Document of The World Bank FOR OFFICIAL USE ONLY Report No: ICR 147538-GH IMPLEMENTATION COMPLETION AND RESULTS REPORT ON AN IDA GUARANTEE IN THE AMOUNT OF US$500 MILLION AND AN IBRD ENCLAVE GUARANTEE IN THE AMOUNT OF US$200 MILLION FOR THE REPUBLIC OF GHANA IN SUPPORT OF THE SANKOFA GAS PROJECT June 26, 2020 Energy Global Practice Africa Region This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. CURRENCY EQUIVALENTS (Exchange Rate Effective {December 31, 2019}) Currency Unit = Ghana Cedis GHS 5.65 = US$1 US$1.0 = SDR 1 FISCAL YEAR July 1 - June 30 Regional Vice President: Hafez Ghanem Country Director: Pierre Laporte Senior Global Practice Director: Riccardo Puliti Practice Managers: Ashish Khanna/ Sebnem Erol Madan Task Team Leaders: Shinya Nishimura/Jukka-Pekka Strand ICR Task Team Leaders: David Vilar Ferrenbach/Zhengjia Meng ICR Main Contributor: Eugene D. McCarthy ii ABBREVIATIONS AND ACRONYMS CPS Country Partnership Strategy DPOs Development Policy Operations ECG Electricity Company of Ghana ERR Economic Rate of Return ESHIA Environmental, Social and Health Impact Assessment FDI Foreign Direct Investment FIRR Financial Internal Rate of Return FPSO Floating Production Storage and Offloading GDP Gross Domestic Product GEDAP Ghana Energy Development and Access Project GNGC Ghana National Gas Company GNPC Ghana National Petroleum Corporation GSDA Ghana’s Shared Development Agenda GoG Government of Ghana GRIDCo Ghana Grid Company GSA Gas Supply Agreement HSE Health, Safety, Environment IBRD International Bank for Reconstruction and Development IDA International Development Association IESC Independent Environmental and Social Consultant IFC International Finance Corporation IMF International Monetary Fund IPP Independent Power Project IRR Internal Rate of Return LCO Light Crude Oil LRP Livelihood Restoration Plan MCC Millennium Challenge Corporation MIGA Multilateral Investment Guarantee Agency MW Megawatt NEDCO Northern Electricity Distribution Company NGO Non-Governmental Organization NPV Net Present Value OCTP Offshore Cape Three Points ORF Onshore Receiving Facilities PAD Project Appraisal Document PAP Project Affected People PDO Project Development Objective TCF Trillion cubic feet VRA Volta River Authority WAGP West African Gas Pipeline WAPCo West African Gas Pipeline Company WBG World Bank Group iii TABLE OF CONTENTS DATA SHEET .......................................................................................................................... V I. PROJECT CONTEXT AND DEVELOPMENT OBJECTIVES ....................................................... 1 A. CONTEXT AT APPRAISAL .........................................................................................................1 B. SIGNIFICANT CHANGES DURING IMPLEMENTATION (IF APPLICABLE) .......................................4 II. OUTCOME ...................................................................................................................... 5 A. RELEVANCE OF PDOs ..............................................................................................................5 B. ACHIEVEMENT OF PDOs (EFFICACY) ........................................................................................5 C. EFFICIENCY .............................................................................................................................7 D. JUSTIFICATION OF OVERALL OUTCOME RATING ......................................................................8 E. OTHER OUTCOMES AND IMPACTS (IF ANY) ..............................................................................8 III. KEY FACTORS THAT AFFECTED IMPLEMENTATION AND OUTCOME ................................ 11 A. KEY FACTORS DURING PREPARATION ................................................................................... 11 B. KEY FACTORS DURING IMPLEMENTATION ............................................................................. 12 IV. BANK PERFORMANCE, COMPLIANCE ISSUES, AND RISK TO DEVELOPMENT OUTCOME .. 14 A. QUALITY OF MONITORING AND EVALUATION (M&E) ............................................................ 14 B. ENVIRONMENTAL, SOCIAL, AND FIDUCIARY COMPLIANCE ..................................................... 15 C. BANK PERFORMANCE ........................................................................................................... 17 D. RISK TO DEVELOPMENT OUTCOME ....................................................................................... 19 V. LESSONS AND RECOMMENDATIONS ............................................................................. 20 ANNEX 1. RESULTS FRAMEWORK AND KEY OUTPUTS ........................................................... 22 ANNEX 2. BANK LENDING AND IMPLEMENTATION SUPPORT/SUPERVISION ......................... 26 ANNEX 3. PROJECT COST BY COMPONENT ........................................................................... 29 ANNEX 4. EFFICIENCY ANALYSIS ........................................................................................... 30 ANNEX 5. BORROWER, CO-FINANCIER AND OTHER PARTNER/STAKEHOLDER COMMENTS ... 36 ANNEX 6. SUPPORTING DOCUMENTS (IF ANY) ..................................................................... 42 iv The World Bank Ghana Sankofa Gas Project (P152670) DATA SHEET BASIC INFORMATION Product Information Project ID: Project Name: Ghana Sankofa Gas Project P152670 Country Financing Instrument REPUBLIC OF GHANA GU - Guarantee Original EA Category Revised EA Category A - Full Assessment A - Full Assessment Organizations Borrower Implementing Agency GOVERNMENT OF GHANA Project Development Objective (PDO) Original PDO The Project Development Objective (PDO) is to increase the availability of natural gas for clean power generation by leveraging private capital investment. FINANCING Original Amount (US$mn) Revised Amount (US$mn) Actual Disbursed (US$mn) IDA Guarantee 500 500 NA IBRD Enclave Guarantee 200 200 NA Non-World Bank Financing 5,1 billion 4,65 billion 4,65 billion 0 0 0 KEY DATES FIN_TABLE_DAT A Approval Effectiveness MTR Review Original Closing Actual Closing 07/30/2015 09/30/2016 N.A. 12/31/2018 12/31/2018 v The World Bank Ghana Sankofa Gas Project (P152670) RESTRUCTURING AND/OR ADDITIONAL FINANCING Date(s) Amount Disbursed (US$M) Key Revisions KEY RATINGS Outcome Bank Performance M&E Quality MS MS Substantial RATINGS OF PROJECT PERFORMANCE IN ISRs No. Date ISR Archived DO Rating IP Rating Actual Disbursements (US$M) 4 6/19/2018 MS MS N.A. 3 8/3/2017 S S N.A. 2 6/26/2017 S S N.A. 1 7/15/2016 S S N.A. SECTORS AND THEMES Sectors Major Sector/Sector (%) Natural Gas Development and Power Generation 100 Themes Major Theme/ Theme (Level 2)/ Theme (Level 3) (%) Private Sector Development 100 ADM STAFF Role At Approval At ICR Regional Vice President: Makhtar Diop Hafez Ghanem Country Director: Henry Kerali Pierre Laporte Director: Anita Marangoly George Riccardo Puliti Practice Manager(s): Sameer Shukla/Pankaj Gupta Ashish Khanna/Sebnem Erol Madan Robert Schlotterer/Sunil Jukka-Pekka Strand Task Team Leader(s): Mathrani Shinya Nishimura David Vilar Ferrenbach ICR Task Team Leader(s): Zhengjia Meng ICR Contributing Author: Eugene D. McCarthy vi The World Bank Ghana Sankofa Gas Project (P152670) I. PROJECT CONTEXT AND DEVELOPMENT OBJECTIVES A. CONTEXT AT APPRAISAL Country Context 1. Ghana returned to multi-party democracy in January, 1993. Since then, successive, peaceful transitions of democratic power provided a supportive environment for continued economic growth. Over the two-decade period from 1995-2015, Ghana’s economy experienced sustained and broadly inclusive growth during which the country progressed steadily to lower middle-income status. Its economy also outperformed most developing Sub-Saharan countries by a large margin. In addition, economic growth was accompanied by significant improvements in poverty and social indicators, with poverty rates (US$ 1.90 per day) falling from 47.4 percent in 1991 to 12.1 percent in 2012. 2. Following this period of sustained growth, the country’s economic performance after 2012 weakened considerably. Expansionary fiscal policies led to large and protracted external imbalances while Ghana’s net international reserves fell sharply. The fuel import bill also rose rapidly, due, in part, to unreliable gas imports from Nigeria, which exerted pressures in Ghana for a secure, domestic source of fuel. In this worsening economic climate, the government embarked on a program to restore fiscal discipline and, in April, 2015, just prior to Board presentation, reached a three-year agreement with the IMF to provide an Extended Credit Facility, aimed at strengthening the country’s overall economic performance. Sector Context 3. Ghana’s energy sector has always been closely linked to the country’s macroeconomic performance- through the real sector, fiscal accounts, and the balance of payments. Studies had estimated the fiscal shock of electricity shortages as well as declines in international oil and gold prices as high as 2 percentage points of GDP. Consequently, the availability of domestic oil production as well as natural gas supplies for power generation had the potential to help mitigate such external price shocks. 4. Historically, Ghana’s energy needs have been met through the development of its hydro resources for electricity generation1, with imported crude oil and petroleum products increasingly meeting the country’s transportation and thermal power needs since the 1990s. Prior to 2007, Ghana had no domestic oil or natural gas production. As a consequence, the country’s economic performance was vulnerable to sharp, unpredictable swings in international oil prices. However, in 2007, following the discovery of a number of commercially viable oil and natural gas deposits in nearby offshore waters, the country was provided with the prospect of meeting a significant share of its growing energy needs through lower cost, oil and natural gas supplied domestically while, at the same time, enhancing Ghana’s energy security. One of these discoveries, the Sankofa oil and gas field, was expected to produce oil in the order of 30,000-45,000 b/d on average per day over the 2017-2022 period, equivalent to approximately US$5 billion at current World Bank Group oil price projections. In addition, non-associated gas production from Sankofa was expected to start in the first half of 2018, continue for fifteen years at average production levels of 170mmscfd, and displace higher cost liquid fuels being imported. As a result, the impact of the Sankofa development on the balance of payments had the potential to reduce pressure on local currency and enhance debt sustainability, given the composition of public debt. 1 The Bank financed a series of hydropower projects, starting during the 1960s, which enabled the construction of the Akosombo hydropower station (over 1,038 MW) in the Upper Volta region. 1 The World Bank Ghana Sankofa Gas Project (P152670) 5. The regional energy context at the time of appraisal provided additional impetus to develop Ghana’s recently discovered gas potential. Natural gas imported from Nigeria through the West Africa Gas Pipeline (WAGP) had been intended to provide a reliable and low-cost source of energy for Ghana’s growing power requirements. However, since 2010 when gas flows through the WAGP began, the volume of gas imports was much lower than contracted and daily supply had been erratic2, which had forced Ghana to purchase higher cost, liquid fuels in the spot market to keep its thermal plants running. Frequent black outs also became a common occurrence, which exerted pressure on the Government of Ghana (GoG) to find solutions to these recurring problems. Consequently, the security of supply of key fuels for thermal power generation while cushioning their cost became an increasingly important consideration in Ghana’s energy supply strategy. It was in this context that the recently discovered offshore gas reserves-both associated gas from the Jubilee and TEN fields and non-associated gas from the Sankofa development-offered the prospect of reducing Ghana’s growing dependence on higher cost fuel imports. Rationale for the World Bank Support and contribution to higher level objectives 6. The project supported the development objectives set out in Ghana’s second Shared Growth and Development Agenda (GSGDA II) for the period 2014-17. The GSGDA II emphasized the need for structural economic transformation to accelerate poverty reduction and achieve a key objective of ‘a stable, united, inclusive and prosperous country with opportunities for all’. Specifically, the project supported three of the seven thematic areas of the GSGDA II- namely, sustaining macroeconomic stability, enhancing the competitiveness of Ghana’s private sector, and developing the recently discovered non-associated, natural gas reserves. 7. The project was also aligned with the World Bank Group Country Partnership Strategy for the period FY13-FY183. Specifically, it contributed to Pillar 2: “Improving Competitiveness and Job Creation”, by helping provide more efficient delivery of infrastructure services in the energy sector. The project also supported the WBG twin goals -namely, ending extreme poverty and boosting shared prosperity by 2030. Finally, the project was in line with the WBG 2013 Energy Directions Paper, which planned for a scale up of WBG involvement by supporting natural gas development as a lower cost and cleaner fuel. 8. The Bank guarantees deployed in this project were one of several WBG instruments used to support the further development of the energy sector in Ghana4. In earlier and ongoing operations, the Bank was already financing investments in distribution reinforcement for the power distribution companies, ECG (Electricity Company of Ghana) and NEDCO. IDA had financed high voltage (HV) grid investments undertaken by GRIDCo, while IFC had invested in thermal power generation. Also, IFC Advisory Services were being used to provide support to GoG on a planned concession for ECG; in addition, a US$60 million IDA operation was planned to improve ECG’s financial performance and operational efficiency. A key goal of WBG support to the sector was to improve the quality of energy services for end-users, increase access to electricity, and restore investor confidence in Ghana’s energy sector. Finally, an IMF Extended Credit Facility was approved in April 2 The ‘contracted’ daily gas supply through WAGP was 133 mmsfcd, of which 123 mmscfd was supplied to Ghana. However, supplies in 2010, 2011, 2012, and 2013 were much reduced and averaged 40, 84, 49, and 38 mmscfd respectively, including days of no gas supply at all; 3 The original CPS (FY 13-16) was Report No. 76369-GH; this CPS was later extended from FY13-18 through a Performance and Learning Review (PLR): Report No. 105606-GH. 4 The World Bank has had a long-standing involvement in Ghana’s energy sector, dating back to the 1960s when it supported the development of the Ghana’s hydro resources in the Upper Volta region. These operations were followed by a series of electric power operations supporting investment in thermal generation, transmission lines, and distribution. 2 The World Bank Ghana Sankofa Gas Project (P152670) 2015 while the first in a series of three planned IDA DPFs was approved in June 2015. Both these operations were intended to reinforce fiscal discipline and provide budget space for the GoG to settle its arrears for electricity consumption. In addition, the first IDA DPF5 had close links with the Sankofa gas project in supporting institutional reform of the main gas institution, GNPC, and helping facilitate its access to long term capital markets. The project’s instruments also included an IBRD Enclave Guarantee in combination with an IDA Payment Guarantee. The IBRD Enclave Guarantee was intended to benefit the project’s private sponsors and allow more headroom for the use of IDA resources. The additional IDA exposure was made available to provide possible support for the development of a number of IPPs in Ghana, which were planned to be the main consumers of natural gas from the Sankofa project but it was dropped. Theory of Change (Results Chain) 9. The key project objective was the development of Ghana’s non-associated gas resources in the offshore Sankofa field for use as a lower cost, domestic fuel in thermal power generation6. The two main activities required to achieve this objective were the following: (i) physical investment: the drilling of wells, laying of a subsea pipeline, and pipeline connections to onshore gas consumers; (ii) mobilization of financing: from oil companies and commercial banks through the use of WBG guarantee instruments and political insurance. These two activities were intended to reduce substantially the need for more expensive liquid fuel imports as well as natural gas from Nigeria, which was not supplying the contracted volumes and supplies were erratic at the time of appraisal7. In addition, the demonstration effect of this first development of the country’s non-associated gas resources was expected to (i) spur further exploration in Ghana’s offshore waters; and (ii) reduce the need for scarce public funds for the development of the country’s energy sector by mobilizing large amounts of financing from private investors and commercial banks. 10. In terms of the project’s longer-term benefits, the proposed investment was expected to (i) improve the reliability of Ghana’s electricity supply to industrial and residential consumers by having a secure, domestic source of gas supply;(ii) provide electricity to consumers at lower cost which, in turn, would drive 5 First DPF, MSCG1, Macroeconomic Stability for Competitiveness and Growth, approved June 2015 6 Alternative fuel supply options such as domestic Associated Gas fields and gas imported through the WAGP, as well as imported liquid fuels and LNG where analyzed at appraisal. Associated gas from the nearby TEN field were also being developed but volumes were small and could not be used for baseload generation (PAD para 28) and WAGP supply was unreliable compromising firm offtake as baseload generation (PAD para 15). The alternatives to firm fuel supply were LNG and liquid fuels imports and these were more expensive options than Sankofa gas at Appraisal. 7 PAD para 15: ‘WAGP has experienced severe supply shortages in Nigeria and interruptions in deliveries have compromised its contractual ability to supply Ghana with 120 mmscfd of firm gas’. 3 The World Bank Ghana Sankofa Gas Project (P152670) further economic growth; (iii) mobilize private sources of finance for the further development of the country’s energy sector; and (iv) lead to climate change benefits by replacing the need for LCO and other petroleum fuels currently being used in thermal power generation. 11. The critical assumptions underlying the Theory of Change and progress from ‘Activities’ to ‘Longer Term Outcomes’ were the following: (i) a gas pipeline infrastructure link between the western and eastern regions was completed; (ii) the financial crisis at the time in the country’s electricity sector, which was to be the main consumer of natural gas, was resolved, (iii)power sector improvement including ECG’s operational and financial performance and, IPPs being procured competitively and based on the least-cost power expansion plan; and (iv) political stability and government creditworthiness was sustained. Project Development Objectives (PDOs) 12. The Project’s Development Objective (PDO), as stated in the Project Appraisal Document (PAD), is ‘to increase the availability of natural gas for clean power generation by leveraging private capital investment’. Key Expected Outcomes and Outcome Indicators 13. The PDO indicators were the following: (i) volume of natural gas supplied to power plants; (ii) private sector capital mobilized (US$ million); and (iii) number of beneficiaries, including female beneficiaries. By providing additional, long term and lower cost gas supplies to Ghana’s thermal power plants, the project was expected to provide indirect benefits to all customers of the Electricity Company of Ghana (ECG). As a result, the number of customers who would benefit was based on the total number of ECG customers, increasing at 2 percent per year. No outcome indicators were specified for the longer-term benefits of incremental gas supplies to the network due to the difficulties of measurement and attribution. Components 14. The project comprised the use of World Bank Group guarantees, together with a MIGA political guarantee and IFC financing, to support the development of a major, offshore, natural gas discovery in Ghana. While the IDA Guarantee and the IBRD Enclave Guarantee covered both project sponsors i.e. ENI-Vitol, IFC financing and the MIGA Guarantee were for the exclusive benefit of Vitol8. 15. The main project components were (i) proven, recoverable natural gas reserves amounting to nearly 1 TCF (equivalent to an average daily gas production of 180mmscfd over 14 years);(ii) drilling of five gas producing wells linked to a floating production and storage unit on an oil tanker; (iii) a 63 km, 22-inch subsea pipeline linked to an onshore receiving facility (ORF); and (iv) a short, onshore connection to an existing GNGC gas pipeline bringing associated gas from the Jubilee and Ten fields to power generation plants at Aboadze. B. SIGNIFICANT CHANGES DURING IMPLEMENTATION (IF APPLICABLE) Revised PDOs and Outcome Targets 16. The PDO was not revised during implementation. Revised PDO Indicators 17. None of the PDO indicators were revised or modified during implementation. 8The Sankofa gas field is being developed by Eni Exploration and Production Ltd (“Eni Ghana”) a subsidiary of Eni S.p.A of Italy (“Eni”) and Vitol Group of the Netherlands (“Vitol”) (together “Private Sponsors”). 4 The World Bank Ghana Sankofa Gas Project (P152670) Revised Components 18. There were no revisions to the project components as described in the PAD. Other Changes 19. The IBRD Enclave Guarantee was withdrawn in September 2018 after the Sankofa project sponsors had successfully mobilized the needed project financing without recourse to this enclave guarantee. 20. Extended Monitoring Period prior to Preparation of ICR: The Project was closed on December 31, 2018. The first gas supplies from the Sankofa field to thermal power plants started in June 2018, close to the forecast date at appraisal9, though in smaller volumes, due to the limited size of the power market in the western region for Sankofa gas, together with delays in the pipeline interconnector link needed to transport Sankofa gas to the eastern region and which was completed in April 2019.. Consequently, this ICR has been prepared in early 2020 to document more fully the impact of the project. However, in line with standard practice for project investment operations and guarantees, the ratings are based on events that took place during the implementation period until closing. To benefit further from the 12-month period following project closure, the ICR also takes into account issues during the post-closure period, which are reflected in the “Performance since Closing date and Prospects” section, Section D (Risk to Development Outcome), and Section V (Lessons and Recommendations). II. OUTCOME A. RELEVANCE OF PDOs Assessment of Relevance of PDOs and Rating Rating: High 21. The PDO continued to be closely aligned with the FY13-18 CPS for Ghana at the time when the project was closed. First, the CPS stated that future growth prospects would be closely linked to petroleum sector developments, both in terms of oil and natural gas production as well as its impact on public expenditure. The CPS placed particular emphasis on the potential fiscal impact of recent oil and gas discoveries in broadening the country’s sources of growth through increased public investment. Second, a key structural pillar of the CPS was ‘increasing competitiveness and job creation’, which entailed, inter alia, improving the ‘quality, availability, and reliability of electricity services in the country’. The prospect of developing domestic sources of natural gas in Ghana for the first time to meet growing demand for electricity, at lower cost than LCO liquids being imported at that time, provided a special opportunity to strengthen the performance of the electricity sector. Third, strengthening regional energy integration in six West African countries, including Ghana, was also identified as a priority activity of the CPS. Ghana’s growing gas supply potential could lead to the export of competitively priced electricity for export to other West African countries via the existing West Africa Power Pool (WAPP). Finally, the ‘demonstration’ impact of a successful development of natural gas reserves for a domestic market would provide additional incentives for exploration by international oil companies in Ghana’s offshore waters. 22. Based on the above considerations, the relevance of the PDO is considered ‘High”. B. ACHIEVEMENT OF PDOs (EFFICACY) Rating: Substantial 23. The assessment of achievement of the project development objectives and outcomes at the project closing 9 February 2018 5 The World Bank Ghana Sankofa Gas Project (P152670) date is presented below, based on progress made during implementation towards the target values of the project indicators. 24. The key PDO indicators are (i) the quantity of natural gas supplied to power plants; (ii) private sector capital mobilized (in US$ million); (iii) the number of indirect beneficiaries; and (iv) percentage of female beneficiaries. 25. PDO Indicator 1: Quantity of Gas supplied to Power Plants10 Gas supply from the Sankofa gas field was available for delivery to onshore power plants in June 2018, only four months behind the original schedule. However, the first gas supplies did not occur until August 2018, and then in much smaller volumes, for the following reasons: (i) the market for gas in the western region of the country close to landfall was limited by two factors: (a) gas demand in the western region (Takoradi) was not sufficient to offtake the contracted volume of Sankofa gas; and (b)associated gas supplies being produced from the nearby Jubilee and TEN fields, and averaging 50 mmscfd in the 2nd half of 2019, were given first preference11; and (c) without the availability of a pipeline interconnection linking gas supplied from Sankofa to the eastern region of the country12, the size of the market was limited to Takoradi. Completion of this gas pipeline interconnection was delayed through mid CY 2019. Consequently, actual gas supplies in 2018 and 2019 were well below the forecast volumes of 158 and 171 mmscf/d for these two years. Nevertheless, based on steadily increasing gas supply volumes in 2019 and early 2020, which exceeded 200 mmscf/d in December 2019 and January 2020 for several days, they are forecasted to reach, on average, 180 mmscf/d in 2021 and in subsequent years13. Overall, the original target of gas supplies remains likely to be achieved although two years behind the original forecast. Rating: Partially achieved. 26. PDO Indicator 2: Private Sector Capital Mobilized The cumulative amount of private capital mobilized was US$4,653 million compared with the target of US$5,100 million estimated at appraisal. It includes US$1.35 billion mobilized from commercial banks and other sources with the remainder financed by equity. The lower cumulative amount reflects cost savings realized during the construction phase of the offshore development. Rating: Achieved 27. PDO Indicator 3: Indirect Project Beneficiaries and PDO Indicator 4: Female Beneficiaries. These two indicators were measured by the number of project beneficiaries, including female beneficiaries, in the client base of the power system. The proportion has not varied significantly recently; nor is it anticipated to change in the medium range. That being said, the choice of these two indicators is better used in electricity distribution investments rather than in a natural gas supply project for thermal power generation where the direct impact of the investment on electricity consumers/beneficiaries is difficult to measure. Rating: Achieved 10 Project Indicator 1 is not fully aligned with the PDO, which refers to ‘increasing the availability of gas supply’, rather than increasing the ‘quantity of gas supplied to power plants’. Gas from Sankofa became ‘available’ for supply in the annual forecast volumes, starting in June 2018 (although only in August 2018 started to be offtaken by GNPC), and currently (i.e. June 2020) can make available up to 220mmscfd if the gas market in Ghana is able to absorb this volume. 11 ‘Jubilee gas’ is associated gas, produced jointly with oil from the Jubilee field and supplied at zero upstream cost during the first years of oil production 12 The onshore gas pipeline interconnector served to reverse gas flow through the existing WAGP to the eastern region of Ghana, where there was a waiting market for the gas supplied from the offshore Sankofa field; 13 March 2020 average offtake was 180 mmscf/d 6 The World Bank Ghana Sankofa Gas Project (P152670) TABLE 1. Assessment of Achievement of Each Objective/Outcome at Project Closing Date Actual Values Original Target Values Formally Revised Target Baseline Value Achieved at Completion or (from approval documents) Values Target Years PDO Indicator 1: Quantity of gas supplied to power plants (mmscf per day) Value 0 2017: 0 n.a. 2017: 0 (quantitative or 2018: 158 2018: 41 qualitative) 2019: 171 2019: 90 2020: 171 2020: 171 (est) 2021: 171 2021: 180 (est) 2022: 171 2022: 180 (est) Comments Achievement of targets: 2018 = 6 percent; 2019 = 54 percent; PDO Indicator 2: Private Sector Capital Mobilized (US$ million) Value 0 2014 800 n.a. 2014 (quantitative or 2015 1,400 2015 qualitative) 2016 2,600 2016 1,600 2017 3,900 2017 3,282 2018 5,100* 2018 4,653 Comments *Figure in PAD; ISRs show US$ 5,000 million PDO Indicator 3: Indirect Project Beneficiaries (million) Value 7.2 2015: 10.6 n.a. 2015: 15.2 (quantitative or 2016: 10.8 2016: 16.5 qualitative) 2017: 11.0 2017: 17.1 2018: 11.3 2018: 17.8 2019: 11.5 2019: NA Comments The increase in the number of indirect beneficiaries is dependent on the number of new connections, which has progressed faster than expected. PDO Indicator 4: Female beneficiaries (percent) Value 49 2015: 50 n.a. 2015: 49.3 (quantitative or 2016: 50 2016: 49.3 qualitative) 2017: 50 2017: 49.3 2018: 50 2018: 49.3 2019: 50 2019 ---- Comments The indicator is based on the proportion of female beneficiaries in the client basis of the power system. The proportion has not varied significantly recently; nor is it anticipated to change in the medium range. The achievement is 100 percent. C. EFFICIENCY Assessment of Efficiency Rating: Substantial 28. Economic Analysis The assessment of efficiency includes a comparison of the economic analysis of the Project at appraisal and at closing. The economic benefits were calculated using the following methodology: (i) the difference between the 'with project' economic cost of Sankofa gas and the 'without project' economic costs, using imported fuel reference prices as follows:(a) 2018-2021: imported light crude oil; and (b) 2022 onwards: imported LNG (including the cost of regasification); and (ii) revenues from oil and condensate sales. In 2018 and 2019, daily average off-take volumes were 41mmscfd and 90mmscfd respectively, and lower than the ‘take or pay’ volumes of 170mmscfd. Oil price forecasts have been revised following the latest Commodity Market Outlook published by the World Bank in April 2020, which forecasts an oil price of US$35.4/barrel in 2020, US$41.8/barrel in 2021, and US$43.5/barrel in 2022, in comparison to US$66, US$69, and US$72 per barrel for these years at the time of appraisal. Using the above 2020 oil price forecasts, the NPV and the ERR of the project were estimated at US$1.3 billion and 14.0 percent respectively compared to the appraisal estimates of US$4 billion and 20.2 percent. Table 2 summarizes the results of the economic analysis at appraisal and at closing. Under a prolonged low oil price scenario, the project remains economically viable with a 13 percent EIRR (see Annex 4). The results indicate a ‘Substantial’ rating for the efficiency of the investment. Further details of the economic analysis are in Annex 4. 7 The World Bank Ghana Sankofa Gas Project (P152670) TABLE 2. Comparison of ERR and NPV at Appraisal and Project Closing ERR Change NPV Change Appraisal 20.3% $4.0bn Capital Cost Savings Effect 21.6% +1.4% $4.3bn +$0.3bn Lower Crude Oil Price Effect 15.6% -6% $1.7 bn -$2.6bn Lower Offtake Volumes Effect 14.0% -1.6% $1.3 bn -$0.4bn Closing 14.0% $1.3 bn 29. Implementation Efficiency: Implementation efficiency assesses the following aspects of project implementation: (i) production of gas at the offshore platform; and (ii) supply of natural gas to develop the onshore power market. The first component, a large complex offshore investment exceeding US$5 billion, was undertaken with ‘high’ efficiency; the Sankofa gas was available in June 2018 for supply to the onshore power market, with a small delay of four months from the original schedule. The second component comprised (a) an offshore pipeline plus a short onshore link to connect to thermal power plants in the western region; and (b) infrastructures including physical interconnection between GNGC pipelines and WAGP to allow reverse flow to transport Sankofa gas from the western region of the country to a larger power market in the eastern Tema region 14. While the offshore pipeline and short onshore link were completed on schedule, the contractual agreement with WAGP on the reverse flow was delayed. As a result, Sankofa gas was not supplied to the Tema region until April 2019. The component was undertaken with only ‘Modest’ efficiency. Overall, taking into account the implementation of both components, implementation efficiency was rated ‘Substantial’. D. JUSTIFICATION OF OVERALL OUTCOME RATING Rating: Moderately Satisfactory 30. The overall outcome rating of the project is ‘Moderately Satisfactory’. The overall rating takes into consideration the following: (i) ‘High’ Relevance of the PDO; (ii) ‘Substantial’ Efficacy; and (iii) ‘Substantial’ Efficiency. In addition, it takes into account the ‘high’ gas payment risk (paragraph 75) as a result of the continuing and unresolved weak finances of the country’s power sector, which led to the drawdown of the Letter of Credit (LC) backstopped by the IDA Guarantee of this Project on April 7, 2020, 15 months after project closing. E. OTHER OUTCOMES AND IMPACTS (IF ANY) Mobilizing Private Sector Financing 31. An important contribution of the project was the mobilization of approximately US$1.35 billion in financing from commercial banks and other sources to support the development of the offshore Sankofa gas field. The project developers financed the remaining costs with equity and shareholder loans. It is doubtful whether a private investment project of this magnitude in Sub-Saharan Africa i.e. almost US$8 billion for both oil and gas investments15, could have occurred without the active involvement of the World Bank Group. A particular feature of the financing plan was the use of different WBG instruments to finance this infrastructure investment, specifically political insurance and risk guarantees. The IDA US$500 million partial risk guarantee was used to cover the main risk concern of partial payment or non-payment for the gas supplies. IFC provided a loan of US$235 million and a further US$65 million in debt from its 14 PAD, paragraph 42: “Rapid expansion of domestic gas supply at Takoradi during the critical 2018 -2019 period is likely to outpace gas demand from IPPs in Takoradi during the same period with the result that Sankofa and the other domestic gas projects might not be fully absorbed. For this reason, it is critical to establish an interconnection between the GNGC pipeline and the WAGP to deliver excess gas in Takoradi via “back-flow” of the WAGP to the existing cluster of power plants in Tema in Eastern Ghana, where a gas deficit is expected to persist.” 15 US$4.65 billion for the gas field project 8 The World Bank Ghana Sankofa Gas Project (P152670) Managed Co-Lending Portfolio Program; MIGA provided political risk insurance16. The broader goal was to provide incentives for increased private sector participation in the country’s energy sector by supporting the development of a domestic gas market in Ghana. 32. An IBRD Enclave guarantee17 for up to US$200 million had also been proposed be used to provide additional protection to lenders i.e. shareholders as well as commercial banks, and to the Sankofa project sponsors against non- payment by either the government or the Ghana National Petroleum Company (GNPC). This guarantee would be triggered once the other security layers were exhausted and a payment default remained under the GSA that would lead to a debt service payment default by the private sponsors to their lenders under the particular tranche. In the end, however, the Sankofa project sponsors were able to mobilize successfully the large project financing without recourse to this Bank enclave guarantee. Consequently, the IBRD Enclave guarantee never became effective18. Improving Energy Security and Stability of Gas Supply 33. In the PAD, Sankofa was forecast to be able to provide sufficient fuel for 1,000MW of thermal power at peak gas production over a 14 -year period extending beyond 2030. Installed thermal power capacity in Ghana in early 2020 was approximately 2,800MW while thermal generating capacity dispatched was 1,617 MW19. Consequently, as a result of the development of Sankofa gas field, Ghana has taken a significant step to improve energy security by project closing in being able to provide nearly two thirds of the fuel needed for the country's thermal power plants from domestic resources, compared to zero at the time of project approval in mid-2015. The first non-associated gas supplies from Sankofa have also provided much-needed stability in Ghana’s gas supply, which was not available when only associated gas from the Jubilee and TEN fields and gas imported from Nigeria were the only gas supplies. As a result, the power sector supply has become more reliable and outages have been substantially reduced. Sankofa gas is also being supplied through a long-term sales contract with a flat price in real terms, which eliminates any gas price market risks. Overall, Ghana has now 15 years of a stable, reliable, affordable domestic gas supplies, which can provide a solid foundation for local development, full energy independence, and an expanding role as an energy supplier for the region. Gender 34. The project did not have a specific gender component and any gender impact of the project would be indirect and longer term. Nevertheless, the supply of a lower cost and cleaner natural gas fuel from the Sankofa gas field for thermal power generation is expected to bring benefits to all electricity consumers, in particular to low income consumers, which include 49 percent females. Institutional Strengthening 35. The project did not include a component for institutional strengthening though project implementation has highlighted the urgent need for clarification of institutional roles in this increasingly important energy sub-sector. Two public sector companies were involved with project implementation: Ghana National Petroleum Company (GNPC) and 16 IFC committed a loan of US$235 million to Vitol and arranged a further US$65 million in debt from the Managed Co-Lending Portfolio Program that enables third-party investors to participate passively in IFC’s senior loan portfolio. IFC financing was part of a US$1.35 billion loan facility provided by several international commercial banks. MIGA committed up to US$217 million in political risk insurance to VITOL lenders. 17 This would have been the second deployment of this instrument, originally authorized by the Bank’s Board of Executive Directo rs in April 1997. The first IBRD enclave guarantees were approved for the Southern Africa Regional Gas Project in 2003. 18 It should also be noted that signing of the IBRD Enclave guarantee was also subject to funding the Reserve Escrow Account in full as well as the implementation of arrangements to deposit funds into the escrow accounts. However, since these requirements were never complied with in full by GNPC and the Government of Ghana, it would not have been possible for the Project Sponsors to enter into the IBRD Enclave Guarantee, irrespective of their ability to mobilize sufficient capital to complete the Project. 19 GRIDCO: Electricity Supply Plan, 2020 9 The World Bank Ghana Sankofa Gas Project (P152670) the Ghana National Gas Company (GNGC). GNGC was established in 2012 with responsibility to act as gas aggregator/marketer and was subsequently incorporated as a subsidiary of GNPC. Both entities benefitted from the development of the country’s first non-associated gas resources from the Sankofa field, in particular GNGC, which has been taking the lead in supporting the development of non-power markets for the Sankofa gas in the country’s mining, ceramics, and phosphate industries. Poverty Reduction and Shared Prosperity 36. The development of non-associated gas from the Sankofa field is expected to have an indirect impact by lowering the overall economic cost of electricity supply. As domestic gas supplies play an increasingly important role in the mix of fuels for thermal power generation, gradually displacing higher cost LCO and other imports, this will lead to a reduction in the overall cost of fuel supplies and reduce the country’s vulnerability to oil price volatility. This reduction in the cost of supply will eventually trickle down to the main electricity consumers in Ghana and help keep electricity prices relatively low, reducing the need for larger Government subsidies for low income residential consumers. Finally, there are fiscal and macro impacts from having a domestic source of fuel. Lower oil imports mean lower trade and fiscal deficits, making room for government programs to support social programs. It would also help reduce past frequent depreciations of the local currency (Cedi), which results in higher import prices and thus contributes to inflation. Other Unintended Outcomes and Impacts 37. None Performance since Closing date 38. The time to the closing date of December 31, 2018, was a relatively short period for a project of this complexity - less than 3.5 years from the Board approval-though its length was aligned with Guarantee procedures. Gas production from Sankofa was only forecast to start in February 2018- it actually started in June 2018- so the need for a longer implementation period should have been anticipated. Payments for the gas supplied also did not begin until 2019 and there have been extended delays in making these payments, which continued into 2020. Payments could not be made by GNPC due to the lack of liquidity, resulting from the increasing volume of arrears and a lack of payment discipline among energy sector SOEs. These delays became an issue of serious concern for both GNPC and GoG as well as for the World Bank during early 2020 since the GoG’s ability to continue to make contractually agreed monthly payments for gas offtake during 2020 became further constrained by parallel obligations to a number of IPPs, placing increased pressure on scarce foreign exchange resources. 39. By end March 2020, GNPC was in arrears by over US$190 million (about 5 months of unpaid gas invoices). One week later on April 7, the Letter of Credit (LC) backed by a US$500 million IDA Payment Guarantee was drawn down for an amount of US$191,576,694 by the Sankofa Project sponsors (ENI and Vitol). Each drawdown becomes a loan to Ghana National Petroleum Corporation (GNPC) that has to be repaid within 12 months. If the loan is not repaid within 12 months by April 7 2021, the two LC Banks (HSBC and Standard Chartered) would be entitled to call on the IDA guarantee at that time, whereupon IDA would be obligated to pay the LC issuing banks within 60 days from the date of demand.20 Any amount paid out by IDA under its guarantee would be subject to the indemnity and reimbursement obligations of GoG to the World Bank under, and in accordance with, the terms of the Indemnity Agreement. If GoG fails to comply with such obligations for 60 days following notice from the Bank, the Bank will be entitled to exercise its 20 In accordance with the terms of the guarantee documents, the amount that was drawn under the LC was automatically converted into a loan of equivalent size owed by GNPC to the LC issuing banks, which will become due 12 months from the date of the LC drawdown and will accrue interest at a rate of LIBOR +0.70 percent. GoG made a first repayment of $20 million in June. 10 The World Bank Ghana Sankofa Gas Project (P152670) remedies at its discretion, in whole or in part, against the country portfolio. 40. The gas payment arrears of the Sankofa Project sponsors had been steadily growing since June 2019, reaching US$100 million by end 2019 and US$192 million by March 2020. Despite several communications sent by the sponsors to the GoG to clear the outstanding arrears, only US$4 million was paid in March 2020 to reduce the arrears. At the same time, following the impact caused by COVID-19, oil revenues to the government from the offshore producing fields fell sharply due to historically low oil prices. This led to a sudden drop in fiscal revenues and forced the government to use the limited fiscal space available for priority public sector expenditures and other priority payments. III. KEY FACTORS THAT AFFECTED IMPLEMENTATION AND OUTCOME A. KEY FACTORS DURING PREPARATION 41. The discovery of offshore, non-associated natural gas reserves at the Sankofa field by an international oil consortium in 2011 provided Ghana with the prospect of developing a lower cost, less polluting, domestic fuel, reduce its vulnerability to external oil price fluctuations, and start to replace the increasing volumes of costly imports of liquid fuels, in particular LCOs, needed to meet its expanding power generation needs. In addition, the development of Sankofa had the potential to provide a secure supply of domestic gas over a 15 to 20-year period at a time when gas imports from Nigeria through the WAGP were proving to be increasingly unreliable and adding to the cost of Ghana’s fuel import bill. The energy context at the time provided additional impetus to develop a domestic gas supply: a long history of unpredictable gas supplies through WAGP; shortages of fuel for power generation; and frequent blackouts. 42. At the same time, the development of the Sankofa gas field also presented some risks, both technical and financial. First, proven gas reserves i.e. one TCF, were modest in size and found in deep, offshore waters, which gave rise to relatively high production costs; in this regard, the cost of Sankofa gas was higher than gas being imported into Ghana intermittently through the WAGP,-though still lower than the cost of liquid fuel imports for thermal power generation. Second, secure payment mechanisms for the gas supplied to the power stations were essential if the large investment needed to develop the Sankofa gas field was to proceed. While non-associated gas reserves had been successfully developed in other developing countries for power generation21, such developments required carefully designed financial instruments and guarantees to the private oil companies in order to mitigate the different risks associated with the development of the gas reserves. 43. The design and effective use of different World Bank Group instruments was the key factor during preparation which enabled this large offshore gas development to move ahead. These instruments provided the needed incentives and risk reduction measures for the private oil company consortium to commit to a total investment, estimated at approximately US$4.6 billion for gas investments. The payment security package, agreed by the government, the oil company consortium, and the World Bank, included a comprehensive set of risk mitigation steps, and only limited government support, aimed at enhancing the creditworthiness of GNPC, which was the off taker of the Sankofa gas for the duration of the gas supply agreement (GSA). The security package itself comprised different layers of recourse and intervention in the following order: (i) the payment mechanism that channels receivables of GNPC into segregated accounts; (ii) a cash reserve escrow account funded by GNPC for an equivalent of 4.5 months of gas sales; (iii) a letter of credit backstopped by an IDA Payment Guarantee; and (iv) a limited sovereign guarantee backed by an IBRD Enclave Loan Guarantee. MIGA guarantees were used to support termination payments for financiers and private equity partners. Overall, this well-designed framework of incentives and guarantees addressed 21 Second Natural Gas Project in Thailand, 1985 11 The World Bank Ghana Sankofa Gas Project (P152670) effectively the financial risks of the project and enabled the Sankofa investment to proceed. 44. The environmental and social impacts of the project were well identified during preparation. Six of the eight performance standards were triggered and plans to address and/or mitigate the impacts of the offshore gas development were carefully prepared. Initial consultations with the affected Sanzule community took place during preparation. These consultations involved the routing of the offshore gas pipeline at landfall to avoid population displacement and to minimize adverse economic impacts on nearby communities. An Environmental, Social, and Health Impact Assessment (ESHIA) was also completed before Board which identified, inter alia, the impact of the investment on farming, including an estimate of the number of farmers affected. In this regard, the need to prepare a Livelihood Restoration Plan was identified. The ESHIA also identified the potential impacts of unplanned events such as oil spills. Overall, consultations with affected communities were extensive. The single shortcoming was the failure to anticipate the need for a longer implementation period to address the different environmental and social impacts, which experience in other large infrastructure investments indicate. B. KEY FACTORS DURING IMPLEMENTATION 45. Implementation covered a period of approximately 3.5 years from July 2015 to December 2018 when the project closed. Project implementation comprised the following main activities: (i) Construction of offshore receiving facility (ORF): (a) drilling of 5-6 development wells; (b) laying of offshore pipeline; and (c) onshore gas pipeline link to onshore power plants; (ii) Mobilization of non-recourse financing; completion of financial security arrangements; financing to be mobilized for the gas development was US$3.9 billion; (iii) Construction of a 2.5 km onshore interconnection link with the existing West Africa Gas Pipeline (WAGP) to enable supply of Sankofa gas to power plants in eastern region; (iv) Activities to ensure compliance with environmental and social safeguards; (i) Factors Subject to Control of Private Oil Consortium 46. Implementation of Component (i) was carried out with high efficiency by the private oil company consortium and completed just four months later than the original schedule of February 2018. The drilling of 5 gas production wells, laying of the subsea pipeline, and the tie-in with the onshore receiving facility (ORF) was completed in June 2018, at which time gas from the Sankofa offshore gas fields was ready to be supplied to the onshore power plants at Takoradi; the only delay incurred from the original schedule was a short one in completing the ORF. The mobilization of the large amounts of financing from commercial banks and other sources for the offshore gas development was also undertaken efficiently by the oil consortium. (ii) Factors Subject to Control of Government or Government Agencies 47. Government had the contractual obligation to put in place several elements of the financial security arrangements including, inter alia, (a) the establishment of a segregated designated account (GDA) into which revenues from GNPC on-sale gas would automatically be deposited; and (b) the funding of a reserve escrow account (REA) that could be tapped for gas payments if amounts in the GDA were insufficient. Such measures were required to be in place before the first gas supplies started under the terms of both the project agreements and GoG/GNPC’s contractual arrangements with IDA. These obligations were either not met or, in the case of REA, met only partially with an initial funding of US$100 million followed by some further deposits, which were subsequently drawn down following first gas (in June 2018) and never adequately replenished. 12 The World Bank Ghana Sankofa Gas Project (P152670) 48. The need to construct an onshore pipeline link to the existing WAGP, in order to enable gas supplies from Sankofa to reach the expanding power market for gas in the eastern region of the country, had been identified at appraisal; however, its investment costs were underestimated. Furthermore, the institutional and contractual arrangements associated with reaching agreement on the gas transport tariffs were also underestimated. By project closing, a contractual agreement on the pipeline carrier arrangements had yet to be agreed and construction of the pipeline interconnection (referred to as the TTIP) had not started. The impact of these delays was to (i) restrict the market for Sankofa gas to the western region in Takoradi during 2018-19 and (ii) prevent increased volumes of Sankofa gas being supplied to an expanding power market in the eastern region. The overall impact was that neither GNPC nor the government could meet its ‘take or pay’ related obligations during this initial production period and a waiting market in the eastern region had to seek alternative fuel supplies at higher cost. 49. Institutional arrangements and responsibilities for gas purchasing and delivery had been developing in an ad hoc way since 2012 and lacked a clear vision as to how the sector could, and should, provide a new source of domestic fuel for an expanding domestic power market. Rather than having clear roles for gas sector institutions across the value chain, there was considerable overlap of responsibilities, causing inefficiencies. For example, GNPC, GNGC, and even the Volta River Authority (VRA) were each acting as gas suppliers in a market structurally not able to sustain a competitive market22. A Gas Master Plan in 2016 had made recommendations to avoid such duplication of roles 23 and decisions were taken to clarify the roles of these gas institutions. However, successive governments did not follow through on these decisions and ended up changing the institutional arrangements of the gas sector mid-way through the implementation period. This prevented decisions being taken on key actions, including the interconnector investment and commercial arrangements with off-takers. 50. Finally, regarding the power sector, the gas supplied from Sankofa was meant to be consumed and paid by thermal power plants (IPPs and VRA) to GNPC, which, in turn, were supposed to be paid by the distribution utilities (ECG and NEDCO). However, only 11 percent of the gas sales has actually been paid by the power sector directly, the rest having been paid by GoG. The lack of liquidity of the power sector was mainly due to large arrears by public sector institutions and all energy SOEs in the energy value chain. A GoG Energy Sector Recovery Plan was not put in place and actively pursued on time to stop a rapidly deteriorating sector financials, which was further aggravated with the approval of numerous IPPs that created an excess power capacity and additional financial obligations for the sector. (iii) Factors subject to Control of World Bank Group 51. Adequacy of Supervision and Reporting. Bank supervision missions took place regularly and were well supported by the Country Management Unit (CMU). These missions addressed not only issues affecting the project but also the financial viability of the power sector, a major risk to the project outcome. However, follow up was ineffective and the Bank lacked the operational instruments to address the deteriorating finances of the power sector, already precarious at Board. Supervision effectiveness was also impeded by other constraints. For example, while government decisions were taken to define the respective roles of GNPC and GNGC, including their corporate governance structure, successive governments did not follow through on these decisions. This caused delays in approving key actions that needed to be taken, including the interconnector investment and commercial arrangements with the gas off-takers. Budget allocations for the supervision of the project were also insufficient, 22 Gas Master Plan (2016): GNPC would sell gas purchased from the Sankofa field; GNGC would sell gas purchased from the TEN/Jubilee fields; while VRA would only sell gas imported from NGas 23 One solution proposed was to eliminate the duplication of roles across the gas value chain caused by past decisions that were no longer justified: GNPC would be operationalized as the gas aggregator and gas shipper; GNGC would be operationalized as the gas transporter; while VRA would continue only as a gas off-taker 13 The World Bank Ghana Sankofa Gas Project (P152670) resulting in the use of budget resources from other operations. Finally, though supervision reports provided detailed accounts of issues and delays, the growing risks to the project outcome were not reflected in the main DO and IP ratings until late in implementation in June 2018. 52. Compliance of payment security package: Since early 2018, the payment security package of the Project was not complied with by either GNPC or GoG. As a guarantor, the Bank also play a role to ensure these arrangements were in place and then remained compliant throughout the project. Given the limited instruments available to the Bank to enforce compliance, these obligations needed to be addressed much earlier in 2018, and at the highest level of government. 53. Safeguards Supervision. Activities and community consultations associated with ensuring safeguard compliance were implemented satisfactorily. Implementation took longer than anticipated at appraisal, due mainly to the need for extensive consultations with families whose economic livelihoods were being affected by the gas pipeline in the onshore landfall area. However, the additional time required for such consultations was not unusual and in line with experience in similar projects which impact the economic livelihoods of small communities. The private oil consortium engaged an Independent Environmental and Social Consultant (IESC) at the World Bank Group’s request. In addition, supervision continued after the formal closing date of December 2018 to ensure completion of the Livelihood Restoration Plan (LRP) and full compliance with the remaining safeguards. (iv) Factors outside of control of Government and/or Implementing Agencies 54. There were not any significant factors affecting implementation outside either government control, the main gas sector agencies, or the private oil consortium during the period July 2015 to December 2018. Over this period, there was political stability in Ghana which was maintained through elections in December 2016. There was also a favorable external macro environment for growth. Oil prices also remained stable, averaging approximately US$65 per barrel (until they fell sharply after closing in April 2020 to below US$20 per barrel). Finally, there were no periods of conflict or civil unrest. IV. BANK PERFORMANCE, COMPLIANCE ISSUES, AND RISK TO DEVELOPMENT OUTCOME A. QUALITY OF MONITORING AND EVALUATION (M&E) M&E Design 55. The M&E framework was soundly designed to monitor progress towards the two key outputs shown in the Theory of Change: (i) volume of natural gas supplied for power generation in Ghana; and (ii) amount of private capital mobilized to finance the Sankofa gas development. Two indicators were selected for this purpose: (i) daily volume of natural gas supplied from the Sankofa gas field for power generation, measured in millions of standard cubic feet per day (mmscfd); and (ii) amount of private sector financing mobilized, measured in millions of dollars-for the Sankofa offshore gas development using the IDA and Bank guarantee instruments. Two other indicators were also included in the M&E design framework: the number of indirect project beneficiaries and the percentage of female beneficiaries. 56. An additional indicator might have been used to differentiate between the (a) availability of natural gas supplies at the offshore platform; and the (b) supply of Sankofa gas to the (i) Western and (ii) Eastern regions of the country. Such an indicator would have also highlighted where delays in the needed onshore pipeline infrastructure were occurring in order to supply natural gas to thermal power plants in different regions of the country. 14 The World Bank Ghana Sankofa Gas Project (P152670) 57. The remaining two indicators in the results framework of the PAD were linked to the project beneficiaries and were, at the time of approval, core indicators for IDA energy investment operations. However, these indicators are more appropriate for an electricity distribution project than a gas supply project for use in thermal power plants M&E Implementation 58. M&E data collection was the responsibility of private oil companies in partnership with the government and was carried out satisfactorily. Over the 3.5-year implementation period from mid-2015 to project closing in December, 2018, the basic M&E data was collected on an annual basis and recorded in the ISRs. However, given the relatively short implementation period to project completion i.e. 3.5 years, and the subsequent infrastructure delays experienced in making available the Sankofa gas for power generation, the actual M&E data recorded in the annual ISRs in terms of implementation progress was limited and related only to the aggregate volume of private sector capital mobilized for the offshore gas investment. The Aide Memoires provided more detailed accounts of implementation progress and identified the main issues impeding progress towards the main project objective. M&E Utilization 59. Overall, M&E data was used effectively during implementation. Progress in implementing the offshore investments was carefully monitored by the Sankofa project sponsors which enabled the first gas supplies to become available in June 2018, close to the original schedule. Mobilization of the forecast amounts of private capital in the results framework was needed to achieve this goal. Corrective action was also taken quickly by the Sankofa project sponsors during implementation when the offshore pipeline from the production platform needed to be re-trenched at landfall. The delays being incurred in the gas supply targets in the M&E results framework enabled supervision teams to address the different onshore gas infrastructure issues impeding the supply of gas to thermal power stations in the Eastern region of the country. Justification of Overall Rating of Quality of M&E 60. The overall rating of the quality of M&E is ‘Substantial’. Despite some shortcomings in the M&E design, the choice of the main project monitoring indicators was sound. In addition, data collection and monitoring of implementation was carried out regularly, compliance with the environmental and social safeguards was carefully monitored throughout and continued beyond project closing date, and the basic M&E data was used during implementation to address the issues impeding the transportation of gas supplies from the Sankofa offshore gas field to the different thermal power stations. B. ENVIRONMENTAL, SOCIAL, AND FIDUCIARY COMPLIANCE Environmental and Social Safeguards 61. The development of the Sankofa offshore gas facility, including the related onshore pipeline and receiving facilities, had significant environmental and social impacts. The project was rated a Category A operation and triggered the following six performance standards (PS): (i) Social and Environmental Assessment and Management Systems (PS 1); (ii) Labor and Working Conditions (PS 2); (iii) Resource Efficiency and Pollution Prevention (PS 3); (iv) Community Health, Safety, and Security (PS 4); (v) Land Acquisition and Involuntary Resettlement (PS 5); and (vi) Biodiversity Conservation and Sustainable Management of Living Natural Resources (PS 6). The Sankofa development also involved a number of risks during the production drilling and operational phases. These risks included accidents such as an explosion, oil spills or a well blowout; worker health and safety risks; and risks to community livelihood associated with economic displacement during construction and operation of the pipeline landfall and shore-based facilities. 15 The World Bank Ghana Sankofa Gas Project (P152670) 62. The main safeguards issues requiring monitoring during supervision were: (i) Environmental, Social, and Health Impacts: A comprehensive assessment (ESHIA) was prepared prior to submission of the project to the Board covering acquisition of land for the Offshore Receiving Facility (ORF), which caused economic displacement of an estimated 205 farmers; it also addressed all aspects of the offshore development-from planning to decommissioning, air quality control, noise, and cuttings dispersal; (ii) Land Acquisition and Involuntary Resettlement: despite the lack of any physical displacement of populations or communities, there was a need for continuous monitoring to help ensure that resettlement activities related to economic displacement and livelihood restoration were implemented with the proper disclosure of information and the participation of stakeholders; 63. Environmental management issues were given careful and sustained supervision during construction of the ORF until its completion in June, 2018. The potential concerns included the (i) arrangements to preserve wetlands; (ii) steps taken to preserve forest tracts and wildlife habitats within the concession; (iii) use of more costly-though environmentally more benign-methods such as horizontal directional drilling (HDD) rather than trenching, in the areas where the offshore pipeline crossed the beach; and (iv) maintaining a well-balanced ratio of one HSE officer for every twenty workers at the site. Air quality, noise, and water quality monitoring started in June 2017 and was carried out satisfactorily over the subsequent 12-month period. For the monitoring of biodiversity, a local NGO was employed. This NGO had responsibility for marine mammal monitoring programs as well as for monitoring programs such as the Avian and the Sea Turtle Biodiversity Action Plans. The IESC conducted semi-annual field visits as well as desk reviews of progress reports in the intervening quarters to evaluate implementation of environmental, social, health and safety management plans and compliance with the performance standards. The IESC remains under contract to the consortium to conduct a completion audit required by IFC, MIGA, and commercial lenders. Overall, environmental management issues linked to the main project investment were addressed satisfactorily. 64. Livelihood Restoration Plan (LRP). The LRP addressed impacts related to onshore economic displacement, and, in particular, to vulnerable groups. It also addressed other issues such as (i) income generation opportunities, including near shore and offshore fisheries; (ii) work-related mobility patterns and practices in the impacted zone: and (iii) population influx. The LRP identified a number of livelihood options, including support for agriculture (crop and livestock production), skills training, and micro-enterprise development. The LRP was prepared in close consultation with project households, local authorities, local Community Based Organizations (CBOs), women, and youth groups. A Grievance Mechanism was also put in place at the Project site, where complaints concerning economic displacements for project affected people (PAPs) and project-related concerns were received and resolved. Implementation of the LRP did not start until mid-2017 and had not been completed by project closing in December 2018. By project-closing, all PAHs had received material and technical support and were undertaking various livelihood options with early indications of ‘success’. Finally, a local Chieftaincy dispute had threatened community peace, fractured consultation with the community, and delayed implementation of some social investments at an early phase of the project. However, it was possible to mitigate the impact of this dispute on the project through broad consultations with the different interest groups. 65. Safety Issues. The overall safety performance of the main contractor during implementation was of high standard. Lost time due to injuries was negligible while adherence to workplace safety rules was exemplary. Some specific safety issues emerged during implementation but were addressed satisfactorily. In particular, the gas pipeline carrying gas from the Sankofa OCTP field was found to be partially exposed on the beach at Sanzule in 2018. A comprehensive risk assessment of the environmental and safety concerns was then carried out for the exposed pipeline as well as for the proposed re-trenchment method, which was to bury the pipeline deeper. The pipeline is now completely buried. 66. There were also some safety concerns at the Tema Metering Station- which is located in a densely populated 16 The World Bank Ghana Sankofa Gas Project (P152670) area, where households are close to the metering station. These concerns were evaluated carefully to assess risks to the surrounding population using different assumptions of gas flows and pressure. The conclusion was that with gas flows up to 60mmscfd and at low pressure, there would be no risk to the surrounding population. Gas flows to Tema during 2019, one year after the project closing date didn’t reach 60mmscfd. Fiduciary and Procurement 67. The project involved only an IDA guarantee; no IDA or IBRD lending was used to finance any of the investments. Consequently, there were no IDA or IBRD-financed procurement activities under the project; there were also no traditional financial management issues. Nevertheless, the World Bank’s operational policies on Investment Project Financing (OP/BP 10.00), which also govern Bank Guarantees, require that procurement of works, goods, and services for a project supported with guarantees be carried out with due regard to economy and efficiency. The large capital investments for the offshore development, estimated at appraisal as US$7.9 billion24, were carried with high efficiency following industry practices and has resulted in a lower overall cost when compared with the original estimates- by about 6 percent or US$500 million. C. BANK PERFORMANCE Quality at Entry 68. Preparation of the project took place over a short period of time-approximately 6 months from project identification in early 2015 until the Board presentation in July 31, 2015. The time period between disclosure of the Project Identification Document (PID) and Board presentation was just two months. Overall, the project was prepared quickly and efficiently. Two considerations guided both the design of the investment and the financing instruments deployed: (i) the investment was a key step in building up a domestic gas industry in Ghana through the development of an offshore, non-associated gas field for use in thermal power generation; and (ii) the deployment of different World Bank Group financing and guarantee instruments helped achieve this investment goal which represented the largest foreign private investment in Sub-Saharan Africa i.e. US$8 billion. A further consideration in the project design was to structure the operation in a way that reduced the government’s fiscal support at a time when macroeconomic uncertainty was affecting the investment climate. Overall, the design of the project was sound and the combination of financing and guarantee instruments deployed to achieve the main objective was particularly innovative. The environmental and social impacts of the offshore gas development on the small communities close to landfall were also well identified. 69. One shortcoming during project preparation was underestimating the investment required and time needed for the contractual agreements with WAPCo for a pipeline interconnector transporting Sankofa gas from Takoradi to Tema. This pipeline interconnector was an essential link to be able to supply Sankofa gas to the eastern region of the country, where power plants either were using higher cost, imported LCO for generation or operating below capacity due to the unreliability of gas supplies from Nigeria. At the time of appraisal, the investment for the pipeline interconnector was estimated at less than US$10 million25 and was expected to be financed by GNPC and/or WAPCo. However, the total investment required to complete the pipeline system was much higher, close to US$100 million, which also included investment in compression and looping to enable higher gas flows-. Overall, the delays incurred in completing the interconnection delayed the supply of Sankofa gas to the eastern region of the country and reduced the overall benefits of the original investment in the first year of its operation. 70. A further shortcoming was the lack of a coordinated power sector supporting strategy to address and 24 Comprising $3.9 billion for the oil and $4.0 billion for the gas developments respectively 25 PAD, paragraph 77 17 The World Bank Ghana Sankofa Gas Project (P152670) monitor progress of the power sector’s financials at the time of board approval despite a careful assessment of its weak financial position and resulting payment risks in the PAD. Specifically, the IMF’s Extended Grant Facility had only been approved in April 2015 and the planned concession for the country’s major distribution company, ECG, under the MCC Compact 2 program, was terminated by GoG only six months after the concession was effective. While the latter outcome could not have been foreseen at Board, the Bank lacked an effective instrument to address and monitor progress of the power sector’s financials before 2019. 71. Finally, leaving key technical investments (such as the TTIP interconnector) and important payment security features (such as setting up the GDA and REA) to be addressed by effectiveness or during implementation reflected the readiness of the project at the time of Board. Quality of Supervision 72. Supervision was carried out over more than three years - from Board approval in July 2015 to project closing in December 2018. Supervision missions visited Ghana on a regular basis, starting in October 2015 and continuing in 2019 after project closing to support efforts by the government to put in place the agreed payment security package for the Sankofa gas. 73. The effectiveness of supervision was constrained at the outset by the following: (i) the main consumer of the Sankofa gas was the electric power sector, whose financial position had been deteriorating as generation costs steadily increased; however, there was no clear parallel strategy to address these concerns while the plans to improve the finances and operational efficiency of the main distribution company, ECG, through a management concession, did not materialize; (ii) the two gas sector institutions, GNPC and GNGC, were relatively weak and unable to exert a strong leadership role, despite efforts during supervision to strengthen their effectiveness; and (iii) within the Bank, budget resources allocated for the supervision of the project were inadequate, constraining further the overall effectiveness of supervision. In this regard, guarantees should have been assigned adequate supervision budgets, similar to those assigned for IDA and Bank financed project investment operations. 74. Within these constraints, successive Bank supervision teams addressed the main sector issues. They endeavored to reach agreement with the government on actions needed to strengthen the financials of the power sector while monitoring progress to help ensure the initial gas supplies from Sankofa, expected in the first half 2018, would be fully utilized. The importance of completing a pipeline interconnector was also given considerable attention by supervision teams to help ensure there would be sufficient offtake capacity in place in early 2018 to absorb the first Sankofa gas supplies. In this regard, the supervision aide memoire of April 2016 and related ISR (#1), highlighted ‘sufficient gas offtake capacity’, and the consequent need for a pipeline interconnector, as the ‘biggest risk to the project outcome’. The same documents highlighted a further risk to the project outcome- namely, power sector reform and the financial turnaround of key offtake utilities such as VRA and ECG. These risks continued to be highlighted in subsequent supervision documents through to project closing in December 2018. The construction of a gas interconnector was eventually completed in April 2019, which expanded the domestic market for Sankofa gas. 75. Early supervision reports provided well written accounts of the main issues affecting implementation of the Sankofa gas project They also highlighted the main risks to the development outcome though these risks were not reflected in the DO and IP ratings until June 2018, close to project closing. Supervision teams were closely involved with ongoing negotiations for the gas interconnector while, in parallel, they continued to be involved on the power sector reform agenda, which presented an important risk to project sustainability. However, by early 2018, with no real progress having been made with the construction of a gas pipeline interconnector, it became clear that the approaching ‘take or pay’ gas payment obligations, starting in June 2018, would require upfront payments for the 18 The World Bank Ghana Sankofa Gas Project (P152670) Sankofa gas since the full, contracted volumes could not be met in terms of gas offtake. 76. In regard to environmental and social safeguards, supervision was thorough, sustained, and effective. The corporate safeguards standards from ENI and Vitol were compliant with WB safeguards standards and, as a result, the companies had a sound framework for monitoring safeguards issues and implementing mitigation measures during the project. Given the several safeguard obligations which needed to be brought into compliance, the main issues addressed during implementation were (i) the economic displacement issues of some 200 families as a result of the onshore pipeline routing, which required preparation of a Livelihood Restoration Plan (LRP); (ii) a chieftaincy dispute in the affected communities which delayed reaching agreement on the LRP; (iii) the need to re-trench the offshore gas pipeline which had surfaced close to the affected communities; and (iv) a number of biodiversity concerns resulting from the gas pipeline routing . Safety and hazard prevention drills took place regularly on the offshore platform. By project closing, most of the above issues and concerns had been satisfactorily addressed. The rating for the Environment Safeguards was Satisfactory and for the Social Safeguards was Moderately Satisfactory, due to delays in completing the Resettlement Plan. Safeguard supervision continued after project closing, using resources from other sources. By end 2019, the remaining resettlement issues had been satisfactorily addressed. Justification of Overall Rating of Bank Performance 77. Overall Bank performance is rated ‘Moderately Satisfactory’. The project design and deployment of different Bank Group lending instruments were innovative and proved successful in attracting large amounts of private capital to develop the country’s first, significant, natural gas resources. At the same time, the specific measures needed, and to be put in place, in order to strengthen the financials of the country’s main power sector institutions, the main market for the Sankofa gas, were not addressed within the framework of this operation, either through other lending instruments of the Bank or of other development partners. Consequently, nearly four years after Board approval, the sustainability of the Sankofa gas development remains uncertain, given the weak and unresolved financials of the country’s power sector. D. RISK TO DEVELOPMENT OUTCOME 78. The main risk to the sustainability of the development outcome is the gas payment risk. The gas supply risk is considered ‘low’, given the adequate proven reserve level and technical performance to sustain gas supplies at a minimum of 170mmscfd for the coming ten years. 79. Gas Payment Risk: The contractual agreements with the Sankofa project sponsors provide for ‘take-or-pay’ arrangements when gas becomes available in minimum volumes of 170mmscfd. These payment obligations first came into effect in 2018 which required government to maintain payments of approximately US$40 million per month. From the outset, GNPC was not able to pay for the gas (neither gas consumed nor the resulting payment obligation for gas not utilized under the ToP obligation). Consequently, the government (GoG) was contractually obligated to make the payments to the gas supplier. Over the period June 2018 until September 2019, GoG was able to pay-although with some delay. Subsequently, however, the payment backlog has steadily accumulated as noted earlier in the paragraph, Performance since Closing Date. 80. The prospects of a re-occurrence of payment arrears remain real without a fundamental restructuring of the country’s power sector. Ghana’s power sector financial situation has been in crisis for some time with recurring revenue shortfalls that, by 2019, amounted to about US$1 billion. The main issues have been (i) over-investment in new power generating capacity, (ii) gas offtake constraints due to delays in infrastructure completion; (iii) competing interests in gas aggregation and nomination between GNPC, GNGC and VRA; (iv) non-payment of GoG and SOE electricity bills and a build-up of arrears among sector SOEs; (v) high system losses and low collections from the 19 The World Bank Ghana Sankofa Gas Project (P152670) distribution utilities; and (vi) tariffs which do not cover cost recovery. These issues have resulted in a weak financial position of the country’s main consumer of power (and therefore gas), Electricity Company of Ghana (ECG), and an inability to pay gas suppliers for the gas being consumed by the sector. The power sector (ECG, and the Volta River Authority, VRA) can only pay 40 percent of the monthly invoices of the gas that is currently offtaken. The gas payment risk is therefore considered ‘High’. V. LESSONS AND RECOMMENDATIONS Successful natural gas developments for a domestic market require not only the risk mitigation instruments needed to attract private capital to develop the gas but also a financially viable power sector that is able to pay for the gas being consumed. 81. The main consumer of natural gas supplied from Sankofa was the power sector; consequently, the project’s longer-term viability was dependent on having financially solvent and liquid power sector in order to pay for the gas. However, Ghana’s power sector had been in crisis for some time. Growing levels of receivables in public sector entities for unpaid electricity consumption were severely impacting the financials of the country’s main distribution company, ECG. At the time of appraisal, the government had started to reduce some payment arrears to ECG for its electricity consumption. Also, under the MCC Compact II program, the government had committed to procure an international distribution utility to become a private concessionaire for ECG and improve its technical and commercial performance (decrease losses and increase collections), eliminate payment arrears, and maintain public sector receivables at less than two months of sales by end-2017. In addition, an IMF Extended Credit Facility was also intended to reinforce fiscal discipline by applying a ceiling to the contracting and guaranteeing of new non- concessional external debt by the government and/or public enterprises. However, none of these measures and/or instruments either materialized or proved sufficient to restore financial health to the power sector. By mid-2018, when the first Sankofa gas deliveries began, the country’s power sector was in deep crisis, with only limited capacity to pay for the gas. 82. A financially viable power sector is critical for the development of a domestic gas industry. Past Bank Group experience in developing natural gas in countries such as Thailand26, and other countries, reaffirm this requirement. Consequently, the use of WBG financing and/or guarantee instruments needs to be accompanied by a careful assessment of the financial health of the power sector and the possible need for parallel lending, such as DPOs, to strengthen sector reform. Gas to power operations therefore need to be part of a comprehensive plan, which addresses not only the supply of gas but also the financial health of the power sector, where most of the gas is consumed. 83. To reduce the payment risks of WB Guarantees, such operations should be designed in parallel with other operations (either IPF, P for R, or DPO) which support improvements in the financial situation of the sector. 84. A WB Partial Risk Guarantee is usually designed to enhance the creditworthiness of the offtaker. At the same time, the Guarantee alone does not address the root cause of the lack of creditworthiness of the offtaker. Consequently, it is important that such Guarantees be designed in parallel with an additional lending and TA operation which target the improvement of the sector’s financials. The current drawdown of the Letter of Credit of the Sankofa Project could have been partially mitigated through the implementation of an investment ‘program for results’, or budget support operation, focusing on improvements of the financial situation of the power sector. In this regard, the World Bank initiated a US$ 20 million TA project (GESTIP) in 2019 (after Project closing) which seeks 26 PPAR, June 28, 1985, Thailand: Second Natural Gas Development Project 20 The World Bank Ghana Sankofa Gas Project (P152670) to provide implementation support to a recently approved Energy Sector Recovery Plan prepared by an inter- ministerial task force. A Program for Results operation is also being developed for 2021 Board approval. These programs, had they been underway in 2016, would have assisted in supporting progress towards a financially viable energy sector in Ghana. 85. Board approval and effectiveness conditions of World Bank guarantees should be both carefully timed and then enforced, taking into consideration the major risk factors of the investment as well as the limited options subsequently available to enforce key agreements, once the guarantee becomes effective. 86. The agreed security package at negotiations in June 2015 required the government to set up a government disbursement account (GDA) and fund a reserve escrow account (REA) with 4.5 months of the estimated gas revenues i.e. approximately US$190 million, by first gas, forecast to occur over 30 months after Board approval. As of April 2020, the GDA has not been used, and the REA was only partially funded to an accumulative amount of US$100 million. In this context, it is important to note that, in a typical project finance operation, commercial lenders usually require security arrangements to be in place by financial closure. In the case of the Sankofa project, however, the legal obligations of GoG and GNPC to activate the GDA and fund the REA after financial close were considered sufficient by the Bank. Despite it being a legally binding commitment, the government did not fulfill its requirement either to set up the GDA or to fully fund the REA. Considering the limited instruments subsequently at its disposal, the Bank must require the government to fulfil its original commitments at the time the guarantee becomes effective. Otherwise the only option available is to suspend the entire IDA lending program to the country. Importance of a Communications’ Strategy in advance of a major natural resource development. 87. The development of the country’s first natural gas resources was accompanied by considerable criticism in the local news media. A common perception was that of an international petroleum consortium, supported directly by the World Bank Group, exploiting a domestic natural resource for its own benefit. The potential benefits for Ghana were not explained to the public while the high cost being paid for the first natural gas supplies, in conjunction with the perception of a punitive ‘take or pay’ contract, left the impression of an investment benefitting only the private oil consortium 88. The World Bank Group, in collaboration with the government, should have anticipated the need for a communications strategy, explaining the economic benefits of the country’s first natural gas development, part of which would not be realized immediately. The fact that associated gas produced from the nearby Jubilee field was not being charged to consumers27 reinforced the urgency for such a strategy to clarify expectations and dispel misunderstandings in advance of the development itself. Since the first non-associated gas supplies started in production in June 2018, there has been continuing criticism in the local communications media of the Sankofa offshore development which has generated pressure for alternative investments-such as LNG imports as an alternative to domestic gas development-further exacerbating the country’s growing indebtedness. 89. In summary, major natural resource developments, whether in energy or in mining, should be accompanied by a carefully thought out communications’ strategy to the public to explain the rationale for such an investment and the anticipated benefits. . 27 However, significant gathering and transfer charges are applied before delivery to the end-user 21 The World Bank Ghana Sankofa Gas Project (P152670) ANNEX 1. RESULTS FRAMEWORK AND KEY OUTPUTS RESULTS INDICATORS A.1 PDO Indicators Formally Revised Actual Achieved at 1. Indicator Name Unit of Measure Baseline Original Target Target Completion Quantity of gas supplied to MMSCFD 0.00 171 N.A. 10 power plants December 31, 2018 Comments (achievements against targets): Gas supplied at closing was below original target. However, gas supplies increased to 90 MMSCFD in 2019 and are projected to reach 120 and 171 MMSCFD in 2020 and 2021, respectively. Formally Revised Actual Achieved at 2. Indicator Name Unit of Measure Baseline Original Target Target Completion Private sector capital US$ million 0.00 5,100 N.A. 4,653 mobilized (US$ million) Comments (achievements against targets): Lower private sector capital mobilized at closing reflects cost savings in overall capital investment. 22 The World Bank Ghana Sankofa Gas Project (P152670) Formally Revised Actual Achieved at 3. Indicator Name Unit of Measure Baseline Original Target Target Completion Indirect Project Beneficiaries (Million) 0.00 10.6 N.A. 17.8 Comments (achievements against targets): Formally Revised Actual Achieved at 4. Indicator Name Unit of Measure Baseline Original Target Target Completion Female Beneficiaries (%) % 0.00 50 N.A. 49.3 23 The World Bank Ghana Sankofa Gas Project (P152670) A.2 Intermediate Results Indicators Formally Revised Actual Achieved at 1. Indicator Name Unit of Measure Baseline Original Target Target Completion Gas production capacity achieved by the project MMSCFD 0 171 N.A. 10 (MMSCFD) Comments (achievements against targets): Formally Revised Actual Achieved at 2. Indicator Name Unit of Measure Baseline Original Target Target Completion Commissioning of project according to budget Yes/No No No N.A. Yes (Yes/No) 24 The World Bank Ghana Sankofa Gas Project (P152670) A. KEY OUTPUTS BY COMPONENT Objective/Outcome 1 1. Quantity of Gas Supplied to Power Plants (MMSCFD) 2. Private Sector Capital Mobilized (US$ million) Outcome Indicators 3. Indirect Project Beneficiaries (million) 4. Female Beneficiaries (%) 1. Gas Production Capacity Achieved by the Project (MMSCFD) Intermediate Results Indicators 2. Commissioning of project according to budget (Yes/No) Key Outputs by Component Project had a single component: supply of non-associated gas to (linked to the achievement of the Objective/Outcome 1) power plants (MMSCFD) 25 The World Bank Ghana Sankofa Gas Project (P152670) ANNEX 2. BANK LENDING AND IMPLEMENTATION SUPPORT/SUPERVISION A. TASK TEAM MEMBERS Name Role Preparation Sunil W. Mathrani Co-Task Team Leader Robert Schlotterer Co-Task Team Leader David Santley Senior Petroleum Specialist Manuel Luengo Senior Energy Specialist Thomas E. Walton Environment Specialist Demba Balde Social Safeguards Specialist Rita Ahiboh Team Member Chita Azuanuka Obinwa Team Member Ayishetu Terewina Team Member Shingira Samantha Masanzu Senior Counsel Monica Restrepo Counsel Vincent Francois Jean Launay Infrastructure Finance Specialist Supervision/ICR Shinya Nishimura Co-Task Team Leader Robert Schlotterer Co-Task Team Leader Jukka-Pekka Strand Co-Task Team Leader David Ferrenbach Vilar ICR Task Team Leader Vincent Francois Jean Launay Infrastructure Finance Specialist Thomas E. Walton Environment Specialist Asferachew Abate Abebe Environment Specialist Alidu Babatu Adam Social Safeguards Specialist Ayishetu Terewina Team Member Marie-Claudine Fundi Team Member Demba Balde Social Safeguards Specialist Gloria Mahama Social Specialist 26 The World Bank Ghana Sankofa Gas Project (P152670) Rita Ahiboh Team Member Chita Azuanuka Obinwa Team Member Shingira Samantha Masanzu Counsel Monica Restrepo Counsel Sunjung Melissa Kim Counsel Charles John Aryee Ashong Procurement Specialist Robert Wallace DeGraft-Hanson Financial Management Specialist Anas Benbarka Team Member Anthony Molle Team Member Collins S. Umunnah Team Member Dante Ariel Mossi Reyes Team Member Lauren Claire Culver Team Member Adwoa Asantewaa ICR Team Member Sunil W. Mathrani Team Member Zhengjia Meng ICR Task Team Leader 27 The World Bank Ghana Sankofa Gas Project (P152670) B. STAFF TIME AND COST Staff Time and Cost Stage of Project Cycle No. of staff weeks US$ (including travel and consultant costs) Preparation FY15 53.66 266,627.28 FY16 38.18 192,746.42 Total 91.84 459,373.7 Supervision/ICR FY17 51.97 255,881.35 FY18 48.31 257,154.76 FY19 26.12 131,573.78 FY20 9.76 54,056.38 Total 136.16 698,666.27 28 The World Bank Ghana Sankofa Gas Project (P152670) ANNEX 3. PROJECT COST BY COMPONENT Amount at Approval Actual at Project Percentage of Approval Components (US$M) Closing (US$M) (US$M) Total 4,800 0.00 97% 29 The World Bank Ghana Sankofa Gas Project (P152670) ANNEX 4. EFFICIENCY ANALYSIS 1. This annex presents the economic28 analysis of the Implementation Completion and Results Report (ICR) of the Ghana: Sankofa Gas Project (SGP). It is a cost-effectiveness analysis to determine whether the project represented the expected least-cost solution to provide gas to the Ghanaian electricity sector. During project appraisal, the Net Present Value (NPV) and Economic Rate of Return (ERR) were used to determine the economic and financial viability of the project and remains the main metrics of assessment in this analysis. These indicators are re-estimated based on actual project costs and benefits and modified with relevant macroeconomic and institutional assumptions. The original economic and financial models used in the appraisal analysis are also used for this completion analysis to allow for an effective comparison of both results. PROJECT RATIONALE AND DEVELOPMENT IMPACT 2. The SGP was a two-phased oil and gas development project with Phase 1 being the development of the Sankofa East oil field and Phase 2 being the development of the Sankofa Gas Field (SGF). Each phase had a separate commercial arrangement, and World Bank support was for the development of the SGF only. The envisioned support was in the form of a US$700 million guarantee package, which comprised a US$500 million IDA Guarantee and a US$200 million IBRD Enclave Loan Guarantee to mitigate against GNPC off-take and payment risks, and long-term political risks in Ghana. This support underpinned an US$7.9 billion private sector investment by Eni of Italy and Vitol of the Netherlands. 3. The Project Development Objective (PDO) of SGP was to “Increase the Availability of Natural Gas for Clean Power Generation by Leveraging Private Capital Investment”. The rationale was that gas from the Sankofa field would replace expensive and polluting liquid fuels currently used for electricity generation in Ghana while enabling new generation capacity to be constructed. The field, which is expected to be in production for twenty years would provide baseload gas to fuel up to 1000 MW of power plant capacity. At the time of appraisal, significant Independent Power Production (IPP) projects were underway in Ghana and an additional IPP capacity of over 2,500 MW was expected to be added to the generation base by 2020. This corresponded to a gas demand of 373 MMscfd in the electricity sub-sector alone. Based on these projections, it was projected that even with the development of SGF, a gas deficit would persist at the national level, making it necessary to develop additional sources of gas supply. 4. In addition to presenting specific benefits to the electricity sub-sector, the SGP had significant macroeconomic merits. At the time of project preparation, the liquid fuel import bill in Ghana had increased by about US$27 million per month when gas was unavailable. In addition to the project being a new source of government revenue, it presented far-reaching benefits to the broader macro economy through multiple channels in the real sector, fiscal accounts and balance of payments. However, these economic benefits are not included in the economic analysis of the project given the complexity of capturing these effects. PROJECT COSTS 5. The economic costs of the project included the total capital costs and the operation costs. During appraisal, the total capital cost of the project was estimated at US$7.9 billion dollars. These costs included capital costs for both the oil and gas infrastructure and was based on tenders already received by the project’s private sponsors for 28The economic analysis is consistent with the following guidelines: (a) World Bank IPF Policy and Directive (b) Power Sector Policy and Investment Projects: Guidelines for Economic Analysis, (c) Discounting Costs and Benefits in Economic Analysis of World Bank Projects 2016. 30 The World Bank Ghana Sankofa Gas Project (P152670) contracts including Floating Production Storage and Offloading unit (FPSO), and Subsea Pipeline System and drilling costs contracts. The total capital costs also included a 20-year lease cost for FPSO of US$2.4 billion, and the investments planned in 2026 and 2028 to maintain gas production at the plateau level. However, the operating and maintenance costs of the FPSO estimated at more than US$100 million per year are recorded as operating expenditures in the income statement. As oil production was expected to start in the third quarter of 2017, the planned investments in 2026 and 2028 would be self-financed by oil revenues. The total capital expenditures for the project was US$4.65 billion for both project phases. World Bank Support 6. To secure this investment, a comprehensive set of risk mitigation and intervention structures were put in place to enhance the creditworthiness of GNPC as the offtaker. These included the (a) payment of GNPC receivables into a segregated account to assure the prioritization of payment for Sankofa gas purchases (b) a cash reserve escrow account, (c) a Letter of Credit (LC) backstopped by a World Bank IDA Payment Guarantee; and (d) a limited Sovereign Guarantee. Table 1 presents an overview of this security package. As at project closing, Layer 1 of the security package had not been triggered as it was never operationalized, and Layer 2 had been triggered and exhausted as GNPC did not fully fund the escrow account. Also, ENI and Vitol drew down US$192 million of the LC on April 7, 2020. Table 1: Overview OF Risk Mitigation and Intervention Structures for Sankofa Gas Project Security Description Remarks Layer 1: Government This Channels all revenues from the on-sale of gas from Sankofa, The GDA has been Disbursement Account Jubilee and TEN as well as GNPC's share of Net CAPI into a set up but not yet (GDA) Government Disbursement Account (GDA) from which the GSA been payments, GNPC's debt service and the replenishment of the operationalized at Jubilee and TEN escrow accounts will be made on a pari passu the Bank of Ghana. basis. Following these disbursements, the remaining funds are expected to be used to replenish the Sankofa reserve escrow cash account, if needed, and residual funds released to GNPC. Layer 2: Reserve This account was to be funded by GNPC to the tune of US$210 GNPC did not fully escrow account million equivalent to 4.5 months of gas sales. These funds would fund the reserve be available if the proceeds from the GDA are insufficient. The escrow. It only required amount in the escrow would be reduced to 3.5 months provided an initial in the absence of default under the GSA during the first 5 years US$100 million for of operation. the account. This was depleted after the first invoices and has not since been replenished. Layer 3: US$500 million The LC would be accessible to the Private Sponsors once the first ENI and Vitol drew Commercial Bank LC two security layers have been exhausted. This covers close to down US$192 backstopped by IDA one year of gas sales under the GSA and is backstopped by an million of the LC in IDA Payment Guarantee to repay the LC issuing bank for April 7, 2020 31 The World Bank Ghana Sankofa Gas Project (P152670) amounts drawn by the private sponsors 12 months after the draw-down. Layer 4: Sovereign The GoG would be liable for gas payments estimated around This is yet to be Guarantee US$600 million per year on average over the life of the GSA once triggered all these layers of security are exhausted. 7. Alternate Supply Options. At appraisal, the gas price negotiated in the GSA was US$9.80/MMBtu but with a levelized net economic cost of US$6.60/MMBtu (2014 US$), i.e. after considering direct and indirect revenues to the GoG. The main alternatives to SGP at the time were imported liquid fuels whose costs were estimated at US$12/MCF and LNG imports at US$10/MCF, both of which had a higher economic and financial cost than Sankofa. Gas imports from Nigeria at US$8/MCF at the time was competitive with SGP but had significant security of supply concerns given growing demand in Nigeria and a past history of unreliability of supply. This made Sankofa gas the cheapest and most reliable additional fuel source for power generation in the country. 8. By project closing, some lessons had been learned about the structure of the Ghanaian gas market which threatened the position of Sankofa as a baseload gas supplier. The institutional framework of the Ghanaian gas sector is not fully developed, and this has created a misalignment of interests among gas sector institutions. There is considerable overlap and inefficiencies in the sub-sector, with the GNPC, GNGC, and VRA each acting as gas suppliers in a market that is structurally not able to sustain a competitive market. GNPC sells gas purchased from the Sankofa field, GNGC sells gas purchased from the TEN and Jubilee fields, and VRA sells gas purchased from NGas. Each of these SOEs make decisions about which gas supply to purchase in line with their individual best interest, i.e. the lowest cost to them. Given that Ten and Jubilee are associated gas sources with an economic marginal cost of zero, it is clearly more competitive in the market as any positive price is viable for this source29. As a result, Sankofa, as the only non-associated gas source in the country is in a weaker competing position in such a market environment. Notwithstanding, with the Take or Pay (ToP) obligations of Sankofa, any displaced ToP quantity by any supply source does in fact impose an opportunity cost of the Gas price in the GSA. Thus, these individual decisions made by these market actors are not necessarily in the best economic and financial interest of the country. 9. Offtake Volumes. Offtake of gas from Sankofa did not begin until August 2018. This was in large part because of delays in the construction of a critical pipeline infrastructure to transport gas from the SGF in Takoradi (in the Western region of Ghana) to Tema in the Eastern part of the country where most operational power plants were located. As of project closing, the infrastructure had been completed successfully and the 450 MW Karpowership, which was a mobile power plant, had been relocated to Takoradi to be closer to the SGF. As a result, off-taken volumes increased significantly at the end of 2019 as can be seen in Figure 130. 10. However, offtake volumes remain very volatile both intra-day and inter-day due to constantly changing offtake nominations from GNPC resulting from the availability of other gas supply sources and the availability of the thermal power plants for dispatch. Since the downstream gas supply shutdown of the other sources in April 2019, offtake of Sankofa gas has increased significantly. However, this cannot be considered as the steady state nominations as this could change again once supply from other sources come back online. 29 However, associated gas volumes from JUB/TEN in 2019 have been limited by the need to re-inject the gas to maintain oil supply volumes. 30 Daily supply nomination shows what was nominated to the buyer, i.e. what the operators are able to deliver to the buyer, on any given day. These quantities were different from contractual quantities. “Taken Quantity” shows what was off tak en by the buyer in any given day (i.e. average volume taken for a given day). This has edged above 150 mmscfd a few times. 32 The World Bank Ghana Sankofa Gas Project (P152670) OCTP Volume 250.000 200.000 150.000 100.000 50.000 0.000 11. In the GSA, there is a provision for any gas supplies not taken in earlier years to be banked for five years. Thus, gas that was not taken in 2018 and 2019 remain available until the end of 2023 and 2024, respectively. However, given that the maximum production capacity of SGF is 220 MM'scfd, Ghana may be able to absorb all its banked gas over the next five years if gas demand is sufficiently high. In the absence of any new policy direction to remove the perverse market incentives that have emerged in the gas supply market, this analysis takes a conservative position that gas supplied from Sankofa over the short to medium term would not exceed the ToP obligations and could in fact be lower given historical precedence. Thus, the economic analysis undertaken is based on the offtake projections made at appraisal from 2020 onwards. 12. Crude oil prices were an important determinant of the project viability given the indexation of LNG prices to crude oil prices and the shared capital assets by both the Phase 1 and Phase 2 of the project. In the appraisal analysis, the 2015 oil price forecast was used but this completion analysis uses the most recent 2020 oil price forecast by the World Bank as presented in Table 2. Table 2: Crude Oil Price Projections in 2015 and 201931 Year 2015 Oil Price 2020 Oil Appraisal Actual offtake Actual offtake Projections Price Offtake Volumes and Volumes and Projections Volumes Appraisal forecast forecasts at closing (US$) (Conservative) (Optimistic) 2018 60 68.3 167 41 41 2019 63 61.7 180 89.5 89.5 2020 66 35.4 180 171 171 2021 69 41.8 180 180 200 2022 72 43.5 180 180 200 2023 76 45.2 180 180 220 2024 76 45.2* 180 180 220 31 * represents extrapolation estimates based on price trend of projections. 33 The World Bank Ghana Sankofa Gas Project (P152670) 2025 76 48.9 180 180 220 2026 76 50.9* 180 180 220 2027 76 52.9* 180 180 220 2028 76 54.9* 180 180 220 2029 76 56.9* 180 180 220 2030 76 59 180 180 220 Methodology 13. The economic analysis follows a standard cost-benefit framework, which compares the present value of incurred costs to the stream of attributable benefits. The EIRR and NPV of the project inform the project’s viability over its economic lifetime. The economic benefits of the project are assessed using the avoided-cost approach. In this, the project benefits are quantified as the economic costs of displaced fuels, i.e. light crude oil until 2021 and LNG from 2022 onwards. An economic discount rate of 10 percent is used in the analysis as was the case at appraisal. The main revisions in the economic analysis in the ICR analysis were the revision of project costs to reflect the project cost savings, the revision in crude oil price forecast by the World Bank (using the April 2020 forecast), and the actual and forecasted offtake volumes. Results 14. Under the 2020 oil price forecast, the NPV and the ERR of the project were estimated at US$1.3 billion and 14 percent respectively. These were 6.2 percent and US$2.7billion lower than the appraisal estimates. Table 3 presents the detailed results of the economic analysis at appraisal and completion. Details of the sources change in the NPV and ERR are in Table 4. Table 3: NPV and ERR of Sankofa at Appraisal and Completion Including Sunk Costs Appraisal Appraisal Closing NPV $4.0bn $1.3 bn ERR 20.2% 14% Table 4: Cumulative change in project NPV and ERR at Project Closing ERR Change NPV Change Appraisal 20.2% $4.0bn Capital Cost Effect 21.6% +1.4% $4.3bn +$0.3bn Crude Oil Price Effect 15.6% -6% $1.7 bn -$2.6bn Offtake Volumes Effect 14.0% -1.6% $1.3 bn -$0.4bn Closing 14.0% $1.3 bn Sensitivity Analysis 34 The World Bank Ghana Sankofa Gas Project (P152670) 15. Economic Impacts of Covid-19 Pandemic. One may reasonably expect a substantial global recession in the aftermath of the Covid-19 Pandemic, but at the time of writing in May 2020, its depth and duration cannot yet be determined. A significant fall in international energy prices is expected, with the spot prices of LNG having already fallen to record low levels, although this cannot be fully attributed to the COVID-19 emergency. A World Bank assessment of the economic impacts of the pandemic categorized Ghana’s exposure as high, based on its commodity trade exposure and its trade with China. However, there is an anticipated decline in economic growth as a result of reduced exports, tourism and FDI– which in turn makes a reduction in electricity demand (and gas) likely. This could lead to a further reduction in off-take volumes and a reduction in the project NPV and ERR. 16. Low Crude Oil Prices. In addition to the ongoing pandemic, the collapse in crude oil prices in Q1 2020 is creating significant challenges for oil and gas projects like Sankofa. Given the uncertain nature of the crude oil price, two oil price sensitivity scenarios were tested. In the low case scenario, oil prices are capped at US$45/bbl till the end of the project life while in the high case scenario, oil prices bounce back to US$50/bbl already in 2021 and prices recover to the base case after 2026. Under the lower oil price scenario, the ERR of the project is reduced to 13 percent and the NPV to US$0.9 million. In the higher oil price scenario, the ERR of the project remains unchanged at 14 percent but the NPV of the project increases to US$1.4 billion. 17. Financial Analysis. SGP was designed to ensure that the financial returns are safeguarded. The multi-layered payment security in place and the fact that the prices are fixed contractually ensures that the revenues of the project are assured over the project life. As a result, the FIRR and the FNPV of the project remain unchanged despite these changes. However, delays in payments by GNPC affects the cashflow of the company. However, we do not have adequate information at this point to assess the actual impacts of these payment delays on the company’s balance sheet. 35 The World Bank Ghana Sankofa Gas Project (P152670) ANNEX 5. BORROWER, CO-FINANCIER AND OTHER PARTNER/STAKEHOLDER COMMENTS32 COMMENTS FROM GOVERNMENT OF GHANA (received on June 22, 2020) 1. Introduction The report represents the World Bank’s (WB) internal evaluation of the Sankofa Gas Project. The measurement criteria and methodology are the World Bank’s and the GoG is unable to provide comments on the evaluation results without the full benefit of appreciating the details of the methodology. All the major issues identified in the ICR which relates to the GoG or its institutions are currently being addressed via the Energy Sector Recovery Program (ESRP), which is a four-year roadmap to reach structural balance in the energy sector. Implementation of the ESRP will ensure that Ghana’s credit rating is not downgraded, and that Ghana continues to attract private investment in its infrastructure projects on competitive terms. Nonetheless, GoG finds the report very useful. The lessons and recommendations should serve as guide for future operations for the benefit of the gas sector in particular and for mega project executions in Ghana, in general. Our comments are in two categories; 1. General comments – this section provides our general comments on the WB’s assessment of some key aspects of the project and 2. this section is specific on some paragraphs, to draw attention to what may be factual inaccuracies. 2. General Comments Project Outcomes The Sankofa gas project was successfully completed below budget and largely on time. It has provided much-needed stability in gas supply, which was not available when only Jubilee and TEN were supplying gas. As a result, the power sector has become more reliable and outages have been substantially reduced. Switching to gas has also reduced fuel costs and the need to maintain liquid fuel inventories. Some non-power sector industrial customers are planning to move from more expensive fuels to gas. Overall, the Sankofa project can be viewed as a success story for Ghana, with Sankofa gas becoming the backbone gas supplier for the power sector. Project Results After completion of the Takoradi Phase of the TTIP project and the relocation of the Karpowership from Tema to Takoradi in 2019, Sankofa gas volumes offtake has seen an increase in 2019 and early 2020. Total gas consumption in Ghana has increased to slightly more than 310 MMscfd as at mid-2020. Project Issues The issues enumerated in the ICR are artifacts of long-standing liquidity issues with origins in the power sector. The Energy Sector Recovery Program (ESRP) was proposed in 2018 and then approved by Cabinet in 2019. It has been actively pursued by the GoG to bring financial balance to the legacy of a structurally challenged energy sector. When completely implemented by 2023, the ESRP will have addressed the financial and structural issues in the energy sector. These activities, when completed, should bring stability to the payment challenges that have arisen on the Sankofa gas project. The implementation of the Cash Waterfall Mechanism (CWM), a recommendation arising from the ERSP, began in April 2020 and has been a significant improvement in disbursement of energy sector revenues in a fair manner to all sector players 32The ICR team has revised some paragraphs in the final version of the ICR to address some of the comments received by GoG and ENI/VITOL 36 The World Bank Ghana Sankofa Gas Project (P152670) Payment Security Package General Disbursement Account now called Government Disbursement Account (GDA) has been set up. GNPC is currently engaging the partners on how to operationalize it. Non-replenishment of Reserve Escrow Account (REA) by GNPC This is largely due to the liquidity challenges of the gas-to-power sector referred to above and will be addressed through the implementation of the ESRP. Drawdown of LC backstopped by World Bank guarantee The Government of Ghana has been making deposits and intends to repay the full loan amount within the 12-month period. Lack of Clarity on Institutional and Offtake Arrangements Government through the Ministry of Energy is currently consulting with the relevant stakeholders to streamline the roles of the gas sector institutions by the end of this year. Underestimation and Delays in WAGP Interconnection This is largely attributable to delays in reaching consensus on certain agreements which were required to allow the contractors to undertake the expansion works in some of the facilities. For example, the Construction Management Agreement (CMA) between Eni and WAPCo and, Eni and GNGC took several months to conclude, whilst, the GTA between WAPCo and GNPC took a couple years. Challenges and Many of the challenges faced by the GoG in the Sankofa gas project stem from the fact Lessons Learned that the country was emerging from a power crisis which disrupted the smooth flow of economic activity. The country has successfully emerged from these crises and is in a stronger position to support projects of this nature in the coming years. 3. Specific Comments Paragraph Statement/ inference from Report GoG’s Comment Para 1 Successive, peaceful transitions of Ghana's return to multi-party democracy was in democratic power in Ghana since ‘1990’ 1993. Elections held in1992 and the new provided a supportive environment for constitution took effect on Jan. 7 1993. continued economic growth. Para 5; The ‘contracted’ daily gas supply through The contracted daily gas supply through WAGP footnote 2 WAGP, most of which was supplied to is123 MMscfd Ghana, was ‘133’ MMscfd. Para 25 “However, the first gas supplies did not First gas supplied from the Sankofa field take place until ‘November’ 2018”. commenced in August 2018. Thus, 6 months behind schedule. Quantity of Gas supplied to Power Plants It should be noted that, the offtake volumes recorded in December 2019 and January 2020 were due to specific challenges with evacuating gas from the Jubilee/TEN Fields (Jubilee planned shutdown, TEN pipeline blockage and remediation) and subsequent GNGC pipeline blockage and remediation. This should have very little impact on the projections for 2020 daily average. 37 The World Bank Ghana Sankofa Gas Project (P152670) Para 27 PDO Indicator 3: Indirect Project The PDO target on female beneficiaries appears very Beneficiaries difficult to measure and marking it as 100% PDO Indicator 4: Female Beneficiaries ‘achieved’ without quantifiable evidence makes it Achievement is 100% appear a bit arbitrary. Para 28 In 2018 and 2019, actual off-take volumes In 2018, the daily average for the year were 126mmscfd and 90.5mmscfd was 40.74mmscfd and the highest recorded daily respectively. rate that year was 90.64mmscfd. Para 66 “The conclusion was that with gas flows up The Quantitative Risk Assessment report led to the to 60mmscfd and at low pressure, there descoping of the Tema RMS capacity from the would be no risk to the surrounding originally planned 345 mmscfd to 235 mmscfd. The population. Gas flows to Tema during 2019, current flows from both Takoradi and Tema are well one year after the project closing date above the 60 mmscfd. didn’t reach 60mmscfd”. 38 The World Bank Ghana Sankofa Gas Project (P152670) COMMENTS FROM ENI AND VITOL (received on June 23, 2020) Eni and Vitol agree with the key contents of the report and thank the World Bank for the productive engagement in its preparation. The fruitful cooperation between the Government of Ghana, the World Bank, Eni, Vitol and Ghana National Petroleum Corporation has been crucial for the implementation of an unprecedented project in Sub-Saharan Africa: the only deep-water non-associated gas development project entirely dedicated to the domestic market in the region, supporting Ghana’s shift from oil-fueled power generation to a cleaner power source, with economic as well as environmental benefits for the country. Thanks to Sankofa, Ghana is able to count on at least 15 years of stable, reliable, affordable gas supplies, as well as providing a solid foundation for local development, energy independence and an expanding role as an energy supplier for the region. The cooperation among the key stakeholders mentioned above has allowed the project to overcome significant hurdles in arriving to its current state. The ongoing vision and commitment of the Government of Ghana are critical to maintain a framework conducive to investment and to ensure the long-term success of the project and its contribution to the achievement of the Sustainable Development Goals. We confirm our commitment to support the Government of Ghana and the World Bank in the resolution of the outstanding issues described by the Report, in order for the project to achieve its full potential. Paragraph 14 and 31 There is a distinction to be made between the different components of World Bank Group’s support to the Sankofa project. While some of them are for the benefit of both Project sponsors (namely the IDA Guarantee and the IBRD Enclave Guarantee, if it had been entered into), others are for the exclusive benefit of Vitol, namely the IFC financing and the MIGA political guarantee. Paragraph 25. PDO Indicator 1: Quantity of Gas supplied to Power Plants has been assessed as partially achieved on the basis of lower than expected consumption of Sankofa gas. It’s important to emphasize that viewed from a supply perspective, the Sankofa project has been an outstanding success and would be considered, in our view, to have achieved the objective. Sankofa has been able to supply gas at and beyond contractual rates from start-up. In fact, in the nearly 2 years since start-up, Sankofa has achieved reliability performance figures exceeding 99% percent, with the project now consistently making available, to the Country, gas at a rate of 220 MMscfd. Ghana’s gas and power system has benefitted enormously from the consistent availability of supply by relying on Sankofa gas to compensate for continued supply volatility of associated gas from Jubilee and TEN fields, and piped gas from Nigeria. Overall consumption of Sankofa gas has, on the other hand, been constrained by several factors principally including delays in completing some of the government infrastructure projects identified at project inception as being critical to support the utilization of Sankofa gas. These projects include installation of additional power capacity in the Takoradi region and completion of pipeline infrastructure to transport gas from the Takoradi region to Tema. DRAFT – SUBJECT TO REVIEW 18/6/2020 CONFIDENTIAL Paragraph 29, 45, 66, 69 The Sankofa project was comprised of upstream activities undertaken by the private sponsors, plus a number of Ghanaian infrastructure projects to connect the Sankofa gas supplies to thermal power plants in both the western and eastern regions. For the sake of clarity, the upstream activities include the complex offshore offshore investment, offshore gas export pipeline, and onshore receiving facilities. The pipeline infrastructure to connect the onshore receiving facility to thermal power plants and to transport gas to gas in eastern markets falls under the scope of the infrastructure projects. 39 The World Bank Ghana Sankofa Gas Project (P152670) Paragraphs 35, 38, 41, 43, 47, 71, 84 The report makes several references to establishment of the GDA and to the roles of GNPC and other state entities in the gas sector. The key point as to the GDA (and the other associated escrow accounts comprised in Layer 1 of the security package) is not establishment, which occurred at project inception, but the fact that GNPC and the Government of Ghana have still not made the arrangements required by the security package contracts to procure the deposit in this account of (a) revenues from the on-sale of gas from OCTP, TEN and Jubilee, and (b) disbursements made to GNPC from the Petroleum Holding Fund (essentially reflecting GNPC’s oil revenues from its carried and participating interests, net of GNPC’s equity financing costs). These contractual obligations have been thoroughly discussed and negotiated between GNPC, the Government of Ghana and the Project sponsors in order to comply with applicable Ghanaian legislation and to reflect the standard practice of GNPC, the Government of Ghana and the Bank of Ghana. Detailed provisions governing the deposit in the GDA and the escrow accounts are reflected in the Agreed Payment Process set out in the Multi Party Deed. Since the security package agreements have been entered into, the Project sponsors have constantly requested GNPC and the Government of Ghana to comply with such contractual obligations and procure that the agreed revenues are deposited in the escrow accounts, highlighting how these arrangements were and are critical to the long-term sustainability of the Project. The report also mentions GNPC’s lack of liquidity and the fact that it does not act as gas aggregator, however there is no clear linkage between these two issues and the problems concerning the escrow accounts and the payment for gas. The report should clarify, firstly, that when the project was sanctioned, it was a fundamental element of the agreed security package that GNPC would act as the gas aggregator in respect of all gas in Ghana. This requirement as to GNPC’s role was reflected in the security package agreements and formed the basis for the establishment of the GDA and the Agreed Payment Process. The report should also clarify that the deposit of GNPC’s gas on-sale revenues in the GDA, which was specifically designed to support the liquidity of GNPC and enhance its ability to pay for the Sankofa gas, hinges on GNPC acting as gas aggregator. The effectiveness of this payment mechanism is affected by GNPC’s reduced role and should be assessed in this context. Addressing these issues remains critical to resolve the ongoing payment issues and mitigate the payment risk relating to the Sankofa gas, and the Private sponsor are constantly engaged in discussions with GNPC and the Government of Ghana to that purpose. DRAFT – SUBJECT TO REVIEW 18/6/2020 CONFIDENTIAL Paragraph 32 According to the report, the IBRD Enclave guarantee was not eventually granted because the Project sponsors were able to mobilize sufficient capital to finance the project without recourse to this guarantee. While we concur with this recollection, we also believe that the report should clarify that the signing of the IBRD Enclave guarantee was subject to the funding the Reserve Escrow Account in full and the implementation of the arrangements to deposit funds into the escrow accounts. As detailed in other sections of the report, these requirements were never complied with in full by GNPC and the Government of Ghana. Because of such breach by GNPC and the Government of Ghana of their contractual obligations regarding the security package, it would not have been possible for the Project Sponsors to enter into the IBRD Enclave Guarantee irrespective of their ability to mobilize sufficient capital to complete the Project. Paragraph 47 We would like to highlight that in addition to the initial funding of US$100 million, other subsequent deposits were made by the Government of Ghana into the Reserve Escrow Account in order to support payments for Sankofa gas. Nonetheless, it should still be noted that none of this deposits was at any time remotely sufficient to meet the funding obligations of GNPC and the Government of Ghana in relation to the Reserve Escrow 40 The World Bank Ghana Sankofa Gas Project (P152670) Account. We therefore still concur with the general statement in the report that “[t]hese obligations were either not met or, in the case of the [Reserve Escrow Account], met only partially”. Paragraph 74 In reference to the following statement, it should be clarified that the gas interconnector has been completed and gas has been flowing from the western region to Tema since April 2019. However, even though these risk continued to be highlighted in subsequent supervision documents through closing, only minimal progress on the construction of a gas interconnector and on improvements to the financials of ECG and VRA had been made by project closing. Paragraph 86 We support the comments around the need for a collaborative communication strategy, to articulate the significant economic benefits of the country’s first natural gas development. In particular, this communication strategy would explain the structure of the Sankofa gas price, and the difference between the headline price and the net cost of Sankofa gas, which as highlighted in Annex 4, paragraph 8 of the report, is lower cost than imported fuels including liquids, LNG and pipeline gas from Nigeria. In reference to the statement “associated gas produced from the nearby Jubilee field was not being charged to consumers…”, while Jubilee foundation gas was supplied at zero upstream cost, significant gathering and transfer charges are applied before delivery to the end-user. 41 The World Bank Ghana Sankofa Gas Project (P152670) ANNEX 6. SUPPORTING DOCUMENTS (IF ANY) 42