IMPLEMENTATION COMPLETION AND RESULTS REPORT (IDA-43530,IDA-43540) ON A CREDIT IN THE AMOUNT OF SDR 9.1 MILLION (US$13.8 MILLION EQUIVALENT) TO THE

Transcription

1 Public Disclosure Authorized Document of The World Bank Report No: ICR Public Disclosure Authorized Public Disclosure Authorized IMPLEMENTATION COMPLETION AND RESULTS REPORT (IDA-43530,IDA-43540) ON A CREDIT IN THE AMOUNT OF SDR 9.1 MILLION (US$13.8 MILLION EQUIVALENT) TO THE REPUBLIC OF MOZAMBIQUE FOR A TRANSMISSION UPGRADE PROJECT Public Disclosure Authorized March 20, 2017 Energy and Extractives Global Practice Country Department Southern Africa 2 (AFCS2) Africa Region

2 CURRENCY EQUIVALENTS (Exchange Rate Effective January 18, 2017) Currency Unit = Mozambican Metical MZN 1.00 = US$ US$ 1.00 = MZN FISCAL YEAR January 1 December 31 ABBREVIATIONS AND ACRONYMS AIDS APL CAS CPS CREE DAM EdM ENDE EIRR ERR ESCOM ESIA ESMAP FIRR FRR GDP GNI GoM GWh HCB HIPC HIV IBRD ICR ICT IDA IEG IP Acquired Immunodeficiency Syndrome Adaptable Program Loan Country Assistance Strategy Country Partnership Strategy Commission for Foreign Economic Relations Day Ahead Market Electricidade de Mocambique Mozambique National Development Strategy Economic Internal Rate of Return Economic Rate of Return Electricity Supply Corporation of Malawi Environment and Social Impact Assessment Energy Sector Management Assistance Program Financial Internal Rate of Return Financial Rate of Return Gross Domestic Product Gross National Income Government of Mozambique Gigawatt hours Hydroelectrica de Cahora Bassa Heavily Indebted Poor Countries Human Immunodeficiency Virus International Bank for Reconstruction and Development Implementation Completion and Results Report Information and Communications Technologies International Development Association Independent Evaluation Group Implementation Performance ii

3 ISR kv kva MARR M&E MGDS MoU MVA MW MWh MZN NEP NEPAD NPV O&M PAD PARPA PDO PIU PPA PPF PSA PVA QAG QEA QSA ROR RPF SADC SAIFI SAPMP SAPP SCADA SE4ALL SIL STEM TTL TUP USD WAPP XDR Implementation Status and Results report Kilovolt kilovolt-ampere Minimum Acceptable Rate of Return Monitoring and Evaluation Malawi Growth and Development Strategy Memorandum of Understanding Mega Volt Amp Megawatt Megawatt hours Mozambican Metical (Mozambican currency) Malawi National Electricity Policy New Partnership for Africa s Development Net Present Value Operations and Maintenance Project Appraisal Document Action Plan for the Reduction of Absolute Poverty Project Development Objectives Project Implementation Unit Power Purchase Agreement Project Preparation Facility Power Supply Agreement Poverty and Vulnerability Assessment Quality Assurance Group Quality At Entry Quality of Supervision Run-of-River Resettlement Policy Framework Southern African Development Community System Average Interruption Frequency Index Southern African Power Market Program Southern Africa Power Pool System Control and Data Acquisition Sustainable Energy for All Specific Investment Loan Short Term Energy Market Task Team Leader Transmission Upgrade Project United States Dollar West African Power Pool Special Drawing Right iii

4 Vice President: Senior Global Practice Director: Global Practice Director: Country Director: Acting Practice Manager: Project Team Leader: ICR Team Leader: ICR Authors: Makhtar Diop Riccardo Puliti Lucio Monari Mark R. Lundell Lucio Monari Zayra Romo Kabir Malik Kabir Malik, Bobak Rezaian iv

5 REPUBLIQUE OF MOZAMBIQUE Transmission Upgrade Project CONTENTS Data Sheet A. Basic Information... vi B. Key Dates... vi C. Ratings Summary... vi D. Sector and Theme Codes... vii E. Bank Staff... vii F. Results Framework Analysis... viii G. Ratings of Project Performance in ISRs... xv H. Restructuring (if any)... xv I. Disbursement Profile... xvii 1. Project Context, Development Objectives and Design Key Factors Affecting Implementation and Outcomes Assessment of Outcomes Assessment of Risk to Development Outcome Assessment of Bank and Borrower Performance Lessons Learned Comments on Issues Raised by Borrower/Implementing Agencies/Partners Annex 1. Project Costs and Financing Annex 2. Outputs by Component Annex 3. Economic and Financial Analysis Annex 4. Bank Lending and Implementation Support/Supervision Processes Annex 5. Beneficiary Survey Results Annex 6. Stakeholder Workshop Report and Results Annex 7. Summary of Borrower's ICR and/or Comments on Draft ICR Annex 8. Comments of Cofinanciers and Other Partners/Stakeholders Annex 9. List of Supporting Documents Annex 10. Map IBRD v

6 A. Basic Information Country: Mozambique Project Name: Mozambique Transmission Upgrade Project Project ID: P L/C/TF Number(s): IDA-43530,IDA ICR Date: 03/20/2017 ICR Type: Core ICR Lending Instrument: SIL Borrower: Original Total Commitment: Revised Amount: XDR 9.10M Environmental Category: B Implementing Agencies: Electricidade de Moçambique (EdM) Cofinanciers and Other External Partners: B. Key Dates Republic of Mozambique XDR 29.60M Disbursed Amount: XDR 9.10M Process Date Process Original Date Revised / Actual Date(s) Concept Review: 11/13/2006 Effectiveness: 04/01/ /01/2008 Appraisal: 04/09/2007 Restructuring(s): 04/29/ /21/ /23/ /17/2016 Approval: 07/17/2007 Mid-term Review: 02/15/ /10/2014 C. Ratings Summary C.1 Performance Rating by ICR Outcomes: Risk to Development Outcome: Bank Performance: Closing: 06/30/ /31/2016 Moderately Satisfactory Moderate Moderately Unsatisfactory Borrower Performance: Moderately Satisfactory C.2 Detailed Ratings of Bank and Borrower Performance (by ICR) Bank Ratings Borrower Ratings Quality at Entry: Moderately Unsatisfactory Government: Moderately Satisfactory Quality of Supervision: Moderately Implementing Unsatisfactory Agency/Agencies: Moderately Satisfactory Overall Bank Moderately Overall Borrower Performance: Unsatisfactory Performance: Moderately Satisfactory vi

7 C.3 Quality at Entry and Implementation Performance Indicators Implementation QAG Assessments Indicators Performance (if any) Potential Problem Project at any time (Yes/No): Problem Project at any time (Yes/No): Yes Yes DO rating before Moderately Closing/Inactive status: Satisfactory D. Sector and Theme Codes Major Sector/Sector Energy and Extractives Quality at Entry (QEA): Quality of Supervision (QSA): Original None None Rating Actual Energy Transmission and Distribution Major Theme/Theme/Sub Theme Economic Policy Trade Trade Facilitation Private Sector Development ICT ICT Solutions Regional Integration E. Bank Staff Positions At ICR At Approval Vice President: Makhtar Diop Obiageli Katryn Ezekwesili Country Director: Mark R. Lundell Mark D. Tomlinson Practice Manager/Manager: Project Team Leader: ICR Team Leader: ICR Primary Author: ICR co-author: Lucio Monari Zayra Luz Gabriela Romo Mercado Kabir Malik Kabir Malik Abdolreza B. Rezaian Yusupha B. Crookes Wendy E. Hughes vii

8 F. Results Framework Analysis Project Development Objectives (from Project Appraisal Document) The APL 2 development objective is to implement the Mozambique-Malawi transmission interconnection (i) to increase access to diversified, reliable, and affordable supplies of energy; and (ii) to expand Malawi and Mozambique s opportunities to benefit from bilateral and regional power trading on the Southern African Power Pool. Revised Project Development Objectives (as approved by original approving authority) To reduce the frequency of electricity outages in Tete province. (a) PDO Indicator(s) Indicator Indicator 1 : Baseline Value Original Target Values (from approval documents) Formally Revised Target Values Actual Value Achieved at Completion or Target Years Average interruption frequency per year in the project area (Number, Core) Value quantitative or Qualitative) Date achieved 12/31/ /30/ /30/2016 Comments (incl. % achievement) Achievement percentage is 719%. This measure is the actual reduction as a percentage of the targeted reduction. A large portion of the decline occurred prior to the commissioning of the transformer due to factors unrelated to the project. Indicator 2 : Customers served in the project area (Number, Core Supplement) Value quantitative or Qualitative) Date achieved 11/01/ /30/ /30/2016 Comments (incl. % 93.8% of the target achieved. achievement) Indicator 3 : Direct project beneficiaries (Number, Core) Value quantitative or Qualitative) Date achieved 11/01/ /30/ /30/2016 Comments (incl. % achievement) 93.8% of the target achieved. Growth in number of electricity customers in the region was marginally below expectation, but the target is expected to be met and exceeded in the near future. viii

9 Indicator 4 : Female beneficiaries (Percent, Core Supplement) Value quantitative or Qualitative) 0 51% 50% Date achieved 11/01/ /30/ /30/2016 Comments (incl. % achievement) Indicator 5 : Malawi: Volume of trading via interconnector (GWh). Value quantitative or Qualitative) Dropped Date achieved 12/31/ /31/ /29/2013 Comments (incl. % achievement) Indicator 6 : Malawi did not sign the credit agreement. Thus, the 1st project restructuring changed the PDO and cancelled the regional interconnector investments. All associated indicators were dropped. Mozambique: Volume of trading with Malawi via interconnector (GWh). Value quantitative or Qualitative) 0 55 Dropped Date achieved 12/31/ /31/ /29/2013 Comments (incl. % This indicator was dropped during the 1st project restructuring. achievement) Indicator 7 : Decreased capacity deficit in case of drought in Malawi (text) Value quantitative or Qualitative) 0 No-off peak load shedding Dropped Date achieved 12/31/ /31/ /29/2013 Comments This indicator was dropped during the 1st project restructuring. (incl. % achievement) EDM incremental earnings from ESCOM payments associated with use-oftransmission Indicator 8 : line Value quantitative or Qualitative) 0 Exceeds EDM debt service for the interconnector investment Dropped Date achieved 12/31/ /31/ /29/2013 Comments This indicator was dropped during the 1st project restructuring. (incl. % ix

10 achievement) Indicator 9 : SAIFI (System Average Interruption Frequency Index) for Matambo customers (%) Value quantitative or Qualitative) 4.5% 3.5% Dropped Date achieved 01/01/ /31/ /21/2014 Comments (incl. % achievement) Indicator 10 : This indicator was dropped during the second restructuring and replaced by the relevant sector core indicator. Customers served in the project area (supplemental) Value quantitative or Qualitative) Dropped Date achieved 01/01/ /31/ /21/2014 Comments (incl. % achievement) This indicator was a supplemental indicator to PDO indicator above. It was thus dropped during the second restructuring along with the main PDO indicator and replaced by the relevant sector core indicator. (b) Intermediate Outcome Indicator(s) Indicator Indicator 1 : Value (quantitative or Qualitative) Baseline Value Original Target Values (from approval documents) ESIA final report completed and approved by GoM No ESIA ESIA final report completed and approved by GoM Formally Revised Target Values Actual Value Achieved at Completion or Target Years ESIA final report completed and approved by GoM Date achieved 07/17/ /30/ /24/2016 Comments (incl. % The final report was submitted to GoM for approval in December GoM approval was finally received in June achievement) New 220/66/33 kv transformer installed and commissioned at Matambo Indicator 2 : substation Value (quantitative or Qualitative) No transformer 80 MVA of new transformation capacity commissioned 130 MVA Transformer delivered on site and commissioned Date achieved 01/01/ /30/ /30/2016 x

11 Comments (incl. % achievement) Indicator 3 : Targeted transformer capacity exceeded by 162.5% as a larger capacity transformer was affordable within the project envelope. The transformer was commissioned in July 2016, prior to project close. Level of charge of the current Matambo transformer (percent) Value (quantitative or Qualitative) 26% 70% 45% Date achieved 07/17/ /30/ /30/2016 Comments (incl. % achievement) Indicator 4 : Value (quantitative or Qualitative) Reduction target exceeded by 256%. The purchase of an additional transformer helped split the load between two transformers and reduce stress on the original ("current") transformer. New 220/33 kv, 25 MVA mobile substation delivered No transformer 50 MVA of new transformation capacity delivered Transformer was delivered and ready for use Date achieved 07/17/ /30/ /30/2016 Target achieved. The new transformer at Tete is on site and is expected to be Comments (incl. % achievement) commissioned in August The mobile substation was commissioned in December (The project funded delivery, but not installation, of this transformer). Indicator 5 : Completion of survey for building interconnector (Malawi, Moz) Value (quantitative or Qualitative) Preliminary survey completed Detailed survey completed Dropped Date achieved 12/31/ /01/ /29/2013 Comments This indicator was dropped during the 1st project restructuring. (incl. % achievement) % of towers installed against planned installation (Malawi, Moz) Indicator 6 : Value (quantitative or Qualitative) 0 Atleast 90% Dropped Date achieved 12/31/ /31/ /29/2013 Comments (incl. % This indicator was dropped during the 1st project restructuring. achievement) % expenditures against planned expenditures for building interconnector Indicator 7 : (Malawi, Moz) Value 0 At least 85% Dropped xi

12 (quantitative or Qualitative) Date achieved 12/31/ /31/ /29/2013 Comments (incl. % This indicator was dropped during the 1st project restructuring. achievement) Site mobilization completed for the construction of Phombeya and upgrading of Indicator 8 : Matambo substations (Malawi, Moz) Value (quantitative or Qualitative) none Site mobilization completed Dropped Date achieved 12/31/ /31/ /29/2013 Comments This indicator was dropped during the 1st project restructuring. (incl. % achievement) % expenditures against planned expenditures at substations (Malawi, Moz) Indicator 9 : Value (quantitative or Qualitative) 0 At least 85% Dropped Date achieved 12/31/ /31/ /29/2013 Comments (incl. % This indicator was dropped during the 1st project restructuring. achievement) Key agreements signed and adhered to (I) Wheeling, (ii) system operating, (iii) Indicator 10 : maintenance, (iv) implementation agreements (Malawi, Moz) Value (quantitative or Qualitative) Agreements substantially negotiated Agreements adhered to; Dropped ESCOM payments to EdM on time Date achieved 12/31/ /31/ /29/2013 Comments This indicator was dropped during the 1st project restructuring. (incl. % achievement) ESCOM becomes operating member of SAPP Indicator 11 : Value (quantitative or Qualitative) None ESCOM signs SAPP agreement among operating members Dropped Date achieved 12/31/ /31/ /29/2013 Comments This indicator was dropped during the 1st project restructuring. (incl. % achievement) xii

13 Indicator 12 : ESCOM signs a power trading agreement Value (quantitative or Qualitative) not signed signed Dropped Date achieved 12/31/ /31/ /29/2013 Comments (incl. % This indicator was dropped during the 1st project restructuring. achievement) ESCOM financial sustainability plan adopted and implemented (Malawi) Indicator 13 : Value (quantitative or Qualitative) None Plan prepared, adopted and implemented xiii Dropped Date achieved 12/31/ /31/ /29/2013 Comments This indicator was dropped during the 1st project restructuring. (incl. % achievement) Cash coverage ratio (Malawi) Indicator 14 : Value (quantitative or Qualitative) 0.81 At least 1.0 Dropped Date achieved 12/31/ /31/ /29/2013 Comments (incl. % This indicator was dropped during the 1st project restructuring. achievement) Collection/generation ratio (Malawi) Indicator 15 : Value (quantitative or Qualitative) 70% at least 75% Dropped Date achieved 12/31/ /31/ /29/2013 Comments (incl. % This indicator was dropped during the 1st project restructuring. achievement) % task completed against planned timelines for studies (Malawi) Indicator 16 : Value (quantitative or Qualitative) 0 at least 90 % Dropped Date achieved 12/31/ /31/ /29/2013 Comments (incl. % This indicator was dropped during the 1st project restructuring. achievement) % completion of feasibility study against planned timelines for extending Indicator 17 : connection to Northern Mozambique (Mozambique)

14 Value (quantitative or Qualitative) none Consultants selected Dropped Date achieved 12/31/ /31/ /29/2013 Comments This indicator was dropped during the 1st project restructuring. (incl. % achievement) Decrease network faults due to failure of breaker operations in areas under Indicator 18 : renovation in Malawi Value (quantitative or Qualitative) 17 At least 10% fewer faults from baseline Dropped Date achieved 12/31/ /31/ /29/2013 Comments This indicator was dropped during the 1st project restructuring. (incl. % achievement) Reduction in number of excitation trips at generation stations Indicator 19 : Value (quantitative or Qualitative) 19 At least 10% fewer faults from baseline Dropped Date achieved 12/31/ /31/ /29/2013 Comments This indicator was dropped during the 1st project restructuring. (incl. % achievement) Increased number of customers served out of Matambo Indicator 20 : Value (quantitative or Qualitative) % yearly increase Dropped from baseline Date achieved 12/31/ /31/ /29/2013 Comments This indicator was dropped during the 1st project restructuring. (incl. % achievement) % expenditures against planned expenditures for building Matambo transformer Indicator 21 : Value (quantitative or Qualitative) 0 At least 85% Dropped Date achieved 12/31/ /31/ /29/2013 Comments (incl. % This indicator was dropped during the 1st project restructuring. achievement) xiv

15 G. Ratings of Project Performance in ISRs No. Date ISR Archived DO IP Actual Disbursements (USD millions) 1 12/20/2007 Moderately Satisfactory Moderately Unsatisfactory /02/2008 Moderately Satisfactory Moderately Unsatisfactory /24/2008 Moderately Moderately Unsatisfactory Unsatisfactory /23/2009 Moderately Moderately Unsatisfactory Unsatisfactory /22/2009 Unsatisfactory Unsatisfactory /18/2010 Unsatisfactory Unsatisfactory /28/2011 Unsatisfactory Unsatisfactory /08/2012 Unsatisfactory Unsatisfactory /22/2012 Unsatisfactory Unsatisfactory /12/2013 Moderately Satisfactory Moderately Satisfactory /30/2013 Moderately Moderately Unsatisfactory Unsatisfactory /30/2014 Moderately Unsatisfactory Moderately Satisfactory /03/2014 Moderately Satisfactory Moderately Satisfactory /15/2015 Moderately Satisfactory Moderately Satisfactory /28/2015 Moderately Satisfactory Moderately Satisfactory /09/2016 Moderately Satisfactory Moderately Unsatisfactory 9.39 H. Restructuring (if any) Restructuring Date(s) Board Approved PDO Change ISR Ratings at Restructuring DO Amount Disbursed at Restructuring in USD millions 04/29/2013 Y MS MS 2.97 IP Reason for Restructuring & Key Changes Made Failure of Malawi to sign Financing Agreement. Restructuring cancelled all activities related to the regional interconnector. The PDO was changed to reflect the reduced scope that focused on transmission upgrade for Mozambique. Closing date for the project extended by one and xv

16 Restructuring Date(s) Board Approved PDO Change ISR Ratings at Restructuring DO IP Amount Disbursed at Restructuring in USD millions 11/21/2014 N MU MS /23/2015 MS MS 6.70 Reason for Restructuring & Key Changes Made a half years to 31 December (Level 1) Utilization of savings on request of Government. Two additional activities were added. PDO indicators were refined to incorporate core sector indicators. Closing date extended by one and a half years to 30 June (Level 2) Reallocation of funds among disbursement categories. (Level 2) 06/17/2016 MS MU 9.39 Closing date extension by 2 months to enable completion of activities. (Level 2) If PDO and/or Key Outcome Targets were formally revised (approved by the original approving body) enter ratings below: Outcome Ratings Against Original PDO/Targets Unsatisfactory Against Formally Revised PDO/Targets Satisfactory Overall (weighted) rating Moderately Satisfactory xvi

17 I. Disbursement Profile xvii

18 1. Project Context, Development Objectives and Design 1.1 Context at Appraisal Regional Context 1. In 2007, when the project was presented to the Board, the Southern Africa Region had experienced sustained economic growth in recent years bolstered by the continued global economic expansion, prudent macroeconomic policies, debt relief and sustained demand for commodities amid favorable commodity prices. Estimated average rate of growth in real Gross Domestic Product (GDP) for the region was about 5.5 percent. Accompanying growth in electricity demand implied that the region was entering a period of power generation capacity shortage and additional generation capacity was required. The new demand required large, regional generation projects that would utilize the variation in energy resource endowments across countries in the region. As a result, regional trade in electricity was expected to increase, highlighting the need to address constraints in the regional transmission infrastructure The Southern Africa region possessed the institutional infrastructure to support regional trade. The Southern Africa Power Pool (SAPP) was created in 1995 by Southern African Development Community (SADC) member countries by concluding an Intergovernmental Memorandum of Understanding (MOU). The utilities of 12 Southern African countries, including Malawi and Mozambique, were the original members of the SAPP. The interconnected national grids of the member countries formed the regional SAPP network. The SAPP aimed to facilitate regional co-operation in the development of energy resources and energy pooling to enable reliable energy supply in an efficient manner. 3. Malawi, Angola, and Tanzania were the only countries not yet interconnected to the SAPP network. Connection to the regional grid of non-connected SAPP member power systems was a priority in terms of SAPP planning. In the 2006 SAPP Annual Report, the Mozambique-Malawi Transmission Interconnection was explicitly noted as a priority for the regional pool. 4. Electricity on the SAPP was largely traded through medium-to-long term bilateral contracts. A Short Term Energy Market (STEM) was in operation since its inception in However due to a shortage of hydropower generation in the region due to the exposure of individual countries to drought and floods, demand in the market was well above supply. Energy traded on the STEM fell from 842 GWh in 2002, to 226 GWh in 2006, and 68 GWh in Therefore, at the time, Mozambique was one of the only systems to have excess capacity due to the 2,075 MW Hydroelectrica de Cahora Bassa (HCB) hydropower plant. In 2006, the Day Ahead Market (DAM) had not yet been established by SAPP, with the trial version only commissioned in 2007, and the operational model completed in January The Mozambique Malawi Interconnector Project was part of the World Bank s wider Southern Africa Power Market Adaptable Program Loan program, which aimed at assisting the Southern Africa region to (i) generate savings through aggregation of loads with different load profiles, (ii) achieve efficient use of energy resources by exploiting large scale power generation schemes that are viable only on the basis of large multicountry markets, (iii) manage the risks of climate-related power shortages in hydro-dependent countries. 2 The Potential of Regional Power Sector Integration Southern African Power Pool (SAPP) Transmission & Trading Case Study. Submitted to ESMAP by Economic Consulting Associates, pp 29,

19 Malawi Context 5. At the time of project appraisal in 2006, Malawi was one of the poorest and most densely populated countries in sub-saharan Africa. The population density in 2006 was about 109 persons per square kilometer, with a total population of about 12.9 million. The Malawi Poverty and Vulnerability Assessment (PVA) at the time of appraisal showed that the percentage of the population living below the poverty line was around 52 percent in 2004/05. The per capita Gross National Income (GNI) was estimated at US$ Economic diversification through an increased share of the industrial sector was constrained by high interest rates, transport and energy costs. 3 Under President Bingu wa Mutharika, who came into office in 2004 and was re-elected for a second term in 2009, improved macroeconomic management placed Malawi on a path for faster economic growth. The growth in real Gross Domestic Product (GDP), which averaged 1.5 percent between 2001 and 2004, increased to an average of around five percent between 2005 and However, the economy remained largely agrarian, with agriculture contributing to 80 percent of the workforce and 80 percent of the foreign exchange earnings. Malawi s economic activities and exports were narrowly concentrated on agricultural commodities (tobacco was the main export) and thus remained highly vulnerable to rainfall and weather shocks, resulting in variable growth rates and low foreign exchange reserves. 7. The Malawi Growth and Development Strategy (MGDS), , explicitly mentioned the Mozambique/Malawi interconnection project as contributing to the objective of improving access to reliable and affordable electricity to stimulate economic growth. The six key focus areas for the Government s development strategy were: energy generation and supply; agriculture and food security; infrastructure development, irrigation and water development; integrated rural development; prevention and management of HIV AIDS; and improved governance. The 2006 Investment Climate Assessment carried out by the World Bank identified the quality of electricity supply as the fourth most serious constraint for businesses. 8. The Electricity Supply Corporation of Malawi (ESCOM) was (and is) the sole electricity utility in Malawi. ESCOM, a vertically integrated, government-owned electric utility, had about 175,000 customers and 284 MW of installed hydropower capacity in 2006 (with less than 260 MW of capacity available for distribution at peak times, of which about 40 MW of capacity was unavailable at the time due to damage resulting from flooding of the Shire river). 4 The peak demand was about 250 MW in 2006 with an expected growth of about five percent annually over the following decade. Only about six percent of the population had access to electricity (this has increased to about 10 percent by 2016). 9. Malawi was also vulnerable to a drought-induced power crisis due to its dependence on run-of-river (ROR) hydropower plants on the Shire River (98 percent of the installed capacity). ESCOM prepared an Integrated Resource Plan in 2005, which recommended further expansions of domestic generation capacity situated on the Shire River and thus would not reduce the droughtvulnerability of the power system. Climate change was expected exacerbate this vulnerability due to increased drought stress in sub-saharan Africa. 5 3 Malawi Growth and Development Strategy (MGDS), Malawi now has an installed capacity of approximately 350 MW. 5 The issue of excess demand and drought-vulnerability has remained unaddressed since the time of the project appraisal and since then the situation has deteriorated as the country continues to rely primarily on 2

20 10. In recognition of these issues, Malawi developed a power sector strategy that was designed to put in place measures to mitigate the consequences of a severe drought, and increasing access to reliable and affordable electricity supply. Key elements of Malawi s strategy in 2006 included: (i) implementation of the interconnector with the SAPP network by 2010 as the least-cost option for mitigating the risk of drought-related power crisis, and facilitating the import of electricity as needed and export electricity when available; (ii) expansion of low-cost domestic generation capacity by 2011; (iii) further addition to available capacity of 30 to 50 MW by Based on the Integrated Resource Plan for the Malawi Power Sector, further regional imports were expected to be the least-cost option for this next increment of supply. 11. While identifying the Mozambique-Malawi interconnector as a key investment, the drive for domestic power generation was strong at the time of appraisal. In the updated Integrated Resource Plan, published in 2011, it is evident that Malawi has been planning to develop new generation facilities based on its domestic hydro and coal resources. 12. Development of the power system was partly constrained by the relatively weak financial situation and commercial performance in the power sector. ESCOM had not been achieving financial sustainability and was unable to meet all cash flow requirements from its cash inflows. In response, a Financial Sustainability Plan that included elements of improved efficiency and tariff adjustments was proposed. This was expected to be funded by the project and was expected to improve the capability of ESCOM for implementing the power sector strategy and for meeting the associated financial obligations. 6 Implementation of the plan was seen as critical to sustaining the benefits of the interconnector and was thus included as a legal covenant in the project at appraisal. Mozambique Context 13. Between 1996 and 2005, following the end of the war in 1992, Mozambique s economy grew at an average of eight percent per year. The poverty headcount index fell from 69 percent in 1996/97 to 54 percent in 2002/03, and GNI per capita was US$389 in Economic expansion had been made possible by overall macroeconomic stability, sound policy reforms, and continuing strong support from development partners. 14. Drivers of economic growth in Mozambique were infrastructure development, mining, energy, manufacturing mega-projects, agriculture and, increasingly, tourism. These activities were expected to facilitate rural development, a central theme of the country s second Action Plan for the Reduction of Absolute Poverty (PARPA II), defined by the council of ministers. The priorities in the PARPA II were improved governance, increased human capital, and economic development. A significant theme of the country s economic development strategy was to ensure access to basic services, such as water and electricity for rural communities, in order to promote economic growth. 15. The power sector in Mozambique was dominated by EdM, a vertically integrated, government-owned electric utility. In 2006, EdM had installed hydropower capacity of 140 MW (of which 86 MW was available) and 109 MW in thermal power stations (of which 82 MW was available). Peak demand in Mozambique was 350 MW. EdM bought most of its power supply (300 hydropower. Certain parts of the country now experience eight hours of load shedding per day, and the country has an estimated suppressed demand of 300 MW. 6 With delays in effectiveness of Malawi s credit (and ultimate cancellation), a Public Private Infrastructure Advisory Facility grant was obtained to support the ESCOM financial sustainability plan in

21 MW) from HCB in Tete province. Cahora Bassa had an installed capacity of 2,075 MW (5 x 415 MW), the majority of which was exported to South Africa and Zimbabwe. 16. Mozambique s load growth was projected at eight percent in 2007, seven percent annually from 2008 to 2010, and five percent thereafter. At the time Mozambique s transmission grid was already interconnected with Zimbabwe, South Africa, and Swaziland. About eight percent of the population had access to electricity. Since then, electricity access in Mozambique has increased to approximately 26 percent. 17. To meet the growth in domestic demand and planned new industrial demand, the Government planned significant new generation and transmission infrastructure with considerable private sector participation. The Government had begun the process of selecting private sector strategic investors for three new generation projects and large transmission projects, which at the time required investments of around US$4 billion. Through some of these projects, EdM had gained technical and institutional capacity to implement large transmission projects, including interconnectors. 18. At the time of appraisal, Mozambique had adopted a power sector strategy that focused on (i) rapid expansion of access to electricity though grid intensification, grid extension, and off-grid approaches; (ii) rehabilitation of existing hydropower plants; and (iii) development of new power generation through private sector and public-private partnerships. These objectives were largely achieved, as demonstrated by the improvements in the power system grid and scale-up of electricity access in Mozambique Mozambique s strategy included a focus on expanding opportunities for power trade to optimize the use of domestic hydro resources as well as bolster foreign exchange earnings. Key interconnection routes included the Mozambique-Malawi transmission interconnector. It was envisaged that, in the future, the Mozambique-Malawi interconnector could be extended east across southern Malawi into northern Mozambique. This extension of the interconnection would significantly improve the quality, quantity, and reliability of supply to the north of Mozambique, where demand for power was growing. The extension of the power transmission line into northern Mozambique was not thoroughly studied during project preparation in This was partly because the rapid load growth in Nacala and Pemba areas in northern Mozambique was not foreseen. This growth began at the time of the later gas discoveries in the north of Mozambique. Therefore, at appraisal, extension of the line was envisaged for the future. 20. Despite the surplus in power generation, the transmission infrastructure in Mozambique was not adequate for the high-volume power flows. Some regions in northern Mozambique experienced low quality of supply. Aging equipment was often overloaded and needed to be replaced. The requirements for grid improvements in Tete province were already known and upgrades at the Matambo substation, serving the province, were envisaged as part of the Mozambique-Malawi interconnector. 7 The national grid is extended to all 152 districts in the country and the peak demand for EdM has risen to 680 MW. Power generating capacity has also increased with the installation of new gas fired independent power projects. The country now has a surplus capacity, which it sells to the region through the SAPP market. 4

22 Rationale for Bank Involvement 21. The Bank had committed to lend long-term support to both SADC and the New Partnership for Africa s Development (NEPAD) initiatives to promote electricity trade in the region, through the approval of the Southern African Power Market Program (SAPMP) Adaptable Program Loan (APL) in In this context, the Bank was requested by the Governments of Malawi and Mozambique to finance the Mozambique- Malawi Transmission Interconnection Project. 22. The project was a key element in Malawi s MGDS for ensuring adequate, affordable electricity to support expanded access and economic growth. The project benefited Mozambique by providing a source of new revenues. It was also a key step in EdM s medium-term plans to improve the reliability of supply to Northern Mozambique. An additional benefit of the interconnector was that Malawi would be able to sell off-peak energy that would otherwise be spilled, either into the SAPP STEM or under short-term bilateral contracts with Mozambique or other SAPP members. It was expected that Malawi would generate new revenues from such power export, and nearby countries in the SAPP would potentially benefit from importing low-cost electricity from Malawi. 1.2 Original Project Development Objectives (PDO) and Key Indicators 23. The project consisted of a US$48 million equivalent Credit to Malawi and a US$45 million Credit to Mozambique. At the time of approval by the Board, the PDO was stated in the Financing Agreement as follows: 24. The objectives of the Project are to support the Recipient s efforts to implement Mozambique-Malawi Interconnector as part of the Southern African Power Pool, and to: (i) increase access to diversified, reliable, and affordable supplies of energy; and (ii) expand opportunities to benefit from Southern African regional power trading. 25. The PDO in the Project Appraisal Document (PAD) was worded slightly differently: To implement the Mozambique-Malawi transmission interconnection (i) to increase access to diversified, reliable, and affordable supplies of energy; and (ii) to expand Malawi and Mozambique s opportunities to benefit from bilateral and regional power trading on the Southern African Power Pool. As regional power trading encompasses bilateral trade, the PDOs were not substantively different. 26. The key indicators for the above objective were the following: Malawi: Volume of trading via interconnector (GWh). Mozambique: Volume of trading with Malawi via interconnector (GWh). Decreased deficit in case of drought in Malawi; EdM incremental earnings from ESCOM payments associated with use-of-transmission line. 1.3 Revised PDO (as approved by original approving authority) and Key Indicators, and reasons/justification 27. In April 2013, project was restructured for the first time, in order to drop the investments associated with the regional (interconnector) transmission line. With the failure of Malawi to sign its Credit Financing Agreement (more in Section 2 below), the focus of the restructured project 5

23 shifted to investments to upgrade Mozambique s transmission infrastructure in the fast-growing Tete province (transformer upgrade Matambo substation in Tete province was already part of the original project). The restructuring was thus both corrective and adaptive in that it reoriented the focus of the project based on the changed circumstances and improved implementation performance. 28. As part of the restructuring, the PDO was revised to the following: To reduce the frequency of electricity outages in Tete province. The project closing date was also extended by 18 months, from June 30, 2013 to December 31, The achievement of the new PDO was directly linked to the transformer upgrade at Matambo substation (the only transmission substation serving Tete province). With the change of the PDO and the project scope, all results indicators for the activities related to the regional infrastructure investments were dropped, and the restructured project only had one PDO level indicator: SAIFI (system average interruption frequency index) for Matambo customers; 8 o Customers served in the project area (supplemental indicator). 30. In November 2014, the project was restructured a second time, with no change to the PDO. This opportunistic restructuring was the result of savings under the main contract for the Matambo transformer upgrade due to the competitive bidding process. This allowed for the upsizing of the transformer at Matambo as well as the purchase of additional critical equipment for the Tete region, namely the purchase of a second transformer for the Tete substation and a mobile substation to enable critical repair and maintenance work without causing a shutdown. As part of the restructuring, the closing date was extended by a further 18 months, to June 30, The PDO indicators were updated to include core sector indicators (the existing PDO indicator was dropped and replaced by a core sector indicator, and other core sector indicators were added). The restructured project in the 2014 thus had the following PDO indicators: Average interruption frequency per year in the project area; 9 o Customers served in the project area (supplemental indicator). Direct project beneficiaries; o Female beneficiaries (supplemental indicator). 31. There were 16 intermediate indicators in the original project. These were all dropped in 2013 at the time of the first restructuring, and two revised intermediate indicators were added. During the second project restructuring another two intermediate indicators were added. 10 Target values for existing indicators were not revised at any point in the project. The list of PDO and intermediate indicators, along with targets and achievements, are presented in section F of the datasheet. 8 SAIFI is a commonly used metric in the electricity sector. It is measured as the customer-weighted average of sustained interruptions in year. This indicator was initially incorrectly defined in the restructuring paper (details in section 2.3). 9 This was a core sector indicator that replaced the earlier SAIFI indicator. 10 The intermediate indicators included post restructuring are further discussed in section 2.3, in addition to section F of the datasheet. 6

24 1.4 Main Beneficiaries 32. As stated in the PAD, the primary beneficiaries of the project included the businesses and citizens [of Malawi and Mozambique] who would have increased and more reliable access to electricity as a result of electricity trade between Mozambique and Malawi and the SAPP. By implication, intermediate beneficiaries would also be the electricity utilities in the two countries ESCOM and EdM who would gain from lower cost of power purchase and increased sales revenues, respectively. 33. The main beneficiaries of the restructured project were not explicitly defined in the restructuring paper. However, the number of customers served and direct project beneficiaries were included in the results framework of the restructured project. As a result of this, the main project beneficiaries are interpreted as: The electricity customers in the Tete province, especially the households and businesses that would benefit from a more reliable power supply. EdM, through the strengthening of its network and a reduction in costly outages. 1.5 Original Components 34. The original components at project approval included the regional interconnection and related transmission infrastructure upgrades. 35. Component A. (US$47.1 million for Malawi, US$43.5 million for Mozambique): Construction of the transmission interconnection from the Malawi electricity grid to the Mozambique electricity grid, thereby interconnecting Malawi with the Southern Africa Power Pool network. On the Malawi side, this would include construction of approximately 76 kilometers of 220 kv transmission line, installation of a new 220 kv substation and the studies, works, engineering, and project management support required to complete the interconnection. On the Mozambique side, this would include construction of approximately 134 kilometers of 220 kv transmission line, the extension of the existing Matambo substation, and the studies, works, engineering, and project management support required to complete the interconnection. 36. Component B. (US$2.9 million for Malawi, US$1.7 million for Mozambique): Technical assistance, capacity building, training and equipment necessary for ESCOM and EdM to strengthen and expand power trading activities. For Malawi the activities included rehabilitation and reinforcement studies for generation, transmission, and distribution; development of revised system operating and maintenance procedures and practices; studies and equipment to support improved ESCOM financial performance, and training in the areas of environmental and social management, project management, electricity trading and system operation. For Mozambique the activities included a feasibility study for extension of the interconnection to the northern region of Mozambique and technical assistance and training in the areas of environmental and social management, loss reduction, project management, electricity trading and system operation. 37. Component C. (US$9.9 million for Malawi, US$4.6 million for Mozambique): Investments to replace worn-out, inadequate, or obsolete equipment to remove critical bottlenecks in the networks which could impede the flow of traded electricity. For Malawi, investments would include replacement of digital excitation equipment, circuit breakers and other switchgear, control and protection equipment, and some critical communication links for better system operation control. For Mozambique the project would support the provision of a new 220/66/33 kv power 7

25 transformer at the Matambo substation and connecting the new transformer to the 66 kv and 33 kv systems. 1.6 Revised Components 38. In the first restructuring in 2013 the components were revised and focused on reducing the frequency of outages in the Mozambique network in Tete province. Under the restructured project, the only infrastructure investment was the installation of the transformer at Matambo substation (part of component C in the original project). 39. Component A: Consulting Engineering Services (US$0.8 million, excluding contingencies). This component included consulting services and provision of technical assistance to assist EdM with engineering, technical, procurement, and environmental and social aspects associated with the Matambo substation works. 40. Component B: Capacity Building and Technical Support to EdM (US$1.6 million, excluding contingencies). This component continued to finance system planning and operation studies for EdM, through the provision of technical advisory services and training, including inter alia in environmental and social management, loss reduction, project management, system operation and other areas. As part of this, Component B continued to support the Environment and Social Impact Assessment (ESIA) for the proposed Mozambique Regional Transmission Development Project (P108934). 41. Component C: Improved Transmission Infrastructure (US$10 million, excluding contingencies). This component ensured that the Matambo substation had sufficient power transformer capacity for the rapidly growing electricity loads in Tete province. The scope of work, which had already been appraised as part of the original project preparation, covered the rehabilitation and reinforcement of the existing 220 kv Matambo substation including the provision of a new 220/66/33 kv (80/60/20 kva) power transformer, the connection of the new transformer to the 66kV and 33 kv busbars at the substation, and the carrying out of associated civil, control and protection works at the substation. 42. As part of the second restructuring in 2014, the amounts allocated to Components A, B and C were revised to US$0.12 million, US$1.58 million, and US$7.85 million, respectively, as a result of savings experienced. A Component D was added to the project, with the savings allocated to it to support: (i) reinforcement of the existing Tete Substation (downstream from Matambo), including the provision of a new 66/33 kv, 50 MVA power transformer and respective switchgear; and (ii) provision of a 220/33 kv, 25 MVA mobile substation to be used whenever there are faults in the overloaded network. 1.7 Other Significant Changes 43. Funds reallocation. A third restructuring (level 2, simple) was implemented in June 2015 to reallocate resources between disbursement categories to meet project commitments related to technical support to EdM on the works being performed at Matambo substation. 44. Project closing date extension. A fourth restructuring (level 2, simple) was implemented in June 2016 to extend the project closing date from June 30, 2016 to August 31, This was in part due to an unexpected delay in the shipment from China of a transformer for Tete because 8

26 the transformer was too heavy for the scheduled ship, which could have endangered the navigability of the ship. In total, the project was extended by three years and two months. 2. Key Factors Affecting Implementation and Outcomes 2.1 Project Preparation, Design and Quality at Entry Preparation 45. The Mozambique-Malawi Transmission Interconnection Project was developed as the second phase of the Southern African Power Market Program (SAPMP). SAPMP was approved in 2003 as a 10-year, three-phase Adaptable Program Loan (APL) in response to the growing demand for regional power grid integration, and the overall objective of facilitating the development of an efficient regional power market and fostering regional integration The overall objectives of the project were closely aligned with Malawi s aim of interconnecting itself to the southern region electricity power pool, 12 and with Mozambique s goal of improving its integration into the regional economy by strengthening power trading in the SAPP and the expansion of access to electricity through grid intensification and grid extension The project was well prepared, and benefited from extensive consultations between EdM, ESCOM, HCB, and the World Bank, with active participation of the Governments of Mozambique and Malawi during bilateral and/or joint meetings throughout project preparation and design. 48. The PAD for the APL-1 (P069258, 2003) had identified specific triggers for the International Development Association Board to consider in the subsequent phases of the APL. The triggers for the Mozambique-Malawi Transmission Interconnection project (APL-2) were: (i) completion and disclosure of the environmental assessments, resettlement action plans, and any required mitigation plans for both Mozambique and Malawi portions of the project; (ii) completion of bidding documents for major contracts; (iii) completion of an implementation agreement between EdM and ESCOM; and (iv) completion of a power purchase agreement (PPA) between ESCOM and the supplier of power. 14 Therefore, satisfying these triggers was an integral part of project preparation. 49. To this end, the Bank provided the Governments of Malawi and Mozambique with Project Preparation Facility (PPF) funding of US$730,000 and US$510,000, respectively. These funds were used for hiring consultants for the preparation of technical design, environmental and social impact assessments, and the tender documents for the major components in each country. The feasibility studies for the interconnector line and the corresponding transmission substations were also completed. 50. On the technical side, project components and activities were agreed upon, and alternative transmission line routes were evaluated. Four other key agreements the project implementation 11 The first phase included the Democratic Republic of Congo-Zambia interconnector, and the third phase envisages the Tanzania-Zambia interconnector. 12 Malawi Growth and Development Strategy , page Mozambique Poverty Reduction Support Strategy (PARPA II). 14 SAPMP APL1 PAD, pages

27 agreement, maintenance agreement, system operating agreement, and wheeling agreement were also substantially prepared during the preparation phase. 51. Lessons from prior regional cross-border transmission projects, such as the Costal Transmission Backbone Project of the West African Power Pool) APL (P094917, 2005) and the SAPMP-APL1 (October 2003), as well as lessons from previous power sector experiences in Malawi were taken into account. 52. An important lesson from SAPMP APL-1 was to ensure that a firm PPA was in place between ESCOM and the energy supplier. 15 Therefore, the signing of a firm power supply agreement (PSA) between ESCOM and a power supplier was actively pursued by all parties throughout project preparation. However, the PSA negotiations between ESCOM and HCB did not lead to an agreement during their discussions. The take or pay clause of the agreement was one of the key outstanding issues, which delayed the project preparation. Therefore, to move the project processing forward, the Bank decided to drop this requirement prior to appraisal. 53. Even in the absence of a PSA, the Bank assessed a strong justification for the project for Malawi based on improved energy security due to a diversification of energy supply and enabling future trade in electricity though the SAPP. 16 The lack of a signed PSA implied that the costs and quantities of available electricity to ESCOM were no longer firmly fixed or predictable and also signaled the reservations of the Government of Malawi from fully committing to the project and the risk of their future withdrawal. Since a PSA represents a commercial contractual agreement between a buyer and seller, its absence meant that any party could pull out without any financial consequences. Design 54. The project design was simple and realistic, and consisted of the required activities for achieving the PDO. The project activities were directly relevant to the achievement of the stated objectives, as they would (a) establish the necessary transmission line interconnector and associated substation links, (b) provide the necessary technical assistance to ESCOM and EdM to support the sustainability of the project outcomes, and (c) help remove infrastructure bottleneck that could impede power trade. 55. Technical design of the project was sound and based on proven technology. The design was based on a feasibility study carried out by international consultants in 2005, and updated in A double circuit 220kV line was proposed as the most suitable alternative that could meet the network stability requirements of such a long interconnection between the relatively vulnerable grid systems of Malawi and Mozambique. The proposed interconnector was designed with adequate transmission capacity to facilitate bilateral and regional power trade as intended. 56. Implementation arrangements were reasonably structured at three levels: (i) a joint Mozambique-Malawi Project Steering committee; (ii) a joint EdM-ESCOM project coordination committee; and (iii) separate PIUs at EdM and ESCOM, respectively. The project monitoring 15 See, for example, Minutes of Pretoria Meeting, November 26, While we still see some clear advantages to having a Power Supply Agreement in place, there is recognition that the Government has reservations in this regard. We view the interconnector as an important project for Malawi both from the point of view of energy security and in terms of promoting trade in electricity, and these considerations provide a strong justification for proceeding with the project even in the absence of a PSA. Excerpt from World Bank letter to the Government of Malawi s Minister of Finance, February 22,

28 arrangements were adequate for monitoring implementation progress, and the key indicators were well designed to capture the project outcomes. 57. The economic case for the project was strong (based on the assumed energy costs). The economic aspects of the proposed interconnector investments in Mozambique and Malawi were analyzed over a 30-year period ( ) resulting in an estimated a net present value (NPV) of about US$361 million, and an economic internal rate of return estimated at 28 percent. Therefore, the rates of return were shown to be robust and satisfactory. 58. However, the economic and financial analysis of the project during appraisal did not adequately reflect the change in expected electricity flows in the absence of a firm PSA. The analysis assumed that the price and quantity of electricity purchased by ESCOM would be fixed as per the draft PSA, rather than variable (and uncertain) based on contemporaneous demand and market prices. The analysis also did not consider the investment costs associated with components B and C of the project, which included the Matambo substation. 17 Government Commitment 59. Throughout preparation, there was adequate evidence of commitment and ownership of the project by both Governments. The Governments of Mozambique and Malawi were signatories to the SAPP inter-governmental MOUs, setting the framework for interconnectors and regional power trade. The two Governments had also signed a bilateral agreement in 1998 to enable Malawi to trade electricity with Mozambique, or any of the SAPP members through a high voltage transmission interconnector. Both Governments had formally requested the Bank s financial support for the preparation and implementation of the proposed project in (for PPFs) and 2006 (for IDA funding), respectively. 60. Ministerial level representatives of both Governments were also active participants in the Joint Project Steering Committee, which had contributed to the progress of project preparation issues requiring Governmental approval. In addition, the Ministers of Finance and Energy of Malawi and Mozambique had consistently expressed their respective Government s support for the project to the World Bank throughout preparation (and after Board approval). 61. However, there were also lingering concerns on the part of some government officials in Malawi regarding (i) the take or pay clause in the PSA between ESCOM and HCB; and (ii) the terms of the Wheeling Agreement between ESCOM and EdM, requiring fixed monthly payments for the use of the Mozambique part of the interconnector. 18 These concerns were initially considered as acceptable at the technical level but, over time, and particularly after the Board approval of the project, they proved to be critical factors in the eventual decision of the Government of Malawi not to sign the Financing Agreement. 17 PAD, Annex 9, footnote 41, page This payment was essentially a financing payment to EdM for the Mozambique portion of the line, which was calculated based on a formula which compensated EdM for the capital investments, plus the five percent on-lending interest paid by EdM to GoM, plus a six percent top-up to compensate EdM other costs incurred during line operations, plus operations and maintenance costs of the Mozambique portion of the interconnector line. 11

29 Adequacy of Risks Assessment and Mitigations 62. Key institutional and operational risks were adequately identified at appraisal and various risk mitigation measures were adopted. The risk of ESCOM s failure to service its debt on the onlent Credit was the only risk rated as substantial. The proposed mitigation was to ensure that both ESCOM and the Government of Malawi were committed to the development and implementation of a Financial Sustainability Plan, which would be designed to enable ESCOM to cover cash operating costs and all debt service requirements. Technical risks associated with operation and maintenance of the interconnector transmission line and power trade were to be mitigated by providing technical assistance, training, and equipment for hot-wire maintenance, as well as capacity development in power trading for ESCOM (EdM was already one of the major traders on the SAPP STEM). Other risks were rated as moderate or low, and the overall project risk was rated as Moderate. 63. The institutional project risks were correctly identified, but some country specific risks were not anticipated. For example, although Malawi s reluctance / refusal to sign the Financing Agreement could not be reasonably foreseen by appraisal, there were some early signs of concern since there had been ongoing debates in Malawi for some time on the priority of developing domestic hydro generation plants on Shire River versus construction of the interconnector link to Mozambique and SAPP. Nevertheless, no risks associated with the political economy issues (e.g., opposition to the project from certain sections due to perceived reliance on foreign sources of energy, potential change in political buy-in to the regional interconnector, pressures from a change in the political environment) were identified or discussed in the project documents Also, as ESCOM and HCB did not sign the PSA by the time of Board approval, the risk to energy availability and financial viability of the project in the absence of a firm PSA, should have been explicitly assessed in the project documentation. 65. The overall risk was rated as Moderate at entry. Had other risks, such as the lack of a substantially agreed upon PSA (or a Term Sheet), and the reluctance of Malawi to enter into a takeor-pay agreement been considered adequately, it may have resulted in a higher rating for the overall risk and prompted appropriate mitigation measures. 66. A Quality Enhancement Review meeting was held on February 15, 2007, which provided useful comments and suggestions to the team. The Quality Assurance Group (QAG) did not review this project at entry. However, QAG did review this project as part of its Learning Review of Regional Projects in October 2010, giving it an overall risk rating of high. 19 Notwithstanding Malawi s specific reasons at the time for eventually giving priority to domestic power generation development as opposed to the proposed interconnector, the sentiment seems to be more prevalent among many countries. For example, a recent study of regional integration in Southern Africa finds that: One of the most challenging aspects of regional energy integration is whether power trade is deemed politically acceptable, especially in importing countries (McKinsey & Company, 2015). For potential importers, the main concern is security of supply. They need to have confidence that exporting countries within a regional power trade arrangement will continue to supply electric power in a predictable and reliable way, or not use it as a political or diplomatic pressure tool. Vanheukelom. Jan and Talitha Bertelsmann- Scott,

30 2.2 Implementation 67. After the Board approval of the project in July 2007, the Malawi portion of the Credit required parliamentary approval before the Government could sign the Credit Agreement. However, shortly after Board approval of the project, the parliament in Malawi was suspended for over 18 months, and was subsequently dissolved until elections in May By this time, various concerns had been raised by some key Malawi officials in relation to the take or pay terms in the PSA between ESCOM and HCB, and the financial terms of the Wheeling Agreement. There were also reservations expressed that the approval of the interconnector may diminish Malawi s chances for mobilizing the necessary financing for the development of domestic generation projects, such as the proposed 64 MW Kapichira II plant on the Shire River. 69. Several attempts were made by the World Bank including a meeting with the President of Malawi in 2009 to address these concerns and to demonstrate the project s importance for Malawi s security of power supply. However, in the end, after three years of mutual effort and several extensions of the signing deadlines, the Government of Malawi informed the Bank that it would not be signing the Financing Agreement. The Bank subsequently withdrew the Credit to Malawi in July While waiting for Malawi to sign the Credit, the implementation of major activities under the Mozambique part of the project (aside from some consultancies) had been effectively on hold due to the following factors: (i) (ii) the disbursement cross-conditionality in Mozambique s Financing Agreement, which required the Malawi Credit to be effective before Mozambique could withdraw any funds allocated for the interconnector part of the project; 20 and the combined-packaging of the interconnector related substation extensions (under Component A) and the local Matambo substation transformer upgrade (under Component C) in a single bidding document, which effectively prevented the implementation of the local portion of Matambo substation works from proceeding independently of the interconnector related activities. 71. Due to these requirements, even though the procurement of the works contractor on the Mozambique side was largely complete by 2009 only awaiting signing EdM had to wait for Malawi s decision to sign the Credit (or not) before moving forward. 72. Following the withdrawal of the Malawi Credit, processing of the restructuring was delayed extensively as a result of internal Bank deliberations during FY11-FY12. The Implementation Status and Results (ISR) just following Malawi s withdrawal, and subsequent ones, stated the intention to restructure the project to focus on the Matambo transformer and EdM capacity building. 21 However, the Bank s internal deliberations on how to proceed in this case took more than two and a half years to agree on and implement the restructuring option. Concerns regarding overall portfolio quality and selectivity in resource allocation underpinned the discussion on whether the project should be cancelled, restructured as a stand-alone project, or incorporated 20 See SAPMP APL-2 Financing Agreement, Schedule 2, part B.1.(b), page See ISRs sequence 6 to sequence 10, dated June 2010 through March

31 into another existing project. During this time (FY2011 and FY2012) no supervision budget was allocated to the project, which also hindered the ability of the Bank team to implement the restructuring (ISR Sequence 8). 73. In parallel with the restructuring deliberations, the Matambo substation contract was revised to remove the interconnector related activities and re-bid. This process also took a considerable amount of time from withdrawal of the Malawi credit to the commencement of the rebidding over two years. 22 This delay too coincided with the lack of allocation of a supervision budget for the project. 74. Up until the first restructuring, implementation progress had been slow and limited in scope, and only about US$2.9 million out of the initial US$45 million Credit for Mozambique (or 6.4 percent) had been disbursed. By the end of this period (prior to the first restructuring), the ISR ratings of the project were mostly unsatisfactory due to the withdrawal of Malawi Credit and consequent inability to achieve the original PDO. Project Restructuring 75. The project went through four restructurings in total, starting with a level one restructuring and followed by three level two restructurings. The first restructuring made significant changes to the project design and financing. It removed the project from the SAPMP APL series and created a stand-alone Specific Investment Loan (SIL) operation for Mozambique. The project name was changed to Mozambique Transmission Upgrade Project (TUP). 76. The restructuring cancelled the US$31.2 million (SDR 20.5 million) Mozambique s regional IDA allocation for the cross-border transmission components of the original project, and retained only the activities unrelated to the cross-border transmission line, particularly the reinforcement of the main transmission substation at Matambo in Tete province. 23 This brought total project commitment amount to US$13.8 million (SDR 9.1 million). The closing date of the project was extended by 18 months, from June 30, 2013 to December 31, The first restructuring paper highlighted the growing electricity demand in Tete province, suggesting that there had been a 22 percent annual demand increase during the preceding four years (equivalent to over 200 percent increase over four years), and emphasized the urgency of the required transformer capacity upgrades to address the frequent network overloads and to meet the growing electricity demand in the region. The focus of the restructured project was, therefore, relevant in view of the demand for electricity The procurement of the contractor for the supply and installation of the Matambo substation transformer and the corresponding supervision consultancy suffered numerous delays, but implementation progress improved after contract signing in May 2013 and September 2013, respectively. These contracts totaled US$9.5 million (69 percent of the restructured amount). A significant source of delays was the long time required for contract approval by the Commission 22 The contract was subsequently awarded in May 2013, almost three years after the withdrawal of the Malawi credit. See section 2.4 (procurement) for more details. 23 The credit offer to Malawi was withdrawn in July 2010, after the Government of Malawi decided not to sign the Financing Agreement. 24 However, adequate background analysis and a cost-benefit analysis of the investments were not provided in the project restructuring paper. 14

32 for Foreign Economic Relations (CREE), 25 resulting in an extension in the project closing date. 26 Project implementation planning could have benefited from identifying this potential source of delay and incorporating it into the procurement clearance schedule of major packages based on CREE s track record. 79. The period between the first and second restructuring coincided with a change of the World Bank s project Task Team Leader (TTL), with interim TTLs taking over during the transition period. This impacted the continuity in supervision and coincided with delays in the signing of the amended legal agreements and disbursements related to the Matambo rehabilitation contract During the Mid-Term Review of the project in March 2014, the Government and EdM requested the use of project savings of about US$4 million to incorporate additional activities that were consistent with the PDO and could be completed within the extended project closing date, including the purchase of an additional transformer for Tete, and a mobile substation for systemwide use. 81. To incorporate new investments from the project savings, the second restructuring of the project took place in November It included (i) a further 18 month extension of the closing date to allow for completion of the delayed Matambo substation upgrades; (ii) addition of a new project component, Component D (using the project savings to purchase a mobile substation and a transformer for Tete substation); (iii) creation of one new disbursement category; (iv) reallocation of funds among components and categories; and (v) alignment of project indicators with the core sector indicators and adjustment of targets. The PDO and other aspects of the project remained unchanged. The changes to project indicators are discussed further in Section 2.3 below. 82. The above considerations notwithstanding, project implementation showed noticeable signs of gradual improvement. With all major contracts for the old and new activities already signed, the main attention of EdM and the Bank team shifted to the monitoring of the manufacturing and delivery schedules for the large transformers and the mobile substation, status of shipping arrangements, and other implementation details. Various factors beyond the control of EdM or the Bank continued to create implementation delays, including an unexpected delay in the shipment of the Matambo substation transformer from Europe to Mozambique, failure of the mobile substation equipment during factory acceptance testing, a fire at the contractor s premises in Beira, and a delay in shipping the transformer for Tete substation from China as the equipment was too heavy for the assigned ship. 83. Given the above factors, a two-month extension of the closing date was approved by the Bank, to allow sufficient time for the shipment of the remaining equipment, for which EdM would otherwise be liable to cover US$2.7 million in costs, which neither EdM nor the Government of Mozambique were in a position to cover if the project were closed. 84. By the project closing date, the new 220/66/33 kv transformer for Matambo substation was installed and commissioned in July 2016 and has been in operation since. The 66/33 kv transformer for the Tete substation was delivered to site as required, and according to EdM s 25 CREE consists of ministers and is chaired by the Prime Minister. CREE approves contracts above US$1 million. 26 ISR sequence 11 (November 2013) mentions the approval delays from CREE and the resulting delays in making the down payment to the contractor to commence works. 27 As noted in the Aide Memoire of October 5,

33 schedule, it is expected to be installed in mid The mobile substation was tested and delivered to site, but it has not been used in actual operation as of the writing of this ICR. All studies and other project outputs were delivered prior to the closing date. 2.3 Monitoring and Evaluation (M&E) Design, Implementation and Utilization 85. The four outcome indicators for the project at appraisal were simple, attributable, and measureable towards the stated PDO. The included indicators were: (1) Malawi: volume of trading via interconnector, (2) Mozambique: volume of trading via interconnector, (3) Decreased capacity deficit in case of drought in Malawi, and (4) EdM incremental earnings from ESCOM payments associated with the use of transmission line.. 28 However, one shortcoming was that while the PDO specifically mentioned access to diversified, reliable, and affordable electricity, there was no PDO level indicator measuring affordability. 86. The intermediate indicators selected were also appropriate for expected outputs and intermediate outcomes of activities under each project component (see section F of the datasheet for details). 87. The designed results monitoring system was simple and based on measurable outputs. The respective Project Implementation Units (PIUs) were responsible for collecting and consolidating the project monitoring data. 29 Thus the proposed data collection and reporting methods and arrangements were appropriate. Post-Restructuring M&E 88. During the first restructuring, the interconnector activities were dropped, the PDO was changed, and the results monitoring framework was modified to reflect the reduction in project scope. All the intermediary indicators were based on the delivery and/or installation of an output. 89. After the first restructuring, the new PDO level indicator was the System Average Interruption Frequency index (SAIFI), which is a standard industry indicator for measuring system reliability. However, this indicator was defined incorrectly in the restructuring paper as the total number of sustained interruptions in a year divided by the total number of customers. The baseline value of was 4.5 percent, and the target value was 3.5 percent. 30 Two intermediate indicators were also added during the restructuring, linked to Components C and D, respectively: (i) completion of ESIA for the proposed regional Transmission Development Project; and (ii) new 220/66/33 kv transformer installed and commissioned at Matambo substation. 28 Similarly, EdM s incremental earnings would demonstrate that Mozambique was benefitting from power trade with Malawi. 29 Such as annual reports, annual audit reports, PIU quarterly interim financial reports, quarterly reports from supervision consultants, etc. 30 The industry standard SAIFI definition is: the average number of outages that a customer would experience during a predefined period of time. The SAIFI is an index calculated using the following formula: SAIFI = (OiNi) / Nt, Where, = Summation function; Oi = the outage rate in area i, Ni = Number of customers interrupted in area i; Nt = Total number of customers served. 16

34 90. The second restructuring replaced the SAIFI PDO indicator with the appropriate core sector indicator 31 : Average interruption frequency per year in the project area (Tete). 32 A second PDO indicator, the required core indicator, Direct Project Beneficiaries was also added. Two additional intermediate indicators were also added for Components C and D, respectively, as follows: (i) level of charge at of the current transformer at Matambo, and (ii) new 220/33 kv 25 MVA mobile substation delivered. The latter was to record the completion of the new activities, while the former indicator was included to help better measure the impact of the new Matambo transformer. Overall, with the inclusion of the core sector indictors and the additional intermediate indicators (especially the measure of charge at the existing transformer) during the second restructuring improved the quality of the results framework. The indicator on the level of charge at the existing transformer both provides an indication on the level of stress on the existing infrastructure prior to commissioning of the new transformer, as well as the new transformer s its immediate impact after commissioning the level of charge fell from 98 percent just prior to commissioning to 45 percent just after. 91. This PDO indicator as defined in the results framework after the second project restructuring was simple and measureable, but not adequately attributable to the project or precisely defined. As the baseline and target values suggest, the indicator measured the number of interruptions in the Tete region. This potentially included interruptions related to both distribution and transmission infrastructure (including the Matambo transformer). Thus improvements in the indicator value could result from downstream distribution system investments by EdM to improve system reliability, which were not related to the project. This shortcoming in the indicator is evident in the fact that the indicator value substantially improved (as reported in the ISRs) even before the commissioning of the transformer. In fact, the value just prior to commissioning was already below the stated target value (the reported value in March 2016 was 191 compared to the original target of 411, see discussion in Section 3.2 below). 92. Intermediate indicators included component-specific output indicators, including that for the new activities added under component D. An intermediate outcome indicator for the impact of Matambo transformer was also added. To facilitate any future performance indicator changes, the Financing and Project Agreements were amended to remove the listings or references to specific indicators. However, while Component B still included capacity building and training for EdM, the only activities monitored were consultancy report outputs, and there were no monitoring indicators (nor any specifically defined activities) for EdM s institutional capacity building or staff training. Implementation 93. Since the Mozambique-Malawi Transmission Interconnection project did not materialize, the M&E system as described in the PAD was effectively never implemented. 94. For the restructured project, while formal quarterly or semi-annual monitoring reports were not submitted to the Bank, updates to the results indicator matrix were collected during missions and recorded in the ISRs and Aide-Memoires. However, the updates to the indicators were not regular; e.g., the value of the interruption frequency PDO indicator was not updated between March 2013 and December This did not significantly impact project implementation given (i) the 31 Core sector indicators were introduced by the Bank after the project s first restructuring. 32 This is essentially equivalent to the SAIFI for customers served by the new transformer at Matambo, where there is only one area, or i=1 in the definition above. 17

35 limited set of (and slow moving) activities under the project; and (ii) the presence of the TTL in the field with almost daily communication with EdM. 95. Overall, the clarification of indicator definitions and list of the required data after the second restructuring, which left the project with two main PDO level indicators and four intermediate level indicators, had a positive impact on improved monitoring data collection during the final stages of implementation. Utilization 96. Aside from reporting purposes in the Bank ISRs, there seems to have been no utilization of the periodically collected monitoring data under the project. This is not surprising, given the limited number of activities and as most of the post-restructuring indicators were related to outputs that materialized only at or near the end of the project. Therefore, these data were not relevant to any decisions or resource allocations during project implementation. 97. No specific M&E arrangements were envisaged for long-term monitoring of project outcomes, as the required data were a sub-set of the overall operational data (financial, technical, outages, etc.) collected by EdM. 2.4 Safeguard and Fiduciary Compliance Social and Environmental Safeguards 98. The original project was classified as environmental category B (Partial Assessment) and triggered three safeguard policies. Environmental Assessment (OP/BP 4.01), Involuntary Resettlement (OP/BP 4.12), and Natural Habitats (OP/BP 4.04) were triggered due to investments in the regional transmission interconnection. The cross-border transmission line corridor between Mozambique and Malawi passed mostly through agricultural land and brush-land of low biodiversity value, and the project was deemed as not causing any significant environmental and/or social/resettlement impacts in either country. The restructured project (2013) was of considerably reduced scope with the removal of the cross-border transmission line and significantly reduced the risks of any negative environmental and social impacts of the project. The restructured project was geographically limited to the northwest region of Mozambique, near Tete, and all physical implementation occurred within the existing substation area. The restructured project was nonetheless maintained as environmental category B. 99. A Resettlement Policy Framework (RPF) and Environmental and Social Impact Assessment (ESIA) were carried out as a part of the original project preparation. The RPF and the ESIA were disclosed in Mozambique and at the World Bank Infoshop on January With project restructuring, no modifications were required to the safeguards instruments as restructured activities were not expected to have any additional environmental and/or social implications and the environmental management aspects were handled according to the standards described in the safeguards instruments. The contractor's bidding documents and contract included the obligation to implement the Environmental and Social Clauses (ESC), related to construction activities, as described in the ESIA and RPF. The Supervising Engineer's contract also included the requirement to supervise the adequate implementation of these ESC, in compliance with the World Bank Group General Environmental, Health and Safety Guidelines. The EdM Environmental and Social Unit, with considerable experience with World Bank safeguards policies and environmental and social 18

36 national legislation, continuously carried out supervision of the implementation of the ESIA and RPF requirements and specifically the contractor ESC and contractor ESMP The project was compliant with all applicable safeguards policies, and the overall safeguards rating remained satisfactory throughout project implementation. Procurement 101. The procurement rating at project close was satisfactory. EdM was able to complete all planned procurement successfully and in full compliance with the World Bank Procurement Guidelines, despite the initial challenges arising from the withdrawal of the Malawi Credit and delayed project restructuring Prior to the withdrawal of the Malawi Credit in 2010, the limited procurement capacity of EdM, ESCOM, and the design consultant, responsible for preparing the bidding documents, resulted in some delays. However, some delays were due to the disbursement cross-conditionality requiring the effectiveness of the project in Malawi as a disbursement condition for the interconnector related procurements in Mozambique (and vice versa). On the Mozambique side, EdM had already completed the bidding process for two contracts by 2009 the Mozambique portion of the interconnection line, and the supply and install package for the Matambo substation transformer (which was combined in a single contract with the interconnector related substation extension). However, EdM could not award the contract until Malawi s credit became effective After the withdrawal of Malawi s Credit in July 2010, EdM cancelled the awarding of the interconnector line contract, and had to remove all the interconnector related parts from the Matambo substation works package. By this time, the validity period of these bids had lapsed (it had been more than 19 months since the selection of the winning bid) and a reduced scope of works could not be negotiated with the lowest evaluated bidder. 33 In light of this, as advised by the Office of the Regional Procurement Manager of the World Bank, EdM requested the cancellation of the existing procurement process for the Matambo substation works and rebidding (International Competitive Bidding) of the modified contract. This process, from the withdrawal of the Malawi Credit until the commencement of rebidding, took over two years. 34 During this period, all project monitoring indicators were rated unsatisfactory by default to reflect the inability to achieve project goals due to the withdrawal of the Malawi Credit (as indicated in the ISRs). This was the case until the project was formally restructured in January 2013 to drop the regional activities Post restructuring in 2013 (and subsequently in 2014), contract procurement performance improved despite suffering some delays as discussed in earlier sections of the ICR. Procurement performance was mostly satisfactory throughout this period until project close. 33 The prices were more than two years old and the offers had been made in a period of market uncertainty (before the global recession). 34 The contract was subsequently awarded in May 2013, almost three years after the withdrawal of the Malawi Credit. The delay is not well documented in the project documents, and since there does not seem to have been formal missions at the time, there are no Aide-memories for this period. 19

37 Financial Management 105. The overall financial management performance was rated Moderately Satisfactory throughout most of the implementation period, except for the period between the withdrawal of the Malawi Credit (July 2010) and the first restructuring of the project (April 2013), when the financial management rating along with almost all other ratings were downgraded to Unsatisfactory (as explained above). Financial management arrangements were implemented and maintained adequately by EdM throughout the life of the project, and were compliant with the Bank s policies and procedures. The project assigned suitably qualified financial management specialists who had the appropriate skills to manage the project s accounting, financial reporting, and disbursement issues. The interim financial reports and annual audit reports were prepared and submitted to the World Bank on a regular basis. The audit reports were unqualified, and there were no major financial management issues during the life of the project. 2.5 Post-completion Operation/Next Phase 106. The Government of Malawi completed the construction of the Kapichira II (64 MW) hydropower generation plant in Nevertheless, the power shortages in Malawi continued as the combination of load growth and drought exerted growing pressure on the electricity supply. Recognizing the need for alternative sources of electricity supply in addition to domestic generation, the Government of Malawi signed a new MoU with the Government of Mozambique in 2013 for the construction of the Mozambique-Malawi interconnector. World Bank support for this project is being pursued The original project s design and preparations, including the PSA and Wheeling Agreement negotiations, the feasibility study, safeguards assessments, technical design, and procurement documents, as well as the lessons learned from this case will be directly relevant for informing the design and preparation of the new project The works performed during the project will facilitate the implementation of the planned interconnection with Malawi. As designed in the original project, the planned interconnector will connect to the Matambo substation. With the extension of the 220kV busbar under the restructured project, the new interconnector can be accommodated at 220kV without significant additional works. 3. Assessment of Outcomes 3.1 Relevance of Objectives, Design and Implementation Relevance of Objectives: Substantial 109. The project objective at appraisal was to implement the regional electricity transmission interconnector between Malawi and Mozambique to increase access to diversified, reliable, and affordable supplies of energy and to benefit from the ability to trade electricity on the SAPP. At the time of appraisal, this was well aligned with both the short and long-term needs of the electricity sector in both countries and also with the priorities of the regional power pool, to which both countries were signatories (SAPP Annual Report, 2006). The overall objectives of the project were closely aligned with Malawi s aim of interconnecting to the southern region electricity power 20

38 pool, 35 and with Mozambique s goal of improving its integration into the regional economy by strengthening power trading in the SAPP and the expansion of access to electricity through grid intensification and grid extension. 36 The MGDS formed the basis of World Bank support to Malawi as mentioned in the Malawi Country Assistance Strategy (CAS) The SAPP interconnector was included as one of the planned activities to support outcome two of the CAS: Put in place foundation for longer term economic growth through improved infrastructure and investment climate. In the case of Mozambique, the interconnector was explicitly mentioned as one of the activities aligned to Pillar 3 (strengthened economic growth potential) of the Country Partnership Strategy (CPS) The relevance of the original objectives to Malawi remain substantial. Energy supply remains a key priority for Malawi and is mentioned as such in the latest MGDS ( ). The energy crisis in Malawi has worsened despite some growth in generation capacity. As Malawi is facing increasing power deficit, the interconnection project is being discussed again between the two countries The development objectives of the restructured project remain relevant to the current needs and priorities of the Mozambique electricity sector. Electricity demand in Tete, a mineral rich province experiencing significant growth in economic activity, has been growing rapidly and has put increasing strain on the electricity supply infrastructure in the region. Though the region experienced a slight slow-down due to a fall in commodity prices, the rapid economic growth driven by energy-intensive mining operations resulted in a 22 percent annual growth in electricity consumption between 2010 and 2014, and continues to grow at a significant rate. As a result, Tete city (the main load served by the Matambo sub-station), is considered a growth pole for the subregion. The upgrade of existing infrastructure to meet this growing demand is of high priority to EdM. EdM s Master Plan (2011, updated 2014) gives special attention to demand growth in the Tete province and the need for transmission (and distribution) infrastructure upgrades. Mozambique s latest National Development Strategy (2014) also stresses the importance of increased rural electrification and provision of infrastructure (including electricity) to industries so as to enable economic growth While the outages in Tete province have reduced in part due to EdM s investments in downstream distribution infrastructure improvements (unrelated to the project), the number of outages remains high. With expected electricity demand growth, the strain on the existing infrastructure, in the absence of the project, would be significant and would likely result in increasing unreliability and outages. A sustained reduction in outages is likely to result in significant benefits to households and businesses in the area. The reduction in electricity outages and thereby an increase in the reliability of electricity supply in the Tete province therefore continues to be of relevance to the electricity sector and the economy as a whole The latest Mozambique CPS ( ) mentions the focus on transmission and increasing electricity access and efficiency under Pillar 1 (Competitiveness and Growth). However, there is no explicit mention of energy reliability issues on account of demand growth for the Tete 35 Malawi Growth and Development Strategy (MGDS) , page Mozambique Poverty Reduction Support Strategy (PARPA II). 37 The interconnector project has once again been resurrected and discussions are ongoing to prepare it. As stated in the latest Malawi CAS ( ), Malawi has signaled that reinstating the Malawi-Mozambique energy interconnector has high priority to overcome the crippling power shortages. 21

39 region, or in general. In part, the relatively small and focused investment is unlikely to be explicitly mentioned in the high-level national strategy Thus, overall, as the Tete region continues to experience rapid growth in demand for electricity, the project objectives of reducing costly outages remain relevant to the current context. Relevance of design and implementation: Substantial 115. The components and activities included in the original project were clearly linked to the stated objectives. For the restructured project, despite initial shortcomings described below, the included activities, together, were directly relevant for the stated objective of the restructured project The investments supported by the original project were substantially relevant to the original PDO, as they would establish the necessary transmission line interconnector and associated substation links, provide the necessary technical assistance to ESCOM and EdM to support the sustainability of the project outcomes, and help remove infrastructure bottleneck that could impede power trade. Further, the technical design of the project was sound and based on proven technology. The implementation strategy involved close coordination between at the project management and ministerial level, which would be critical to ensure achievement of regional power trade opportunities The restructured project cancelled the regional investments and retained only the Matambo substation expansion. The purchase of the transformer for the Tete substation (and the mobile substation) was added during the second project restructuring in The Matambo substation is the only 220/66/33 kv substation serving the Tete province. The downstream Tete substation (66/33 kv) serves a substantial proportion of the load in the province at Tete city. Both substations are thus critical links in the supply chain for reliable electricity service to the rapidly growing region. The upgrade and extension works supported by the project were relatively simple in design and based on standard technology. The additional transformers at Matambo and Tete substations, along with the mobile substation, increased the reliability (through redundancy: n-1 reliability ) and the capacity to supply electricity The shortcoming in the design of the restructured project was that the relevance of the selected investment (the Matambo substation) to the stated objectives (the reduction in frequency of outages) was not sufficiently justified or backed up by an updated assessment of the key substation infrastructure bottlenecks in Tete province and their prioritization in relation to outages. 38 During restructuring (and in the original project) an assessment of the main source of outages in Tete province, the split between distribution outages and transmission outages, or the adequacy of related (downstream) system infrastructure in the region was not documented. According to the EdM Master Plan s demand forecast (2011) for Tete, the transformer at Tete substation was already overloaded, suggesting that an upgrade at Tete substation may have been of greater priority in order to reduce outages. In the absence of crucial background analysis for project activities, it is hard to determine whether selected investments were the most relevant choice for 38 The Matambo substation was not evaluated as a part fo the economic analysis in the original PAD; and there was no economic analysis in the restructuring paper either. 22

40 reducing outages. 39 An additional shortcoming in the design of the restructured project was that the main results indicator was not well defined and did not provide easily attributable information to judge the achievements of the project (as discussed in Section 2.3) or whether the Matambo transformer expansion, on its own, would have achieved the PDO target Project savings allowed a second restructuring to include the purchase of the additional transformer for Tete substation and a mobile substation. The expansion in transformation capacity was aligned with EdM s Masterplan that identified inadequacy of transformer capacity at Tete substation as a critical constraint to electricity supply in the province. The enhancement of the capacity both eases the constraint on electricity supply and increases reliability by building redundancy into the system and reducing the strain on existing infrastructure. The mobile substation, similarly, can be deployed to back up the existing substations during planned maintenance or breakdowns. The inclusion of the additional transformer seems to have had a significant impact on prospective benefits delivered by the project, especially in terms of ability to meet growing electricity demand (see Annex 3) Overall, despite issues with project design during the first restructuring, especially with regards to the justification for Matambo transformer, the utilization of project savings for the purchase of the Tete transformer and mobile substation considerably increased the relevance of the project activities to the stated objectives. Therefore, the overall relevance of project design and implementation is rated as Substantial. 3.2 Achievement of Project Development Objectives Rating: Substantial Original PDO 121. The original PDO was not achieved as a result of Malawi withdrawing from the project without signing the credit agreement. All activities associated with the regional interconnection were dropped and the project was restructured to focus on transformer upgrade for Mozambique. The achievement of the original PDO was thus negligible. Restructured PDO 122. The restructured PDO, to reduce the frequency of electricity outages in Tete province, was substantially achieved. All project investments were successfully completed by project close: installation of the 130 MVA (220/66/33kV) transformer, the upgrade of the System Control and Data Acquisition (SCADA) system and 220kV busbar at Matambo substation, the procurement of the 50 MVA (66/33 kv) transformer for Tete substation, and the procurement and delivery of the 25 MVA (220/33kV) mobile substation. The new transformer at Matambo was commissioned in July The mobile substation was commissioned in December The new transformer at Tete is on site and is expected to be commissioned in August For example the Master Plan also mentions that some transmission lines were also expected to become a constraint in the foreseeable future. Though additional line bays were added at Matambo to enable connection of additional lines in the future. 23

41 123. In the results framework, the reduction in outages was measured by the PDO indicator: average interruption frequency per year. By project close, the indicator had significantly declined (as seen in the table below) implying that the average interruption frequency had reduced significantly below the project target. Baseline at restructuring Target Current Achievement rate (compared to restructuring baseline) Baseline Average interruption frequency per year in the project area %* (2007) (2013) (2016) (2016) * This measure the actual reduction as a percentage of the targeted reduction. The majority of the decline seemed to occur prior to the commissioning of the transformer However, as discussed in Section 2.3, this indicator value is not attributable to this project. The average outages had already decreased substantially a few months prior to the commissioning of the Matambo transformer and reduced further in the period after commissioning. 40 This may have been due to upgrades to the downstream distribution network in the province as a part of other projects that EdM was implementing, or a temporary decline in electricity demand from large mining customers who may have lowered their production activities in response to depressed commodity prices The reduction in the Tete province outages as demonstrated by the results indicator is also consistent with the SAIFI indicator that EdM collects for the Central region (where Tete province is situated). The Central region indicator was consistently increasing between 2011 and 2015 and then registered a decline in While this could add to the indicative evidence of the impact of Matambo, it is difficult to estimate the exact contribution of the Matambo transformer upgrade to these results. Central Region SAIFI Despite the shortcomings in the results indicator and very short window of observable performance, it is possible to make an assessment of the real and potential impacts of project investments since, (i) the causal link between the delivered equipment and the short and long term reduction in outages related to the substation upgrades has a strong technical justification; and (ii) alternative data on outages at Matambo and the Central Region (which includes Tete province) was obtained for the ICR, coupled with the intermediate results indicator for the Matambo substation ISR seq. 16 shows that the average interruption frequency was 199 in March 2016 and then fell to 151 in October 2016 as shown in the Borrower s ICR. 41 The World Bank-supported Mozambique Energy Development and Access Project (P108444) was concurrently supporting distribution infrastructure, including distribution lines and transformers, in the Tete province. 42 This is with the caveat that only a couple of months of data are available after commissioning. This is a recurrent issue in estimating the impact of infrastructure projects, which require a longer timeframe of observational data than an ICR affords. 24

42 127. The additional transformers at Matambo and Tete, along with the mobile substation, both add to the transformation capacity at critical supply bottlenecks in the Tete province, and also create n-1 redundancy. Matambo is the only substation serving the Tete province (connecting it to the 220 kv transmission network) and the Tete substation downstream from Matambo, serves a significant majority of the load in the province at Tete city. By project closing, the existing Tete transformer was already overloaded, and the Matambo transformer too was operating at very high charge levels, putting stress on the system (as shown by the intermediate indicator below). The increased redundancy and capacity at both substations directly reduces both planned and unplanned outages linked to the substations and eases capacity constraints on the supply of electricity in Tete province Data collected by EdM and the intermediate results indicator further show the prospective impact of project investments on outages. Data on outages at Matambo substation obtained from EdM shows that between 2007 and 2015 there were an average of nine outages per year with an average duration of three hours. This data can be used to estimate the reduction in outages that are directly attributable to project investments. 43 Furthermore, the intermediate indicator linked to the transformer at Matambo shows that the target was exceeded substantially. The level of charge at the existing transformer signifying the amount of electricity passing through as compared to its maximum capacity fell from 98.1 percent in March 2016 to 45 percent in August 2016, immediately after commissioning of the new transformer. This clearly shows that the existing transformer was operating close to or even beyond its technical limit and the commissioning of the new transformer has eased the stress on the existing transformer. This implies a lower likelihood of future breakdowns leading to outages. Baseline at restructuring Target Current Achievement rate (compared to restructuring baseline) Indicator Baseline Level of charge of the current Matambo transformer 26% 86% 70% 45% 256% The core indicator on the direct project beneficiaries was also 94 percent achieved. The baseline was defined as zero prior to the commissioning of the Matambo transformer. After the commissioning of the transformer, the indicator measured the number of electricity customers in the Tete province (served by the Matambo substation). The project supported the installation and purchase of transmission infrastructure, and did not directly support electricity connections. However, by easing the capacity constraint on the electricity transformation capacity for the Tete province, the project investments have enabled, and will enable an acceleration in new electricity connections in the province. Thus while the beneficiary numbers fell marginally short of the target at 94 percent, the target is expected to be achieved, and exceeded, in the near future. 43 While this is probably a conservative estimate of the impact of Matambo on the outages in the Tete province, it was the only data made available that clearly isolates the impact of Matambo substation. 25

43 Baseline at restructuring Target Current Achievement rate (compared to restructuring baseline) Indicator Baseline Direct Project Beneficiaries ,628 81,276 94% *The direct project beneficiaries are measured as the number of customers served in the Tete province. The indicator is the same as the indicator on the number of customers served in the project area which is a supplemental indicator to the core PDO indicator on the average interruption frequency The short and long term benefits of the Matambo substation upgrades can be more specifically summarized as follows: The previously existing 60 MVA transformer was overloaded and the situation was deteriorating due to the increasing demand. Before the project, the Matambo substation (the only transmission substation in Tete province) had only one 60 MVA transformer. This transformer was significantly overloaded, with a nominal load of 58 MVA (98 percent). With the addition of the 130 MVA transformer, the load of the 60 MVA transformer has dropped to 27 MVA (45 percent). It is expected that with regular maintenance and frequent inspections the transformer should not be overloaded during its remaining useful life. The reduced load on the old transformer is an immediate benefit of the new transformer installation. The new transformer has provided improved system reliability through n-1 configuration. The 60 MVA transformer was the only transformer in the Matambo substation. Any faults on the transformer or auxiliary equipment, meant that all the customers would incur electricity outages. The addition of a 130 MVA transformer gives the system the n-1 configuration it requires, to provide enhanced reliability in the case of faults and the ability for the utility to perform offline maintenance without causing any outages. This configuration is expected to benefit the area for the next 20 years, the period in which the two transformers are expected to operate in parallel. The additional transformer provides the ability to accommodate the estimated load growth until The demand forecast completed in 2012, as part of the Master Plan, foresees a maximum demand of 82 MVA by the year 2026, for the Matambo substation, taking into account all of the industrial projects forecast for the area. With the additional transformer the substation is able to serve the province s growing demand. This is an important factor to consider as the project assisted the utility in achieving a significant part of the long-term transformation requirements in the Tete province, which is of importance to the country due to the growing mining activity. The works performed during the project form a basis for the planned interconnector with Malawi. The Mozambique-Malawi transmission interconnector is a project under preparation with assistance from the Bank. With the extension of the 220 kv busbar under the project, the new interconnector can be accommodated at 220 kv without significant additional works. The installed SCADA system at Matambo substation gives the EdM remote control and supervisory capabilities that they did not have before at that key substation. With this recently installed system, the utility has better access to data and knowledge of the events 26

44 occurred in the substation. EdM will be able to integrate the substation and its equipment into the SCADA system at their future National Control Centre, which is in the feasibility stage at the moment Overall, all project activities were successfully completed and the technical impact of the new equipment on current and future outages is well understood. As shown above all key results indicator targets were also largely met and exceeded. After the commissioning of the new transformer at Matambo substation, the level of charge at the existing transformer after commissioning reduced substantially, signifying an immediate reduction in the strain on the transformer capacity. Furthermore, additional data collected by EdM, on the substation specific outages at Matambo, further indicate a reduction in outages due to project investments. Thus despite the incomplete data for the outage reduction in the results framework, it is clear that the project will lead to the reduction of substation related outages. For these reasons the achievement of the PDO is rated as Substantial. Overall rating 132. The overall rating for the project combines the evaluation of overall project relevance, efficacy and efficiency. Based on the split evaluation methodology for restructured projects, the project outcomes have been assessed for each of the three phases determined by the two significant project restructurings. The performance in each phase is weighted by the proportion of actual project disbursements to derive an overall rating for the project. Based on the outcome of the split evaluation, the achievement of the PDO is rated as Moderately Satisfactory. 44 The explanation is provided in section 3.4 below. (XDR) Pre-restructuring Post First Restructuring (2013) Post Second Restructuring (2014) Overall Unsatisfactory Unsatisfactory Satisfactory Moderately Satisfactory Rating Disbursement Amount ($m) Weight 20.7% 9.6% 69.6% 100.0% Score Efficiency Rating: Modest 133. There was no economic analysis conducted at the time of the first restructuring, and thus it is difficult to evaluate potential improvements in project efficiency compared to that expected at project design stage. Furthermore, shortcomings in the results indicators along with the short time frame since commencement of new equipment operation prevent a comprehensive evaluation of ex-post project efficiency. Additional data obtained from EdM has been used to determine the economic net benefits of the project. 44 This is consistent with the Modest rating for project relevance and efficiency. 27

45 134. Project investments were estimated to be economically viable with positive economic returns accruing over the economic lifetime of the equipment installed. Benefits accrued from the reduction in costly outages and the easing of the supply constraint to the fast growing demand in the Tete region. Overall the project had an estimated Economic Internal Rate of Return (EIRR) of 36 percent and an NPV of US$147 million (at a six percent discount rate). The EIRR for the Component C (Matambo substation) was estimated at seven percent (NPV of US$0.75 million), and the EIRR for the Component D (Tete transformer and mobile substation) was 97 percent (NPV of US$155 million). 45 The project was estimated to be financially unviable from the perspective of EdM as cost of electricity supply exceeds the electricity tariffs for domestic customers (for details see Annex 3) The project as initially planned (at appraisal and first restructuring) was to install an 80 MVA transformer at Matambo substation in Tete province. However, cost efficiencies resulting from the competitive procurement process resulted in savings. This allowed both the upsizing of the transformer at Matambo to 130 MVA, and the purchase of another 50 MVA transformer at Tete and a 25 MVA mobile substation. The addition of these investments added considerably to the project benefits without raising total cost above the initially planned amount. This represents an improvement in project cost efficiency compared with what was expected at appraisal (for this activity) or the first restructuring In contrast to the efficient use of financial resources, the repeated delays and project extensions negatively impacted overall efficiency in the achievement of project objectives. 46 While some of the delays may have been outside the control of the World Bank team or implementing agency, an aggregate 38 month extension in the project closing date, for a relatively straightforward investment project, is difficult to justify. In the end, the relatively simple project activities were realized nine years after project approval in Due to inordinate delays in completion of project activities, the project efficiency is rated as Modest. 3.4 Justification of Overall Outcome Rating Rating: Moderately Satisfactory 138. The PDO of the restructured project was to reduce the frequency of electricity outages in the Tete province. The overall outcome rating takes into account the relevance, efficacy, and efficiency of the investments and achievement of outcomes. The split evaluation methodology is applied to assess each stage of the project marked by the first and second restructuring. 45 Based on available data, the benefits from the Matambo transformer upgrade were quantified considering only a reduction in outages at the substation (from data obtained from EdM). This is thus a conservative estimate of the net benefits as it does not account for the reduction in other associated outages e.g. incidence of load shedding of customers in Tete province caused by transformer capacity constraints. As the Tete substation is downstream of the Matambo substation, benefits from capacity expansion at Tete assumed no capacity constraints at Matambo. Therefore, the quantified benefits from Tete expansion are attributable, in part, to Matambo transformer investment. 46 As future benefits are discounted at the social discount rate, delays in completion of project activities and thus accrual of associated benefits reduces the present value of the stream of benefits. All else equal, this implies a reduction in project efficiency as measured by the NPV. 28

46 139. The project at appraisal was well designed and highly relevant for both countries, despite certain shortcomings discussed in previous sections. However, with the withdrawal of Malawi from the project without any progress towards the PDO, this phase is rated as Unsatisfactory After the first restructuring, the PDO was relevant to the country context; however, implementation progress was slow with limited progress towards the PDO, requiring further project extensions. The relevance of the selected project investments to the PDO was inadequately justified in the project paper. Furthermore, the results framework had shortcomings, with the main PDO indicator being incorrectly defined and not fully attributable to project investments. The phase is thus rated Unsatisfactory After the second restructuring close implementation supervision helped make substantial progress towards the PDO and all project activities were successfully completed, albeit with some delays requiring a final, two-month project extension. Project savings were utilized for additional infrastructure investments which increased the benefits delivered by the project. Certain shortcomings (inadequately attributable results indicators) led to challenges in comprehensively judging the impact of the investments. However, once all the equipment is operational, prospective future impacts are clearly understood from a technical perspective. Therefore, this phase is rated as Satisfactory In the end, the project which was initially designed as a regional interconnector had to be restructured into a modest but locally significant substation upgrade and equipment procurement project. The use of project restructurings to utilize project savings ensured the delivery of all project investments Combing the ratings from each of the phases and weighting by the disbursement proportions in each phase gives an overall rating of Moderately Satisfactory. 3.5 Overarching Themes, Other Outcomes and Impacts (a) Poverty Impacts, Gender Aspects, and Social Development 144. The investment supported by the project helped (and will continue to help) ease critical supply constraints to a fast growing region of Mozambique. While none of the investments were consumer-facing, new and existing electricity consumers were the intended final beneficiaries of the improved electricity service in the region. The number of electricity consumers in the region has increased since the project baseline from 20,527 (2007) to 81,276 (2016). 47 The increased transformation capacity and subsequent reliability resulting from the project, according to EdM, has also accelerated electricity connections in the region There is no background data or impact evaluation available on poverty and gender impacts of the project. However, based on numerous studies that link improved access to reliable electricity to growth and job creation (e.g. Dinkelman 2011, Fisher-Vanden et al 2015, and IEG 48 ), similar 47 The rate of electricity access has steadily increased in the past years and had reached about 15 percent as of 2014 (EdM Annual Report, 2014). 48 Lack of access to electricity is a major constraint to economic growth and increased welfare in developing countries. This has been reemphasized by the United Nations and the World Bank Group as co-chairs of the global Sustainable Energy for All (SE4All) initiative, which was launched in 2011, with the goal of achieving 29

47 long-term impacts are expected from the project. A World Bank literature review of the impact of power outages finds that the reduction in outages is especially beneficial for productive users such as firms and irrigation (World Bank, 2016). Based on this and other studies, it is expected that increased access to reliable electricity supply from the project will improve productivity of firms, lead to job creation, and improve the well-being of households previously relying on low quality, expensive, and unhealthy alternative sources for lighting and phone charging According to the results framework, an expected 50 percent of the beneficiaries are women. But there is no specific data available to judge the accuracy of this number or estimate differential impacts on women or female-headed households in the project target areas. (b) Institutional Change/Strengthening 147. Although Component B of the project included capacity building and training for EdM, there is no indication that any specific institutional capacity building or staff training activities were carried out under the project. The project monitoring arrangements also did not include a specific indicator for monitoring capacity building and training for EdM. It is therefore difficult to assess the impact of Component B on EdM s capacity improvement efforts. However, the ICR team recognized that under the Bank-funded Electricity Development and Access Project (P108444), which was being implemented in parallel with this project, EdM had prepared a training program for capacity building and for developing specific technical skills. This program covered business process management, planning, design and construction of transmission lines, operation and maintenance of switchgear, and environmental and safeguards training. 3.6 Summary of Findings of Beneficiary Survey and/or Stakeholder Workshops 148. No survey or workshop was held as a part of the ICR. The ICR benefitted from extensive discussion with the implementing agency staff and management and current and former World Bank TTLs. 4. Assessment of Risk to Development Outcome Rating: Moderate 149. The objective of the restructured project was: to reduce the frequency of electricity outages in Tete province. The key indicator was: Average interruption frequency per year in the project target area The new transformer and other equipment at Matambo substation have significantly improved the reliability of the power system in Tete province. Furthermore, the new transformer at Matambo provides n-1 redundancy for the substation, thereby decreasing the risk (and impact) of any potential outages. The 25 MVA mobile substation further improves the reliability by enabling the backup of 220/33 kv transformers at Matambo or elsewhere. The 50 MVA transformer at Tete substation will play a complementary role in reducing outages in its service area in Tete city. universal access to energy within the next 15 years, along with improving energy efficiency and increasing the use of renewable energy. Providing access to electricity is also integral to the Bank Group s corporate goals of increasing shared prosperity and ending extreme poverty by IEG: World Bank Group Support to Electricity Access, FY , page xiv,

48 151. Considering the significant transformer capacity increase and the introduction of n-1 redundancy at Matambo substation, the main risks to sustaining the reduction in outages in the Tete province are (i) delays or shortcomings in the installation of the transformer at the Tete substation, and (ii) the financial inability of EdM to undertake adequate investments in maintenance, refurbishment and upgrade of project investments and complementary downstream infrastructure The project supported the purchase of the transformer at Tete substation from project savings. Due to inadequate time and resources, the project only supported the purchase of the transformer and not its installation. It was agreed that EdM would use its own resources to fund the installation of the transformer. The contract for the engineering services consultant under another ongoing Electricity Development and Access Project was amended to include support for the procurement and implementation of the supply and installation of the transformer at the Tete substation and the mobile substation. This arrangement will allow continuity in implementation and the timely installation of the equipment purchased under this project The weak financial position of EdM could constrain timely investments in the necessary sub-transmission and distribution system maintenance and refurbishments in Tete province. 49 Sustaining the decrease in outages in the Tete province depends on (i) the adequacy of capacity compared to the growth in electricity demand in the province; (ii) the timely maintenance of installed infrastructure; and (iii) the adequacy of complementary and downstream infrastructure. The equipment financed by the project is standard technology that EdM has the capacity to operate and maintain. Routine maintenance of the equipment is within the existing maintenance budgets for EdM. However, given the overall poor financial situation of EdM, there is a risk that there may be insufficient funds if replacement or significant repairs are needed for older equipment. Similarly, resource constraints may prevent timely investments in refurbishment or capacity expansion of critical complementary and downstream infrastructure. If the downstream infrastructure is poorly maintained or inadequate for meeting the growing demand, the system may become constrained despite the adequate transformer capacity at Matambo, resulting in increased outages On the demand side, while the outage reduction targets may be achieved in the short term, EdM s fast expanding customer base in Tete region (reportedly around 22 percent over the past several years) as well as the growing demand from the returning mining and other industries could quickly increase the demand and go beyond the additional capacity provided under this project. Recently, there has been a slight reduction in peak demand due to reasons unrelated to this project (e.g., closure of mines and businesses as a result of declining commodity prices in Tete, etc.). The projected demand growth in Tete province for the next five to 10 years is also well within the excess capacity of the new installed transformers at Matambo (see table below). However, a recovery in commodity prices, continuing economic growth, population increase due to new residents migrating into the province for economic reasons, could increase pressure of the electricity infrastructure. 49 ISR (Sequence 16) mentions a deterioration in the financial position of EdM as a potential risk to the timely installation of the transformer at Tete substation. 50 According to EdM s Master Plan: The 66 kv line from Matambo to Tete has a thermal limit of about 60 MVA, thus limiting the long term supply capacity and reliability of supply. The load forecast for Tete is 50 MW in 2026 for the base case and in addition there is the load to be supplied in the Manje area. There is hence a need for new lines coming in to the area. 31

49 Forecast Peak Load (MW) Low Medium High Sub-station Matambo Manje Tete Total (Tete+Manje) Coinc. Peak Total (Tete+Manje+Matambo) The addition of the new 130 MVA transformer and the refurbishment of the switchgear and other improvements at Matambo substation financed by this project have already provided the necessary (although in the long run not sufficient) condition for achieving the project s development objective to reduce the frequency of electricity outages in Tete province. Other risks beyond the scope of this project (such as increased demand, which could overload the weaker parts of the distribution infrastructure, leading to unplanned outages) could, however, affect the achievement of the development outcome in the long run Therefore, based on these considerations, the risk to development outcomes of the project is assessed as Moderate. 5. Assessment of Bank and Borrower Performance 5.1 Bank Performance (a) Bank Performance in Ensuring Quality at Entry Rating: Moderately Unsatisfactory 157. The appraised project design was well aligned with stated country goals and priorities (as described in Section 3.1 above). The financing of the cross-border interconnector and associated investments was underpinned by strong economic justification as shown in the economic analysis. The supply of electricity from Mozambique to Malawi through the proposed interconnector was seen as a least-cost alternative for Malawi to meet the growing electricity demand and mitigating the significant drought-related risk of a large supply shock The project design also included measures to mitigate potential risk of implementation delays, coordination risks, potential infrastructure bottlenecks, and financial risks. Given the need for joint implementation between Malawi and Mozambique, the implementation plan that was a part of the project design, put in place a three-tiered coordination mechanism to mitigate coordination risks a joint project steering committee (Ministry level), joint project coordination committee (utility management level), and regular interaction at the project management level (PMU level). 51 In addition, drawing on lessons learned, project preparation advance was utilized 51 The Financing Agreement explicitly mentions each of these committees along with the requirement that the Recipient and respective utilities shall maintain active participation in the respective committees. 32

50 so that the technical designs were almost complete and major procurement packages for the interconnector were in an advanced stage of readiness at the time of project effectiveness. 52 Risk assessment at the time of appraisal clearly identified financial risk, especially for the intended buyer of the electricity ESCOM (Malawi). The risk mitigation measures included in project design were in the form of a financial sustainability plan for ESCOM (Malawi), and financial covenants for both the utilities. Given the dependence of project investments in one country on the timely completion of the investments in the other, cross-conditionality of Credit effectiveness was built into the disbursement conditions of the project. Lastly, the signings of four key agreements that would underpin the trade of electricity between the two countries were included as conditions of effectiveness The PAD did not discuss the risk associated with the failure of the two countries to reach an agreement on key parameters of power trading. It also did not list the risk of political opposition to the interconnector stemming from the prioritization of the development of domestic energy resources over perceived dependence on foreign energy sources. In the end, these were the reasons why Government of Malawi did not sign the Financing Agreement. 54 While the political turmoil in Malawi shortly after project approval may not have been foreseen, there did seem to be indications of Malawi s reservations regarding the wheeling agreement and the PSA. Late in project preparation, reservations were expressed regarding the take-or-pay clause of the PSA. There were similar concerns about the terms of the wheeling agreement. These risks should have been adequately discussed in the PAD and appropriate mitigation measures identified The PDO at appraisal was stated as increased access to diversified, reliable and affordable energy. However, the PDO level results indicators did not include measures for reliable and affordable which would have made it difficult to adequately judge the achievement of the PDO. Reliability indicators were however included as a part of the intermediate indicators, linked to Component C The project economic analysis at appraisal did not adequately reflect the justification of the project based on its value as an insurance against drought. Despite ESCOM s unwillingness to sign a firm PPA, the baseline economic analysis was rested on the assumption of firm power purchase. Furthermore, the cost-benefit analysis of Matambo substation upgrade (part of Component C in the appraised project) was not included in the economic analysis. Despite the stated criticality of Matambo substation upgrade in meeting the rapidly growing demand in the Tete region, the benefits for this component were not presented in the economic analysis in the PAD. While it may be the case that costs associated with this component were considerably smaller than those for the regional transmission interconnector, the cost and benefits from the investment should have been explicitly modeled. With the benefit of hindsight, this is an important omission as it was the only investment component that was retained in the restructured project and an economic analysis was not conducted at that time either. As a result, there is no baseline that the economic analysis in the ICR can be compared to. 52 The finalization of technical designs was seen as a cause of delay in the earlier project in the same project series (i.e., APL-1, P069258). 53 The four agreements were (i) Implementation Agreement, (ii) Maintenance Agreement, (iii) System Operating Agreement (included the PSA), (iv) Wheeling Agreement. These were later converted form condition of effectiveness to a disbursement condition for the regional investments, on request of Mozambique, to allow progress on other components. 54 In a letter to the World Bank in November 2009 the President of Malawi stated the interconnector was the country s second priority behind the development of own energy production capacity. 33

51 162. Finally, implementation delays may have been avoided by splitting the procurement package for Matambo substation into separate lots for the interconnector (regional) and transformer expansion (national). As designed, the procurement package for the Matambo substation combined the expansion works (additional transformer) with the extension works for the interconnector. As a result, and due to the cross-conditionality of any interconnector related works in one country commencing only after project effectiveness in the other, work on substation expansion critical for Mozambique s Tete province was stalled while waiting for Malawi to sign the Financing Agreement this led to considerable delays in the completion of the substation expansion Taking into consideration the above factors, the Bank Performance at entry is rated as Moderately Unsatisfactory. (b) Quality of Supervision Rating: Moderately Unsatisfactory 164. Regular project supervision was undertaken throughout the duration of the project. A total of 16 ISRs, over the roughly nine-year project life, documented the progress made and the key issues to be addressed at regular intervals. The task team leader was based in Maputo for the most of the implementation period and this allowed regular communication with the client Prior to the withdrawal of Malawi from the project the task team proactively followed up with ESCOM and EdM to address outstanding issues regarding associated key legal agreements. 55 The task team along with World Bank senior management also held high-level discussions with senior government officials to present the merits of the interconnection project in the context of Malawi s energy needs and address stated concerns. 56 While waiting for Malawi to sign the Credit, the task team ensured that activities on Mozambique s part of the project were able to progress to the extent possible, including a timely amendment to convert effectiveness cross-conditionality for Mozambique into a disbursement condition for the regional component disbursement category The project was restructured multiple times to enable adaptations to the changing circumstances so as to maximize development impact of available funds. As a result, the project was able to support more investments than planned during the first restructuring, and all activities were successfully completed by project close A review of the project documentation and timelines suggests that there were some shortcomings in certain phases of implementation The decision on the course of action following the Malawi s withdrawal in 2010 was very protracted as different options were discussed between the task team and regional senior management. The final decision to proceed with restructuring was made in January 2013, almost 2.5 years later. In parallel, while progress was made on procurement for the Matambo substation, the time taken to decide to restart the procurement process, the lack of allocation of Bank budget during Fiscal Year 2012 (documented in the ISRs), the uncertainty regarding the restructuring, all 55 Numerous updates are documented through official correspondences, internal memos, aide-memoires and ISRs. 56 The sector manager of the WB energy team met with the President of Malawi in September 2009 to address stated concerns regarding the project. 34

52 potentially contributed to the delayed start of the substation works (contract award was in May 2013) which later required multiple project extensions After restructuring, the new PDO of the project was To reduce the frequency of electricity outages in Tete province. A more outcome oriented PDO definition, such as improved reliability of electricity services in Tete province may have been more appropriate and would have better lent itself to meaningful evaluation. Also, the project beneficiaries of the restructured project were not defined explicitly in project documentation In addition to the delays, the first restructuring paper did not re-appraise the retained activities as they were appraised as a part of the original project. No thorough background analysis was presented to justify the choice of investments based on the latest available information. Also there was no economic analysis presented on the Matambo substation extension. As the economic analysis in the PAD did not present a separate cost-benefit analysis for this activity, the analysis should have been undertaken to justify the investment based on most recent data (it had been almost six years since appraisal). Related to this, the associated results indicator which was added during the first restructuring was both incorrectly defined and not appropriate for measuring the attributable impact of the investment The World Bank performed well to deliver all outputs under the restructured project and make use of restructuring to successfully utilize savings in a timely manner, However, despite substantially achieving intended outcomes of the restructured project, the significant delays in deciding on the first restructuring, inadequate analytical underpinning for the selection of investment, and inadequate monitoring framework reduced the overall effectiveness of World Bank implementation performance. Taking this into consideration, the Bank Performance of Supervision is rated Moderately Unsatisfactory. (c) Justification of Rating for Overall Bank Performance Rating: Moderately Unsatisfactory 172. The overall Bank performance is rated as Moderately Unsatisfactory. The Bank team provided supervision support to adapt to changing circumstances and utilize project savings. However, there were significant shortcomings in terms of delays, inadequate analytical underpinning, and inadequate monitoring framework, which reduced the overall effectiveness of World Bank implementation performance. 57 This is discussed in detail in Section 2.3. The impact of this during project supervision was minimal as impacts would only materialize after the transformer commenced operations, which was expected after project close. As utilities collect the relevant information as a part of their day-to-day operations, this data was requested and used for the analysis in the ICR. 35

53 5.2 Borrower Performance (a) Government Performance Rating: Moderately Satisfactory 173. As the Government of Malawi never signed the Credit Agreement, they were never officially a borrower for the project. Therefore only the performance of Mozambique is evaluated The Government of Mozambique supported the project throughout. The Government expediently moved forward after project approval. The project was approved in July 2007, and Mozambique signed the Financing Agreement in September 2007 and achieved effectiveness in February With delays on the Malawi side in signing, the Government of Mozambique requested a waiver of the effectiveness cross-conditionality to enable the Mozambique Credit to become effective and to progress on activities on their side. All other effectiveness conditions were met in a timely manner. The Government also participated in the steering committee meetings with Malawi counterparts, but failed to reach an agreement on the key commercial parameters of the interconnector which resulted in the project not going forward as originally designed and the original PDO not being achieved The Government of Mozambique had a limited role under the restructured project, but was committed to the restructured PDO. The Government requested and supported the project restructurings and the utilization of savings which were important in making sustained progress towards the PDO. The delays in obtaining approval from CREE and the slight delay in signing the amended subsidiary loan agreement as a result of the addition of activities under Component D resulted in some project delays Overall, while the original project never got off the ground due to a breakdown of commercial negotiations, the Government of Mozambique actively supported the restructured project and the timely utilization of project funds. Thus, on balance, the Government performance is rated as Moderately Satisfactory. (b) Implementing Agency or Agencies Performance Rating: Moderately Satisfactory 177. The sole implementing agency of the restructured project was the electricity utility of Mozambique, EdM was effectively able to implement all planned activities by project closure, and was also able to utilize project savings to undertake additional investments. Project reporting requirements were mostly met, and interim financial reports and audited reports were submitted in a timely manner for the most part While project activities were completed, there were some shortcomings in procurement and implementation that led to delays in completion. The procurement for the Matambo contract suffered several delays (due to delays on the part of both the World Bank and EdM) and works only commenced in January This resulted in the project closing date requiring extension. There were also issues raised regarding contract management and supervision. There were documented delays in clearing invoices, which led to further delays. One issue was the absence of a centralized or dedicated PIU within EdM, for the project. It is the standard practice of EdM to have designated project teams fully integrated into the EdM staffing structure. As a result, the designated project manager also is responsible for other projects/tasks. While this has certain 36

54 advantages in general, it reduced the ability to focus on project supervision. It is also not clear whether EdM was able to meet the financial covenant of the Current Ratio threshold (at least 1.3) during the last two years of the project Due to the reasons outlined above the Implementing Agency Performance is rated as Moderately Satisfactory. (c) Justification of Rating for Overall Borrower Performance Rating: Moderately Satisfactory 180. At completion the project substantially achieved its targets albeit with delays. In light of this and the issues outlined above, the Overall Borrower Performance is rated as Moderately Satisfactory. 6. Lessons Learned 181. The following key lessons were learned from the project: Mutually agreed and transparent terms of electricity trading should be established as a prerequisite for the implementation of a dedicated transmission interconnector Development of dedicated transmission interconnectors such as the Mozambique-Malawi interconnector (in the absence of further connections between Malawi and other countries) requires that the primary beneficiary country (Malawi) and the primary supplier of electricity reach agreement on a fair, transparent, and firm PPA to ensure the long-term availability of supply at predictable prices. This is critical for ensuring that both parties can agree on the rules of the game and be able to estimate the total cost of delivered power (and the cost effectiveness of the interconnector alternative for the country buying the power) along with the revenue resulting from the project for the seller. A thorough assessment of political economy context, especially with regard to perceived trade-off between self-reliance and imports is essential for regional electricity trade projects The interconnector project was assessed to be strong on technical grounds and the counterparts at the technical level seemed convinced of the project s merits. This was expected to be sufficient for the project to be supported by all stakeholders (or at least not opposed). However, ultimately, Malawi withdrew from the project due to the concerns raised about the project by stakeholders in the highest levels of the Government and also given the internal abrupt changes on the administration of the country right after the project was brought to the Board. An assessment of the political economy risks opposition of some stakeholders to perceived risks of reliance on foreign sources of energy versus domestic resource development, potential change in political buyin to the regional interconnector from a change in the political environment could have led to the identification of some stakeholder concerns and potentially helped mitigate them in a timely 58 The documents do not report on this adequately. The last couple of Aide Memoires say that the information was not provided by EdM and a one of the later ISRs claims that the ratio of 1.3 was not met but it was above 1, as stipulated in a parallel Bank project that was ongoing at the time. 37

55 manner. In general, for regional projects involving cross-border coordination and commitments, an assessment of the political environment in each country along with national priorities and the regional context within which they are developed is critical to the success of the project. To the extent that it is technically feasible, the project design should maximize opportunities for mutually beneficial trade A strong value proposition for both countries can improve the buy-in for both sides. With both countries having greater skin in the game, the negotiations on the terms of electricity trading can take place on a more equal footing. Greater balance in the accrual of benefits, increases the likelihood of being able to reach a mutually agreed upon cost-sharing/pricing arrangement. In this case, with the failure to sign the PSA, the benefit was too uncertain. The other expected benefits from the sale of power during off-peak periods by Malawi to Mozambique, and the extension of the line onward into Northern Mozambique, were also too uncertain to tangibly impact the negotiations. A more balanced accrual of benefits, e.g., from more certainty in the onward line extension, might have changed the negotiating positions, and allowed reaching a mutually agreeable sharing of costs. Timely restructuring can significantly improve implementation outcomes Experience of this project, as also confirmed by many other Bank projects in the energy and other sectors, shows that timely restructuring of the project can make a significant improvement in project implementation and delivery of results, as demonstrated by the two significant back-toback project restructurings under this project. Advance preparation of key technical studies, safeguards assessments, and major procurement packages can significantly speed up implementation As demonstrated by this project, the advance preparation of the interconnector feasibility study was a key input for the project s successful and timely technical design and its subsequent modifications. The preparation and timely disclosure of the safeguards assessments were critical in identifying areas in need of further attention (such as mine clearing) or reassuring all stakeholders of the limited scope of the project s environmental and social impact. Advanced readiness of major bidding packages for the project could help jump start early progress of procurement activities under any project. The procurement package for the largest component of the interconnector in Mozambique was in an advanced stage of readiness by project approval and was progressing fairly quickly, but was delayed for reasons discussed in earlier sections of the ICR. Adequately planning for government clearance requirements for large procurement contracts could speed up implementation Potential procurement and implementation delays can be mitigated through the ex-ante assessment of government procurement clearance requirements and procedures for large contracts and, if and when feasible, by negotiating arrangements for a fast-track review and clearance of World Bank-funded procurement packages. Such an assessment should preferably be carried out at the country portfolio level to ensure that any bottlenecks affecting the timely clearance of major contracts financed by the Bank are identified and addressed. In the absence of this, a sector-specific assessment could be conducted and a project-specific agreement on a feasible timeframe for the clearance of major procurement packages could be a part of project appraisal and formalized in the Project Manual. This could help set a benchmark that can be useful during supervision. 38

56 7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners (a) Borrower/implementing agencies The Borrower s ICR is attached in Annex 7. (b) Cofinanciers There were no co-financiers for the project. (c) Other partners and stakeholders Not applicable. 39

57 Annex 1. Project Costs and Financing (a) Project Cost by Component (in USD Million equivalent) 59 Components Appraisal Estimate (USD millions) Estimate at 2013 Restructuring (USD millions) Percentage of Appraisal Malawi A - Transmission Interconnector NA - B Capacity Building for ESCOM 2.78 NA - C Improved Transmission Infrastructure 9.10 NA - D Tete Substation Upgrades and Mobile Substation NA NA - Total Baseline Cost Physical Contingencies Price Contingencies Total Project Costs Front-end fee PPF Front-end fee IBRD Total Financing Required Mozambique A Transmission Interconnector B Capacity Building for EdM C Improved Transmission Infrastructure D Tete Substation Upgrades and Mobile Substation NA NA - Total Baseline Cost Physical Contingencies Price Contingencies Total Project Costs Front-end fee PPF Front-end fee IBRD Total Financing Required From Annex 5 of the PAD (includes government contributions) 60 This sums to US$60.4 million, however is stated as US$59.9 million in the PAD. The difference is probably due to rounding error. 40

58 (b) Project Cost by Component post-2014 restructuring Estimate at Estimate at Components restructuring restructuring (USD millions) (USD millions) Actual/Latest Estimate (USD millions) Percentage of Appraisal (at 2014 restructuring) Mozambique A Consulting Engineering Services B Capacity Building for EdM C Improved Transmission Infrastructure D Tete Substation Upgrades and Mobile Substation NA Total Baseline Cost Physical Contingencies Price Contingencies Total Project Costs Front-end fee PPF Front-end fee IBRD Total Financing Required (b) Financing Source of Funds Type of financing Appraisal Estimate (USD millions) Estimate at 2013 restructuring (USD millions) Estimate at 2014 restructuring (USD millions) Actual/Latest Estimate (USD millions) Percentage of Appraisal Malawi Borrower 11.9 NA International Development 48.0 NA Association (IDA) Total 59.9 NA Mozambique Borrower International Development Association (IDA) Total

59 Annex 2. Outputs by Component Original Component Component A: Mozambique- Malawi interconnection Component B: Capacity building and technical support for upgrade and expansion to support power trading Component C: Improved infrastructure to support electricity trading Original cost 2013 Restructuring (1) 2014 Restructuring (2) Restructured Component 39.3 Component A: Consulting engineering services 1.68 Component B: Capacity building and technical support to EdM 4.11 Component C: Reinforcement to support Matambo substation Restr. Cost Restructured Component 0.8 Component A: Consulting engineering services 1.6 Component B: Capacity building and technical support to EdM 10.0 Component C: Reinforcement to support Matambo substation Restr. Cost Final Status Remarks 0.12 Completed The interconnection component was dropped due to the failure of Malawi to sign the Credit. The component was revised during the first restructuring in 2013 to include only consultancy services. After project restructuring, this component included just a single contract for the Consulting Engineer to supervise the installation of the power transformer at Matambo substation Completed. The principal activity undertaken via this component was the ESIA and RPF for the proposed Mozambique Regional Transmission Development Project (P108934), which is now under preparation. The final report was submitted to the Government of Mozambique for approval in December Government approved in June Final disbursements were made and the contract was closed Completed Under this component, a new 130 kva transformer was commissioned on July 10, The targeted level of reduction in average outages was exceeded. However, the reduction in outages started prior to commissioning of the new transformer. The latest available indicator data suggests that outages may have declined after 42

60 commissioning of the transformer, but the data is not available for a long enough period after the transformer was commissioned to make a definitive assessment. Component D: Reinforcement of the Tete Substation and Mobile Substation 3.9 Completed Both the additional 50 MVA transformer and 25 MVA mobile substation had passed tests and were shipped to site. The transformer was delivered in June 2016 and is expected to be commissioned in August The mobile substation was delivered in July 2016 and was commissioned in December The late addition of the investments and limited resources available from project savings, meant that the project only funded the acquisition of the equipment but not the installation. 43

61 Annex 3. Economic and Financial Analysis Economic Analysis At Appraisal 1. The economic analysis at appraisal for this project assessed the benefits to both Malawi and Mozambique (evaluated over a 30 year period) as a result of the envisioned transmission interconnection funded by US$105 million in IDA financing. The economic benefits were derived from the incremental two-way flow of energy between the two countries. During times of peak demand in Malawi, energy would be imported from Mozambique and during off-peak hours, a limited amount of energy would be exported from Malawi into Mozambique. The incremental energy flow between the countries was valued at the cost of unserved energy in 2007, estimated to be USc16.0/kWh. The economic cost was taken as the value of the capital expenditure funded by IDA over the three-year implementation period and the operations and maintenance (O&M) cost, assumed to 2.5% of total investment costs annually. Based on the approach and assumptions, the NPV was estimated to be US$361 million at the economic cost of capital of 10 percent and the ERR to the project was 28 percent, (NPV of US$69 million and EIRR of 26.1% for Mozambique, and NPV of US$292 million and EIRR of 29.3% for Malawi). 2. The economic analysis done at appraisal only modeled the benefits from electricity trade from the interconnector and did not explicitly present the benefits from transmission infrastructure strengthening in Mozambique (Component C in the appraised project). As the project was eventually restructured and the interconnection component dropped, this economic analysis is thus no longer relevant. No economic or financial analysis was included in the restructuring paper either. Revised 3. The PDO of the restructured project was the reduction of outages in the Tete province. The economic analysis at completion identified that the main benefits of the project would be: (i) reduction in outages at the Matambo substation in Tete province and (ii) incremental local demand served in Tete province. Matambo Substation 4. The economic analysis at completion identified that the main benefit of this component would be reduction in outages in Tete province, as a result of a new 130 MVA transmission level transformer, valued at the opportunity cost of electricity interruptions of USc25/kWh Outage data of Matambo substation from was used to determine an average annual outage duration of 25.4 hours. The number of outages was assumed to steadily increase from 2015 at the demand growth rate. The amount of energy not served due to outages was based on the demand forecast for Tete province derived from the EdM Masterplan, discussed under the Tete substation component below. 61 Cost of power supply interruptions for the different parts of Mozambique EdM s 2014 Final Master Plan Update Report 44

62 6. Once the transmission level transformer at Matambo is commissioned, it is assumed that outages at this substation would cease for ten years and then begin to slowly increase as electricity demand growth starts to exceed the transformer capacity and equipment starts to deteriorate. On the cost side, the economic capital cost, discounted taxes from the final contract amount and totaled, US$ 11.7 million. O&M was assumed to 2.5% of the capital cost. 7. Based on the approach and assumptions, at a discount rate of 6%, the baseline NPV was estimated to be US$ 0.75 million and the ERR was 7%. It is important to note that the existing transformer at Matambo, installed in 1982, is past the typical life of a transformer, however, benefits from increasing resiliency of the grid by adding a transformer and avoiding the adverse consequences of equipment failure have not been captured here. Furthermore, this is a highly conservative estimate as it only considers the impact on outages occurring at the substation but does not account for reduced incidents of load shedding for Tete customers related to the increase in electricity supply capacity to the province. Year Cost (USD Million) Benefits Net Benefits Value of Increased Demand Economic Served Due to Reduction in Total Capital Cost O&M Cost Total Cost Opp. Cost Benefits % 0.75 EIRR 7% MARR (6%)

63 Tete Substation 8. The economic analysis at completion identified that the main benefit of this component would be the incremental local demand served in Tete province, as a result of a new 50 MVA transformer, valued at the willingness to pay (WTP) for electricity of USc15.2/kWh The benefits of the economic analysis at completion are based on the demand forecast for Tete province, estimated using actual demand data from EDM s annual reports and demand projections from EDM s Master Plan for Tete province. Demand was forecasted from at a compound annual growth rate of 11%. In addition, based on the capacity of the equipment installed and the forecasted peak demand, it was estimated that demand served due to the addition of a new transformer at Tete substation would become constrained after a certain number of years. From 2021 onwards the installed equipment would be overloaded and not be able to meet increasing peak demand (MW), however, the volume of demand met (MWh) would keep increasing, as the substation would be able to serve an increasing volume of demand that is below the threshold of the peak capacity of the substation (during off-peak hours). It is important to note that the benefit from decreased outages due to the addition of Tete transformer are not captured due to lack of data. 10. On the cost side, the economic capital cost, discounted taxes from the final contract amount and totaled, US$ 3.3 million. It is important to note that this cost also includes procurement of a mobile substation which will be used to decrease outages, as power can be routed through the mobile substation while a transformer is undergoing maintenance. O&M was assumed to be 2.5% of the capital cost, and the cost of supplying incremental demand was assumed to the average cost of electricity supplied in Mozambique in 2016, USc7.0/kWh. EDM s snapshot half year report for 2016 determines that the average cost of electricity generation is USc4.46/kWh with an additional USc2.5/kWh of transmission costs, as noted in EDM s Master Plan, to yield a delivered cost of electricity supply at the substation level of USc7.0/kWh. 11. Based on the approach and assumptions, at a discount rate of 6%, the baseline NPV was estimated to be US$ 155 million and the ERR was 97%. 62 Willingness to pay for electricity weighted by the number of customers in each category EdM s 2014 Final Master Plan Update Report 46

64 Year Cost (USD Million) Benefits Net Benefits Economic Cost of Incremental Value of Increased Demand Total Capital Cost O&M Cost Supply Total Cost WTP Benefits % EIRR 97% MARR (6%) 2018 Project Economic Analysis 12. At the project level, combining the Matambo and Tete components, yields a baseline NPV of US$ 147 million and an ERR of 36%, with the project reaching the hurdle rate, also referred to as the minimum acceptable rate of return, of 6% in This implies that the project is already economically viable, just based on the benefits delivered in the first 5-6 years of operation. These results are favorable and demonstrate that the project had a positive impact even after a significant change in the scope of the project. 47

65 Year Cost (USD Million) Benefits Net Benefits Value of Increased Economic Cost of Incremental Value of Increased Demand Served Due Demand to Reduction in Total Capital Cost O&M Cost Supply Total Cost WTP Opp. Cost Benefits % EIRR 36% MARR (6%)

66 Financial Analysis Methodology, Assumptions and Results Original 13. The financial analysis at appraisal assessed the financial performance of the utilities in Mozambique and Malawi. The financial analysis at appraisal was reconstructed based on a simpler cost- benefit approach to assess the financial performance of the project instead of the sector. Similar to the original economic analysis, the original financial analysis could not be revised at completion due to restructuring. Revised 14. The reconstructed financial analysis at completion assessed the revenues for EDM, from incremental electricity sales because of the project, at the average electricity sales price in 2016 of USc6.6/kWh. On the cost side, the capital cost was assumed to be the final contract amount for the Matambo and Tete components, US$13.4 million. O&M and cost of electricity supply were the same as in the economic analysis. 15. Based on the approach and assumptions, at a weighted average cost of capital for EdM of 5%, the baseline NPV was estimated to be negative and the FRR could not be calculated. The negative financial net benefits is due the electricity tariffs for domestic consumers being lower than the cost of electricity supply. 49

67 Annex 4. Bank Lending and Implementation Support/Supervision Processes (a) Task Team members Names Title Unit Responsibility/ Specialty Lending Wendy E. Hughes Task Team Leader AFTG1 Team Leader Dan R. Aronson Consultant (Social Development Specialist) AFRDE Social Development Sylvester Kofi Awanyo Lead Procurement Specialist OPSPF Procurement Arbi Ben Achour Consultant (Environment Safeguards Specialist) GSU11 Environment Jyoti Bisbey Infrastructure Finance Special GCPPP Infrastructure Finance Antonio L. Chamuco Sr. Procurement Specialist GGO07 Procurement Adelia Ninete Matias Chebeia Temporary (Team Assistant) GPSJB ACS Support Sunil Kumar Khosla Lead Energy Specialist GEE02 Energy Marius Koen Lead Financial Management Specialist GGO21 Financial Management Esther Angellah Lozo Executive Assistant AFMMW ACS Support Joel J. Maweni Energy Adviser GEE09 Energy Reinaldo Goncalves Mendonca Consultant (Power Engineer) GEE05 Power Engineering Robert Mills Senior Economist AFTG1 Economics Fanny Kathinka Missfeldt- Ringius Senior Energy Economist GEE06 Energy Economics Donald Herrings Mphande Lead Financial Management Specialist GGO31 Financial Management Somin Mukherji Sr Financial Analyst GEEDR Financial Analysis Edith Ruguru Mwenda Senior Counsel LEGAM Legal Diep Nguyen-Van Houtte Consultant (Operations Officer) AFCMW Quality Review Samuel A. O'Brien-Kumi Senior Energy Economist AFTG1 Energy Economics Robert A. Robelus Consultant (Environment Safeguards Specialist) GEN05 Environment Joao Tinga Financial Management Specialist GGO26 Financial Management Augustine Kudawoo Wright Temporary (Program Assistant) GEEES ACS Support Supervision/ICR Wendy Hughes Task Team Leader AFTG1 Team Leader Robert Mills Task Team Leader AFTEG Team Leader Reto Thoenen Energy Specialist AFTEG Team Member Sara Nso Consultant (Energy Specialist) AFTEG Energy Antonio L. Chamuco Senior Procurement Specialist GGO07 Procurement Salma Chande Program Assistant AFCS2 ACS Support Mtchera J. Chirwa Infrastructure Specialist AFTU1 Team Member Elvis Langa Financial Management Specialist GGO26 Financial Management Jonathan Nyamukapa Sr. Financial Management Specialist AFTME Financial Management Arbi Ben Achour Consultant (Social Development Specialist) GSU11 Social Development Reinaldo Goncalves Mendonca Consultant (Power Engineer) GEE05 Power Engineering 50

68 Robert A. Robelus Consultant (Environment Specialist) GEN05 Environment Augustine Kudawoo Wright Temporary (Program Assistant) GEEES ACS Support Maria Isabel Neto Task Team Leader GEE01 Team Leader Zayra Romo Task Team Leader GEE01 Team Leader Gulgoren A. Cansiz Consultant GEE01 Team Member Francesca Fusaro Consultant GEE01 Team Member Maria Meer Program Assistant GEEX2 ACS Support Kabir Malik Economist, ICR Lead Author GEE01 ICR Author Abdolreza Bobak Rezaian Consultant (Energy Specialist & ICR co-author) GEE01 ICR co-author Claudio Buque Energy Specialist GEE01 Team Member Allison Berg Sr. Operations Specialist GEE08 Team Member Asad Ali Ahmed Operations Analyst GEESO Team Member Raima Oyeneyin Sr. Program Assistant GEE08 Team Member (b) Staff Time and Cost Staff Time and Cost (Bank Budget Only) Stage of Project Cycle USD Thousands (including No. of staff weeks travel and consultant costs) Lending FY FY FY FY FY Total: Supervision/ICR FY FY FY FY FY FY FY FY FY FY Total:

69 Annex 5. Beneficiary Survey Results Not Applicable 52

70 Annex 6. Stakeholder Workshop Report and Results Not Applicable 53

71 Annex 7. Summary of Borrower's ICR and/or Comments on Draft ICR Southern Africa Power Market (APL2): Extension of Matambo 220kV Substation Project IMPLEMENTATION COMPLETION REPORT Submitted to the World Bank by EDM s Project Implementation Unit Introduction. 2 nd November 2016 This report concerns the Southern Africa Power Market (APL2): Extension of Matambo 220kV Substation Project. The Project Agreement between IBRD and The Government of Mozambique (GoM), through Electricidade De Moçambique EP EDM as implementing Entity, dated 21 st September 2007, for the original project and 15 th November 2013 after project amended restated financing agreement, sets out the requirements for this report. Specifically, Section II.A.2.(c) provides that EDM prepare and furnish to the Ministry of Finance and the World Bank not later than four (4) months after the Closing Date: (i) (ii) a report of such scope and in such detail as the Bank shall reasonably request, on the execution of the Project, the performance by EDM of its obligations under this Agreement and the accomplishments of the purpose of the Loan, and a plan designed to ensure the substantiality of the Project s achievements. 1. Assessment of objective, design, implementation, and operational experience. The Project s development objective was to reduce outages in the Tete province. Th project investments in transformer capacity expansion, sought to increase supplies to the Town and Province of Tete to cope with load growth as well to refurbish certain switchgear and control and protection equipment within the substation to improve operational efficiency and flexibility. 2. Assessment of outcome against the objectives. Our overall assessment of the extent to which the operation's major relevant objectives were achieved, (or are expected to be achieved) efficiently, is: Satisfactory. There were minor shortcomings in the operation s achievement of its objectives, in its efficiency, or in its relevance. This rating is based on the following criteria. 2.1 Achievement of objectives. The objectives were fully achieved as follows. Frequency of electricity outages in Tete province were reduced from 414 outages in 2006 to 151 outages in October This was an assessment of a shorter duration. In the longer term, the outages are expected to reduce to the minimum possible by summarizing it to planned shutdowns for network maintenance. The new130mva transformer capacity was added in additional to the existing 60MVA capacity transformer. This required to enable increasing of supplies to the Town and Province of Tete to cope with load growth. The new transformer capacity of 130MVA was energized on 3 rd July 2016 and loaded on 10 th July

72 Specific plant such as certain switchgear as well as Substation control and protection equipment were refurbished while New 66kV and 33kV Busbars were constructed and full SCADA system installed within the substation to improve operational flexibility, efficiency and reliability; 2.2 Efficiency. The operation has achieved, or is expected to achieve a return higher than the opportunity cost of capital, and is the least cost alternative. The initial plan was for the installation of an 80MVA transformer; however, a bigger 130MVA transformer was supplied for the same cost (except minimal additional installation cost); in this case there was an assessment of the demand for electricity made to determine the size of the transformer needed. There was also an assessment of the Tender prices for the 80MVA and 130MVA transformers. The assessment showed that the price for the 130MVA transformer was not much higher and based on the load forecasts in the Master Plan a 130MVA transformer would be required to accommodate the load growth until From the assessments made it was concluded that 130 MVA was the appropriate size for the growth of consumption that occurred in the city and province of Tete. The optical fiber in the OPGW from Matambo to Tete had breaks and had not been operational; this was repaired in the process to allow Tete to be fully controllable from Matambo; The 33kV feeder to Changara had not been operating with its own breaker; it was additionally providing independent switch bay from one of the spare, thereby improving flexibility; 2.3 Relevance. The operation s objectives, design, and implementation are fully consistent with the country s current development priorities and with current country and sectoral assistance strategies and corporate goals. The relevance of the initial regional project was aligned with Mozambique electricity strategy in order to facilitate the exchange of power between the two countries and making Mozambique to be able to sell energy to Malawi and create a redundancy in the supply of energy to the northern region of Mozambique in the event of a failure of the current north - central transmission line from the Cahora Bassa dam through the Matambo substation to the north of Mozambique. The relevance of the initial regional project for the Mozambique-Malawi interconnection project was seen as a solution to Malawi's current challenges in supplying electricity by encouraging and developing new sources and opening up the market to potential investors. The relevance of the project post restructuring to the electricity plans was to respond to the great demand of electric energy in the city and province of Tete facing the appearance of several industries due to the discovery of mineral coal mines. The relevance of the project post restructuring to the development plans was to enable the construction of medium voltage lines that could not be built before due to the limitation of the source, in this case the Matambo substation that has medium voltage outgoing feeders and also sub-transmission outgoing feeders that feed other substations. 55

73 The impact of the Project is very large in the Tete region because EDM is already making more new customer connection since this new transformer is in service. Other overhead MV distribution lines that could not be built due to limited availability are being designed and extended to more neighborhoods in the city of Tete. New district headquarters are already benefiting from the power supplied from the new transformer. The Mozambique Electricity Master Plan has an objective to facilitate implementation of power projects on a rational basis. Work was carried out in an economically efficient and environmentally sustainable manner. The design implemented included some spare bays for connection of more feeders to increase access to efficiently priced electricity to the affected communities and to be able to cater for future demand; 3. Evaluation of own performance during preparation and implementation, with special emphasis on lessons learned that may be helpful in the future. Our assessment of the extent to which the Government and implementing agencies ensured quality of preparation and implementation, and complied with covenants and agreements, toward the achievement of development outcomes, is as follows. 3.1 For the Government overall: Highly Satisfactory. There were no shortcomings in Government performance. There was commitment at Government level for providing enabling environment including supportive macro, sectorial, and institutional policies (legislation, regulatory and pricing reforms, etc.). The government played a very important role in the preparation phase, such as the mobilization of local governments for public consultation regarding project implementation issues, issuance of project environmental impact licenses, among other activities. In the preparation phase the government also provided a matrix of the strategic energy plan and the overall plan for the energy sector. In the implementation phase it also played an important role in the issuance of entry visas to the foreign technicians of the Consultant as well as the Contractor. Issuance of DIRE for foreign resident technicians of both firms. In the restructuring of the regional project after Malawi withdrew. 3.2 For implementing agency Satisfactory. There were minor shortcomings such as getting the clearance on the duty exemption for the materials/equipment s imported under the project due to low performance of others stakeholders involved in the implementing agency s performance. Throughout the implementation of the project, EDM exhibited; Ownership and commitment to achieving development objectives (timely meetings); 56

74 Adequacy of beneficiary/stakeholder consultations and involvement (timely meetings with HCB, VALE, ICVL, and other stakeholders); Readiness for implementation, implementation arrangements and capacity, and appointment of key staff; Timely resolution of implementation issues by resolving issues related to provision of information and coordination of shut downs; Fiduciary (financial management, governance, provision of counterpart funding, procurement, reimbursements, compliance with covenants); Adequacy of monitoring and evaluation arrangements, including the utilization of M&E data in decision-making and resource allocation; Relationships and coordination with donors/ partners/stakeholders; Adequacy of transition arrangements for regular operation of supported activities after Loan/Credit closing. 3.4 Lessons learned The need for carrying more detailed studies during project initiation to avoid scope creep and emergence of numerous variations. During the project numerous outages were planned for installation of substation equipment. These outages were not in line with the planned substation outages for the year. This caused some delays in implementation as the Contractor had to wait for the dates of the planned outages to continue with the Works Including appropriate clauses in the EPC contract to be able to deal with issues of delays in schedule by Contractor. There were significant delays in the installation, caused by a fire in the Contractor s warehouse. This lead to equipment damages, this equipment had to be re-manufactured as the Insurer s cost. EDM had to pay and incur the delays related to additional Factory Acceptance Tests and delivery of equipment. 4. Evaluation of the performance of the World Bank, including the effectiveness of their relationship, with special emphasis on lessons learned. 4.1 World Bank: Quality at entry. Our assessment of the extent to which services provided by the Bank ensured quality at entry of the operation is as follows. Highly Satisfactory. There were no shortcomings in identification, preparation, or appraisal. The World Bank team was always ready to help implementation agency both with its technicians based in Mozambique as well as from abroad. It has always been helpful in overcoming a delicate subject matter of the Procurement project, clarifications regarding addenda and variations orders submitted by the Consultant or Contractor. Always warned about the financial status of the project. The World Bank local team made possible EDM's financial department technicians be linked to the Project, financial status via Client Connection system so that the technicians could best follow the financial health of the Project. 57

75 The World Bank team traveled to the site on service missions and warned about the risks the project was exposed. These risks included delays in 130MVA transformer delivery, due to additional logistic requirements in transporting equipment from Richards Bay Harbour in South Africa, through Zimbabwe and into Mozambique. There was a long waiting period at Richards Bay, caused by the Xenophobic crisis, because police escorts were not available to escort the transformer from South Africa to Zimbabwe border. 4.2 World Bank: Implementation. Our assessment of the extent to which the Bank supported effective implementation through appropriate supervision (including ensuring adequate transition arrangements for regular operation of supported activities after loan/credit closing) is as follows. Highly Satisfactory. There were no shortcomings in the proactive identification of opportunities and resolution of threats. 5. Description of proposed arrangements for future operation. Plan to ensure the sustainability of the Project s achievement. 5.1 Future arrangements. Arrangements to achieve or maintain the same development outcome include: - The infrastructure and equipment installed during the TUP project will meet the electricity demand in Tete province for the next 10 years. The Matambo substation extension is sufficient to maintain the development outcomes until In addition to the new equipment installation, maintenance of the equipment in accordance to international best practices will be done in order to maintain the network operational and prevent unplanned outages. - To maintain the objectives beyond the foreseeable 10 years, the Utility will have to regularly update their development plans for the province and monitor the demand growth to ensure that unforeseen growths are addressed before they reach the capacity of the existing equipment. 5.2 Risk to development outcome. Our assessment of the risk, at present, that development outcomes (or expected outcomes) will not be maintained (or realized) is: Negligible to Low; Changes that may occur that would be detrimental to the ultimate achievement of the operation s development outcome include: - Rapid demand growth due to construction activity in large infrastructure projects; - Inability of existing transmission lines to support the forecasted load growth. The likelihood is negligible to low that some of these changes may occur. The equipment sizing was done based on the recommendations from a recent transmission and distribution master plan, which covered the domestic, commercial and industrial load growth in the province up to Other factors, such as equipment failures pose a low risk, since the extension was done using a new transformer and auxiliary equipment, procured through an international competitive process. Transmission infrastructure constraints may be experienced if the load grows beyond their capacity, this infrastructure was not upgraded at this stage, EDM will have to monitor the load growth in future to ensure that the capacity of the lines do not hinder the performance of the network. The impact on the operation s development outcomes of some or all of these changes materializing is significant. The Matambo substation is essential for the supply of electricity to Tete City and the surrounding industrial areas. If these negative changes materialize the network will be constrained and the quality of supply will be reduced. High load growths and insufficient transmission capacity in the region can cause outages which negatively impact the population and businesses in Tete province. 58

76 6. Results Project Development Objective Indicators PHINDPDOTBL Average interruption frequency per year in the project area (Number, Core) Baseline Actual (Previous) Actual (Current) End Target Value Date 17-Jul Mar Aug Aug-2016 PHINDPDOTBL Customers served in the project area (Number, Core Supplement) Baseline Actual (Previous) Actual (Current) End Target Value PHINDPDOTBL Direct project beneficiaries (Number, Core) Baseline Actual (Previous) Actual (Current) End Target Value Date 01-Nov Dec Aug Aug-2016 Comments PHINDPDOTBL Female beneficiaries (Percentage, Core Supplement) Baseline Actual (Previous) Actual (Current) End Target Value Overall Comments 59

77 PHINDIR ITBL PHINDIR ITBL PHINDIR ITBL PHINDIR ITBL Intermediate Results Indicators ESIA final report completed and approved by GoM (Text, Custom) Baseline Actual (Previous) Actual (Current) End Target Value No ESIA -- ESIA final report completed and approved by GoM ESIA final report completed and approved by GoM Date 17-Jul Nov May Jun-2016 New 220/66/33 kv transformer installed and commissioned at Matambo substation (Text, Custom) Baseline Actual (Previous) Actual (Current) End Target Value No transformer Transformer on site fully and fully assembled 130 MVA Transformer delivered on site and commissioned 80 MVA of new transformation capacity commissioned Date 01-Jan Dec Aug Aug-2016 Level of charge of the current Matambo transformer (Percentage, Custom) Baseline Actual (Previous) Actual (Current) End Target Value Date 17-Jul Nov Aug Aug-2016 New 220/33 kv, 25 MVA mobile substation delivered (Text, Custom) Baseline Actual (Previous) Actual (Current) End Target Value No transformer Transformer undergoing tests Transformer was delivered and ready for use 50 MVA of new transformation capacity delivered Date 17-Jul Dec Aug Aug

78 Annex 8. Comments of Cofinanciers and Other Partners/Stakeholders None 61

79 Annex 9. List of Supporting Documents 1. Project Appraisal Document: Southern African Power Market Program (SAPMP) APL-1, Sept Project Appraisal Document: Southern African Power Market Program (SAPMP) APL-2, June Project Appraisal Documents: Costal Transmission Backbone Project of the West African Power Pool (WAPP) APL Program, June Electricity de Mozambique (EdM). Annual Statistical Report, Malawi Growth and Development Strategy Malawi Growth and Development Strategy Malawi National Electricity Policy (NEP, 2003) 8. Mozambique Poverty Reduction Support Strategy (PARPA II) 9. Mozambique National Development Strategy 2014 (ENDE) 10. Malawi CAS: Malawi CAS: Mozambique CPS: Mozambique CPS: SAPP Annual Report, Quarterly Progress Report, Extension of Matambo 220KV substation project, April June Quarterly Progress Report, Extension of Matambo 220KV substation project, December Norconsult, Final Master Plan Update Report, Master Plan Update Project, , Technical Assistance for Electricity de Mozambique (EdM). 18. Dikelman, Taryn, The Effects of Rural Electrification on Employment: New Evidence from South Africa, American Economic Review, Vol. 101, No. 7, (pp ) 19. Karen Fisher-Vanden, Erin T. Mansur, Qiong (Juliana) Wang, Electricity shortages and firm productivity: Evidence from China's industrial firms, Journal of Development Economics, Volume 114, May 2015, Pages ICR Guidelines: August 2006, (updated 2011). 21. Project Restructuring Papers, 2013, 2014, 2015, and ISR Seq1-Seq16: Mozambique Transmission Upgrade Project. 23. Other Project Documentation: Financing Agreements, Aide Mémoires, Office memos, Official Correspondences. 24. Vanheukelom. Jan and Talitha Bertelsmann-Scott, The Political Economy of Regional Integration in Africa, The Southern African Development Community (SADC). 25. World Bank, Background Paper for the Guidance Note on Energy Project Appraisal Documents: Literature Review 62

80 Annex 10. Map IBRD