR has been calculated to be 11.5% based on the water demand projection of 1.5% per annum while PAD's target had 15.9% ERR based on the projected water demand growth of 3%. This figure still puts the project above other alternatives. Water demand projection figures are still inconclusive, and are still be- ing reworked, especially in the light of the start of the LHWP Phase 2. 3.4 Institutional development Impact The institutional impact of the project has been positive with respect to LHDA as an implement- ing organization. The 1997/98 restructuring of LHDA injected the necessary capacity to imple- ment Phase 1B, with local staff in management positions, while concurrently operating and main- taining the Phase 1A facilities. This was successful and Lesotho attained the local capacity to sus- tainably develop its water resources (future phases of the LHWP and Lowlands Water Supply). Expatriate advisors, specialists, as well as expert panels for engineering and environment boosted that local capacity. However, the 2003 restructuring (downsizing) aimed at refocusing the organi- zation's core business towards operation and maintenance, saw the loss of some of those skills (except operation and maintenance). This has had the negative effect on the LHDA's organiza- tional efficiency, skills retention, and capacity with respect to future water projects implementa- tion. While the new governance set-up brings long term improvements, the short term dearth of - 66 - requisite operational skills is constraining performance e.g limited Board's technical leadership, restructuring outcome, and prompt approvals. In summary, the institutional impact was positive for construction related purposes only, but as most skills (including institutional memory) left for the neighbouring RSA, future projects are likely to require re-skilling. 4.0 Major Factors Affecting Implementation and Outcome Factors Outside of Government and Implementing Agency Control: as far as construction of major works is concerned, delays of more than 10 months on the tunnel were mainly caused by frequent TBM breakdowns as a result of unexpectedly difficult tunneling conditions and the unre- liability of the TBM as well as with subterranean water. The time needed to ensure effective par- ticipation of project affected people in the development of sustainable livelihood interventions, resulted in delays in implementing the social and environmental mitigation when compared with the engineering components. Factors Generally Subject to Government and Implementing Agency Control: ensuring that LHDA continues to improve performance and capacity for the sustainable development of Leso- tho's water resources is a factor within the control of LHDA and the GOL; in particular, skills retention for operation and maintenance, as well as for developing subsequent phases. Another factor within control is providing resources for effective implementation of the EAP (Treaty compliance), and yet Lesotho's financial contribution has been less than required, which in turn limits the RSA counterpart contribution, and this will require rectification. Tardy attention to complaints lodged by the affected is within LHDA's control; and if left unattended, it runs the risk of alienating the complainants and hence discouraging their effective participation on sus- tainable livelihoods programmes (EAP). Costs and Financing: at Appraisal the total project cost was US$ 884.4 million, with the Bor- rower having secured a loan from the World Bank in the amount of US$ 45 million for Phase 1B. The Loan agreement was signed on 18th December 1998 and became effective on 9th June 1999. However, fulfillment of conditions precedent necessitated some use of bridging finance from the Common Monetary Area. The World Bank's Phase 1B loan accounted for about 5.1% of the total project capital cost, with the Phase 1A loan providing an additional 0.9% that was ultimately can- celled. LHDA requested several cancellations which ultimately totaled US$ 23.34 million and thus reducing the loan amount (at closure) to just US$21.66 million. Final project costs now amount to US$ 624.3 million, while cancellations meant that the World Bank loan only contrib- uted 3.5% of final financing. 5.0 Sustainability The project is expected to deliver water to RSA and generate power for Lesotho for its designed life of 50 years. Sustainability in terms of revenue generation is assured through the provisions of the Treaty. Operation and maintenance of major project facilities for the life-time of the project is the responsibility of the Treaty Parties (RSA and GOL). Repayment of all the water transfer debt relies on the revenue stream from water sales to the Vaal River system water users, while on the Lesotho side it relies on the Royalties and hydropower sales. The LHWP Phase 2 feasibility stud- ies started end of 2005, with implementation depending on the RSA's water demand. The institu- tional arrangements could always change in line with future challenges, but skills retention re- mains a challenge. The technical risk due to sedimentation and hydrology was considered very low and integrated catchment management is being implemented to address resource manage- ment and conservation. For downstream water releases, periodic reviews of the IFR Policy and Procedures should ensure their adequate responsiveness to climatic and other variations. Within Lesotho, the sustainable use of project revenues to fight poverty continues to pose an operational challenge. Sustainability of mitigation measures in maintaining livelihoods of the affected as the - 67 - Treaty requires is another challenge facing the LHWP Authorities, as compliance is a lifetime challenge. Due to the pre-existence of Phase 1A facilities under operation and maintenance by LHDA from the 1998 start of Phase 1B major works implementation activities, the take-over of completed facilities for operation and maintenance purposes posed minimal problems. 6.0 Bank and Borrower Performance The Bank: Overall, the Bank's performance during the preparation and implementation stage of the LHWP Phase 1B has been helpful in many ways, especially the following: · During the preparation stage, the Bank rightly advised of the need to conduct detailed geo- technical investigations of the project sites to reduce implementation related risks. · The Bank financed the engineering Panel of Experts at an early stage to advise LHDA. It proved useful as technical problems encountered during project planning and implementation of the contracts were discussed jointly with LHDA and supervising Consultants, sometimes at site level. · The Bank also financed the Environmental and Social Panel of Experts, overlapping with Phase 1A, from planning to construction and post construction stages as social and environ- mental issues take much longer. The POE input has been useful in assisting LHDA to navi- gate problems and effectively implement the mitigation programmes to the extent it has. Their usefulness has persuaded LHDA to extend their tenure beyond the Bank's loan closure. · The Bank assigned a Supervision Team which visited the Project bi-annually. This supervi- sion identified and flagged strategic issues for follow-up action by LHDA. The missions proved even more helpful when coinciding with the POE visits (unlike in 1A) as all advice coincided on site where conflicts could be resolved to enable LHDA to act on one set of ad- vice. · The Bank also helped LHDA to formulate and implement the IFR Policy and Procedures from December 2003 despite operational constraints caused by: i) design limitations of the dam structures facilities, ii) contractual obligations and requirements to meet construction completion schedules, iii) operating non-fully operational reservoir systems; and iv) commis- sioning as well as impoundment schedules of dam structures driven by natural occurrences such as rainfall. Borrower: The Borrower's performance is covered under Sections 3 and 4 above (objectives and outputs achievement). It is also covered under Section 5 (Sustainability), Section 7 (Lessons learned) as well as in the annexes. 7.0 Lessons Learned Given that Phase 1B was built on lessons from Phase 1A which were important in informing the design of Phase 1B, the Phase 1B lessons learned will in turn be important in informing the sub- sequent phases of the LHWP. There were 19 Phase 1A lessons that were incorporated into the Phase 1B design, and all were successful except the following: 1. "Responsive complaint management based on operational field teams, a conflict resolution mechanism and local project offices", as intimated earlier, tardy response to the lodged complaints as well as smart mobilization of the affected through the local NGOs called in the Ombudsman's intervention who ended up finding in favour of the complainants ­ which means that LHDA still has to effectively implement that lesson. 2. "Methods to devolve housing construction to owners and local contractors, to minimize costs (cost savings accrue to households), generate local employment, and maximize owner involvement and satisfaction", as intimated earlier, the LHDA decided instead, to establish in-house construction team using local labour that did the work rather well in terms of qual- - 68 - ity, time and lower costs, while still maximizing local employment. Phase 1A lessons in fact showed `owner built' housing to be impractical on account of limited construction and su- pervision skills, owner availability versus resettlement destination, quality and time versus main works construction schedule. 3. "Strengthened management and project selection procedures, for Development Fund, as well as enhanced direct community participation ­ based on international best practice", as men- tioned earlier, the Community Fund failed like its LHFP predecessor due to `political inter- ference in project selection' resulting in poor quality outputs; to the extent that GOL is con- templating moving the fund to the newly elected community council structures. This means that fund operation set-up is still to be mastered. In addition to the above lessons which should be learnt, Phase 1B has the following lessons that could inform planning of the LHWP further phases: 1. Future tunnel contracts should include TBM specifications for a full range of possible condi- tions, and any proposed or required refurbishment should be fully inspected to minimize risks of delays and related claims. 2. A single contract awarded for both design and supervision work should be preferred in order to avoid procurement delays, risks of non-ownership, as well as possible misinterpretation of logic and/or assumptions made by designers. 3. Coordination between various construction contracts could best be made the responsibility of one of the main Consultants, for the purpose of standardizing the general conditions of contract clauses, examination of program interdependencies, evaluation of programme risks, and more realistic liquidated damages calculation. 4. Lump-sum compensation should be preferred over the annuity (for 50 years) in order to leave the compensation register with only the vulnerable households (as informed by con- tinuous monitoring and evaluation). This will cut the current `umbilical cord of dependency' on compensation and hence minimize the number of potential complainants. `Prompt com- pensation' is also the Lesotho constitutional requirement where land is expropriated for pub- lic purpose. 5. LHDA has mastered construction of feeder roads and bridges by balancing the machine use with labour intensive mode whose outcome has been good quality roads (and bridges), much lower costs (than outsourcing) while optimizing local labour engagement and hence boost- ing local economy. 6. A similar lesson is that of establishing in-house teams to do work alone or complemented by outsourced consultancy skills, and this has been utilized in the cases of: Resettlement hous- ing construction, WATSAN provision, IFR Biophysical Monitoring, Impact Monitoring, and Eco-tourism (starting in Phase 1A); this has had the benefit of keeping the costs down, im- proving internal and local skills, maximising local employment, and enhancing sustainability chances. 7. Post implementation assets transfer process would require a comprehensive strategy to en- sure that all those agencies/institutions earmarked for eventual take-over of "ready to trans- fer" facilities or projects are involved from their planning stages, in order to ensure `buy-in', preparedness and operational resourcing. 8. In the organisational restructuring process, a skills retention strategy is a prerequisite. Any institutional changes need to be implemented quickly and efficiently to minimize the period of uncertainty that also impacts negatively on staff performance and productivity. 9. The requirement for Monitoring and Evaluation is another lesson LHDA is learning the hard way that without it, it is unable to know as well as to assure stakeholders that the standard of living of the affected is being maintained, improved or otherwise in compliance with the Treaty requirement to the effect. A single M&E contract should be defined during project preparation to cover the entire project period. This would ensure continuity and consistency in methods and data and obviate long-drawn consultancy procurement process. - 69 - Annex 8. Comments of Cofinanciers and Other Partners/Stakeholders Building on the assessment of Phase 1B of the LHWP undertaken by stakeholders during the ICR workshops and the individual interviews, a range of stakeholders were invited to comment on the ICR. These included the Borrower and Implementing Agencies, along with co-financiers and other national and international stakeholders. In addition to the comments received on the project performance during the workshops and interviews, detailed written comments on the ICR were received from the LHWC and LHDA, the Development Bank of Southern Africa (DBSA), and a joint comment from the International Rivers Network (IRN) and the Environmental Defense (ED). The DBSA was a cofinancier and both NGOs have a long-standing interest in the LHWP. The Borrower's evaluation is provided in Annex 7 and their comments have been addressed in the revisions to the ICR. The comments from DBSA and IRN/ED follow. 1. Comments from the Development Bank of Southern Africa The Report correctly reflects the objectives, process and achievements realized in the project. You may recall that a consultant also interviewed me some time ago with regard to stakeholder views. Just a few comments from my side: Technical Assistance With reference to paragraph 36, it is sad that a relatively low percentage of the amount available for capacity building was utilized - it was one of DBSA's main issues of concern at the time. In our interactions with field staff the desire for training, especially with regard to the evaluation of projects within the context of the revised Phase 1B compensation strategy was expressed many times - it just seems that it never materialized. Institutional Arrangements & Socioeconomic Issues Paragraph 38 correctly states that many of the institutional problems persisted in 1B. Allocation of responsibilities and the coordination thereof was a problem, not only within the LHDA but also between various government departments ­ much of this could possibly be ascribed to an inherent resentment because of the "privileged" status of the LHDA people and their perceived arrogance in government circles. I recall it was a major problem to get maintenance issues sorted out, in par- ticular with regard to the roads and communal infrastructure. Another problem was that commu- nal infrastructure was perceived to belong to the LHDA. Somehow the beneficiaries were not able to accept ownership, even in 1B ­ one wonders if the foundations were properly laid, and whether the preparatory phase (community liaison etc.) before engineering commences should not be much more extensive and much longer than current practice suggests. All of this links up with paragraph 49, where mention is made of the persisting dependency on compensation rather than on developed, alternative means of livelihood. The role of the LHDA is another issue, as indicated in paragraph 57. Environmental & Social Components Paragraph 55 states that the innovative approach to compensation and livelihood reestablishment was "acknowledged as world class" (Ps 31 to 33), and correctly states that it is ongoing work. I agree that the approach is world class, but the outcome, regrettably, is not world class. We have to acknowledge this. Somebody will have to ensure that a similar experience such as Kariba (Gwembe Tonga) does not repeat itself, where people are even more destitute than before ­ does the GoL have the capacity & will to do so? Transition Arrangements The DBSA would strongly support the proposal that ongoing WB support would be required, over and above mere monitoring. Social matters will have to get long-term attention. The ration- - 70 - ale of scaling down of the LHDA's activities may not have been the way to go, although it is un- derstood that the GoL should ultimately be responsible. The WB as an independent body can have significant clout in assuring that the less than world class outcome of resettlement and com- pensation, and socio-economic development in general can be addressed. Achievement of Project Objectives Paragraph 60 refers to the benefits SA achieved and to the fact that Lesotho's revenues were in- creased (M1,918m) and to governments "solid budget allocation framework", but also states that the impact of royalty revenues on poverty "is difficult to determine". 9,000 temporary jobs were created and M400m was earned. What about permanent jobs? What about the conversion rate? What about capacity building? We all know that temporary employment creates expectations and inevitably leads to disillusionment if permanent employment does not follow. Box 3 refers to the fact that Lesotho is spending a significant amount on poverty reduction. This must be maintained ­ the WB and other DFIs can play a significant role in this regard. Sustainability is the issue. Paragraph 66 says that impact of compensation is difficult to measure ­ therefore the need of maintained, hand-on support to the communities ­ for a long time to come. The fact that HIV/AIDS programs fell flat upon the failure of GoL to take over these initiatives (P67) supports the need for support to the entire GoL, including the restructured LHDA. Maybe the downscaling was not best practice, as is referred to in P68, with key staff being lost to SA? Somehow the most critical functions were done away with. Risk Rating Having been involved in the project for more than 5 years, I fully support the sentiments ex- pressed by stakeholders in P71 ­ technically the project was world class. As far as addressing socio-economic issues in a sustainable manner it was less successful, although a significant im- provement over Phase 1A is acknowledged. I largely agree with the risk rating on Borrower per- formance (moderately satisfactory), although more could have been achieved had more time been allowed to provide a sound social and institutional framework on which to build the project. Lessons Learned & Best Practice Paragraph 79 & 80 address best practice & lessons learned ­ this is accurately reflected. The sen- timents expressed in P81 (social objectives) are strongly supported and should be expanded upon to include recommendations concerning ongoing, hands-on support by field P 82 refers to the adoption of adaptive management practices ­ this is not only relevant with regard to environ- mental issues but especially when it comes to resettlement and compensation staff ­ one of the issues the DBSA picked up during field visits was the lack of delegation and empowerment at field staff level, together with the lack of training which would enable delegation (also refer to your P36 which states that TA was not utilized). With regard to P84, institutional arrangements ­ there is an inherent risk in establishing an elite implementing agency (jealousy, rivalry, weal co- ordination with other departments, capacity constraints etc.) All of these need to be considered and lessons learnt from the problems experienced especially with regard to the hand-over of as- sets and the required sustainability of development and compensation-related issues. I would plead with the WB to actively engage with the stakeholders to address sustainability is- sues, as highlighted under 5.5 on page 66 ­ it is indeed a lifetime challenge which in all probabil- ity, only an organization like the WB can help with. You and your team did an excellent job, overall, given the many constraints. - 71 - 2. Comments from the International Rivers Network and Environmental Defense We appreciate being given a chance to comment on this report. The following summarizes our major concerns with the report's analysis. The World Bank's central goals are to reduce poverty and promote sustainable development. In order to offer meaningful lessons for future World Bank lending, the Implementation Completion Report (ICR) has to respond to the question of whether the Bank has achieved these basic goals and to explain the conditions that contributed to the success or failure of the project. While the ICR attempts to answer these questions, the answers are deeply buried within the report and sur- rounded by contradictory statements. This format is more prone to promote confusion than to be conducive to learning. Related to this, we question how the ratings of "satisfactory" for the project as a whole and for Bank performance were determined. The project fared poorly on two of its four main goals (pov- erty reduction and capacity building), and the report contains many examples of poor perform- ance and unacceptable outcomes concerning the poverty impacts and sustainability of the project. Yet the simplified rating system implies a project that is on track for meeting its goals and a suc- cess for all parties. Thus, the rating system appears to serve primarily to allow the casual reader to take away a false impression from the report that the LHWP has been a success. Over the course of the project, the ratings moved from 'satisfactory' during the implementation period, to 'moderately unsatisfactory' during the two extension periods and to 'moderately satis- factory' at closing (p. 8, #2.2). If project implementation was 'satisfactory', one may ask, why did there have to be extension periods at all? Clearly, the social and environmental aspects ­ the ele- ments most critical to poverty reduction ­ were not implemented in a satisfactory fashion. Below are more specific comments on major issues described in the report. POVERTY ALLEVIATION Although one of the project's four key goals is "Maximizing the Poverty Reducing Impact of LHWP," there are major shortcomings in the project's contribution to poverty alleviation. "Box 3: Poverty Alleviation and the LHWP" (p. 19) states that Lesotho has gotten poorer since the project began ("Since 1995 the Human Development Index has declined and Lesotho now ranks 28th lowest on the HDI"), and notes that "off-budget" programs failed to help project-affected people recover from the project's impacts to them (these programs were thus rated "highly unsatisfac- tory"). The report implies, however, that an increase in GDP from the project makes up for this failure, and is leading to poverty reduction programs and expenditures. We urge the Bank to reconsider this kind of accounting. When key programs for poverty reduction fail, it is insufficient to use a macro-economic picture to justify the project's poverty alleviation capabilities. It is critical to consider benchmarks that show how directly affected people have fared as a result of increased government revenues. The case is not made in the completion report; in fact, the excerpts below paint the opposite picture: · "The impact of royalty revenues on poverty is difficult to determine." (p.18, #3.2) · "Agricultural production and household nutrition in areas affected by the LHWP have followed national trends of decline." (p. 21, # 3.5) - 72 - · "Delays in the implementation of a number of important social and environmental activi- ties (e.g., ICM, compensation payments, Maluti Minnow) contributed to shortcomings in achieving the project's social and environmental development objectives, re-affirming the challenges of addressing resettlement and downstream impacts with a development framework." (pp. 21-22) · "[C]ommunal payments have been problematic because the institutional apparatus for payment and utilization of compensation have proven extremely difficult to establish" (p.10, #32). LHDA and GoL to address this in the future." · "The GoL exhibited limited commitment to addressing the project's broader social and environmental objectives and failed to capitalize on a number of significant opportuni- ties" (p.24, #76). · Monitoring and evaluation on environmental, social and poverty reduction was not car- ried out in a satisfactory manner. "These problems have complicated determination of project impacts on Highlands communities and the extent to which the Treaty provisions have been met." (p. 13, # 2.3). · Large-scale construction projects in remote areas are well-known as a major factor in the spread of HIV/AIDS. Yet, no adequate safeguard measures were taken or appropriate M&E systems set up. While HIV/AIDS has become a very serious problem in the High- lands, there is inadequate health care: "The loss of public health staff after the re- organization and the failure of other GoL agencies to take up these programs have un- dermined their long-term contribution." (p.21, #67). SOCIAL AND ENVIRONMENTAL COMPLIANCE According to the ICR: "The lessons learned in LHWP Phases 1A & 1B and in similar projects are that intricate and detailed work needs to be done in sensitive areas such as working with affected communities and assessment of environmental impacts at the front end of the projects in order to avoid problems later on, and yet this crucial stage of project preparation does not fit well within existing instruments and project cycles of the Bank" (p. 28), While we appreciate the acknowledgement of the complexities involved, a more detailed analysis is required if project-affected people are not to be negatively affected and to actually benefit from Bank projects. The report reiterates that "no affected person should be left worse off" as the result of a Bank-financed project. It would have been laudable if one of the lessons emerging from the project would be the need to advocate an improvement in this standard to ensure that affected people are left BETTER off by World Bank projects. The report reveals some of the project's serious impacts on the environment; for example, para- graph 47, under OP/BP 4.01, Environmental Assessment, states: "Although it was originally re- stricted in scope, the EIA identified 140 impacts, including 29 downstream and 48 other impacts. Of these, 31 were of high significance and two of very high significance: the loss of life through HIV/AIDS and extinction of the Maluti Minnow. Failure to implement all of the measures which would have better protected the Maluti Minnow demonstrates a lack of commitment from the GoL and raises broader questions as to the long-term sustainability of many environmental meas- ures." Further, the report's "lessons learned" section states: "Large components of the implementation of environmental and social programmes lag by years" and "There are major shortcomings in red data species protection, resulting in the potential extinction of the Maluti Minnow." - 73 - On sustainability, the ICR states: "Failure to implement all of the measures which would have better protected the Maluti Minnow demonstrates the lack of commitment of GoL and raises broader questions of long-term sustainability of many environmental issues." (p.15, #47). LHDA is now expected to improve management and record keeping to provide adequate com- pensation in the future (p.13, # 42). The ICR claims that Katse Gardens are able to propagate the rare and endangered spiral aloe, while the last mission report (Jan/Feb 2007) expressed concerns about the lack of financial sup- port for the Katse Gardens. These flaws, combined with the poor performance of maintaining adequate downstream flows (see below), make this project less than satisfactory in addressing the social and environmental impacts of this huge scheme. The "lessons learned" section at the end of the report only begins to touch on some of the issues for this performance issue. INSTREAM FLOWS On the IFR study, the report states: "The audit found that implementation has been approximately 60% fully or partially compliant with the Policy and Procedures and has identified issues likely to affect the sustainability of the IFR. It must be noted that the IFR work described here is emerging science with many of the elements of the Policy and Procedure having been developed during and through the project." The report also states: "The contribution of the project to the determination of environmental flows is of global significance, although long-term commitment and sustainabil- ity remain to be proven." (p. 20) We applaud the commission of the world-class IFR study and the Bank's championing of this re- port on the project. However, the devil is in the details, and a good report cannot substitute for a commitment to protect downstream communities' health and well-being in this case. CORRUPTION For the sake of accuracy, we believe the ICR should be more straightforward about the World Bank's slow and limited response to the corruption that plagued the project. For example, the in- troduction (see parag. 4, p. 2) is simplistic: "The Bank cooperated in the investigations and has debarred one individual and two companies, Acres International (in July 2004) and Lahmeyer International GmbH (in November 2006)." Later, the report ranks the Bank's performance on cor- ruption as an area where the Bank performed especially well, stating: "Support for good govern- ance: The Bank supported good governance and dealt with corruption effectively, e.g. disbarment of fraudsters." (p. 56) On page 26, the report does offer areas of improvement for the Bank, stating: "The Bank has strong preventative mechanisms, including procurement procedures and legal frameworks, and supports capacity-building to prevent and pursue corruption, but lacks judicial powers to gather evidence and prosecute. More support is needed at the project level to develop indicators of po- tential corruption, including resources to enable Task Teams to pursue allegations or suspicions of corruption." This is a start, but more could be added to address the slowness in responding to convictions, and the fact that only a very limited segment of project contractors would qualify for debarment. We agree with the 2001 findings of a UK parliamentary committee report, which faulted the Bank for only disbarring a company if corrupt activity could be proved within a specifically Bank- funded part of the project. The report says, "That position is based on the narrowest legal inter- pretation of the Bank's guidelines and a singularly selective view of the Bank's involvement in the - 74 - project." The situation can only be improved by a broader interpretation of the Bank's key in- volvement in a project of this magnitude. As the ICR itself states, "The Bank provided interna- tional credibility to the project with its sound policies, professional approach and expertise. This helped in ensuring project-funding security." The project would not have been built (or pro- ceeded) without Bank support; in such cases, the Bank should take responsibility for corruption across the entire project, not just on the specific sections it has funded. Further, this section (p. 2, parag 5), seems to contradict official statements about the offer of fi- nancial assistance: "To support the GoL in meeting the costs of persecution, the Bank offered supplementary financing for an equivalent amount of eligible expenditures under any IDA Credit in Lesotho's World Bank portfolio in order to free up Government resources for the cost of the prosecutions. This offer was never acted upon, nor was the potential financial support offered by the EU and UK." It is our understanding that the Lesotho government did indeed ask for support; see for example http://www.lesothoemb-usa.gov.ls/news/Speech_Lehohla_Corruption.htm LEARNING FROM PHASE 1A The report emphasizes the importance of phasing the project, stating: "Phasing large projects en- ables the lessons from early phases to influence the design of latter stages and the incorporation of changes in the global context for example processes such as the World Commission on Dams. This is clearly evident from the experience of Phases 1A and 1B of the Lesotho Highlands Water Project." Yet Phase 1B seemed to move forward of its own weight, and clear lessons from the first phase went unlearned. To give just two examples as described in the ICR itself: · The ICR states that significant sunk investments and cost implications made it necessary to proceed with Phase 1B without delay (p.6, #18). While preparations for Phase 1B may have tried to consider lessons from poor resettlement, inadequate compensation, failure of programs intended to restore livelihoods and lack of adequate environmental protec- tion, the Bank should not have moved forward with Phase 1B before there was demon- strated capacity on the part of LHDA and GoL to find a satisfactory resolution of the problems created by Phase 1A. · A 1994 Back-to-Office report by a Bank staff person reported that the LHWP had failed to address the needs of the people negatively affected by the project and recommended a reconsideration of further disbursements if there were no improvements. Instead Phase 1B was approved in order to avoid construction delays. BEST PROJECT TO MEET THE NEED? The lower-than-expected ERR was a direct consequence of moving forward before learning the lessons of Phase 1A. At the time the second dam was approved, IRN reported to task manager John Roome that DWAF's Willie Croucamp stated in a meeting in Johannesburg (attended by IRN) that "Phase 1B could 'easily' be delayed seven years under the current state of demand man- agement without threatening South Africa's water supply, and could be delayed up to 11 years or more with the successful implementation of new demand management strategies." A delay of as much as 18-20 years was estimated to be feasible by Rand Water conservation experts. At that time, demand projections were incomplete and the ability of demand-side management efforts to effectively put off the need for the project had not been thoroughly studied. And yet, the ICR states: "Phase 1B had a lower than expected ERR due mainly to much lower water demand growth than projected at appraisal: from 15.9% in 1998 to 11.5% in 2006. An in- creased allocation to environmental flows also reduced the project yield and hence the ERR. ... it was clearly not anticipated that water demand growth would decline so drastically shortly after - 75 - implementation. And lastly, the project helped bring about the reductions in water demand growth as it contributed significantly to the development of water demand management policies in South Africa." (p. 19). (Despite this, the project was rated to have a satisfactory rate of return.) It seems clear that the project did not contribute "significantly" to the development of demand- management guidelines; in fact, World Bank staff at the time of approval argued against attempt- ing DSM in order to promote Phase 1B. A 1998 paper by then task manager John Roome, titled "Economics of Delaying Phase 1B," stated "It is not clear what the scope is for further demand management" and "Demand management capabilities and their impact in South Africa are theo- retical and have not yet been tried and tested." Mr. Roome stated in a March 28, 1998 email to IRN that delaying Phase 1B "must be weighed against the costs of demand management that will be needed to get these delays and/or the increased risks of water restrictions" in South Africa. Another lesson learned on this point that is not reflected in the ICR is that large, "lumpy" projects such as large dams are much more prone to such over-estimations of their profitability than smaller scale, decentralized additions to water and energy supply, and that better understanding of actual needs and the variety of ways to meet those needs is a critical component of project plan- ning. These are the lessons of the World Commission on Dams, and it would be especially useful to see such lessons described in detail in this completion report. CAPACITY BUILDING Another of the project's four key goals is "Improving Institutional Capacity". The project did poorly on this count. See for example this section: · "Institutional arrangements for the implementation of large infrastructure projects need to be clearly defined. Such projects often involve the formation of well-resourced extra- governmental implementing agencies. This needs to be regarded as a developmental op- portunity for the government with the potential to strengthen capabilities in related gov- ernment departments and build agencies which will be better able to replicate similar pro- jects. This potential was largely unrealized [emphasis added] in the case of LHWP Phases 1A & B due to a range of factors including political influence. A related issue is the transfer of assets from the implementing agency - the LHDA - to the respective gov- ernment departments. Generally the departments were poorly prepared and under re- sourced which may undermine the sustainability of the assets. Project design should in- clude how asset transfer will be achieved and consider budget allocation frameworks and clearly defined institutional responsibilities. Provisions should be made to track post- project asset performance and ensure they maximize potential benefits." (p. 27) Given that performance on two key goals ­ capacity-building and poverty alleviation ­ were poor, it is surprising that the project was rated satisfactory. THE BIG PICTURE According to the ICR, the institutional arrangements in Lesotho were inadequate because of po- litical influence (p.27, # 84). A critical lesson to draw from this is that a realistic assessment of the political economy of a country be undertaken before launching into large-scale infrastructure development projects. We very much appreciate the insights in this section: · Design and Implementation. "Current World Bank operational policies and procedures for project preparation - the Lending stage - do not provide sufficient time or resources for the preparation of large, complex projects such as the LHWP. Elements such as de- tailed design, resettlement and compensation planning, engagement of stakeholders, full environmental impact assessment including IFR design, building of client capacity etc. - 76 - are too complex, lengthy and expensive to fit within the Lending stage of the project cy- cle. There is no formal mechanism to accommodate this. As the Bank increasingly reen- gages in such large infrastructure projects, different, often ad hoc, mechanisms are being used to undertake the initial preparation phases including incorporating detailed prepara- tion work as components of other projects. One problem with this ad hoc method is that the host project of which the preparation is a component is often not a water resources development project and the project team is generally not aware of the complexities of large infrastructure works and their preparation. The lessons learned in LHWP Phases 1A & 1B and in similar projects are that intricate and detailed work needs to be done in sen- sitive areas such as working with affected communities and assessment of environmental impacts at the front end of the projects in order to avoid problems later on, and yet this crucial stage of project preparation does not fit well within existing instruments and pro- ject cycles of the Bank." (p. 28) Our concern is that this critical paragraph will not be used to re-evaluate the Bank's approach to how projects are planned and monitored. The critical lessons from projects such as this one, is that the Bank refrain from financing projects under some conditions and completely change the planning and design process to ensure that poverty reduction objectives are met under others. A central question that must be answered is whether a massive engineering scheme was the best approach for poverty reduction and sustainable use of natural resources for Lesotho. Unfortu- nately, the ICR avoids these issues, which are at the very core of development effectiveness. - 77 - Annex 9. List of Supporting Documents Lessons Learned Ambrose, J. (2006) A Dam Decade: Lessons Learned From Ten Years as an Environmental Specialist As- sociated with the Lesotho Highlands Water Project. Unpublished Report, October 2006. Devitt, P. & Hitchcock, R.K. (2007) Who Drives Resettlement? The Case of Lesotho's Mohale Dam. Un- published Report, Michigan State University. ECA Watch (2005) A Trojan horse for large dams: how export credit agencies are offering new subsidies for destructive projects under the guise of environmental projection. Hoover, R. (2001) Pipe Dreams: The World Bank's Failed Efforts to Restore Lives and Livelihoods of Dam-Affected People in Lesotho. International Rivers Network, Berkeley, California. Johannesson, P. (2007) Modified zoning and compaction requirements for future high CFRDs. Interna- tional Symposium on Key Techniques for Dam Construction, China, 2007. Johannesson, P. (2007) Slab performance on a few CFRDs and suggested improvements in slab design. International Symposium on Key Techniques for Dam Construction, China, 2007. Johannesson, P. and Tohlang, S. (2007) Lessons learned from the cracking of Mohale CFRD slab. Interna- tional Symposium on Key Techniques for Dam Construction, China, 2007. LHDA (2004) Macroeconomic Impact Study of Phase 1B. Lesotho Highlands Development Authority, Maseru, Lesotho. LHDA (2005) The Economic Impacts of Phase 1 of the Lesotho Highlands Water Project. Lesotho High- lands Development Authority, Maseru, Lesotho. Ombudsman (2003) Ombudsman's First Report on the Public Hearings in the Mohale catchment. Govern- ment of Lesotho. Maseru, Lesotho. 71pp. Ombudsman (2006) Ombudsman's First Report on the Public Hearings in the Katse catchment. Govern- ment of Lesotho. Maseru, Lesotho. 472pp. Thamae, M.L. and Pottinger, L. (2006) On the Wrong Side of Development: Lessons Learned from the Lesotho Highlands Water Project. Maseru, Lesotho: Transformation Resource Centre. Transformation Resource Centre (2004) The Irony of the `White Gold.' Maseru, Lesotho: Transformation Resource Centre. Transformation Resource Centre (2005) Lives of Resettled Communities After Resettlement: A Survey on the Lives of Communities Affected by the Lesotho Highlands Water Project. Maseru, Lesotho: Trans- formation Resource Centre. Watson, P.L. (2006) Managing the River, as well as the Dam: Assessing Environmental Flow Require- ments - Lessons Learned from the Lesotho Highlands Water Project. Unpublished Report. Related Bank Documents World Bank (1986) Lesotho Highlands Water Engineering Project (Credit 1747-LSO) Staff Appraisal Re- port. Report No. P-4286-LSO. World Bank (1993) Lesotho Highlands Water Engineering Project Implementation Completion Report. Report No. 11976. World Bank (1991) Lesotho Highlands Water Project Phase 1A: Staff Appraisal Report. Report No. 8853- LSO. World Bank (1999) Lesotho Highlands Water Project Phase 1A Implementation Completion Report. Re- port No: 19169. World Bank (1999) Community Development Support Project, Project Appraisal Document. Report No: 18387 World Bank (2004) Community Development Support Project Implementation Completion Report. Report No: 27944-LSO World Bank (2002) Lesotho Highlands Water Project Phase 1B Quality of Supervision Assessment (QSA5). LHDA Project Reports GoL / RSA (1986) Treaty on the Lesotho Highlands Project between the Government of the Kingdom of Lesotho and the Government of the Republic of South Africa, and subsequent amendments. GoL (1986) Lesotho Highlands Development Authority Order. Lesotho Government Gazette Extraordi- nary, Vol XXXI, No 54: 429. Published by the Authority of the Chairman of the Military Council, Mas- eru, Lesotho. GoL (2000) Lesotho Highlands Development Authority Amendment. Lesotho Government Gazette Ex- traordinary, Vol XLV, No 92: 842. Published by the Authority of His Majesty the King, Maseru, Leso- tho. - 78 - LHDA / LHWC (2004) Lesotho Highlands Water Project: Governance Manual regulating the relationship between Lesotho Highlands Water Commission and Lesotho highlands Development Authority. Mas- eru, Lesotho. LHWC () Principles and Procedures for the Release of Water from the Lesotho highlands Water Project into the Caledon River for use by Lesotho or South Africa. LHDA (2004) Organisational and manpower Study Completion Report. Lesotho Highlands Development Authority, Maseru, Lesotho. LHDA (2000) Mid-Term Review Report. Lesotho Highlands Development Authority, Maseru, Lesotho. Mohale Consultants Group (1997). Emergency preparedness plan for Mohale Dam, Lesotho Highlands Development Authority, Maseru, Lesotho. Resettlement and Compensation LHDA (1997a) Compensation Policy. Lesotho Highlands Development Authority, Maseru, Lesotho. LHDA (1997b) Compensation Rates. Lesotho Highlands Development Authority, Maseru, Lesotho. LHDA (1997c) Compensation Procedures. Lesotho Highlands Development Authority, Maseru, Lesotho. Phillips, R. (2000). Public Health Assessment (Contract LHDA 648), Lesotho Highlands Development Authority. Tshabala, Mavuso and Wilson Kisubi (2003) Socioeconomic Monitoring of Households Affected by Phase 1 of Lesotho Highlands Water Project: A Synthesis of 2000/01 Results. Lesotho Highlands Develop- ment Authority, Maseru, Lesotho. LHDA (2006) Compensation Audit Report. Internal unpublished report. LHDA, Maseru, Lesotho. LHDA (2006) Report on the LHDA Annual Stakeholders' Conference. LHDA (2006) Socio-economic and Epidemiological Impact Survey Upstream of Phase 1 dams. Report prepared by the Human Sciences Research Council. Contract No. 1204. Environment LHDA (1990) Environmental Action Plan. Lesotho Highlands Development Authority, Maseru, Lesotho. LHDA (1997). Lesotho Highlands Water Project Phase 1B Environmental Impact Assessment, Lesotho Highlands Development Authority, Maseru, Lesotho. LHDA (2005) Conservation of the Maloti Minnow (Pseudobarbus quathlambae) (Contract LHDA 1041), Lesotho Highlands Development Authority, Maseru, Lesotho. Instream Flow Requirements King, J. and Brown, C. (1999). IFR Methodology: Volume 1 (Contract LHDA 648), Lesotho Highlands Development Authority, Maseru, Lesotho. King, J. and Brown, C. (1999). IFR Methodology: Volume 2 (Contract LHDA 648), Lesotho Highlands Development Authority, Maseru, Lesotho. King, J. Sabet, H., Brown, C. and Hirst. H. (2000). Monitoring protocol (Contract LHDA 648), Lesotho Highlands Development Authority, Maseru, Lesotho. LHDA (2002) Analysis of the Minimum Degradation, Treaty, Design Limitation and Fourth Scenarios for Phase 1 Development. Metsi Consultants. Contract LHDA 678. LHDA (2002). Lesotho Highlands Water Project Phase 1: Policy for Instream Flow Requirements (1st Edi- tion), Lesotho Highlands Development Authority, Maseru, Lesotho. LHDA (2002-2003). Lesotho Highlands Water Project Phase 1: Policy for Instream Flow Requirements (2nd Edition - 30 July 2002; Incorporating Corrigenda - 30 July 2003). Lesotho Highlands Development Authority, Maseru, Lesotho. LHDA (2003-2004). Report on implementation of the Instream Flow Requirements policy (January 2003 - September 2004), Lesotho Highlands Development Authority., Maseru, Lesotho. LHDA (2005). Procedures for the implementation of the LHWP Phase 1 Instream Flow Requirement pol- icy (Final) (2nd Edition), Lesotho Highlands Development Authority, Maseru, Lesotho. Turpie, J. et al. (2006). A study of the intangible resource losses and gains in Phase 1 areas of the Lesotho Highlands Water Project, Lesotho Highlands Development Authority, Maseru, Lesotho. LHDA (annually). Flow releases downstream of LHWP structures, Lesotho Highlands Development Au- thority, Maseru, Lesotho. LHDA (2006) Final Project Evaluation Report. Implementation of the Instream Flow Requirements Bio- physical Monitoring Procedures Downstream of Phase 1 Dams. Report No. 1237-06. LHDA (2007) Additional Task: Input to Decision Rules. Implementation of the Instream Flow Require- ments Biophysical Monitoring Procedures Downstream of Phase 1 Dams. Report No. 1237-07. LHDA (2007) Instream Flow Requirements Audit for Phase 1 Dams of the Lesotho Highlands Water Pro- ject. Lesotho Highlands Development Authority, Maseru, Lesotho. - 79 - - 80 -