Kingdom of Eswatini Energy Masterplan 2034

1.1 BACKGROUND  18

1.2 OBJECTIVE AND SCOPE OF THE ENERGY MASTERPLAN 18

1.3 IMPLEMENTATION OF THE STUDY FOR THE MASTERPLAN DEVELOPMENT 19 2 PRESENT ENERGY POLICY SETTING 20

2.1. NATIONAL ENERGY POLICY OF 2003 20

2.2 ELECTRICITY LEGISLATION  20

2.3 SUSTAINABLE ENERGY FOR ALL GOALS 21

2.4 ESWATINI NATIONALLY DETERMINED CONTRIBUTION  22

2.5 REGIONAL ENERGY CONTEXT 22

2.6 CURRENT INSTITUTIONAL FRAMEWORK FOR ENERGY AND ELECTRICITY PLANNING  23

2.6.1. Energy statistics 24

2.6.2 Electricity sector 24 3 ENERGY BALANCE: RETROSPECTIVE AND GENERAL OVERVIEW 26

3.1 ENERGY BALANCE FOR 2014 26

3.1.1 Total primary energy supply 27

3.1.2 Transformation of primary energy 28

3.1.3 Total final energy consumption 28

3.2 TRENDS IN ENERGY SUPPLY, TRANSFORMATION AND USE 30

3.2.1 Total primary energy supply  30

3.2.2 Transformation of primary energy 30

3.2.3 Total final energy consumption  32 4 KINGDOM OF ESWATINI PART II: ENERGY DEMAND ASSESSMENT 35 4 ENERGY DEMAND ASSESSMENT 2015–2034: METHODOLOGY AND ASSUMPTIONS 36

4.1 BRIEF DESCRIPTION OF THE METHODOLOGY 36

4.2 KEY SCENARIO ASSUMPTIONS  37

4.2.1 Climate 38 4.2.2 Demography and major demographic policy goals  38 4.2.3 Macroeconomic situation and major economic policy issues 39

4.2.4 Energy efficiency goals  44 5 ENERGY DEMAND ASSESSMENT 2015–2034: RESULTS 45

5.1 OVERALL ENERGY DEMAND DEVELOPMENT: 2015–2034 45 5.1.1 Energy demand development by 2034 in the reference scenario  45 5.1.2 Energy demand development by 2034 in alternative scenarios  47

5.2 ENERGY DEMAND DEVELOPMENT IN KEY SECTORS  48

5.2.1 Agriculture  48

5.2.2 Sugar industry 48

5.2.3 Other industry  49

5.2.4 Commerce and government services 49

5.2.5 Transport  50

5.2.6 Residential  51 PART III: ENERGY SUPPLY ASSESSMENT 55 6 DOMESTIC ENERGY RESOURCES AND ENERGY TECHNOLOGY OPTIONS 56

6.1 SUGARCANE  56

6.1.1 Bagasse 57

6.1.2 Molasses 57

6.2 WOODY BIOMASS 59

6.2.1 Wood chips 59

6.2.2 Fuelwood 59

6.3 COAL 60

6.4 WIND 61

6.5 SOLAR PV 65

6.6 HYDROPOWER  67 ENERGY MASTERPLAN 2034 5 7 ENERGY SUPPLY INFRASTRUCTURE 68

7.1 FOSSIL FUEL PRODUCTS  68

7.2 ELECTRICITY SUPPLY SYSTEM 68

7.2.1 Generation 69

7.2.2 Imports 70

7.2.3 Transmission and distribution 72

7.3 BIOFUEL SUPPLY SYSTEM  75 8 ENERGY SUPPLY PLANNING SCENARIOS FOR 2014–2034 76

8.1 BRIEF DESCRIPTION OF THE METHODOLOGY 76

8.2 SCENARIO DEFINITION: KEY POLICY ISSUES TO BE ADDRESSED 76

8.2.1 The Base Case  77

8.2.2 Forced Coal scenarios  77

8.2.3 Domestic Resources scenarios 77

8.2.4 No Import Enhancement scenarios  78

8.2.5 Low Import Price scenario 78

8.2.6 Natural Gas scenario 79

8.3 MODELLING OF THE ESWATINI ENERGY SYSTEM 79

8.3.1 Model configuration 79

8.3.2 Reference energy systems 80 8.3.3 Definitions of seasons and daily load representation 83

8.3.4 Key technology parameters 86

8.4 MODELLING OF ESWATINI ENERGY SYSTEM: KEY MODEL RESULTS  90

8.4.1 Base Case results: power sector 90

8.4.2 Alternative case results: power sector 97 8.4.3 Comparative assessment of scenarios: power sector 107

8.4.4 Primary energy requirement  111 9 VISION FOR 2034 113

9.1 KEY INSIGHTS FROM THE SCENARIO ANALYSIS  113

9.2 RESULTS OF THE STAKEHOLDER CONSULTATIONS  117

9.3 WAY FORWARD: KEY POLICY GAPS 118

9.4 ENHANCING NATIONAL ENERGY PLANNING CAPABILITY  119 ANNEX121 ANNEX A MEMBERS OF THE NATIONAL WORKING TEAM; TRAINERS AND KEY CONTRIBUTORS 121 ANNEX B DETAILED RESULTS OF THE DEMAND ASSESSMENT 122 6 KINGDOM OF ESWATINI FIGURES Figure 2.1 Planning process flow 25 Figure 3.1 Total primary energy supply in 2014 27 Figure 3.2 Fuel mix for electricity and heat production in 2014 28 Figure 3.3 Final energy use by sector in 2014 29 Figure 3.4 Trends in total primary energy supply 30 Figure 3.5 Trends in input fuels for heat and electricity generation  31 Figure 3.6 Trends in the electricity mix 31 Figure 3.7 Trends in installed capacity 32 Figure 3.8 Trends in final energy consumption by sector 32 Figure 3.9 Trends in final energy consumption by fuel 33 Figure 4.1 Population projections, 2010 to 2034 38 Figure 4.2 Trends in GDP between 2000 and 2016 (in 2014 constant prices) 39 Figure 4.3 Trends in GDP by sector between 2000 and 2016 (in 2014 constant prices) 40 Figure 4.4 Assumptions regarding the composition of GDP 41 Figure 4.5 Three GDP scenarios for 2014 to 2034 41 Figure 4.6 Electrification projections for 2014 to 2034 42 Figure 4.7 Share of household fuel used for cooking 43 Figure 5.1 Final energy demand by sector, 2014–2034 45 Figure 5.2 Final energy demand by fuel type, 2014–2034 46 Figure 5.3 Energy demand growth under the three scenarios 47 Figure 5.4 Energy demand in the agriculture sector 48 Figure 5.5 Energy demand in the sugar industry 48 Figure 5.6 Energy demand in other industry 49 Figure 5.7 Energy demand in commerce and government services 49 Figure 5.8 Energy demand in the transport sector 50 Figure 5.9 Energy demand in the residential sector 51 Figure 5.10 Energy use in non-electrified households 52 Figure 5.11 Energy end-use in electrified households 53 Figure 5.12 Energy use in electrified households 53 Figure 6.1 Process of ethanol production from sugarcane molasses 58 Figure 6.2 Locality map of the coal field of Eswatini 60 Figure 6.3 Location of the five priority zones for wind power 62 Figure 6.4 Average day wind capacity factors for each month, zone J (values from 2015) 63 Figure 6.5 Average day load profiles for each month (values from 2013) 64 Figure 6.6 Solar PV zones in Eswatini 65 Figure 6.7 Average day solar PV capacity factors for each month, zone B (values from 2004) 66 Figure 7.1 Eswatini electricity industry market structure 68 Figure 7.2 Eswatini generation and import profile 70 Figure 7.3 Time-of-use tariffs for high season (left) and low season (right) 71 Figure 7.4 Transmission system map 73 Figure 7.5 Electricity coverage map 74 ENERGY MASTERPLAN 2034 7 Figure 8.1 Reference energy system (power sector part) 81 Figure 8.2 Reference energy system (non-power sector part) 82 Figure 8.3 Load profile representations in each season and in each day type 84 Figure 8.4 Assumptions on electricity import price development 86 Figure 8.5 Base Case: Electricity generation 91 Figure 8.6 Base Case: Electricity generation capacity by type 92 Figure 8.7 Base Case: Capacity expansion of renewable energy 93 Figure 8.8 Base Case: Fuel consumption of electricity production 94 Figure 8.9 Base Case: Transmission and distribution needs 95 Figure 8.10 Base Case: System cost summary 96 Figure 8.11 Forced Coal scenarios: Electricity generation 97 Figure 8.12 Forced Coal scenarios: Capacity mix 98 Figure 8.13 Limited Import 1 and 2 scenarios: Electricity generation 99 Figure 8.14 Limited Import 1 and 2 scenarios: Capacity mix 100 Figure 8.15 Limited Import 3 scenario: Electricity generation and capacity mix 101 Figure 8.16 No Import Enhancement scenarios: Electricity generation 102 Figure 8.17 No Import Enhancement scenarios: Capacity mix 103 Figure 8.18 Low Import Price scenario: Electricity generation 104 Figure 8.19 Low Import Price scenario: Capacity mix 105 Figure 8.20 Natural Gas scenario: Electricity supply mix and capacity mix 106 Figure 8.21 Cumulative carbon dioxide levels from 2014 to 2034 for all scenarios 107 Figure 8.22 Comparative share of renewable energy for all scenarios 108 Figure 8.23 Share of import electricity in total electricity production for all scenarios 108 Figure 8.24 Undiscounted cumulative lump-sum investment costs by type over the period 2014 to 2034 109 Figure 8.25 Average generation costs for all scenarios 110 Figure 8.26 Primary energy requirements for all scenarios 111 Figure 8.27 Primary requirements for ethanol, molasses and bagasse production 112 8 KINGDOM OF ESWATINI TABLES Table 2.1 Sustainable Energy for All goals 21 Table 3.1 Eswatini energy balance for 2014 26 Table 3.2 Oil product use in 2014 29 Table 4.1 Key scenario assumptions 37 Table 5.1 Final energy demand by sector and by fuel type in 2034 (in PJ) 47 Table 6.1 Overview of sugarcane area and production, 2012/13 to 2016/17 56 Table 6.2 Planned energy generation capacities by the sugar industry 56 Table 6.3 Bagasse production, 2012/13 to 2016/17 57 Table 6.4 Molasses production, 2012/13 to 2016/17 57 Table 6.5 Potential power generation capacities in the timber industry 59 Table 6.6 Mineable coal reserves, annual coal production and projected life 61 Table 6.7 Characteristics of identified wind zones 63 Table 6.8 Characteristics of identified solar PV zones 66 Table 7.1 List of power plants in Eswatini 69 Table 7.2 SEC power generation stations 69 Table 7.3 Eskom Megaflex customer categories 71 Table 7.4 Transmission lines 72 Table 8.1 Seasons defined in the SPLAT-SW model 83 Table 8.2 Fuel price assumptions 86 Table 8.3 Techno-economic parameters for industrial ethanol plants 87 Table 8.4 Technology assumptions for existing power plants 87 Table 8.5 Technology assumptions for planned power projects 88 Table 8.6 Generic technology assumptions 88 Table 8.7 Levelised cost of electricity (USD/MWh), 2014 and 2034 89 Table 8.8 Techno-economic parameters for transmission and distribution 90 Table 9.1 Summary comparison table for the scenarios 117 Table A1 Energy use by fuel and by sector in 2014 (PJ) 122 Table A2 Demand projection under the Reference scenario (PJ), 2014–2034 122 Table A3 Energy use by fuel and by sector in 2034 under the Reference scenario (PJ) 122 Table A4 Demand projection under the High Growth scenario (PJ), 2014–2034 123 Table A5 Energy use by fuel and by sector in 2034 under the High Growth scenario (PJ) 123 Table A6 Demand projection under the Low Growth scenario (PJ), 2014–2034 124 Table A7 Energy use by fuel and by sector in 2034 under the Low Growth scenario (PJ) 124 ENERGY MASTERPLAN 2034 9 ABBREVIATIONS AGOA Africa Growth Opportunity Act CHP Combined heat and power CO₂ Carbon dioxide CSO Central Statistics Office CSP Concentrated solar power GDP Gross domestic product GIS Geographic information system GWh Gigawatt-hour INDC Intended Nationally Determined Contribution IPP Independent power producer IRENA International Renewable Energy Agency km Kilometre Kt Kilotonne kV Kilovolt kW Kilowatt kWh Kilowatt-hour LCOE Levelised cost of electricity LEAP Long-range Energy Alternatives Planning System software LED Light-emitting diode LPG Liquefied petroleum gas LUSIP Lower Usuthu Smallholder Irrigation Project m/s Metres per second MESSAGE Model for Energy Supply Strategy Alternatives and their General Environmental Impact MNRE Ministry of Natural Resources and Energy MW Megawatt MWh Megawatt-hour NDC Nationally Determined Contribution NDS National Development Strategy NEP National Energy Policy O&M Operation and maintenance PJ Petajoule PPA Power purchase agreement PV Photovoltaic R South African Rand REEESAP Renewable Energy and Energy Efficiency Strategy and Action Plan REPN SADC Regional Energy Planning Network RERA Regional Energy Regulators Association RIDMP Regional Infrastructure Development Master Plan RISDP Regional Indicative Strategic Development Plan RSSC Royal Swaziland Sugar Corporation SACREEE SADC Centre for Renewable Energy and Energy Efficiency SACU Southern African Customs Union 10 KINGDOM OF ESWATINI ABBREVIATIONS SADC Southern African Development Community SAPP Southern African Power Pool SEC Swaziland Electricity Company SE4ALL Sustainable Energy for All SERA Swaziland Energy Regulatory Authority SPLAT-SW SPLAT-Swaziland model SWADE Swaziland Water and Agricultural Development Enterprise SZL Swazi Lilangeni TFEC Total final energy consumption TJ Terajoule TPES Total primary energy supply UNDP United Nations Development Programme USD United States dollar W/m 2 Watts per square metre ENERGY MASTERPLAN 2034 11 FOREWORD Minister for Natural Resources and Energy In the era of renewable energy, long-term energy planning This Energy Masterplan projects outcomes of the current and is imperative for the transformation of the energy system of planned policies in the energy system to 2034 and shows com- the Kingdom of Eswatini and its liberation to sustainable en- parative scenario results that assume deployment of diverse ergy growth. For the country, energy planning is key for the technologies and measures to meet the country’s energy de- achievement of universal electrification, to improve modern mand. The drivers of the policies suggested by the alternative energy access and energy efficiency, and to foster technologi- scenarios bring a sustainable energy future within reach, which cal breakthroughs, all by way of well-informed policy making. clearly supports the commitment to accelerate renewable energy deployment. The key driver of renewable energy is its The key challenge facing the country’s energy system is a lack economic feasibility, which the Energy Masterplan presents in of security of supply. Consequently, electricity import tariffs the country’s context. As renewable power generation costs will be exacerbated over the coming years, and a trend has fall rapidly, policy makers and investors need to confront the been set for continually escalating tariffs for end-users. This economic opportunities, as well as challenges, arising from a has helped make the harnessing of indigenous energy resourc- scale-up of renewable energy. This report shows that renew- es to produce electricity become more competitive. For in- able energy costs are at the lower end of the fossil fuel cost stance, falling technology costs have made renewable energy range by 2034. In addition, several solar PV and wind power increasingly competitive, and the likes of domestic bagasse for projects will provide some of the lowest-cost electricity from co-generation from the sugar industry, solar PV, wind and hy- any source. Informed decision making about the role of renew- dropower could meet the country’s electricity demand sustain- ables in future electricity systems will depend on continually ably. Meanwhile, the harnessing of domestic resources should updated reliable cost and performance data, as provided in not undermine the electricity trade facilitated by existing in- this report from current assessment. terconnections with Mozambique and South Africa. Rather, the country should find an adequate balance between harness- ing domestic resources and maintaining electricity trade with neighbouring countries. 12 KINGDOM OF ESWATINI The Energy Masterplan assesses the cost of the energy system It is therefore evident that renewable energy increasingly from the deployment of various energy technologies, show- makes business sense for the country and for investors. For ing renewables to be cost-competitive in a growing array of this reason, renewables will be the driving force of the King- conditions. This is in accordance with the observed combina- dom of Eswatini’s energy transformation, while permitting af- tion of expertise, purchasing power and access to international fordable energy, enabling universal access to modern energy, financial markets that drives down project costs and risks of increasing energy security and diversifying energy supply. renewables deployment globally. This Masterplan’s modelling Lastly, the Masterplan also recognises the opportunities that output will enable the country to structure policy advice and can be harnessed from the discovery of natural gas in neigh- planning for the development of renewable energy projects in bouring Mozambique. a robust manner. All in all, I would like to express my utmost gratitude to all The overall trend for the Eswatini energy system is clear: de- the stakeholders who contributed to this interactive exercise pendency on electricity imports will remain above 50 % in total and provided valuable insights and directions in the energy electricity production to about 2019, then gradually decrease sector at large. A special thanks to the International Renew- until 2034 to less than 10 %. By 2020 renewable power genera- able Energy Agency (IRENA) and its experts for their training tion technologies can be expected to provide 40 % of all elec- support, guidance, expertise and financial assistance for the tricity generation, which increases to an average of 70 % across development of this Masterplan, that will help usher the coun- all the scenarios assessed in this Masterplan, by the end of the try into the era that is about more than just resilience but also planning period (2034). As costs for renewables go head-to- inclusiveness for all. The core national working team for this head with or even out-compete fossil-based power solutions to Energy Masterplan has gained good experience in long-term provide new capacity without financial support, key opportuni- energy planning through working with this group of IRENA ties exist to pursue cost-effective technology pathways. experts, and the resulting skills will benefit the country as a whole for the future. Honourable Jabulile Mashwama (Senator) Minister of Natural Resources and Energy ENERGY MASTERPLAN 2034 13 EXECUTIVE SUMMARY The development of this Energy Masterplan has been long over- The planning tool used for development of the Masterplan is due in the Kingdom of Eswatini. The last comprehensive energy called SPLAT-Swaziland (SPLAT-SW; based on MESSAGE mod- policy document was published in 2003, and since then the elling software), and the review of energy demand scenarios global and regional energy landscape surrounding the coun- is generated from the LEAP model. MESSAGE is a dynamic, try have changed dramatically. In particular, the opportunities bottom-up, multi-year energy system software applying linear for developing indigenous resources are opening up rapidly as and mixed-integer optimisation techniques; it allows for assess- renewable energy costs become increasingly competitive, and ing the least-cost technically and economically feasible combi- pro-active planning is called for to better evaluate the needs nations of energy supply options to meet a specified demand and options in the future. under a set of conditions. On that basis it selects the least-cost optimal energy mix, taking into consideration the investment Against this background the Kingdom of Eswatini, in partner- needs and other costs for new infrastructure, energy supply se- ship with the International Renewable Energy Agency (IRENA), curity, energy resource utilisation, rate of introduction of new initiated a capacity-building programme on long-term strategic technologies, and environmental and other constraints. planning. The programme included a data preparation meeting, an online training and two 2-week national training courses pro- The Masterplan supports a growth in gross domestic product vided to the national working team as well as frequent working (GDP) of 1.8 % to 3 % between 2014 and 2020, and of 3 % from team meetings and technical interactions with IRENA experts 2020 to 2034. Under the preferred future scenario (Limited in the process of developing the Energy Masterplan. Long-term import scenario 3 – see below), 676 MW of domestic capacity energy planning in the energy sector is essential for finding a are required to meet the projected demand and to provide ad- balance between divergent legitimate expectations and views equate reserves in 2034. The electricity demand increases from of the variety of stakeholders – including government, consum- 1,270 gigawatt-hours (GWh) in 2014 to 2,648 GWh in 2034, liq- ers and private sector – for development of the sector. This bal- uid fuels from 10 petajoules (PJ) to 15 PJ and solid remain con- ance is attained through scenario planning using policy inputs stant at around 24 PJ over the planning period. This assumes that are modelled under specific constraints. the implementation of demand-side management programmes, and a gradual reduction in energy intensity due to technology Consequently, the Energy Masterplan of the Kingdom of Es- advancement. watini seeks to provide national decision makers with the quan- titative basis for planning future energy sector development Electricity demand forecasting established the basis for nation- through identifying and addressing the country’s distinct bar- al generation, transmission and distribution expansion planning, riers in the supply of energy. It is also nestled in analysis of and for assessing the impact of specific large energy-intensive the available energy resources and a need to maximise energy projects on existing resources and infrastructure. An energy use for sustainable energy development. In addition, it seeks to demand accounting model for Eswatini was created, as well as help shape appropriate policy and regulatory choices through three energy demand pathways driven by different assumptions the determination of suitable energy sources and technologies of economic growth, population growth and the electrification that are relevant and consistent with national priorities. rate. The sectoral energy demand was formulated by taking into account the useful energy needs of energy consumers and The Energy Masterplan was developed to cover the use and their choice of technologies and appliances. The output level of supply of both electricity and primary fuels, which include each sub-sector was modelled separately and was combined biofuels, petroleum, coal, biomass and fuelwood. The scenario with projections of its respective energy intensity and end-use modelling is based on the planning horizon 2014–2034. The shares to derive the expected consumption of each energy Masterplan provides suggestive guidance on the development source by sub-sector. and assessment of energy supply scenarios to meet domestic future energy demand, and outlines sector-specific develop- ment pathways through to 2034. 14 KINGDOM OF ESWATINI Based on the projected demand, the modelled scenarios are agreement with South Africa combined, and is contrasted with built on policies such as those that will encourage participation the case of no import enhancement but with the demand as in in the generation of electricity through promoting the deploy- the Base Case. The Low Import Price scenario assesses condi- ment of renewable energy technologies as well as the goals of tions under no limit on imports of electricity and the assump- Sustainable Energy for All. The model incorporates the latest tion of a lower tariff for imports from South Africa. Finally, the information on energy technologies and costs, with some re- Natural Gas scenario assesses the possible replacement of the finements to adequately characterise options available to Es- coal options with natural gas imports, for demand-balancing watini. The modelling approach has been expanded to cater power and as a way to reduce carbon emissions. to both the power and non-power sectors to take into account the demand for fuels other than electricity. A Base Case was Throughout the planning horizon, electricity imports decline created to depict the business-as-usual future, and serves as a in all of the assessed scenarios, and new generation technolo- point of reference for the alternative scenarios. gies that are increasingly economically feasible include solar PV, wind and biomass. The natural gas and coal technologies The accuracy of the Base Case is improved by regular reviews prove to not be economically feasible unless they are forced. and updates as and when things change or when new infor- The share of renewable energy shows an upwards trend in all of mation becomes available. In view of this, the long-term plan the scenarios except for the coal scenarios. presented here is considered as indicative rather than definitive. The Masterplan aims to achieve a balance between affordability All 10 scenarios were assessed against key long-term goals, of energy in the competitive market and a move towards a sus- namely 1) the provision of sufficient energy supply; 2) the provi- tainable energy supply. sion of access to clean energy technologies; 3) fostering indus- trialisation and 4) ensuring security of energy supply. Careful The Base Case includes all other technology options such as analysis of the modelling results indicates that the Limited Im- coal power generation, natural gas, biomass and wind tech- port 3 with renewables scenario possesses the most balanced nologies, to name a few. The deployment of technologies is not combination of the desirable policy features. This analysis co- forced in the model solution. incides with the feedback from the national stakeholder con- sultations held as part of the Masterplan development process, The nine alternative scenarios include two Forced Coal sce- where the importance of the country’s commitment to the Paris narios, which aim to provide insights into the impact of coal Agreement and to much stronger endorsement of renewable capacity in providing baseload power. These scenarios explore energy were highlighted. the possibility of reducing the country’s reliance on imports by exploiting domestic coal endowments through constructing a The key recommendations of this Masterplan are the increased new coal power plant. Coal is one of the resources that Eswatini development of renewable energy options involving the pri- has in abundance and has never exploited. vate sector; the creation of a comprehensive distributed power generation plan; and the need to develop policies and regula- Three Domestic Resources scenarios assess the implications tions to facilitate private sector engagement in rural electrifi- of becoming less import-dependent and maximising the use cation. These recommendations need to be complemented by of domestically endowed energy resources. These scenarios strengthening renewable energy targets, examining the con- (Limited Import 1, Limited Import 2 with forced hydro and Lim- straints of existing grid infrastructure and exploring the best ited Import 3 without coal and with renewables) envision an mechanisms to reduce costs and ease technical challenges. increase in renewable energy deployment as a result of the coupling of rapid reductions in renewable energy costs and Finally, throughout the process of developing this Masterplan, the country’s renewable energy promotion policy with the re- national energy planning capacity has been significantly en- duction of reliance on imports. Two “No Import Enhancement” hanced, while areas of further enhancement were identified. scenarios assume that no investment into expanding the im- To keep the Masterplan relevant, and to maintain and further port capacity is allowed; one of the scenarios assumes a higher strengthen the country’s planning capability, it is recommended local electricity demand, above the maximum electricity avail- that processes for regular updates of this Masterplan are insti- able to Eswatini from domestic sources and the current supply tutionalised and that the next round of updates start immedi- ately, building on the learning from this exercise. ENERGY MASTERPLAN 2034 15 PART I: ESWATINI ENERGY CONTEXT 1 INTRODUCTION

1.1 BACKGROUND To enhance energy security, affordability and environmental sus- tainability, Eswatini needs to develop domestic energy supply The Kingdom of Eswatini imports all of its petroleum product capacity and to diversify its national energy mix towards renew- requirements and around 70 % of its power from South Africa able supply options. Specifically, the long-term goal of energy and Mozambique, despite being well endowed with conven- sector development is a well-diversified energy mix that will: tional and renewable energy resources, including coal, solar, hydro, wind and biomass residues from the sugar and forestry • provide sufficient energy supply to support the country’s de- industries. These resources are potentially able to meet the en- velopmental goals, as contained in the National Development tire national electricity demand if fully exploited, as well as po- Strategy (NDS); tentially provide for the export of excess energy to Eswatini’s • contribute to poverty alleviation and improvement of quality neighbours. The high dependency on power imports contrib- of life by providing access to clean energy technologies; utes greatly to the country’s account deficit and increases Es- • foster industrialisation to bring about linkages with other watini’s exposure to energy supply risks, in terms of both sup- sectors; and ply security and price shocks. • ensure security of energy supply to support economic growth and attract foreign direct investment. Almost 100 % of the electricity generated in Eswatini is from hydropower and sugarcane-based co-generation. Biomass 1.2 OBJECTIVE AND SCOPE OF THE ENERGY (fuelwood and agricultural waste) is used mainly for household MASTERPLAN cooking and heating, as well as for co-generation in the sugar industry. Petroleum products are used mostly for transport, and This document is a national Energy Masterplan that addresses paraffin and liquefied petroleum gas (LPG) are used for cook- Eswatini’s distinct challenges and is designed to set the energy ing and heating. sector on firm footing for a sustainable future. The Masterplan aims to provide national decision makers with the quantitative Energy development in Eswatini is guided by the National En- base for planning future energy sector development. ergy Policy of 2003. Since then, the country’s energy sector has been undergoing rapid transformation with the liberalisation of The objectives of this Energy Masterplan are to: the electricity sector to encourage private sector investment. With the recent promising development of locally sourced re- • review the current national energy policy setting and energy newable energy, Eswatini has an opportunity to reduce its ex- balance; cessive reliance on imports for energy products and to improve • assess future national energy demand up to 2034; energy security. • identify all of the energy sources needed to meet this demand, including untapped potential for domestic energy The country previously did not have in place an energy mas- supply from renewable sources; ter plan to help to identify least-cost energy supply options • provide quantitative paths for future energy sector develop- to meet future national energy demand. With a master plan in ment to 2034; and place it will be easier for Eswatini to follow a clear energy sec- • assess energy supply options in terms of their implications tor development plan to ensure that energy policies govern- for financial requirements, energy security, energy access ing the country’s sub-sectors are developed consistently, while goals, carbon dioxide (CO₂) emissions and other related pol- bringing long-term policy stability that would help to reduce icy goals. uncertainty for potential energy investors. The present Master- plan identifies the priorities, time frame and costs associated with the energy sector development goals. The time horizon of this Masterplan is 20 years, up to 2034. 18 KINGDOM OF ESWATINI The national energy planning in this document covers the use The working team and steering committee members were and supply of both electricity and primary fuels, which include drawn from this stakeholder list. As part of the process for de- biofuels, petroleum, biomass and fuelwood. The scenario mod- veloping this report, IRENA provided the national working team elling is based on a 20-year planning horizon (2014 to 2034) (consisting of 12 national practitioners – see full list of working that is consistent with the national development targets. The team members in Annex A) with training on modelling tools Masterplan provides suggestive guidance on the development and scenario development for energy planning, to equip these and assessment of energy supply scenarios to meet domestic practitioners with the tools and means to manage and adapt future energy demand, and outlines sector-specific develop- national energy planning in line with developments and to ment pathways through 2034. The methodologies used to de- serve as custodians of the country’s sustainable energy future. velop energy scenarios are quantitative, traceable and consen- The training aimed at addressing specific gaps and building the sus-based. capacity of the national working team in the areas of long-term strategic energy planning, the development and implementa-

1.3 IMPLEMENTATION OF THE STUDY FOR THE tion of the planning tool SPLAT-Swaziland (SPLAT-SW; based MASTERPLAN DEVELOPMENT on MESSAGE modelling software) and the review of energy demand scenarios generated from the LEAP model. This Masterplan has been developed with technical support from the International Renewable Energy Agency (IRENA), which provided training and data analysis support. Input from stakeholders was drawn from government agencies, the na- tional utility, the national regulatory authority, civil society, aca- demia and the private sector. A steering committee was formed to give policy direction to the national working team whose role was to collect and analyse data, model different energy scenar- ios and prepare the Masterplan report. The following stakehold- ers (listed in alphabetical order) were identified and consulted during development of the Masterplan. • Biomass Group • Central Statistical Office • Coordinating Assembly on Governmental Organizations • Department of Meteorology • European Union • Ministry of Economic Planning and Development • Ministry of Finance • Ministry of Natural Resources and Energy • Oil Marketing Companies • Renewable Association of Swaziland • Royal Swaziland Sugar Association • Swaziland Cane Growers Association • Swaziland Economic Policy Analysis and Research Centre • Swaziland Electricity Company • Swaziland Energy Regulatory Authority • Swaziland Sugar Association • Swaziland Ubombo Sugar Limited • The Climate Change Unit • United Nations Development Programme • University of Swaziland ENERGY MASTERPLAN 2034 19 2 PRESENT ENERGY POLICY SETTING This Chapter reviews existing energy targets and other relevant 2.2 ELECTRICITY LEGISLATION development goals currently in place in Eswatini. The Energy Masterplan is built on these goals, and the comprehensive use Implementation of the NEP of 2003 has led to liberalisation of of quantitative tools ensures the consistency of objectives in the energy market, aimed at improving services offered to con- the future expansion of the energy sector. sumers and encouraging private sector investment. The Minis- try of Energy developed new legislation to govern the electric-

2.1. NATIONAL ENERGY POLICY OF 2003 ity sector in order to liberalise the electricity supply industry in Eswatini. The following pieces of legislation were promulgated The Swaziland National Energy Policy (NEP) of 2003 was de- as part of the unbundling process: veloped to address the challenges of transforming the energy sector and enhancing the overall development of the country. • Electricity Act of 2007 An aim of the NEP is “ensuring that the development goals of • Swaziland Electricity Company Act of 2007 the country are met through the sustainable supply and use of • Swaziland Energy Regulatory Authority Act of 2007 energy for the benefit of all citizens of the country”. The promulgation of the Electricity Act of 2007 and the Swazi- The vision of the NEP underlines the importance of the avail- land Electricity Company Act of 2007 transformed the Swaziland ability and accessibility of energy to cater to the development Electricity Board into the Swaziland Electricity Company (SEC). needs of the Eswatini nation. Energy should be affordable, and Further, the Swaziland Energy Regulatory Authority Act of 2007 national energy resources should be harnessed with optimum created the Swaziland Energy Regulatory Authority (SERA) to efficiency while ensuring due attention to environmental con- regulate developments in the liberalised energy sector. The cerns. The main objectives of the NEP are to: mandate of SERA is to administer the Electricity Act of 2007, and its primary and core responsibilities are to exercise control • ensure access to energy for all over the electricity supply industry and to regulate the genera- • enhance employment creation tion, transmission, distribution, supply, use, import and export • ensure security of energy supply of electricity in Eswatini. SERA also is responsible for regulating • stimulate economic growth and development electricity tariffs and the quality of supply and services. • ensure environmental and health sustainability. To date, the SEC is still wholly owned by the Government of The NEP was structured according to the demand sectors, the Kingdom of Eswatini. It is semi vertically integrated since supply sectors, rural electrification, cross-cutting issues and the SEC still enjoys monopoly ownership of the transmission implementation. and distribution of the energy supply chain, due mainly to the lack of technical and financial capacity to promote participa- tion in these sectors. Municipal councils, particularly from the country’s two cities, have expressed interest in participating in the distribution of electricity. Although transmission and most of the distribution networks are owned and operated by the SEC, prospective energy companies are allowed by law to use the existing infrastructure for electricity trade; however, such companies are expected to pay wheeling charges to the SEC. 20 KINGDOM OF ESWATINI The driving force behind the enactment of this electricity leg- 2.3 SUSTAINABLE ENERGY FOR ALL GOALS islation was to ensure participation of the private sector in the generation, transmission and distribution of electricity. The gen- In response to the NEP’s calls for ensuring access to energy for eration component of the electricity supply industry in Eswatini all and ensuring security of energy supply, Eswatini embraced is already fairly deregulated, as is apparent from the existence the United Nations’ Sustainable Energy for All (SE4ALL) initia- of two independent power producers (IPPs) and a range of pri- tive, that reflects the goals linked to long-term development of vate entities that are in the process of developing projects. a well-diversified energy mix (see above). As a result, a country gap analysis exercise (SE4ALL Gap Analysis) was done, from Recently, further work has been done to enhance IPP participa- which an Action Agenda and Investment Prospectus reports tion in the generation of electricity. The Government has com- were developed in 2016. The SE4ALL targets are outlined in the pleted a coherent and practical IPP Policy that will be vital in following table adapted from the Swaziland Sustainable Energy addressing many issues faced by potential IPPs, including in for All Action Agenda, compiled by the Ministry of Natural Re- the renewable energy sector. The policy provides much-needed sources and Energy (MNRE) with the help of the United Nations guidance to facilitate the growth and development of power Development Programme. generation projects in Eswatini. A key objective was to increase use of the country’s extensive local energy resources including coal, biomass, solar, wind and geothermal. TABLE 2.1 SUSTAINABLE ENERGY FOR ALL GOALS 1. Ensuring universal • 5,000 households connected to grid electricity per annum to reach 18,500 additional households access to modern by 2018, reaching 75 % electricity access at household level energy services • Build a 300 megawatt (MW) coal power plant to enhance security of supply by 2018 • 100 % access to electricity at the household level attained by 2022 • Effective policies and incentives for improved cook stoves established by end-2016 • Regulatory structures and mechanisms for LPG established by end-2016 • Penetration of improved cook stoves increased by 10 % per year from 2017 • Penetration of LPG for cooking increased by 10 % per year from 2017 • Geographic information system (GIS) database for rural energy access developed by 2017 • Rural energy master plan and implementation strategy to attain universal access to energy including LPG, improved cook stoves, solar home systems and biogas developed by 2018 • 100 % access to clean energy at the household level attained by 2030 2. Renewable energy • Solar and wind maps developed by 2016 • 46 MW solar photovoltaic (PV) power plant (IPP) online by 2017 • 20 % of all public buildings to have solar water heating by 2018 • 134 MW of hydropower (SEC) by 2019 • 130 MW of co-generation from bagasse (IPP) by 2020 • 80 MW from wood waste from IPPs (with 40 MW online by 2019) • 10 % ethanol blending by 2016 • 50 % renewable electricity penetration in the electricity mix by 2030 3. Energy efficiency • Attain 60 gigawatt-hours (GWh) of energy savings per year by 2018 • Attain 180 GWh of energy savings per year by 2025 • Energy efficiency policy and regulations developed by 2019 • 500,000 light-emitting diodes (LEDs) distributed to households by 2020 • Design and adopt an energy efficiency appliance labelling and standards programme ENERGY MASTERPLAN 2034 21

2.4 ESWATINI NATIONALLY DETERMINED With regard to the second mitigation action in the energy do- CONTRIBUTION main, the ethanol to be blended is intended to be sourced from sugarcane molasses, a renewable source. This commitment will In 2015 Eswatini submitted to the United Nations Framework reduce emissions in the transport sector, which accounted for Convention on Climate Change (UNFCCC) its Intended Nation- 9 % of the nation’s greenhouse gas emission inventory in 2010. ally Determined Contribution (INDC) towards global green- Transport-related emissions are anticipated to increase expo- house gas emission reductions. Of Eswatini’s four mitigation nentially as more vehicles are purchased; currently the number actions outlined in the INDC, two are in the energy domain: of registered vehicles in Eswatini is growing at an average rate of 7 % per year. • Doubling the share of renewable energy in the primary energy mix, relative to 2010 levels1, and Eswatini does not currently blend ethanol in petrol, although a • Introducing an ethanol blend in petrol of at least 10 % (E10) successful pilot project was undertaken. The E10 blending com- by 2030. mitment is likely to have a positive influence on the country’s agricultural sector, particularly in the sugar industry as it has Eswatini recognises that access to renewable forms of energy the potential to increase the production of sugarcane. By-prod- plays a significant role in improving the livelihoods of its people ucts from this industry are bagasse and molasses, which can be – in terms of both increasing social equity and improving eco- used as feedstock for the production of ethanol. nomic growth. The doubling of the share of renewable energy in the primary energy mix would drastically reduce emissions 2.5 REGIONAL ENERGY CONTEXT from Eswatini’s energy sector, which represented more than 20 % of the nation’s greenhouse gas emission inventory in 2010. Since the adoption of the SADC Energy Protocol in 1996, the Southern African Development Community (SADC) has enact- The contribution covers on-grid and off-grid applications, ed several strategic instruments for the energy sector, including where Eswatini will: the following: • Implement small-scale, decentralised renewable energy tech- • SADC Energy Cooperation Policy and Strategy (1996) nologies to improve energy access in rural areas. This will also • SADC Energy Action Plan (1997) reduce unsustainable wood harvesting practices in the country. • SADC Energy Activity Plan (2000) • Increase the use of grid-connected renewable energy tech- • Regional Energy Access Strategy and Action Plan (2010). The nologies fuelled by sources such as waste, solar, bagasse plan defines the following objectives for improving energy (from the sugar industry) and wood chips. This contribution access, which is a major goal of energy policy in all SADC has the further co-benefits of improving universal energy ac- Member States: cess and security of supply, which are particularly relevant in the context of the challenges experienced in the regional –– Harnessing regional energy resources to ensure, through electricity system. national and regional action, that all people in the SADC region have access to adequate, reliable, least-cost, envi- ronmentally sustainable energy services. –– Halving the proportion of people without electricity within 10 years for each end-use and halving this again in succes- sive year periods until there is universal electricity access for all end-uses. ¹ According to the Swaziland INDC, the share of renewable energy in the national energy mix in 2010 was 16 %, including both grid-connected renewable energy and sustainable/renewable biomass. 22 KINGDOM OF ESWATINI • SADC Framework for Sustainable Biofuels (2010). The objec- Eswatini also is represented, via the SEC, in the Southern Af- tive is to provide a set of basic guidelines for the develop- rican Power Pool (SAPP), where it benefits from regional co- ment of sustainable biofuel strategies. The regional guide- ordination, energy planning and cross-border trading. The Gov- lines can, in turn, be adapted to meet the requirements of ernment is obliged to implement policies and measures that SADC Member States, e. g., recognising different legislative will most favourably take advantage of market conditions and regimes, national development priorities, land issues and regulatory structures with regard to the SAPP. Connection to specific conditions affecting biofuels crop production. and participation in the SAPP affords Eswatini the opportunity • Regional Infrastructure Development Master Plan (RIDMP) to export and trade electricity generated locally into a competi- (2012). The RIDMP can serve as a key guide for the crea- tive market with other interconnected SAPP members. tion of efficient, cost-effective transboundary infrastructure connecting all SADC Member States in the energy, water, To improve regulation and governance in the power sector information and communications technology, and transport Eswatini joined the Regional Energy Regulators Association sectors. It will be implemented over a 15-year period that (RERA) with a view to benefiting from regional best practices, began in 2013. benchmarking information and shared experiences in address- • Revised Regional Indicative Strategic Development Plan (RIS- ing common power sector challenges. DP) (2015). This is a guiding document for the SADC regional integration and development programme over the period 2.6 CURRENT INSTITUTIONAL FRAMEWORK 2015 to 2020. The priorities are industrial development and FOR ENERGY AND ELECTRICITY PLANNING market integration, infrastructure in support of regional in- tegration, peace and security co-operation and special pro- The Energy Department of the Ministry of Natural Resources grammes with a regional dimension. and Energy is the custodian of policy and operational activi- • Renewable Energy and Energy Efficiency Strategy and Ac- ties pertaining to the energy sector. Its mission is to effectively tion Plan (REEESAP) (2016). The REEESAP has been devel- manage the national energy resources and to work towards af- oped as a regional framework from which SADC Member fordable and sustainable energy provision for all people in the States will develop their own renewable energy and energy country, while ensuring the international competitiveness of the efficiency strategies and Action Plans to accelerate efforts energy sector. to promote renewable energy and energy efficiency in their State, hence contributing to energy security and energy The Government leads the national energy planning process, access in the SADC region. with input from different energy sector stakeholders drawn from government agencies, the national utility, the national SADC’s common goal is to achieve regional economic and social regulatory authority, civil society, academia, the private sector development in an environmentally sound manner by increas- and IPPs. This serves to leverage the expertise and resources ing access to modern energy services and improving energy of other agencies and to promote diverse energy supplies, security in SADC Member States. This goal is pursued through while integrating various energy sector priorities and goals and the newly established SADC Centre for Renewable Energy and building consensus. Energy Efficiency (SACREEE), which is mandated to promote markets for renewable energy and energy efficiency technolo- gies and services in the SADC region. ENERGY MASTERPLAN 2034 23 The Government develops national policies and programmes, The Department is further introducing GIS software for map- while private energy companies have a key role in implementa- ping energy infrastructure and renewable energy sources. For tion and investment. Apart from the electricity sector, no quan- future projections of demand, trend analysis has been prac- titative long-term energy sector planning has been practiced. ticed. National energy statistics are disseminated internally as This Energy Masterplan is the first attempt to devise a com- well as to external partners; however, the information is not prehensive long-term framework for the entire energy sector. widely published. Integrated resource planning is becoming important to allow key stakeholders to participate in the identification and opti- 2.6.2 ELECTRICITY SECTOR misation of the appropriate mix of energy resources to meet short- and long-term energy needs in a sustainable way. The national utility SEC, among its functions, is entrusted with the safe operation and control, expansion planning and main-

2.6.1. ENERGY STATISTICS tenance of the electricity infrastructure, as well as energy trad- ing. The electricity infrastructure expansion planning framework Energy data are crucial for planning and monitoring the im- involves all relevant stakeholders, from the Government (in its plementation of policies. The national Energy Department function as policy maker) to customers and to IPPs and the utility. collects, analyses and disseminates energy statistics, in close collaboration with the Central Statistics Office (CSO). The CSO The SEC plans for all expansion in the electricity sector. This is instrumental in collecting data through the national census, includes grid expansion plans for the transmission and distribu- the Eswatini household income and expenditure survey, and tion networks, generation expansion plans and demand fore- other means. casting. Typical planning time horizons are 10 years. The Energy Department conducts annual surveys on energy sta- Demand forecasting is key in establishing the basis for national tistics to collect primary data from major industries (e. g., sugar- generation, transmission and distribution expansion planning as producing companies, timber industries, hotels). The Depart- well as for assessing the impact of specific large energy-inten- ment also uses secondary data from various companies such as sive projects on existing resources and infrastructure. The aim the SEC and oil companies, to cross-check the primary data. of demand forecasting is to determine the present and future electricity requirements of electricity end-users in order to rec- The Department compiles annual energy balance tables using oncile these with the available resources and electricity infra- Excel-based software designed by the SADC Regional Energy structure. Spatial consideration of electricity needs and require- Planning Network (REPN). However, technical support from ments of various user sectors are considered, as determined by SADC for operating the REPN software has long ceased, and the characteristics and trends of each and highlighting where Department staff have experienced technical difficulties. and when imbalances between electricity requirements and supply are most likely to occur. Broad socio-economic charac- teristics of study areas are assessed to enable the collection of data that provide a sketch of a holistic picture of the varying realities contained within the study areas as well to portray the impacts created by various factors. 24 KINGDOM OF ESWATINI Influential planning strategies at the town and village level are Figure 2.1 demonstrates the planning process flow used by the assessed, taking note of residential, commercial and industri- utility. The components shown in broken lines are suggested to al developments in order to develop electrical characteristics be included when the Energy Masterplan is completed. around proposed developments to predict future needs. Differ- ent methods based on the characteristics of customer sectors The generation expansion plan will in future be driven from the are applied to forecast the future demand, where demand and output of this Energy Masterplan. The generation expansion plan projected growth of the industrial sector differs dramatically programme is currently implemented by the utility and IPPs. from the residential sector, which is based on the growth of ru- ral electrification. The types of forecast methods include spatial forecasts, point load forecasts and household forecasts. FIGURE 2.1 PLANNING PROCESS FLOW Socio-economic Municipal Dev. Plans Framework Customer Applications Government Infrastructure Dev. Plans Rural Electrification Program Demand Forecast EMP from Government (Leap, PowerGLF) IPP Initiatives Network Capacity Assessment General Expansion Plan (PSSE/DigSILENT)) Transmission Network Transmission Network Expansion Plan Expansion Plan ENERGY MASTERPLAN 2034 25 3 ENERGY BALANCE: RETROSPECTIVE AND GENERAL OVERVIEW The development of this Energy Masterplan was initiated in 3.1 ENERGY BALANCE FOR 2014 2016. The latest available year for energy sector statistics at that time was 2014, and Section 3.1 discusses the energy bal- The Eswatini energy balance for 2014 is shown in Table 3.1. ance for that year. For time-series data, the statistics go back to The table summarises the country’s primary energy supply, trans- 1994. Where available, data for more recent years beyond 2014 formation of primary energy, and final consumption. The struc- are presented in the time-series analysis in Section 3.2 ture of each of these items is discussed in the sub-sections below. TABLE 3.1 ESWATINI ENERGY BALANCE FOR 2014 Eswatini Energy Balance Renewables Hard coal Petroleum Hydro Electricity Total 2014 (TJ) and waste Indigenous production 5,554 – 1,157 25,796 – 32,507 From other sources – – – – – 0 Import 2,449 10,412 – 151 3,097 16,109 Export –5,554 – – – – –5,554 International marine – – – – – 0 bunkers Stock changes – – – – – 0 Total primary energy 2,449 10,412 1,157 25,947 3,097 43,062 supply (TPES) – – – – – 0 Statistical differences – –1 –68 3 – –66 – – – – – 0 TRANSFORMATION –542 – –1,089 –2,780 2,061 –2,350 Losses – – – – 631 631 FINAL CONSUMPTION 1,907 10,411 – 23,170 4,527 40,015 Industry 1,264 88 – 11,194 1,771 14,317 Residential – 356 – 11,976 1,218 13,550 Agriculture – 394 – – 1,168 1,562 Commerce and government 643 234 – – 370 1,247 services Transport – 9,339 – – – 9,339 26 KINGDOM OF ESWATINI

3.1.1 TOTAL PRIMARY ENERGY SUPPLY from domestic bagasse. Hydropower contributed 3 % of TPES, and imported electricity contributed 7 % of TPES. Coal (imports) Sources of total primary energy supply (TPES) in Eswatini in- contributed 6 % of TPES, and petroleum (imports) contributed clude hard coal, renewables and waste, hydropower, electricity 24 %. Eswatini is reliant on imports for petroleum products that and petroleum. The contribution of each energy source to the include petrol, diesel, paraffin, LPG and lubricants. country’s energy supply is shown in Figure 3.1 In 2014 renewa- bles and waste contributed 60 % of the TPES; of the total supply of renewables and waste, 46 % is from domestic wood and 54 % FIGURE 3.1 TOTAL PRIMARY ENERGY SUPPLY IN 2014 7 % Electricity 6 % Hard coal 24 % Petroleum 3 % Hydro 60 % Renewables and waste ENERGY MASTERPLAN 2034 27

3.1.2 TRANSFORMATION OF PRIMARY ENERGY to generate combined heat and power (CHP, via co-generation) from coal and from renewables and waste, particularly bagasse Eswatini has a vibrant energy industry, which produces, trans- (36 %) and wood chips (27 %). Coal accounts for 12 % of total forms and supplies energy, mainly to provide heat and electric- fuel use for power generation, and the remaining 25 % is from ity. In 2014 an estimated 4,411 terajoules (TJ) of fuel was used hydropower (Figure 3.2). FIGURE 3.2 FUEL MIX FOR ELECTRICITY AND HEAT PRODUCTION IN 2014 25 % Hydro 36 % Bagasse 27 % Wood chips 12 % Hard Coal The total installed generation capacity in Eswatini in 2014 was 3.1.3 TOTAL FINAL ENERGY CONSUMPTION around 180 MW, including 106 MW of biomass, 61.5 MW of hy- dropower, 9 MW of diesel and 2.2 MW of coal power plants. Total final energy consumption (TFEC) in the country is pre- Due to the high operational costs, it is no longer financially or sented in the main sectors of transport, industry, residential, ag- economically viable to run the diesel power plants. Approxi- riculture, and commerce and government services. The industry mately 304 GWh of electricity was generated in 2014 from the sector represents the sugar sector and other industrial sectors. public utility’s hydropower plants and biomass. Biomass plant Final energy use in 2014 by sector is shown in Figure 3.3 The generation is based on bagasse and wood chips that are used transport sector used only oil products, mainly petrol and die- in combination as feedstock in the existing CHP plants. Sugar sel. The sugar industry consumed electricity, and the residential mills use waste produce (bagasse) and wood chips to run their sector used mainly wood, oil products in the form of paraffin mills during the season and generate steam to run the boilers and LPG, as well as electricity. The agriculture sector consumed and turbines (see Section 6.1 for further detail). Sugar mills are electricity and oil products in the form of diesel. The commerce self-producers and export excess power to the grid. and government services sector comprises the hospitality in- dustry and government departments. The use of coal and re- Although Eswatini has proven domestic reserves of coal, the newables and waste in this sector primarily represents inputs to contribution of coal in the current energy mix remains small; in cooking, as well as water and space heating in the health and 2014, 15 GWh of electricity were generated from the 2.2 MW of hospitality industries. distributed coal capacity. 28 KINGDOM OF ESWATINI FIGURE 3.3 FINAL ENERGY USE BY SECTOR IN 2014 16 14 12 10 8 PJ 6 Electricity 4 Renewables and waste 2 Petroleum 0 Hard coal Industry Residential Transport Agriculture Commerce and government services For oil products, the majority (89 %) was consumed by end- were from the industry sector (1 %), commerce-related services users in the transport sector. The agriculture sector accounted (2 %) and households (4 %) (Table 3.2). for nearly 4 % of oil product consumption, and other end-uses TABLE 3.2 OIL PRODUCT USE IN 2014 Sector Energy (TJ) % of use Agriculture 394 3.8 % Commerce and government services 234 2.2 % Industry 88 0.8 % Residential (households) 376 3.6 % Transport 9,339 89.5 % Total 10,431 100 % ENERGY MASTERPLAN 2034 29

3.2 TRENDS IN ENERGY SUPPLY, 3.2.1 TOTAL PRIMARY ENERGY SUPPLY TRANSFORMATION AND USE As shown in Figure 3.4, the profile for Eswatini’s TPES has Eswatini official energy statistics reach back to the year 1994. not changed much between 1994–2016. However, apparent The following sub-sections discuss structural changes in the trends include the increased use of imported electricity and country’s energy supply and demand between 1994 and re- the decreased use of imported coal as a result of fuel switch- cent years. ing from coal to electricity. The supply of wood increased over the years, as did the use of bagasse. The supply of electricity from hydropower increased slightly in 2011 due to a capacity increase of 20 MW. FIGURE 3.4 TRENDS IN TOTAL PRIMARY ENERGY SUPPLY 50 45 40 35 30 PJ 25 Imported coal 20 Bagasse 15 Oil Products 10 Other wood 5 Hydro Imported electricity 0 1994 1998 2002 2006 2010 2014 Year

3.2.2 TRANSFORMATION OF PRIMARY ENERGY Use of wood waste decreased in recent years. Use of hydropow- er similarly first increased and then declined in recent years; this Trends in the transformation of input fuels for heat and electric- reflects droughts that happened during that period. The fluctua- ity generation since 1994 are shown in Figure 3.5. Overall trends tion in the use of bagasse reflects a similar pattern, as it too is include the increased use of bagasse and decreased use of coal. dependent on rainfall that affects sugarcane production. 30 KINGDOM OF ESWATINI FIGURE 3.5 TRENDS IN INPUT FUELS FOR HEAT AND ELECTRICITY GENERATION 6 5 4 PJ 3 Diesel 2 Wood waste Bagasse 1 Coal Hydro 0 1994 1998 2002 2006 2010 2014 Year Figure 3.6 shows the trends in electricity output by source. No- generation has more than doubled in recent years. Correspond- tably, the electricity supply has increased more than 50 % since ing to the fluctuation of the inputs, hydropower generation also 2000. The dominance of imported electricity continued over fluctuates throughout this period, reaching a high of 1,200 GWh this period, and the supply from domestic CHP and hydropower in 2011 as a result of the 20 MW of additional capacity. FIGURE 3.6 TRENDS IN THE ELECTRICITY MIX 1,600 1,400 1,200 1,000 800 GWh 600 Imports 400 CHP Generation (bagasse, wood 200 chips + coal) Hydropower 0 2000 2002 2004 2006 2008 2010 2012 2014 Year ENERGY MASTERPLAN 2034 31 The country’s installed generation capacity from 1994 to 2016 in electricity output as depicted in Figure 3.6. Diesel capacity is shown in Figure 3.7. In 2012 new biomass capacity of about also has remained constant, with no production of electricity

39.5 MW from bagasse and wood chips was added. Hydro- because the power plant is mothballed. power capacity has been constant since 2011, with fluctuations FIGURE 3.7 TRENDS IN INSTALLED CAPACITY 200 180 160 140 120 Coal 100 MW Solar 80 60 Biomass 40 Hydro 20 Diesel 0 1994 1998 2002 2006 2010 2014 Year

3.2.3 TOTAL FINAL ENERGY CONSUMPTION Figure 3.8 shows the development of TFEC from 1994 to 2016. Over this period TFEC increased slightly, driven primarily by increased fuel use in the residential sector. FIGURE 3.8 TRENDS IN FINAL ENERGY CONSUMPTION BY SECTOR 45 40 35 30 Transport 25 PJ 20 Commerce and govt. services 15 Agriculture 10 Residential 5 Industry 0 1994 1998 2002 2006 2010 2014 Year 32 KINGDOM OF ESWATINI Figure 3.9 shows the development of TFEC between 2011 and stant, except for a slight increase in oil use in the transport sec- 2016, by fuel. The use of biomass shows an upward trend, and tor as a result of increased car ownership. the use of the remaining types of fuels remains relatively con- FIGURE 3.9 TRENDS IN FINAL ENERGY CONSUMPTION BY FUEL 45 40 35 30 25 PJ 20 15 Biomass 10 Oil 5 Electricity Coal 0 2011 2012 2013 2014 2015 2016 Year ENERGY MASTERPLAN 2034 33 PART II: ENERGY DEMAND ASSESSMENT 4 ENERGY DEMAND ASSESSMENT 2015–2034: METHODOLOGY AND ASSUMPTIONS

4.1 BRIEF DESCRIPTION OF THE Population dynamics are also an important driver of energy METHODOLOGY trends and the assumptions. An increase in population increas- es the number of households. This factor is the key driver of Detailed energy demand assessment for the period between energy demand in the residential sector. 2015 and 2034 was performed using the Long-range Energy Al- ternatives Planning System (LEAP) software. An energy demand The Eswatini energy demand model formulated with LEAP accounting model for Eswatini was created, and three demand (hereafter referred to as the LEAP-Swaziland model) explicitly projections were formulated. The model used for the assess- assesses the demand for the following sectors and energy ser- ment was initially developed in 2015 by the Ministry of Tour- vices. Note that the numbers in brackets indicate the number of ism and Environmental Affairs with technical support from the technology options modelled under each category. Energy Research Centre of the University of Cape Town, as part of an exercise to identify CO₂ emission mitigation options for • Industry sector sugar production [2: electricity and oil] Eswatini’s INDC. The model was further reviewed and updated • Industry sector other [2: electricity and oil] for the purpose of assessing the overall energy needs used as a • Agricultural sector [2: electricity and oil] basis for this national Energy Masterplan. Note that the updated • Mining [1: electricity] version greatly simplified the original assessment done in 2015, • Commerce and government services mainly in terms of the number of technology options assessed [3: electricity, coal and LPG] and the representation of end-use energy services. • Residential sector electrified cooking [4: electricity, LPG, wood and paraffin] The LEAP software2 is a tool to generate a detailed accounting • Residential sector electrified water heating model to develop energy demand projections in a bottom-up [5: geyser, solar water heating without back-up, solar water manner, by taking into account energy consumers’ useful ener- heating with back-up, other electricity and wood] gy needs and their choice of technologies and appliances. The • Residential sector electrified refrigeration tool can be used to describe individual policy measures that [1: electric refrigerator] can then be combined into alternative scenarios. This approach • Residential sector electrified air conditioning allows policy makers to assess the impact on an individual pol- [1: air conditioning] icy as well as the interactions that occur when multiple policies • Residential sector electrified lighting [1: electricity] and measures are combined. • Residential sector electrified television/ironing [1: electricity] • Residential sector electrified other [1: electricity] The increase in energy demand is driven mainly by economic • Residential sector non-electrified growth, population growth and the electrification rate. The [3: wood, LPG and paraffin] model takes into account the energy intensity of short- and • Transport sector [1: oil] long-term efficiency improvements. End-user prices were mostly not taken into account, but historical data on the con- The industry is separated into two sub-sectors, the “sugar in- sumption of coal, oil products, electricity and biomass were dustry” and “other industry”, allowing for a more detailed anal- used. The intensity of energy consumption per unit of each ysis of trends and drivers in the industrial sector by fuel. sub-sector’s output and the share of each energy source are projected on an econometric basis. The output level of each The residential sector’s energy consumption is split into electri- sub-sector is modelled separately and is combined with projec- fied and non-electrified households, and the energy demand of tions of its energy intensity and end-use shares to derive the electrified households is further divided into seven end-uses: consumption of each energy source by sub-sector. cooking, water heating, refrigeration, air conditioning, lighting, television/ironing (appliances) and other. The energy consump- ² For details on the LEAP software, see www.energycommunity.org/default.asp?action=introduction. 36 KINGDOM OF ESWATINI tion related to each end-use is computed as the product of an 4.2 KEY SCENARIO ASSUMPTIONS intensity variable. For each end-use, the intensity variable is projected econometrically and is linked to the impact of policies Three demand scenarios were developed: the Reference, High and measures to reduce energy intensity and promote energy Growth and Low Growth scenarios. They are differentiated by efficiency. The fuel shares are also projected econometrically assumptions regarding the GDP growth rate and the electrifica- and are linked to a variable representing the impact of policies tion rate in the residential sector. and measures to promote fuel switching. For example, in the case of fuelwood, fuel share is the function of variables taking The overall energy demand development of the country is into account the impact of policies such as rural electrification, based on a GDP growth rate in the range of 1.8 % to 3 % be- clean cooking and cost-reflective tariffs. tween 2014 and 2020, and 3 % thereafter. This growth rate drives energy demand in the industrial and residential sectors, The division of electrified/non-electrified households is made which is driven mainly by economic expansion. The growth in based on historical data from sample surveys, in order to bet- energy demand in Eswatini lags behind the global average of ter evaluate the differences in these households. As a result 4.9 %. The High Growth scenario assumes a 4 % increase in of the bottom-up analysis, estimated aggregate biomass use revenue from the mining sector. in the residential sector was revised downwards compared to the model from the Ministry of Tourism and Environmental Af- With regard to electrification rates, in the High Growth scenario fairs, which did not assume energy efficiency improvements. universal access is assumed to be achieved by 2022, in the Ref- Energy consumption for both electrified and non-electrified erence scenario by 2030, and in the Low Growth scenario by households was calculated for each fuel as a function of gross 2034. Population growth is assumed to overall decrease, from domestic product (GDP), the electrification rate and past con- 1.2 % per year in 2014 to 0.8 % per year in 2034, according to sumption levels. Fuel demand is projected per household. official government projections. Key assumptions for the three scenarios are summarised in Table 4.1 and are discussed further Economic activity is the main driver of electricity demand in in the sub-sections below. all sectors. On average, electricity demand rises faster than in- come, hence electricity demand is income-inelastic. This differ- ence reflects the penetration of various energy services (satu- ration effects) in the country, and it also reflects changes in the structure of economic activities. Electricity-intensive heavy industry has contributed more of the increase in GDP. The en- ergy efficiency of electrical equipment and appliances is also generally lower, boosting electricity intensity. TABLE 4.1 KEY SCENARIO ASSUMPTIONS Reference High Growth Low Growth 1.8 % to 3 %: 2014 to 2020 1.8 % to 3 %: 2014 to 2020 0.9 % to 1.8 %: 2014 to 2020 Annual GDP growth rate 3 % to 3.5 %: 2020 to 2025 3 %: 2020 to 2034 1.8 %: 2020 to 2034 3.5 %: 2025 to 2034 Electrification rate 100 % achieved in 2030 100 % achieved in 2022 100 % achieved in 2034 Annual population growth Decrease from 1.2 % (2014) to 0.8 % (2034) ENERGY MASTERPLAN 2034 37

4.2.1 CLIMATE 4.2.2 DEMOGRAPHY AND MAJOR DEMOGRAPHIC POLICY GOALS In general, the climatic conditions of Eswatini vary from tropical to near temperate. Winter in the country is generally dry and Eswatini’s population was estimated to be around 1.1 million cold, with an average temperature of around 17 degrees Celsius people in 2014, with about 76 % living in rural areas and 24 % (62.4 degrees Fahrenheit). During the winter time, most house- in urban areas. The distribution of the population among the holds require space heating. four regions of the country is uneven, with Manzini accounting for the highest share at 32 %, Hhohho at 29 %, Shiselweni at The main source of energy used for this purpose is fuelwood; 20 % and Lubombo at 19 %. The migration of people between thus, larger quantities of fuelwood are consumed during the regions is relatively low; however, it is very high between rural winter at the household level. Alternative sources of energy and urban areas as people search for job opportunities and for used are LPG, paraffin and electric heaters. In commercial means of earning a living. Rapid growth in the urban population buildings both air conditioners and heaters are used. is attributed largely to rural-urban migration. Given that the mi- grants are usually young and able-bodied persons, this has the The SEC reports that during summer months the demand for effect of impoverishing the rural areas from where they origi- electricity is attributed to an increase in the need for household nate. More than half of the urban population is concentrated in and industrial cooling systems as well as irrigation systems for the two cities of Mbabane and Manzini. the sugar industry. The country’s official population projections (Figure 4.1) indi- cate that the population will increase at an annual rate of 1.2 % from 2014; this growth rate decreases to about 0.8 % annually by 2034. The decline in population growth is driven primarily by the declining fertility rate, which is expected to drop from 3.5 children per woman in 2014 to below 3.0 children per woman by 2030. FIGURE 4.1 POPULATION PROJECTIONS, 2010 TO 2034 1,400 1,200 1,000 Thousands 800 600 400 200 0 2010 2012 2014 2016 2018 2020 2022 2024 2026 2028 2030 2032 2034 Year 38 KINGDOM OF ESWATINI 4.2.3 MACROECONOMIC SITUATION AND MAJOR Eswatini’s macroeconomic imbalances are also driven in part ECONOMIC POLICY ISSUES by the global economic decline and shocks. The country’s loss of eligibility under the Africa Growth Opportunity Act (AGOA3) As shown in Figure 4.2 Eswatini’s economy grew modestly in points to the importance of enhancing economic diversification the period 2000–2016. From 2000 to 2007 the average GDP and competitiveness. Stepped-up policy efforts in multiple ar- growth rate was 4.6 %. This growth then declined slightly, at an eas are needed, including improving the business climate and average of 2.5 %, to 2016. Significant development challenges export diversification. continue, including high unemployment and poverty. In recent years, Eswatini’s economy has geared towards an expansionary fiscal policy, while revenues from the Southern African Customs Union (SACU) have declined as a share of GDP since 2013/14. As a result, the fiscal balance turned into a deficit in 2014/15. FIGURE 4.2 TRENDS IN GDP BETWEEN 2000 AND 2016 (IN 2014 CONSTANT PRICES) 60 50 40 Billion SZL 30 20 10 0 2000 2002 2004 2006 2008 2010 2012 2014 2016 Year ³ The AGOA is a trade privilege scheme enacted by the United States to enhance access to the US market for qualifying sub-Saharan African countries. To qualify and remain eligible for the AGOA, each country must be working to improve its rule of law, human rights and respect for labour standards. The Act originally covered an eight-year period, from 2000 to 2008, but was later extended to 2015. ENERGY MASTERPLAN 2034 39 Figure 4.3 shows the sectoral composition of GDP and its de- product, although its contribution is expected to fall drastically velopment over the years. In 2016, the main contributor to GDP in the coming years due to the drought experienced in recent was commerce and government services at 53.1 %, followed by farming seasons. The effects of drought go beyond food inse- other industry at 25.2 %, the sugar industry at 14.3 %, agricul- curity and also adversely impact the production of energy. The ture at 7.3 % and mining at 0.1 %. Sugar is the key agricultural contribution to GDP from the mining sector is negligible. FIGURE 4.3 TRENDS IN GDP BY SECTOR BETWEEN 2000 AND 2016 (IN 2014 CONSTANT PRICES) 60 50 40 Billion SZL Sugar 30 Agriculture other 20 Other industry 10 Mining Commerce and 0 govt. services 2000 2004 2008 2012 2016 Year In the coming years, several key challenges lay ahead. They Despite the challenges ahead, the Government aims at GDP include: growth of 1.8 % per year for the coming years. Given the histori- cal trends and policy priorities discussed above, the assump- • reduced GDP growth rate and forecasted deceleration in the tions regarding the development of the country’s economic coming years structure are shown in Figure 4.4 for the Reference scenario. • reduced demand for Eswatini’s major exports The commerce and government services sector is assumed • lingering fiscal imbalances with unprecedented budget short- to be the key driver of economic growth, with average annual falls and unhealthy accumulation of domestic arrears growth of 1.8 % in 2014, rising to 3 % by 2020 and 3 % thereafter. • reduced external reserves (current account) • increased dependence (over-reliance) on SACU revenues with high volatility. 40 KINGDOM OF ESWATINI FIGURE 4.4 ASSUMPTIONS REGARDING THE COMPOSITION OF GDP 90 80 70 60 Billion SZL 50 Sugar 40 Agriculture other 30 Other industry 20 Mining 10 Commerce and 0 govt. services 2014 2018 2022 2026 2030 2034 Year For the High Growth scenario, all of the sectors have the same mining sector is zero), and growth in the entire demand sector GDP growth rate as in the Reference scenario, except that is equal to economic growth, similar to the Low Growth scenar- GDP growth increases to 3.5 % from 2025 to 2034. In this sce- io. The GDP assumptions for future years in the three scenarios nario the mining sector is assumed to contribute SZL 4 billion are shown in Figure 4.5. (USD 286 million4) by 2034 (in the Reference case, GDP in the FIGURE 4.5 THREE GDP SCENARIOS FOR 2014 TO 2034 100 90 80 70 60 Billion SZL 50 40 30 20 Low Growth 10 High Growth 0 Reference 2014 2018 2022 2026 2030 2034 Year ⁴ An Exchange Rate of 14 SZL = 1 USD was applied for this modelling analysis across the planning period 2014–2034. ENERGY MASTERPLAN 2034 41 Electricity For this particular exercise, modelling for the rural electrifica- tion targets has taken into account the slow injection of funds In 2014 the national electricity access rate (i. e., the share of to support the rural electrification programme. Hence the tar- households with electricity access) was 65 %, with the urban get for universal access to electricity is projected to be reached rate being 84 % and the rural household rate being 60 %. Rural later, in 2030 (and 2034 under the Low Growth scenario). To electrification continues to be a priority issue in Eswatini. Ro- provide an outlook for electricity access until 2030, a model bust implementation of the rural electrification programme in able to generate projection of electrification rates has been de- recent years has since increased household electricity access veloped under the Reference scenario assumptions. The model to 74 % in 2017. was developed using the same economic and demographic as- sumptions, taking into account variables such as the rural elec- The objective of the Government of Eswatini is to ensure that trification programme as well as the broad policy commitments all citizens of the country have access to electricity by the year and plans that have been announced around the world to tackle 2022. This ambitious target can only be realised with high in- environmental and energy security concerns in the era of sus- vestment support from the Government to extend the electrici- tainable energy for all. ty grid to non-electrified areas. In recent years, the Government has collaborated with donor funding agencies to support im- plementation of the country’s rural electrification programme. FIGURE 4.6 ELECTRIFICATION PROJECTIONS FOR 2014 TO 2034 100 95 90 Electrification rate (%) 85 80 75 70 65 60 Low Growth 55 High Growth 50 Reference 2014 2018 2022 2026 2030 2034 Year The results in Figure 4.6, however, assume that no addition- (USD 8.6 million) per year. The total number of households was al policies to expand energy access, over and above those in estimated by dividing the total population by the estimated av- place today, are enacted over the projection period. The model erage household size (i. e., persons per household). Currently also assumes a 1.2 % increase in population. The model takes the average household size is estimated to be 4.2 persons per into account the required investment in limited funding for the household. Together with the electrification rate, the number of rural electrification programme, estimated at SZL 120 million electrified households was computed. 42 KINGDOM OF ESWATINI Modern cooking fuels in Figure 4.7, even though fuelwood has remained the major energy resource for cooking over the years, electricity use for In addition to electrical energy, modern energy resources in- cooking is increasing steadily and has remained higher than clude hydrocarbon fuels (e. g., LPG, candles and coal). Despite LPG and paraffin. Use of electricity in cooking is expected to a national electricity access rate of more than 60 %, the use of continue to grow in the future. electricity for cooking in rural areas is not common. As shown FIGURE 4.7 SHARE OF HOUSEHOLD FUEL USED FOR COOKING 90 80 70 60 Percentage (%) 50 40 30 Fuelwood 20 Electricity 10 Paraffin 0 LPG 1997 2007 2013 Year Source: Swaziland Households Energy Access Report, 2014 Fuelwood remains the main energy resource for cooking and To assess the future demand for traditional biomass, assump- space heating, particularly in rural areas. However, according to tions were made regarding the share of households using bio- the 2016 Wood Efficient Cook Stoves survey conducted by the mass for cooking, and the efficiency of biomass cook stoves. In Renewable Energy Association of Swaziland on behalf of the line with SE4ALL goals and other policies to promote alterna- Energy Department, fuelwood is generally used inefficiently in tive modern fuels for cooking, it was assumed that by 2030 rural areas. Following a pilot project aimed at promoting the the share of households using fuelwood would be 20 % (the efficient use of fuelwood, the Government is planning to ex- same as in 2014), the share using LPG would be 20 % (up from tend the use of wood-efficient cook stoves, which can aid in 14 % in 2014), and the share using electric stoves would be 80 % the worldwide fight against deforestation and climate change. (up from 69 % in 2014). The policy also promotes efficient cook stoves, and a 50 % efficiency improvement in wood cook stoves is assumed to 2034. ENERGY MASTERPLAN 2034 43 Residential water heating 4.2.4 ENERGY EFFICIENCY GOALS Increased penetration of efficient water heating is expected in The National Energy Policy of 2003 promotes improving en- the residential sector, in line with various policy goals and tech- ergy efficiency in all sectors. According to the policy, energy nology developments on appliance energy efficiency stand- efficiency and savings aim to reduce energy consumption at ards. Current shares of water heating devices are estimated at the end-user level as well as in the supply system. Energy ef- 82 % for wood, 9 % for geysers, 5 % for solar water heaters with ficiency is about reducing losses within a technical energy sys- back-up, 0 % for solar water heaters without back-up and 4 % tem by optimising energy use, whereas energy savings covers a for other electric devices. broader conservation methodology, including behavioural and operational issues. Energy efficiency initiatives in Eswatini in- The most inefficient wood-based water heating is replaced by clude awareness raising and information dissemination, promo- other more efficient options, reducing its share by 13 % by 2034. tion of energy management in all sectors, building standards, Reflecting its promotion under the SE4ALL initiative, the share energy efficiency within government institutions, energy appli- of solar water heating is assumed to reach 50 % of households ance labelling and demand-side management. Ongoing energy (25 % with back-up and 25 % without back-up). efficiency initiatives include: • installation of solar water heaters in government buildings; • awareness and information dissemination through live radio programmes, roadshows and outdoor advertising; • implementation of time-of-use tariffs; • energy audits of public institutions as well as corporate customers; • roll-out of prepaid electricity metering; • distribution of compact fluorescent lamps (CFLs) and LEDs to the public; • installation of solar PV and LED lighting demonstration projects; and • adoption of energy efficiency standards. The SE4ALL initiative includes some specific efficiency im- provement targets (see Section 2.3), which collectively aim to double the rates of improvements in energy efficiency by 2030. Across all of the scenarios in LEAP-Swaziland, the following ef- ficiency improvements are taken into consideration: • Coal use in the commerce and government services sector • Wood cook stoves in the residential sector. 44 KINGDOM OF ESWATINI 5 ENERGY DEMAND ASSESSMENT 2015–2034: RESULTS

5.1 OVERALL ENERGY DEMAND DEVELOPMENT: half, from 14.5 PJ in 2014 to 6.4 PJ in 2034. As a result of the 2015–2034 electrification of non-electrified households, residential sector energy demand drastically decreases by 2030; once the resi- 5.1.1 ENERGY DEMAND DEVELOPMENT BY dential sector switches to other energy sources, the demand 2034 IN THE REFERENCE SCENARIO starts to pick up again slightly between 2030 and 2034. The sugar industry, agriculture and other industry sectors are ex- The overall energy demand of the country is projected to in- pected to see the largest increase in energy demand of all sec- crease from 39.4 petajoules (PJ) in 2014 to 48.5 PJ in 2034 (Fig- tors, with 75 % expected growth respectively over the planning ure 5.1). This projected increase is driven by the shift from inef- period. The energy demand in the transport and commerce and ficient traditional biomass use to more efficient modern fuels government services sectors is projected to grow by 48 % and – in particular, electricity – in the residential sector. The demand 22 % respectively over the planning period. for traditional biomass is projected to decrease by more than FIGURE 5.1 FINAL ENERGY DEMAND BY SECTOR, 2014–2034 60 50 40 Mining PJ 30 Sugar industry Agriculture other 20 Industry other Commerce and govt. services 10 Residential Transport 0 2014 2018 2022 2026 2030 2034 Year ENERGY MASTERPLAN 2034 45 Figure 5.2 shows the same final energy demand projections, Other industry (excluding sugar but including, e. g., manufactur- segregated by fuel type, between 2014 and 2034. Table 5.1 ing, construction, food and tobacco) is currently the fourth larg- shows the final energy demand projected for 2034, presented est end-user of electricity and its demand grows 80 % by 2034, by fuel type and by sector. when industry contributes around 15 % of electricity demand. This increase is due to economic growth that boosts energy de- The demand for oil products experiences strong growth of mand, but is partly balanced off by more efficient energy use around 50 % by 2034, with 95 % of this growth coming from the by industries over time, which constrains their growth in energy transport sector. Coal has a relatively constant share of total demand. Residential electricity demand increases from 1.2 PJ in energy demand at around 5 % from 2014 to 2034, and remains 2014 to 3.7 PJ in 2034; it contributes the largest sector share of concentrated in the sugar and commerce and government ser- total electricity demand by 2034. The increase attributable to vices sectors. Electricity demand more than doubles, from 1 270 the gain in electricity access represents about 50 % of the total GWh in 2014 to 2 648 GWh (9.5 PJ) in 2034. growth in electricity demand. Note that in Figure 5.2, “Coal unspecified” refers to coal use in commerce and government services that does not include coal- fired power plants. FIGURE 5.2 FINAL ENERGY DEMAND BY FUEL TYPE, 2014–2034 60 50 40 Coal unspecified PJ 30 Bagasse Wood 20 Oil LPG 10 Kerosene 0 Electricity 2014 2018 2022 2026 2030 2034 Year 46 KINGDOM OF ESWATINI TABLE 5.1 FINAL ENERGY DEMAND BY SECTOR AND BY FUEL TYPE IN 2034 (IN PJ) Sector Electricity LPG Oil Wood Bagasse Coal Total Sugar 2.3 – 0.1 4.7 14.8 2.1 24 Agriculture other 1.5 – 0.7 – – – 2.2 Industry other 1.4 – 0.1 – – – 1.5 Mining – – – – – – – Commerce and govt. services 0.6 0.4 – – – 0.5 1.5 Residential 3.7 0.6 – 1.7 – – 6.0 Transport – – 13.3 – – – 13.3 Total 9.5 1.0 14.2 6.4 14.8 2.6 48.5 Note that an overview of detailed results of the demand assess- ment to 2034 is provided in Annex B. 5.1.2 ENERGY DEMAND DEVELOPMENT BY as a result of changes in the residential sector, driven largely by 2034 IN ALTERNATIVE SCENARIOS population growth coupled with increased modernisation. As liv- ing standards improve, people tend to consume more modern In all economic sectors, energy demand grows at the same rate energy sources, which are more energy efficient; interventions in as GDP. Figure 5.3 charts the energy demand growth forecasts energy efficiency also tend to slow total energy demand. for the Reference, High Growth and Low Growth scenarios. Using the GDP growth rate as one of the key drivers, energy demand Until 2028, the Reference scenario has the highest energy de- projections were conducted for each energy demand sector. mand. Because it represents the most likely GDP growth for the country, the corresponding demand under this scenario was The demand projections indicate a slight increase in total energy selected as the projected demand to be used as input to the demand by 2034, as compared to 2014 levels. Under the High modelling analysis of the energy supply strategy (Part III). Growth scenario, there is a steep decline in energy consumption FIGURE 5.3 ENERGY DEMAND GROWTH UNDER THE THREE SCENARIOS 60 50 40 PJ 30 20 High growth 10 Low growth 0 Reference 2014 2018 2022 2026 2030 2034 Year ENERGY MASTERPLAN 2034 47

5.2 ENERGY DEMAND DEVELOPMENT IN KEY well as consumption patterns of agricultural products. Principal SECTORS agricultural products in Eswatini include sugar, maize, cotton, forestry, citrus and livestock. GDP in the agriculture sector is

5.2.1 AGRICULTURE assumed to be stimulated by economic growth over the next 20 years, as is consistent with trends in the last few years (Fig- The growth of the agriculture sector is highly dependent on ure 5.4). Reflecting this, energy demand in the agriculture sec- favourable weather conditions, especially rainfall patterns as tor is assumed to increase nearly two-fold. FIGURE 5.4 ENERGY DEMAND IN THE AGRICULTURE SECTOR 2.5 2.0 1.5 PJ 1.0 0.5 Oil Electricity 0 2014 2018 2022 2026 2030 2034 Year

5.2.2 SUGAR INDUSTRY 20 years, in line with past trends. Nevertheless, energy demand in the sugar industry is increasing by more than 2 % annually, Similar to the agriculture sector, GDP in the sugar industry is which is replicated for the period 2014–2034. assumed to be stimulated by economic growth over the next FIGURE 5.5 ENERGY DEMAND IN THE SUGAR INDUSTRY 30 25 20 Coal unspecified PJ 15 Bagasse 10 Wood 5 Oil Electricity 0 2014 2018 2022 2026 2030 2034 Year 48 KINGDOM OF ESWATINI

5.2.3 OTHER INDUSTRY is expected to see modest growth in energy demand of about 3 % annually between 2014 and 2034. Electricity accounts for Other industry refers to all industries except for the sugar in- the vast majority of current energy use in this sector, and the dustry, including construction, food and tobacco, manufactur- future growth in energy demand is expected to be met primar- ing, textiles, etc. Reflecting the Government’s policy on diver- ily by electricity. sification of the economic structure, the other industry sector FIGURE 5.6 ENERGY DEMAND IN OTHER INDUSTRY 1.6 1.4 1.2 1.0 PJ 0.8 0.6 0.4 Oil 0.2 Electricity 0.0 2014 2018 2022 2026 2030 2034 Year

5.2.4 COMMERCE AND GOVERNMENT SERVICES at a slower pace, of around 1.0 %, due to efficiency improve- ments in coal use in this sector. While overall coal use in the The commerce and government services sector is expected to sector declines over time, from 52 % of total sectoral energy de- be the biggest economic sector contributing to GDP by 2034. mand in 2014 to 29 % by 2034, the demands for electricity and GDP in the sector is expected to grow by about 2.9 % annually LPG increase in proportion to the GDP growth in this sector. in the Reference scenario. Energy demand is expected to grow FIGURE 5.7 ENERGY DEMAND IN COMMERCE AND GOVERNMENT SERVICES 1.8 1.6 1.4 1.2 PJ 1.0

0.8 Coal unspecified 0.6

0.4 LPG 0.2 Electricity 0.0 2014 2018 2022 2026 2030 2034 Year ENERGY MASTERPLAN 2034 49

5.2.5 TRANSPORT fuels, specifically biofuels. A constraint to increasing the fuel demand for road transport is the severe lack of paved roads. A Only 10.5 % of the population of Eswatini owns a passenger second and more positive factor is vehicle fuel efficiency, which light-duty vehicle, compared with 70 % in the United States, for new vehicles is expected to improve by more than 20 %, on 50 % in Europe and 6 % in China, indicating huge growth po- average, to reach 7.2 litres per 100 kilometres by 2034. Eswatini tential. While the number of vehicle registrations is projected relies heavily on imports of second-hand vehicles from Japan to double by 2034, this implies that only around 16 % of the and Europe, both of which have in place comprehensive fuel population is expected to own a vehicle by 2034. As urban ar- economy standards. To a degree, these standards are progres- eas grow significantly in size, urban development policies play sively imported, helping to improve the average fuel efficiency an important role in guiding users of transport services towards of vehicles in Eswatini. private or public forms of transport, and therefore influence future transport energy demand. Diesel accounts for 60 % of the total demand for oil in the trans- port sector, and petrol accounts for 40 %. Energy demand in The transport sector remains almost entirely reliant on oil prod- the sector is expected to increase by almost 50 % in the pro- ucts, with policies in place to promote the use of alternative jected planning horizon. FIGURE 5.8 ENERGY DEMAND IN THE TRANSPORT SECTOR 14 12 10 8 PJ 6 4 2 0 Oil 2014 2018 2022 2026 2030 2034 Year 50 KINGDOM OF ESWATINI

5.2.6 RESIDENTIAL The modelling of access to modern energy has undergone sig- nificant development to better reflect access to electricity and The choice of energy use in households is influenced by fluc- access to clean cooking facilities, and played a role in shap- tuations in relative prices, but also by self-reliance in the physi- ing energy demand in the residential sector. These are priority cal availability of the alternatives to solid biomass. Where new areas in SE4ALL’s Agenda 2030. Urban areas dominate in the sources of energy use in households are adopted, it does not share of electricity supply, accounting for 70 % of total electric- mean that the use of older sources is discarded. Until distribu- ity consumption. Electricity access is improving in both urban tion networks are sufficiently well established to ensure a reli- and rural areas, but urban electrification rates continue to be able supply of alternative energy carriers at reasonably predict- higher, and urban households consume more electricity. Grid- able prices, consumers prefer to retain the option to switch. based systems provide electricity to rural populations when those communities are in close proximity to transmission lines, Fuelwood is adjusted to reflect relative scarcity in non-electri- and extending the grid to them is a viable option for rural areas fied and electrified households. In non-electrified households to gain access to electricity. with extensive forestry, fuelwood has a smaller price increase where fuelwood remains untraded and has a zero price. Elec- The mix of technologies is distinct in the residential sector, with trified households, on the other hand, see a more pronounced electricity and fuelwood being the dominant sources of ener- price increase. Average end-user electricity prices are generally gy supply. The use of wood in the sector decreases, mainly in assumed to adjust gradually over time to reflect the average cooking. This decline is driven by changes in both the fuels and cost of electricity supply, including domestic generation and the cook stoves used: traditional use of solid biomass is gradually cost of electricity imports, network, retail and other costs. As a reduced on a per household basis as households switch to more result, overall energy demand in the residential sector decreases efficient, less polluting cook stoves, such as improved biomass by more than half due to higher efficiency in electric stoves. stoves, or to fuels such as LPG or electricity. FIGURE 5.9 ENERGY DEMAND IN THE RESIDENTIAL SECTOR 16 14 12 10 8 PJ 6 Wood 4 LPG 2 Kerosene Electricity 0 2014 2018 2022 2026 2030 2034 Year ENERGY MASTERPLAN 2034 51 Nationwide, 26 % of households do not have access to electric- attempts to address energy deficits among non-electrified ity (as of 2017). These households often utilise multiple sources households through its successful household electrification pro- of energy such as wood, LPG and paraffin for cooking, heating gramme. The rate of non-electrification starts at 35 % in 2014, and lighting. Over the planning horizon of 2014 to 2034, the declines to 0 % in 2030 and remains constant at 0 % until 2034. consumption of fuelwood declines gradually. The Government The absence of fuelwood use in 2034 is shown in Figure 5.10. FIGURE 5.10 ENERGY USE IN NON-ELECTRIFIED HOUSEHOLDS 12 10 8 PJ 6 4 Wood 2 LPG Kerosene 0 2014 2020 2025 2030 2034 Year Figure 5.11 shows energy use in electrified households. These Although progress in electrification has contributed greatly to- households require sufficient energy to conduct domestic wards achieving improved access to safer, cleaner and more chores such as cooking, heating, lighting, refrigeration, etc. affordable energy services for households across the socio- The figure for 2034 represents 100 % of households, as it is economic spectrum, energy use continues to be characterised assumed that a 100 % electrification rate is achieved by then. by the use of multiple sources of energy. LPG consumption The increase reflects both the increased number of households develops in line with the overall trends for electricity demand as well as the increased consumption per household. In the in electrified households, rising from 0.2 PJ in 2014 to 0.6 PJ in absence of accessible or affordable modern energy services, 2034. Wood consumption increases slightly from 1.45 PJ in 2014 many households turn to alternative sources of energy, includ- to 1.65 PJ in 2034, whereas kerosene consumption declines ing wood, LPG and kerosene. Fuelwood is the widely used al- gradually until 2034, to zero. ternative source due to its “free” availability from nature. While some of these sources provide practical alternatives for house- The perception is that the transition from wood to modern hold energy requirements, they are costly. fuels, particularly electricity, is associated with economic im- provements and improvements in development. Low-income The analysis of energy use in electrified households over time households tend to depend on wood for cooking and heating, shows that the energy usage patterns of households are chang- while electricity is often limited to lighting, refrigeration, tel- ing (Figure 5.12). These changes are driven primarily by elec- evisions, ironing and other appliances such as radios and mo- trification, particularly in non-electrified households. Electric- bile phone charging. As the years progress, the fuel used for ity consumption in the residential sector increases more than cooking switches from biomass to electricity; electricity use for three-fold, from 1.2 PJ in 2014 to 3.7 PJ in 2034. cooking increases 61 % from 2014 to 2034. 52 KINGDOM OF ESWATINI FIGURE 5.11 ENERGY END-USE IN ELECTRIFIED HOUSEHOLDS 7 6 5 4 Other PJ 3 TV/Ironing Lighting 2 Air conditioning Refrigeration 1 Water heating Cooking 0 2014 2020 2025 2030 2034 Year FIGURE 5.12 ENERGY USE IN ELECTRIFIED HOUSEHOLDS 7 6 5 4 PJ 3 2 Wood LPG 1 Kerosene Electricity 0 2014 2020 2025 2030 2034 Year ENERGY MASTERPLAN 2034 53 PART III: ENERGY SUPPLY ASSESSMENT 6 DOMESTIC ENERGY RESOURCES AND ENERGY TECHNOLOGY OPTIONS

6.1 SUGARCANE Table 6.1 provides an overview of the area under cultivation, area harvested and sugarcane production in Eswatini between Eswatini operates three sugar mills, Mhlume Mill, Simunye Mill 2012/13 and 2016/17. Sugarcane production is increasingly and Ubombo Mill, all situated in the eastern part of the country. becoming an area of investment by the private sector and The first two mills are operated by the Royal Swaziland Sug- community-based farmers. Despite the falling price of sugar, ar Corporation (RSSC), and the third is operated by Ubombo sugarcane production is anticipated to increase. Sugarcane is Sugar Limited. The three mills produce sugar from sugarcane the main crop produced in Eswatini, with annual production grown in Eswatini. Sugar processing has two major by-prod- of 5.8 million tonnes in 2015. Sugarcane production was an- ucts, bagasse and molasses. Bagasse is used with wood chips ticipated to decrease to 4.6 million tonnes by the end of the in boilers to produce process steam that is also used for power 2016/2017 fiscal year as a result of El Niño-related drought in generation. Molasses is used for ethanol production. the country. TABLE 6.1 OVERVIEW OF SUGARCANE AREA AND PRODUCTION, 2012/13 TO 2016/17 2012/13 2013/14 2014/15 2015/16 2016/17* Area under cultivation (ha) 57,263 58,979 59,586 59,924 61,798 Area harvested (ha) 53,666 55,478 56,438 57,685 58,520 Cane production (tonnes) 5,648,542 5,591,830 5,639,193 5,836,553 4,633,641 * Estimated figures Source: Swaziland Sugar Association Sugarcane production is expected to increase in the near fu- stated an interest in increasing their power generation capacity ture, providing an opportunity to increase co-generation ca- from 106 MW to 160 MW. Table 6.2 shows some of the planned pacity. Co-generation also provides an opportunity for diversi- capacity by selected independent power producers. fication in the sugar industry’s revenue. Sugar companies have TABLE 6.2 PLANNED ENERGY GENERATION CAPACITIES BY THE SUGAR INDUSTRY Independent New total capacity Internal use Grid supplied Projected year of power producer (MW) (MW) (MW) completion Mhlume Sugar Mill ~50 22 ~28 2024 Simunye Sugar Mill ~50 28 ~22 2022 Ubombo Sugar Mill 25 15 2020 56 KINGDOM OF ESWATINI The RSSC has plans to increase its power generation capacity completion of the project, sugar production has the potential at the Mhlume and Simunye mills, which requires mill efficiency to increase by an estimated 1,150,000 tonnes, providing an op- improvements and the installation of high-pressure boilers. portunity to also increase related power generation. The Swaziland Water and Agricultural Development Enterprise 6.1.1 BAGASSE (SWADE) has begun implementing the second phase of the Lower Usuthu Smallholder Irrigation Project (LUSIP), which in- Sugar millers use bagasse, a by-product of the sugar produc- volves the construction of three dams to form an off-river stor- tion process, as a fuel for boilers that produce process steam, age reservoir to impound 155 million cubic metres of water that which also can be used in power generation. From experience, will be diverted from the wet/rainy season flood flows into the an estimated 100 tonnes of sugarcane can produce 27 tonnes Lower Usuthu River to irrigate a total area of 11,500 hectares of bagasse. On average, local sugar millers have the capacity to of land, primarily sugarcane plantations. One hectare is esti- produce 1,539,000 tonnes of bagasse annually. Table 6.3 shows mated to yield about 100 tonnes of sugarcane. Therefore, upon bagasse production from 2012/13 to 2016/17. TABLE 6.3 BAGASSE PRODUCTION, 2012/13 TO 2016/17 2012/13 2013/14 2014/15 2015/16 2016/17* Bagasse production (tonnes) 1,525,106 1,509,794 1,522,582 1,575,869 1,251,083 * Estimated figures

6.1.2 MOLASSES also is being considered for E10 biofuels production. RSSC and USA Distillers are the two major ethanol producers in Eswatini, Molasses is another useful by-product of sugar processing, and making 134,000 litres per day and 120,000 litres per day, re- a total of 253,073 tonnes was produced in the 2015/2016 fis- spectively. The combined total production capacity is approxi- cal year (Table 6.4). Molasses is used mainly for ethanol pro- mately 60 million litres annually, which is currently used for duction for the beverage and pharmaceutical industries, but it non-energy purposes. TABLE 6.4 MOLASSES PRODUCTION, 2012/13 TO 2016/17 Producer 2012/13 2013/14 2014/15 2015/16 2016/17* Simunye Mill (tonnes) 72,700 76,338 76,179 84,180 61,616 Mhlume Mill (tonnes) 63,300 60,385 68,664 68,480 49,823 Ubombo Mill (tonnes) 111,267 101,041 117,641 100,413 92,383 Total industry (tonnes) 247,267 237,764 262,484 253,073 203,822 * Estimated figures ENERGY MASTERPLAN 2034 57 The flow chart in Figure 6.1 depicts the main steps in the pro- duction of ethanol from sugarcane molasses. FIGURE 6.1 PROCESS OF ETHANOL PRODUCTION FROM SUGARCANE MOLASSES Sugar production Extraction of juice Molasses Sugar from sugarcane (waste product) Production of alcohol (ethanol) Anhydrous ethanol – 99.9 % (fuel grade) Dehydration through fermentation & distillation Denaturing Denatured ethanol Blending process (90 % unleaded petrol + 10 % ethanol) Biofuel (E10) The biofuels programme for Eswatini is preparing for the full programme will support the Government’s ambition to reduce roll-out of ethanol blending (E10). The objective is to introduce carbon emissions, particularly in the transport sector, which an alternative, locally produced and environmentally friendly currently contributes more than 70 % of the country’s overall fuel into the fuel mix and to enhance the country’s energy energy sector emissions. security in terms of petroleum supply. The ethanol blending 58 KINGDOM OF ESWATINI

6.2 WOODY BIOMASS and sugar industries have established an association that in- cludes among its activities the exploration of power generation

6.2.1 WOOD CHIPS opportunities using available biomass resources. The timber in- dustry has the potential to increase the fuel supply to meet new The use of wood chips in Eswatini is on the rise due to increased biomass power generation through the provision of wood chips demand by sugar companies for use in co-generation. Currently and forestry waste. the average annual demand for wood chips by the sugar in- dustry is approximately 206,000 tonnes, and so far the timber Table 6.5 shows additional biomass power generation projects industry is able to meet this demand. Key players in the timber under consideration by the timber saw mills. TABLE 6.5 POTENTIAL POWER GENERATION CAPACITIES IN THE TIMBER INDUSTRY Independent power producer Total capacity (MW) Internal use (MW) Grid supplied (MW) Piggs Peak 15 10 5 Nhlangano 10 5 5 Bulembu 1 1 0 Usuthu Saw Mill 37 4 33 Montigny Investment, which operates the Usuthu Saw Mill (for- ing the future availability of wood resources. The unsustainable merly Sappi Usuthu), has publicised its intention to develop a gathering of fuelwood along with the clearing of woodlands for 33 MW power generation project. The company has attracted agricultural production and the construction of homesteads has interested financiers for the project and is working through ne- resulted in deforestation and land degradation. The availabil- gotiations with the SEC to start developing the project. ity of fuelwood in the country averages 50 %, with most of the fuelwood obtained freely from non-commercial communal for-

6.2.2 FUELWOOD ests. Approximately 14 % of fuelwood is obtained from commer- cial supply, which is not regulated and is traded uncontrolled. Biomass accounts for over 60 % of Eswatini’s total primary energy supply and comprises mainly traditional biomass and Eswatini has, however, controlled the exploitation of indigenous agro-industrial waste for co-generation. According to statistics species of wood for fuelwood and other non-energy uses. Al- from the Food and Agriculture Organisation of the United Na- though desertification in the country is limited to soil erosion, tions, fuelwood production in Eswatini has increased 25 % in biomass productivity and bush encroachment, Eswatini has a the last decade, reaching 1,093,333 cubic metres in 2012. Char- policy that safeguards against commercial use of fuelwood due coal production increased 50 % over the same period, reaching to over-exploitation of the biomass resource. The magnitude of 44,933 tonnes. The country’s total forested area of 624,000 desertification in the country is not comparable with other sub- hectares represents 36 % of the land area; of the forested total, Saharan African countries, where large portions of the territory 464,000 hectares is indigenous forest and woodlands and the have been rendered unsuitable for human occupation. rest is commercial forest plantations. Furthermore, Montigny has a Wattle Extension programme According to the National Forestry Policy, over the years, the focused on combatting the unsustainable use of biomass and increase in population, coupled with the rising demand for fuel- indigenous forest deforestation by promoting wattle manage- wood and poor management of indigenous forests, has placed ment and land optimisation. The project aims to train landown- a high burden on Eswatini’s indigenous woodlands and forests. ers on how to control the spread of wattle by rehabilitating it As a result biomass resources have been depleted in certain ar- into a commercial timber plantation while preserving some land eas. The demand for fuelwood continues to increase, threaten- for domestic use, easing the burden on indigenous forests. ENERGY MASTERPLAN 2034 59

6.3 COAL Coal has been known to exist in Eswatini for a long time. Fig- ure 6.2 depicts the concession map for the locality drawn in the late nineteenth century accurately showing the coal field of Eswatini. FIGURE 6.2 LOCALITY MAP OF THE COAL FIELD OF ESWATINI The coal field of Eswatini, in the eastern part of the country, country’s total area of 17,364 square kilometres. The mineable forms a north-south continuous area with an average width of coal reserves, annual production and projected mine life for ar- 15 kilometres. Located within the Lowveld, the coal field com- eas shown in Figure 6.2 are summarised in Table 6.6. prises Karoo sediments and occupies about one-sixth of the 60 KINGDOM OF ESWATINI TABLE 6.6 MINEABLE COAL RESERVES, ANNUAL COAL PRODUCTION AND PROJECTED LIFE Mineable coal Estimated annual In-situ reserves Area (million tonnes) production (tonnes) (years) Mhlume 18.4 665,000 27 Area 1 9.1 – – Area 2 Mpaka Mine 41.2 500,000 82 Area 3 20.6 500,000 41 Lubhhuku 18.9 510,000 37 Maloma 35.3 300,000 58 Total 143.5 2,475,000 – The coal field of Eswatini consists of semi-anthracitic or lean making it necessary for new townships to be constructed for coals and anthracite. Coals of the Upper Coal Zone are inferior- any new coal mines. Coal’s relatively low cost remains an as- quality anthracite with a high ash content compared to some set in regions concerned about the affordability of electric- of the high-quality anthracite of the Lower Coal Zone. Maloma ity. The coal currently consumed domestically is imported and Colliery, Eswatini’s sole active coal mine, is producing anthra- includes bituminous coal. In 2014 a total of 208 kilotonnes of cite coal for export purposes with a production capacity of ap- anthracite was exported, and 255 kilotonnes of bituminous proximately 300 kilotonnes per year. coal was imported. There is great potential for new developments in coal mining. 6.4 WIND The country is considering developing a coal-fired thermal power station that would result in the establishment of a new The Ministry of Natural Resources and Energy participated in mine to generate feedstock for the power station. The produc- the renewable zoning study under IRENA’s Africa Clean Energy tion capacity of the mine is yet to be determined. Discussions Corridor initiative (IRENA and LBNL, 2015)5, which resulted in are ongoing to resuscitate the defunct Mpaka Coal Mine for inventories of possible wind and solar project zones. power generation as well as for export. The coal field of Eswati- ni is served by an adequate system of tarred roads designed to The zoning study inventory identified a total of 17 potential serve both the coal and sugar industries. Eswatini has direct au- zones for wind power generation, which collectively amount tomatic dialling telecommunication with all neighbouring states to 4,300 MW of possible generation capacity. The total dis- and the world at large. counted generation potential of the 17 zones far exceeds the projected national power demand in 2030. Hence, applying the Adequate electricity distribution networks already exist for projected national power demand as the supply limit, and by any new coal mining development in the coal field. It will be means of the multi-criteria analysis tool developed for the zon- necessary to develop adequate water storage facilities, as the ing study (see below), the Eswatini working team selected for coal field is liable to drought. The coal field is remote from further analysis 5 priority zones from the 17 potential zones for the country’s present administrative and industrial centres, wind power generation. ⁵ IRENA and LBNL (2015). Renewable Energy Zones for the Africa Clean Energy Corridor — Multi-Criteria Analysis for Planning Renewable Energy. International Renewable Energy Agency and Lawrence Berkeley National Laboratory, Abu Dhabi, United Arab Emirates. ENERGY MASTERPLAN 2034 61 The multi-criteria analysis tool in the zoning study applies user- Figure 6.3 shows (in red) the five priority zones screened from defined weighting factors to come up with a cumulative score this process – namely, zones J, I, B, K and M, in order of prior- for screening zones. The national working team decided to ap- ity. The properties of these zones are summarised in Table 6.7. ply the following factors and respective weights for wind power The five priority zones jointly correspond to about 880 MW of zones: 70 % for the levelised cost of electricity (LCOE) of wind output, with a mean capacity factor of 28 % to 31 %. The LCOE generation (including investment mark-up to account for the of wind farms in these zones is expected to be in the range of road and transmission infrastructure needs), 10 % for the capac- USD 95 to USD 106 per megawatt-hour (MWh), including trans- ity value (i. e., production potential at times of high demand)6, mission and the associated road investment. Across the five and 5 %, respectively, for distance to load centre, co-location zones, the cost mark-up is 3.4 % to 6.7 % for transmission and potential with solar PV, population density and human footprint7. 0 % to 0.3 % for road infrastructure investments. FIGURE 6.3 LOCATION OF THE FIVE PRIORITY ZONES FOR WIND POWER ⁶ “Capacity value” can be defined as “the contribution of a power plant to reliably meeting demand” and “the contribution that a plant makes toward the planning reserve margin” (Madaeni, S.H., Sioshansi, R., Denholm, P., (2012). Comparison of Capacity Value Methods for Photovoltaics in the Western United States (No. NREL/TP-6A2º874704). National Renewable Energy Laboratory (NREL), Golden, Colorado, US). ⁷ The factor “human footprint” provides a proxy for the degree of human disturbance. More information on the zoning approach and the weighting factors is available at http://mapre.lbl.gov/. 62 KINGDOM OF ESWATINI TABLE 6.7 CHARACTERISTICS OF IDENTIFIED WIND ZONES Annual Installed Distance to Distance to Average Mean electricity Distance to Zone name capacity substation load centre wind speed capacity generation road (km) (MW) (km) (km) (m/s) factor (MWh) B 226 584,000 81 0 77 4.7 0.29 I 44.4 116,000 25 0 67 4.9 0.30 J 121 327,000 6 0 50 4.6 0.31 K 206 514,000 21 1 55 4.6 0.28 M 283 687,000 8 1 22 4.4 0.28 While the capacity factors differ somewhat across zones, over- is higher throughout the second half of the year. Across the all generation is assumed to follow a consistent pattern. Fig- whole year production is highest at night time, which some- ure 6.4 shows the average day capacity factors for each re- what complements solar PV that does not generate power dur- spective month for wind power in zone J (i. e., the best-ranking ing the night (see Section 6.5 below). zone, as per the cumulative zone score). On average production FIGURE 6.4 AVERAGE DAY WIND CAPACITY FACTORS FOR EACH MONTH, ZONE J (VALUES FROM 2015)8 Average day wind power capacity factors for each month, zone J Months Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 0.6 0.5 Fraction of theoretically 0.4 0.3 possible production 0.2 0.1 0 6 12 6 12 6 12 6 12 6 12 6 12 6 12 6 12 6 12 6 12 6 12 6 12 18 18 18 18 18 18 18 18 18 18 18 18 Hours ⁸ N ote that these average values are derived from hourly data for one year. It may reflect peculiarity of that year. The generating patterns that were assessed are likely to hold in the long term. ENERGY MASTERPLAN 2034 63 Figure 6.5 shows the load curves for Eswatini, of average days All months have, on average, a peak in demand in the evening per respective month. Demand is lower during the southern at around 6:00 p.m. This is a common pattern since, at that hemisphere summer (especially in December and January), time, people generally arrive home and use most electricity for whereas the maximum is reached during winter time (August). lighting and other appliances. FIGURE 6.5 AVERAGE DAY LOAD PROFILES FOR EACH MONTH (VALUES FROM 2013) Average day load profile for each month Months Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 210 190 170 150 MW 130 110 90 70 50 6 12 6 12 6 12 6 12 6 12 6 12 6 12 6 12 6 12 6 12 6 12 6 12 18 18 18 18 18 18 18 18 18 18 18 18 Hours 64 KINGDOM OF ESWATINI

6.5 SOLAR PV months (June-September) but is still adequate for both solar PV and solar water heating. No ground measurements have yet Eswatini is well endowed with solar energy resources. Accord- been carried out in Eswatini to validate satellite data. ing to the Swaziland Renewables Readiness Assessment report (IRENA, 2014), Eswatini has relatively abundant solar potential For solar PV, the zoning study identified five zones for utility- throughout the country with an estimated global horizontal ir- scale solar PV farms. These zones are shown (in red) in Fig- radiance of 4–6 kilowatt-hours (kWh) per square metre per day. ure 6.6. The key characteristics of each zone are summarised The highest irradiation occurs during the summer months (De- in Table 6.8. The estimated capacity potential of these zones cember-March); the lowest irradiation occurs during the winter would total about 587 MW. FIGURE 6.6 SOLAR PV ZONES IN ESWATINI ENERGY MASTERPLAN 2034 65 TABLE 6.8 CHARACTERISTICS OF IDENTIFIED SOLAR PV ZONES Installed Distance to Distance to Mean Zone Annual electricity Distance to Mean resource capacity substation load centre capacity name generation (MWh) road (km) quality (W/m²) (MW) (km) (km) factor A 45 78,600 75 2 89 216 0.199 B 39.8 70,400 2 2 33 219 0.202 C 288 503,000 75 4 71 216 0.199 D 127 222,000 51 6 46 217 0.2 E 87 153,000 3 0 48 218 0.201 The average LCOE for the zones ranges from USD 158 to USD 165 month for zone B9. The capacity factors are higher during sum- per MWh, including the transmission line and road infrastructure mer months (particularly November-January) than during winter investment costs associated with each zone; transmission line months (June and July). The highest daily capacity factors are and road infrastructure investments would add up to 4 % and reached in the morning between 10:00 a.m. and 11:00 a.m. 2.5 % of plant investment costs, respectively, depending on the distance to the nearest substations/to the road infrastructure. Contrary to the solar PV capacity factor, demand is lower dur- ing the southern hemisphere summer and highest during winter Solar PV outputs change over the course of a day and across sea- time (Figure 6.5). A daily discrepancy of production peak and sons, reflecting the changing levels of solar irradiation. Figure 6.7 demand peak, as well as an annual discrepancy in production shows the capacity factor (i. e., the ratio of actual output to full- and demand variation, are noted. load output) for each hour of an average day for each respective FIGURE 6.7 AVERAGE DAY SOLAR PV CAPACITY FACTORS FOR EACH MONTH, ZONE B (VALUES FROM 2004) Average day solar PV capacity factors for each month, zone B Months Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 1 0.8 Fraction of theoretically 0.6 possible production 0.4 0.2 0 6 12 6 12 6 12 6 12 6 12 6 12 6 12 6 12 6 12 6 12 6 12 6 12 18 18 18 18 18 18 18 18 18 18 18 18 Hours ⁹ Similar to the analysis for wind power zones, the national working team determined a weighting of factors to come up with a cumulative score for screening solar PV zones by means of the multi-criteria analysis tool: total LCOE (weight factor of 75 %), co-location potential with wind power (10 %), distance to load centre (5 %), population density (5 %) and human footprint (5 %). Zone B features the best cumulative score of the solar PV zones and is therefore considered as priority zone, and further analysed. 66 KINGDOM OF ESWATINI

6.6 HYDROPOWER In 2015 MNRE conducted a feasibility study for the Lubovane Dam that identified three potential sites; however, only one Eswatini has good experience with small hydropower gen- site, with a total capacity of 850 kW, was found to be finan- eration. Since 1970 numerous studies have been conducted to cially viable. assess hydropower potential, which has led to an estimated theoretical potential of 440 MW and an estimated technical po- The utility is developing the 13.5 MW Lower Maguduza Hydro tential of 110 MW10. Currently Eswatini has an installed capac- project, which is an extension of the existing Maguduza Hy- ity of 61.5 MW, which is used for peaking. As part of its objec- dro Scheme. Preparations to commence construction are at tive to expand the hydropower sector, the Ministry of Natural an advanced stage. The country also conducted a prefeasibil- Resources and Energy (MNRE) built a database of potential ity study to ascertain the hydro potential that may exist along sites that initially identified 35 micro and mini hydro schemes, the Ngwempisi River. The study indicated potential of about ranging from 32 kW to 1.5 MW11. This was further reduced to 120 MW to 140 MW, but further analysis in the form of a feasibil- 26 schemes, based on their potential for electricity genera- ity study would need to be undertaken. MNRE received finan- tion. Four have been identified as viable schemes and are be- cial assistance from the African Development Bank to conduct ing promoted by MNRE: the Mbuluzi River (120 kW minimum), a prefeasibility study for the 120 MW Ngwempisi Hydro Scheme. Lusushwana River (300 kW), Mpuluzi River (155 kW) and Great Upon commission of the 20 Maguga Hydro Project, it was indi- Usuthu River (490 kW). cated that a potential exists downstream of the existing power station and recommended that a feasibility study be conducted. ¹⁰ O ne example is the Swaziland Hydro Power Reconnaissance Study that was undertaken by the engineering consulting firms GIBB and Knight Piesold in 2007 and focused on the Ngwempisi Cascade. ¹¹ By classification, micro hydro ranges from 0 kW to 100 kW, mini hydro from 101 kW to 2 MW and small hydro from 2.1 MW to 10 MW. ENERGY MASTERPLAN 2034 67 7 ENERGY SUPPLY INFRASTRUCTURE

7.1 FOSSIL FUEL PRODUCTS In the petroleum sector, the Government is in the process of finalising the development of the Petroleum Bill to govern the Eswatini imports all of its petroleum products, which are trans- sector. Eswatini has entered into a Memorandum of Under- ported by road and rail from South Africa and Mozambique. standing with Mozambique that covers the possible future sup- The supply and distribution of these products is in the hands of ply of natural gas that could be used for electricity generation, private oil companies. Due to the strategic nature of petroleum cooking, space heating, lighting and transport. products, the Ministry of Natural Resources and Energy regu- lates the prices and quality of the petroleum products. The oil 7.2 ELECTRICITY SUPPLY SYSTEM companies own commercial depots for storage of the products. The petroleum products used in Eswatini include diesel and Eswatini’s power market structure, the conceptual market struc- petrol for vehicle transport, paraffin and LPG for domestic ap- ture/model proposed for the future, and the associated trading plications, and heavy fuel oil for industry. Paraffin is also used arrangements for the different market players are set out un- in some industrial boilers. The Government intends to establish der the proposed stages as shown in Figure 7.1. This structure 90-day strategic stocks for petrol and diesel in order to im- largely corresponds to that outlined in the recently published prove the security of supply for these petroleum products. Swaziland Independent Power Producer Policy. FIGURE 7.1 ESWATINI ELECTRICITY INDUSTRY MARKET STRUCTURE Same legal entity OR Load (remote) non grid-connected Imports Co-generation Load (co-located) SEC Transmission Distribution & Generation (incl. SO & Customer (captive) Supply central buyer) Other Distribution / Supply Customer Exports IPPs (contestable) Legend (power flows): Current allowed transactions Stage II allowed transactions Stage I allowed transactions Stage III allowed transactions Source: Swaziland Department of Energy, 2015 68 KINGDOM OF ESWATINI In the current stage, the SEC acts as a single buyer procuring 7.2.1 GENERATION all electricity imported from South Africa, Mozambique and the SAPP, as well as electricity generated by IPPs and excess power The country’s installed capacity by technology for the indi- from CHP (co-generation). Sugar millers are the only entities vidual companies licensed to undertake generation is shown in that are currently undertaking co-generation and supplying Table 7.1. The SEC generates about 60.5 MW of peaking power loads within their estates (self-consumption), in particular irri- through hydropower generation, from four hydropower sta- gation and households for the RSSC. Ubombo Sugar Limited tions: Ezulwini (20 MW), Edwaleni (15 MW), Maguduza (5.6 MW) further sells excess power to the national utility (SEC). The sec- and Maguga (19.8MW). The bulk of the power, including base- tor is expected to consent to bilateral agreements in the near load supply, is imported from South Africa. The SEC also pur- future where generators will be permitted to sell directly to chases power from IPPs. Currently only two IPPs sell power to certain large customers (contestable). Once the country has the SEC: USL (up to 40.5 MW) and Wundersight (0.1 MW). All of achieved security of electricity supply, IPPs will be allowed to the other IPPs generate for self-consumption. Table 7.2 shows export power to the SAPP. SEC’s existing generation plants, including rated capacity, com- mission year and current status. TABLE 7.1 LIST OF POWER PLANTS IN ESWATINI TABLE 7.2 SEC POWER GENERATION STATIONS Company Technology Capacity (MW) Rated Year of Name capacity Status commission (MW) SEC Hydropower plants 60.5 Edwaleni USL Hydropower plant 1 Unit 1 1964 Unit 2 1964 15 Operational Unit 3 1964 USL Biomass 40.5 Unit 4 1965 Unit 5 1969 SEC Diesel 9 Ezulwini USA Distillers Coal 2.2 Unit 1 20 1985 Operational Unit 2 1985 Wundersight Solar PV 0.1 Maguduza 5.6 1969 Operational RSSC Biomass plant 65.5 Maguga Unit 1 19.8 2007 Operational Unit 2 2010 Total 178.8 Mbabane 0.5 1954 Mothballed Hydro Edwaleni 9 Diesel Efforts are being made to increase local generation through The development of a 300 MW coal-fired power station that private sector participation in the form of IPPs. The utility also would reduce the country’s reliance on imports is still at the has signed Memorandums of Understanding and power pur- inception stage. Coal prospecting to ascertain the coal type and chase agreements (PPAs) with a number of potential IPPs for amount of reserves has been undertaken. A full feasibility study short-term power generation plans. Only one IPP-owned solar for the project is yet to be undertaken. PV plant, a 100 kW plant, is operational at this stage. A 12 MW hydropower plant to be developed by an IPP is expected to be commissioned in June 2019. Short-term IPP plans that include plants at the inception stage include generation using solar PV of about 60 MW, biomass (bamboo and wood waste) of 35 MW and coal using coal finds from mine dumps of 30 MW. ENERGY MASTERPLAN 2034 69

7.2.2 IMPORTS remain even in the longer term due to the small size of the na- tional economy. About 70 % of power is imported from Eskom As the industry matures and progresses according to the stages (South Africa) and other SAPP countries, while the balance is depicted in Figure 7.1 the utility (SEC) is expected to continue met through local generation. Figure 7.2 indicates local produc- to play a central role in the importation of power, energy wheel- tion and the import trends from 2010 to 2014. The country’s ing where bilateral agreements exist and the export of power reliance on imports has increased steadily over the years due to to the SAPP. The SEC owns and operates the entire transmis- rising national demand. sion network in the country, and the status quo is expected to FIGURE 7.2 ESWATINI GENERATION AND IMPORT PROFILE 1,200 70 % 1,000 60 % 50 % 800 40 % SEC 600 GWh 30 % RSSC 400 USL 20 % Imports 200 10 % Total Local Generation 0% Imports 0 percentage 2012/13 2013/14 2014/15 2015/16 Period 70 KINGDOM OF ESWATINI The SEC imports the bulk of its supply from Eskom according to use periods are generally the same during the two seasons ex- the customer categories shown in Table 7.3. The utility pays for cept for the peak demand, which occurs an hour earlier during both energy and demand charges according to the terms and the high season. Figure 7.3 shows the time-of-use periods for conditions outlined in the parties’ PPA. The high-demand sea- the high and low seasons, respectively. The same seasons and son occurs during the winter period from June to August, while time-of-use periods apply to customers in Eswatini. the balance of the year falls under the low season. The time-of- TABLE 7.3 ESKOM MEGAFLEX CUSTOMER CATEGORIES Megaflex tariff (no local authority) Active energy charge [c/kWh] Transmission network charge Trans- High Demand Season [June–Aug] Low Demand Season [Sep–May] in South African mission Voltage Rand VAT Stand- VAT Off VAT VAT Stand- VAT Off VAT VAT Zone Peak incl ard incl Peak incl Peak incl ard incl Peak incl [R/kVA/m] incl < 500 V 272.34 310.47 82.86 94.46 45.24 51.57 89.18 101.67 61.54 70.16 39.23 44.72 R7.79 R8.88 ≥ 500 V & <66 kV 268.06 305.59 81.21 92.58 44.10 50.27 87.44 99.68 60.19 68.62 38.18 43.53 R7.12 R8.12 ≤ 300 km ≥ 66 kV & <132 kV 259.58 295.92 78.63 89.65 42.71 48.69 84.69 96.55 58.28 66.44 36.98 42.16 R6.94 R7.91 ≥132 kV 244.85 278.90 74.11 84.49 40.25 45.89 79.82 90.99 54.92 62.61 34.85 39.73 R8.76 R9.99 FIGURE 7.3 TIME-OF-USE TARIFFS FOR HIGH SEASON (LEFT) AND LOW SEASON (RIGHT) 24 24 23 1 23 1 22 2 22 2 Weekdays Weekdays 21 3 21 3 Saturday Saturday 20 4 20 4 19 5 19 5 18 Sunday 6 18 Sunday 6 17 7 17 7 16 8 16 8 Peak 15 9 Standard 15 9 14 10 14 10 11 Off-peak 11 13 12 13 12 ENERGY MASTERPLAN 2034 71

7.2.3 TRANSMISSION AND DISTRIBUTION TABLE 7.4 TRANSMISSION LINES A 400 kilovolt (kV) line is the major interconnector between Number of Capacity Distance (km) South Africa, Eswatini and Mozambique and is used to facilitate transmission lines power imports and exports among these countries. A 132 kV 400 kV 2 – line also extends from Normandie in South Africa to the south 132 kV 9 296 of Eswatini and is used by Eswatini to import power from South 66 kV 64 970 Africa and to control the import maximum demand on the 400 kV line for the country. The power is then transmitted through 132 kV/66 kV substations where it is transformed from 132 kV to 66 kV. From these substations the power is further stepped The maps in Figure 7.4 and Figure 7.5 depict the transmission down to the 11 kV distribution network. The SEC is currently not system and electricity coverage, respectively, in Eswatini within exporting power. a 1-kilometre radius. Transmission lines are maintained regularly. Most of the old wooden pole structures are gradually upgraded to steel struc- tures. A total of fourteen 66 kV lines are on steel structures, and the other fifty 66 kV lines are still on wooden poles. Some of these lines are partially on steel structures and wooden poles. 72 KINGDOM OF ESWATINI FIGURE 7.4 TRANSMISSION SYSTEM MAP ENERGY MASTERPLAN 2034 73 FIGURE 7.5 ELECTRICITY COVERAGE MAP 74 KINGDOM OF ESWATINI No major policies guide grid extension and rural electrification The Government, in collaboration with the Royal Swaziland in Eswatini, except for national needs or the government tar- Sugar Association, conducted a pilot project on E10 blended gets stated in Chapter 4. The utility acts on instruction by the biofuel. The pilot project ran for two years, from 2008 to 2010, Government and based on available funding to meet the set and was considered a huge success. The RSSC anhydrous etha- targets. However, significant ground has been covered in ex- nol plant has a production capacity of 10,000 litres per day, tending the network countrywide, as shown in Figure 7.5. “Net- and expansion of this capacity is being considered to meet the work coverage” represents transmission infrastructure that is country’s demand for petrol, which averages around 1.2 million within a 1 kilometre radius of the point of connection, while “ac- litres per month. The RSSC has estimated that it would take tual access” depicts the actual connections. Network coverage about 12 months to expand the plant’s existing capacity four- in rural areas is estimated at 67 %, compared with actual access fold. Preparations to roll out the E10 blended biofuel are at an of 60 %. The low access rate compared to the availability of in- advanced stage while finalising the logistics for the E10 fuel frastructure signifies the need to accelerate rural electrification, blending programme. The Government intends to gradually in- including through the use of distributed systems in areas where troduce blending up to a 10 % ratio; according to the target set grid extension is not expected in the foreseeable future due to under the SE4ALL action plan, this was to be achieved by 2016. high capital costs. Liquid biofuels do not need any special equipment for transport The RSSC also undertakes distribution in its estate, which except that ethanol has a very high affinity for water, so tanks accommodates mainly employees and third-party customers must be efficiently dried. Silica gel or any drying chemicals can providing services to the estate. be used in tankers transporting ethanol or ethanol-blended fuel by rail or road. However, pipelines are not suitable for ethanol

7.3 BIOFUEL SUPPLY SYSTEM transport because the fuel is highly corrosive, a property that is not favourable for the joints normally found in pipelines. Pipe- The Government plans to introduce ethanol-blended fuel lines also are not suitable because they typically transport more through a blending mandate for all unleaded petrol in the than one product, which may contaminate the ethanol with wa- country. This is incorporated in the draft Petroleum Bill, and ter particles. the relevant standards that have been developed will be turned into technical standards as part of the mandatory blending regulation. ENERGY MASTERPLAN 2034 75 8 ENERGY SUPPLY PLANNING SCENARIOS FOR 2014–2034

8.1 BRIEF DESCRIPTION OF THE 8.2 SCENARIO DEFINITION: KEY POLICY ISSUES METHODOLOGY TO BE ADDRESSED The System Planning Test – Swaziland (SPLAT-SW) model is a This section provides information on the parameters and related planning tool developed by the Eswatini team, expanding on assumptions that informed inputs into the SPLAT-SW model the SPLAT – Southern Africa (SPLAT-S) model originally devel- for the Base Case and alternative scenarios. The parameters oped by IRENA. The SPLAT-SW model was developed using the are predetermined by known technical characteristics such as Model for Energy Supply Strategy Alternatives and their Gen- costs and efficiencies. The Base Case is the baseline for all of eral Environmental Impact (MESSAGE) platform. MESSAGE is the alternative scenarios, meaning that the scenarios are created a dynamic bottom-up, multi-year energy system software that based on the Base Case, with alteration in specific parameters to applies linear and mixed-integer optimisation techniques, used define the relative policies that deviate from the Base Case. The for medium- to long-term energy system planning. The soft- scenarios are designed for the purpose of addressing different ware analyses the least-cost technically and economically feasi- aspects of the existing energy policy objectives. They are: ble combinations of energy supply options to meet a specified demand under a set of conditions. On that basis it selects the • self-sufficiency in energy supply least-cost optimal energy mix, taking into consideration the in- • universal access to clean and affordable energy vestment needs and other costs for new infrastructure, energy • optimal use of domestic resources supply security, energy resource utilisation, rate of introduction • diversification of the energy supply mix to meet energy of new technologies, and environmental and other constraints. security concerns • climate change goals, as captured in Eswatini’s Nationally The SPLAT-SW model contains a set of demand projections and Determined Contribution (NDC). a catalogue of energy supply technology options characterised by economic and technical parameters as well as data on the Based on the base year (2014), one Base Case scenario and prevailing investment stock and remaining life span. Within this nine alternative scenarios were developed. set of assumptions, the model performs an assessment of dif- ferent scenarios for a given energy demand. The model results The Base Case provides the least-cost supply technology mix to include projections on, among others, investments, production, meet the projected energy demand over the planning horizon, fuel use and trade, and CO₂ emissions. given some existing policies and macroeconomic trends. The model incorporates the latest information on energy tech- The alternative scenarios considered different pathways that nologies and cost, with some refinements to adequately char- could be pursued to achieve the above-mentioned energy poli- acterise options available to Eswatini. The modelling approach cy objectives. The alternative scenarios have been informed by has been expanded to cater to both the power and non-power the policy imperatives of high-impact policies within the ener- sectors to take into account the demand for fuels other than gy sector that include, among others, a large coal power plant electricity. project, the SE4ALL targets and renewable energy targets in the National Development Plan. Key drivers of the alternative As described in Chapter 4, the demand projections were made scenarios are to enhance energy security, to decarbonise the using a LEAP model, which, among other things, takes into ac- energy sector and to achieve self-sufficiency in energy supply count universal electricity access goals as well as the goal of using domestic resources. providing the country with access to modern energy for cooking and eliminating sole reliance on the use of fuelwood. The SPLAT- SW model takes as inputs the demand projections from LEAP. 76 KINGDOM OF ESWATINI

8.2.1 THE BASE CASE Given that the domestic electricity-grade coal resource is vir- tually untapped, it is possible that the country would be in a The Base Case is based on existing policy objectives and de- position to negotiate long-term supply contracts for this coal, picts the business-as-usual future. It therefore serves as a refer- hence lowering fuel prices. Affordability is a key consideration ence point/baseline for the alternative scenarios. Proposed and in the coal scenarios. It was assumed that the coal-fired genera- existing projects were taken into consideration; however, pro- tion would be explored only with domestic coal and not with jects that have reached maturity in terms of bankability were imported coal. Additional coal options would be undermined by included as committed in the model. These projects include the a carbon tax regime, which would render local industries less

13.5 MW Lower Maguduza Hydro Scheme and the 10 MW solar competitive and put economic value, jobs and country growth PV project at Lavumisa. at risk. However, a carbon tax was not taken into account in these scenarios. The Base Case includes all other technology options such as coal power generation, natural gas, biomass and wind technol- The expected commissioning date of the power plant in the ogies, to name a few. The deployment of these technologies is model, of 2025, has been set to coincide with the expiration of not forced in the model solution. the existing bilateral electricity trade agreements with Eskom, while also allowing for sufficient lead time for project prepara- The biomass-based generation options were modelled as tion and construction. Two Forced Coal scenarios are considered: follows: Forced Coal 1 (300 MW) 1) Technologies that use bagasse as a main fuel and coal and wood chips as supplementary fuels; they are used for self- This scenario assumes that a 300 MW plant is built and com- consumption for the sugar industry and also for selling it to missioned in 2025. The objective of this scenario is to meet the the main grid. country’s peak demand capacity through the exploitation of lo- cal abundant coal, while meeting the country’s emissions tar- 2) Technologies that use bagasse as a main fuel and wood chips get. However, using local coal presents a great risk considering as supplementary fuels; they are used for self-consumption the high resource cost. for the sugar industry and also for selling it to the main grid. Forced Coal 2 (600 MW) 3) Technologies that use wood chips as the only fuel for self- consumption in the industry, and for selling the reminder to This scenario assumes that a 600 MW plant is built and com- the grid. missioned in 2025. This scenario has similar assumptions to the Forced Coal 1 scenario (300 MW). However, this scenario 4) Technologies that use bagasse as a main fuel and wood chips looks into the possibility of meeting the growing demand and as supplementary fuel; they are used for self-consumption for long-term cost reduction in developing a bigger coal-based the sugar industry only. power plant. Also, this scenario considers the potential benefits of using the domestic coal resource to generate electricity for

8.2.2 FORCED COAL SCENARIOS exports to the regional market (SAPP) in order to generate in- come from exports. The Forced Coal scenarios use all technology assumptions from the Base Case, with additional alterations only on coal technol- 8.2.3 DOMESTIC RESOURCES SCENARIOS ogy. The purpose of these scenarios is to provide insights into the impact of coal technology in providing baseload power. The domestic resources from hydro, wind, solar, biomass and These scenarios explore the possibility of reducing Eswatini’s coal are distributed across Eswatini and could provide afford- reliance on imports by exploiting the country’s coal endow- able and secure supplies of energy and meet the country’s ments through constructing a new coal power plant. Coal is electricity demand. The Domestic Resources scenarios assess one of the resources that the country has in abundance but the implications of becoming less import-dependent and max- has never exploited. A feasibility study is ongoing to assess imising the domestically endowed energy resources. The sce- the technical and economical parameters of developing a coal narios envision an increase in renewable energy deployment as mining operation for power generation. The plan for developing a result of the rapid reductions in renewable energy costs and 300 MW coal-fired plants by 2018 was included in the SE4ALL the country’s renewable energy promotion policy as well as re- Action Agenda. duced reliance on imports. The appetite for renewable energy ENERGY MASTERPLAN 2034 77 in the market is growing rapidly. By developing renewables, the These scenarios serve the objective of increasing the share of country envisages being able to meet its international climate renewable energy, as enshrined in the SE4ALL agenda, and and sustainability obligations and to access finance. It is against ensuring that electricity consumed in the country is generated this background that renewable energy was considered when locally. Eswatini currently meets about 25 % of its electricity establishing the Energy Masterplan to ensure effective energy requirements from renewable energy resources. Several re- infrastructure planning and to achieve the targets of sustain- newable energy resources were considered in the scenarios, able development goals. based on the technology assumptions used in the Base Case, which assumes a ceiling on electricity imports of 35 % by 2020 According to the IRENA-LBNL Renewable Energy Zoning and 25 % from 2025, with up to 35 MW from biomass power study12 for Southern and Eastern African countries, done in the plants as of 2022 and up to 120 MW from hydropower plants context of IRENA’s Africa Clean Energy Corridor programme, as of 2025. the technical potential for wind and solar PV generation in Es- watini is approximately 880 MW, and the economically viable 8.2.4 NO IMPORT ENHANCEMENT SCENARIOS potential is 586 MW. Biomass (wood chips and wood waste) and sugarcane bagasse are the two main sources of biomass Two “No Import Enhancement” scenarios assume that no in- considered for power generation in the country. Domestic vestment into expanding the import capacity is allowed. The biomass has a technical potential of 100 MW. Hydropower is “High Demand with no import enhancement” scenario assumes growing rapidly, with an installed capacity of 62 MW in 2014 higher local electricity demand that exceeds the maximum and additional potential of 133.5 MW deemed economically electricity available to Eswatini from domestic sources and from feasible. In light of the existing potentials, these scenarios the current supply agreement with South Africa combined. The seek the implications of becoming less import-dependent and current bilateral agreement with South Africa allows maximum maximising the use of domestically endowed energy resourc- energy imports of 185 MW a year. The objective of the scenario es, primarily renewables. is to evaluate a situation in which the growth in local demand is so high that it can lead to insufficient supply from current Domestic Resources scenarios electricity import options. As a reference, another scenario was developed with the same restriction on import enhancement, Three scenarios were developed to determine the role of us- but with the demand as in the Base Case. ing domestic resources in meeting future energy requirements while increasing local capacity. Three policy choice options are • High Demand with no import enhancement: with higher de- identified. mand, but no investment into expanding the import capacity is allowed. • Limited Import 1 seeks the implications of becoming less • Base Case with no import enhancement: with Base Case de- import-dependent and maximising domestically endowed mand, but no investment into expanding the import capacity energy resources (including all resources). is allowed. • Limited Import 2 (with forced hydro) seeks the implications of becoming less import-dependent and maximising domes- 8.2.5 LOW IMPORT PRICE SCENARIO tically endowed energy resources, with an additional 120 MW of hydropower generation forced into the scenario. There is a long-standing ambition to increase electricity trade • Limited Import 3 (with renewables, without coal) seeks the across Africa through co-operative planning and improved implications of becoming less import-dependent and max- transboundary transmission. However, trade within Southern imising domestically endowed energy resources, with only Africa is currently constrained by limited supply. Eswatini is a renewable energy sources and without coal. net importer of electricity, which poses a risk for energy secu- rity. Electricity prices vary greatly, with much of the variabil- ity reflecting the relative ease of energy supply and the extent to which electricity prices are subject to government control. South Africa was recently affected by the shortage in supply, which led to high import prices for the country. ¹² See: http://www.irena.org/publications/2015/Oct/Renewable-Energy-Zones-for-the-Africa-Clean-Energy-Corridor. 78 KINGDOM OF ESWATINI In the Low Import Price scenario, there is no limit on imports of The modelling includes determination of the types of existing electricity. The import of electricity is charged according to the energy technologies and their properties. Possible future op- Eskom tariff at the time of use. This scenario assumes a lower tions based on the country’s needs are assessed. This includes import tariff, with no general cost escalation and an annual im- determining properties such as the efficiencies, capacity fac- port tariff increase of 3 %. tors and operational life of all of the technologies. The domestic availability of natural resources such as coal, natural gas and

8.2.6 NATURAL GAS SCENARIO biomass are considered as well as the costs of importing these resources. The model is then populated with the datasets, and This scenario assumes that a natural gas plant is built locally. It the model runs for the Base Case and alternative scenarios ex- must consider the cost of importing the resource from neigh- ecuted and results analysed. bouring Mozambique. The scenario allows a comparison of nat- ural gas imports with the coal options. Replacing coal-based The primary energy supply includes extraction of coal and all generation with natural gas provides a way to reduce carbon forms of biomass (wood chips, bagasse, wood waste) and im- emissions. This scenario also considers natural gas to be used ports of natural gas and petroleum products. Renewable en- for demand-balancing power to cushion the intermittency of ergy technical potentials of wind power and solar PV are also renewable energy plants. included. The technology options include a variety of electricity generation technologies for coal, wind, solar, hydro, biomass

8.3 MODELLING OF THE ESWATINI ENERGY and natural gas resources. The cost of the whole energy system SYSTEM is determined by summing the costs related to the investments and activity of all of these technologies. The fuel prices are de- The Eswatini energy system is modelled for analysing energy termined from the cost of using the technologies. technology choices. In view of the close correlation between energy sector policy and technology choices, the model consid- 8.3.1 MODEL CONFIGURATION ers how the energy system can be used to inform policy. The parameters defining the model are arranged according to The key assumptions that inform the modelling of the energy the fuel and technology configurations and are grouped into system include projected future energy demand, assumptions technology, demands, costs, storage, activity and capacity. A about the current and future energy technologies, energy number of characteristics define each parameter attributable to prices and policy constraints. While demand is the key input the energy system. The technology, fuel used (input and output into the model, other assumptions such as macroeconomic and of the technology), year and time slice are all dimensions con- demographic assumptions inform the projected demand. The sidered within the energy system. Each parameter has a num- demand projections for other various sectors are also signifi- ber of constraints defining it. The Base Case has information on cant. The main purpose of this modelling is to find the least- parameters and assumptions that inform inputs into the model. cost development path for the Eswatini energy system to the The assumptions can be technical attributes such as costs and year 2034. Such a path is required to meet the predetermined efficiencies, while others are determined on estimates such as demands for energy services within the given policy, environ- potential for wind power. mental and economic constraints. ENERGY MASTERPLAN 2034 79 In order to maintain an energy system approach, dummy tech- in USD per kilowatt (kW) per year. This also applies to fixed nologies are defined for energy carriers in demand technologies costs. Capital costs for new plants are also converted to USD for final demand sectors such as agriculture, commercial, indus- per kW. Inputs for existing plants are based on performance trial, etc. The function of these dummy technologies is to link characteristics. New plants for renewable energy have decreas- the production of energy to final consumption within the model. ing capital costs going into the future. There are no costs associated with these dummy technologies. A constant electricity reserve margin of 10 %, including imports, The technologies for electricity generation are means to convert is used in the model. Six time slices were used to calibrate the energy carriers into usable electricity. They are defined by costs, electricity demand profile. The sum of solar PV and wind power primary sources, efficiency factors and emissions from trans- does not exceed 70 % of the system peak in order to limit the formation. These are split between existing and new generic instantaneous penetration of variable renewable energy. technologies. Each technology is defined by certain technical parameters that determine the investment and operation costs 8.3.2 REFERENCE ENERGY SYSTEMS and performance. Other parameters are less constraining, such as the maximum/minimum limits of the respective technology. The primary purpose of the energy system is to ensure that all demands for energy services are met. A reference energy For the model’s installed capacity, SEC power plants are domi- system reflective of the country’s energy flows and projected nated by hydropower plants, with one diesel power plant and developments is defined in Figure 8.1 and Figure 8.2. The ref- bagasse/coal power plants operated by the sugar industry. The erence energy system presents the country’s different energy “committed”13 power plants are all under development, with forms, whether imported or available locally. The vertical lines signed PPAs. The existing and committed plants are modelled represent energy commodities and the rectangles represent as residual capacity, and variable costs collected are converted technologies, while the horizontal lines represent commodity flows among these. ¹³ “Committed” means that projects are under construction or are at an advanced stage of preparation in their development. This is in contrast to existing projects or “candidate” projects. The latter includes two categories: site-specific candidates and generic candidate projects, that may be built in future years. 80 KINGDOM OF ESWATINI 81 Primary Secondary Tertiary Sugar production other industry Agr com Trans Urban residential Rural ENERGY MASTERPLAN 2034 Wood Chips Oil products Molasses Fuel Wood Electricity Electricity Electricity export Transport oil IndWoodChips Bagasse Coal Gas Ethanol SugElc Sugoil IndElc Indoil Indcoal Indbag AgrElc AgrOil ComElec ComOil Traoil UrbElc Urboil Urbfuw RurElc Ruroil Rurfuw EPBWCNU Wood PP EPBAGNUreg DELCInd Sugar mill CHP MIBAG Distribution to Prod industry Bagasse EPBAGNUhi Sugar mill CHP efficient MICOA DELCCOM Prod Coal EPCOANUSClub / EPCOANUSC Distribution to IMCOA commercial Coal PP Import Coal DELCUrban TELC Distribution to urban IMGAS EPGASNUOC Transmission Import Gas lines DELCRural Gas-Open Cycle PP IMODS Distribution to rural Import Oil EPGASNUCC / EPGASNUSO EPSOLNUPVCOM Gas-Combined Cycle Rooftop PV Com. REFERENCE ENERGY SYSTEM (POWER SECTOR PART) PP EPSOLNUPVrt Rooftop PV urban EPODSEX / EPODSNU Diesel PP EPHYDNUmic Mini-hydro EPODSNUind Diesel PP industry EPODSNUurb Diesel PP Urban EPODSNUrur Diesel PP Rural EPSOLNUPVrtru Rooftop PV rural IMIelc EPSOLNUPVstru / EPSOLNUPVstru2 Electricity trade Rooftop PV with 2 hr storage rural EPHYDEX / EPHYDNU Hydro PP EPWNDNUA / EPWNDNUB / … Wind PP EPSOLNUPVutA / EPSOLNUPVutB / ... Solar PV Utility PP EPTSONUTH CSP EPTSONUTHst FIGURE 8.1 CSP + storage FIGURE 8.2 REFERENCE ENERGY SYSTEM Figure 8.1 and Figure 8.2 illustrate significant components of (NON-POWER SECTOR PART) the energy system. The process starts with primary forms of energy, which are converted into secondary forms of energy. Certain secondary energy carriers such as electricity and liq- uid fuel are imported directly into the energy system. The final MIBWC MIBFW MIBMO processes in the energy value chain are performed by demand Prod Fuelw Prod Molas. Prod technologies that transform energy carriers into services. Energy commodities represented in the SPLAT-SW model include: • Primary energy: bagasse, coal, gas, wood chips, oil products, molasses, fuelwood • Secondary energy: electricity, ethanol • Tertiary energy: electricity, electricity export, oil products industrial Ethanol BOBMONU Energy demand is divided into six categories, with the following production fuel demands: • Sugar production: electricity, oil • Other industry: electricity, oil, coal, bagasse, wood chips • Agriculture: electricity, oil • Transport: oil BOOGSNU Ethanol blending • Urban residential: electricity, oil, fuelwood • Rural: electricity, oil, fuelwood The cost of the whole energy system is measured by adding up all of the costs associated with the investments and activity of all technologies. DCOA DBAG DBWC DBFW DODS DOGS Oil product distribution Distribution of coal Distribution of bagass Blended gasoline Distribution of Distribution of distribution to woodchips fuelwood to other sectors transport 82 KINGDOM OF ESWATINI 8.3.3 DEFINITIONS OF SEASONS AND DAILY resent the output pattern of generations that are seasonal (all LOAD REPRESENTATION renewables) or that have the diurnal production patterns (solar and wind). Based on the analysis of the seasonal demand pat- In the SPLAT-SW model, the daily load curves are defined for tern, rainfall, solar and wind availabilities, and crop production, each season and for each day type (weekday and weekends). a total of five seasons was identified, each of 2–3 months, as The load curves are defined with simplified blocks of hours. The presented in Table 8.1. Note that for the import tariff, season 3 same block definition and seasonal definitions are used to rep- corresponds to the “high-demand season”. TABLE 8.1 SEASONS DEFINED IN THE SPLAT-SW MODEL Season Time of year (in months) Description 1 1–3 High: hydro, solar, demand; Low: wind, crop 2 4–5 High: crop; Low: wind, hydro, solar, demand 3 6–8 High: wind, crop; Low: hydro, solar, demand 4 9–10 High: wind, solar; Low: hydro, crop, demand 5 11–12 High: hydro, wind, solar; Low: demand, crop For the definition of the blocks within a day to represent the • Tariff: weekdays 22:30–6:00 off-peak, 6:00–9:00 and load profiles, the following observations were made: 17:00 –19:00 peak (low-demand period) and 7:00–10:00 and 18:00–20:00 peak (high-demand period). • Solar daily output pattern: 16:00–5:00 (nearly zero); 6:00 –8:00 (between zero and peak); 9:00–12:00 (peak pro- Given these observations, the daily blocks were defined in duction); 13:00–15:00 (between zero and peak) the following manner. Note that the 24 hours of the day are • Wind daily output pattern: no clear pattern exists divided into five to nine blocks, respectively, separated by • Demand: 6:00–7:00 morning peak; 8:00–10:00 weekend the colour code. morning peak; 19:00–20:00 evening peak The detailed load demand approximated for each season and each day type is presented in Figure 8.3. Weekday (seasons 1–2, 4–5) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Weekday (season 3) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Weekend (seasons 1 and 5 – summer) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Weekend (season 3 – winter) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Weekend (seasons 2 and 4) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 ENERGY MASTERPLAN 2034 83 FIGURE 8.3 LOAD PROFILE REPRESENTATIONS IN EACH SEASON AND IN EACH DAY TYPE Daytype1, season 1 Daytype2, season 1 250 200 150 MW 100 50 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Daytype1, season 2 Daytype2, season 2 250 200 MW 150 100 50 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Daytype1, season 3 Daytype2, season 3 250 200 MW 150 100 50 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Actual Average 84 KINGDOM OF ESWATINI Daytype1, season 4 Daytype2, season 4 250 200 MW 150 100 50 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Daytype1, season 5 Daytype2, season 5 250 200 MW 150 100 50 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Actual Average ENERGY MASTERPLAN 2034 85

8.3.4 KEY TECHNOLOGY PARAMETERS (imported), kerosene (imported), natural gas (imported), ba- gasse (domestic), wood chips (domestic), molasses (domestic) Fuel price assumptions and fuelwood (domestic). The cost assumptions for some of these energy sources are summarised in Table 8.2. The fuel Primary energy sources used in the models are coal (im- prices are assumed to stay constant in this analysis, except for ported, domestic), LPG (imported), diesel (imported), petrol imported gas. TABLE 8.2 FUEL PRICE ASSUMPTIONS Parameter Unit Assumption for 2014 Assumption for 2034 Domestic coal USD/GJ 2.7 2.7 Imported coal USD/GJ 2.4 2.4 Imported natural gas USD/GJ 18.9 25.6 Imported oil USD/GJ 22.5 22.5 Domestic wood chips USD/GJ 2.1 2.1 Domestic bagasse USD/GJ 1.1 1.1 Fuelwood USD/GJ 0.0 0.0 Molasses USD/GJ 0.0 0.0 Import and export potential of fuels Eskom’s time-of-use tariff (Figure 7.3). The electricity import price is assumed to increase by about 30 % to 80 % between There is no limitation on fuel imports, and fuel exports are 2014 and 2022 based on the Eskom projections (Figure 8.4). not considered. Import of electricity is charged according to FIGURE 8.4 ASSUMPTIONS ON ELECTRICITY IMPORT PRICE DEVELOPMENT 25 20 USD cents/kWh 15 10 5 0 2014 2015 2016 2017 2018 2019 2020 2021 2022 Low demand - Off peak Low demand - Standard Low demand - Peak High demand - Off peak High demand - Standard High demand - Peak 86 KINGDOM OF ESWATINI Ethanol production plans The characteristics of current and future industrial ethanol production plants are given in Table 8.3. TABLE 8.3 TECHNO-ECONOMIC PARAMETERS FOR INDUSTRIAL ETHANOL PLANTS Overnight Current Fixed operation and investment capacity Start year Main fuels Efficiency maintenance costs Life (yr) costs (USD/ (TJ/yr) (USD/TJ/yr) TJ/yr) 368.5 TJ/yr in 1998, Electricity/ 665 and 368.5 TJ/yr in bagasse/wood 88 % 35,535 355.45 20 2006 chips Power generation The existing power plants that are currently in operation and modelled in the SPLAT-SW are summarised in Table 8.4. In the SPLAT-SW model, existing technologies and future tech- Assumptions on the site-specific planned power projects are nologies that constitute the reference energy system are mod- given in Table 8.5. elled. Two types of future technologies are modelled, namely site-specific and generic technologies. Site-specific technolo- gies may be characterised with project specific techno-eco- nomic parameters, while generic technologies are represented by generic techno-economic parameters. TABLE 8.4 TECHNOLOGY ASSUMPTIONS FOR EXISTING POWER PLANTS Description Capacity (MW) Start year Capacity factor (%) Main fuel SEC existing diesel power plant 9 27 Diesel USA Distillers coal power plant* 2.2 2011 28 Coal 40.5** 2001 Hydro SEC existing hydropower plants 47 20 2007 Hydro USL existing hydropower plant* 1 2011 38 Hydro 15.5 2000 Bagasse USL existing biomass power plant 28 25 2012 17 1989 RSSC existing biomass power plant* Bagasse 18.5 2000 27 (decentralised, for self-consumption only) 30 2011 Wundersight existing solar power plant 0.1 2016 – Solar * represents own consumption ** All plants refurbished between 1998 and 2001. ENERGY MASTERPLAN 2034 87 TABLE 8.5 TECHNOLOGY ASSUMPTIONS FOR PLANNED POWER PROJECTS Overnight Description Capacity (MW) Start year Capacity factor Main fuel costs (USD/kW) SEC Lower Maguduza power plant 13.5 2021 52 Hydro 4,500 Wundersight solar power plant 0.85 2018 – Solar 3,000 Lavumisa 10 2019 – Solar 1,400 The generic technology parameters used in this study are sum- able energy technologies were assumed to decline in line with marised in Table 8.6. Overnight investment cost assumptions as global trends. well as operation and maintenance (O&M) costs for the renew- TABLE 8.6 GENERIC TECHNOLOGY ASSUMPTIONS Start year Capacity Overnight investment cost Description Efficiency Years (capacity online) factor (USD/kW) 2014 2034 Diesel/Gasoline – 90 % 35 % 25 650 650 Diesel/Gasoline 1 kW system – 90 % 35 % 20 758.7 758.7 (residential and commercial) Diesel/Gasoline 1 kW system (rural) – 90 % 16 % 10 728 728 Open-cycle gas turbine 2025 94 % 30 % 25 635 635 Combined-cycle gas turbine 2025 94 % 48 % 30 1,125 1,125 Coal 2025 90 % 33 % 35 2,812 2,812 Hydro 2026 95 % 100 % 50 4,500 4,500 Small hydro 2018 80 % 100 % 30 4,211 4,211 Biomass (wood chip) 2020 100 % 30 % 30 3,500 3,500 Bagasse (high efficiency) 2019 90 % 30 % 30 3,500 3,500 Bagasse (regular efficiency) 2022 90 % 28 % 30 3,200 3,200 Wind (zone J) 2018 33 % 100 % 25 2,458 1,200 Solar PV (zone C, utility) 2018 20 % 100 % 25 2,344 842 Rooftop PV (urban) 2017 18 % 100 % 20 2,414 1,023 Rooftop PV (rural) 2017 18 % 100 % 20 3,748 1,224 Rooftop PV (rural, two-hour battery) 2017 26 % 100 % 20 5,465 1,936 Concentrated solar power (CSP) – 18 % 100 % 25 3,000 1,556 without storage CSP with storage – 24 % 100 % 25 7,000 3,198 88 KINGDOM OF ESWATINI The utility-scale wind and solar plants are assumed to be built Levelised cost of electricity in the geographic zones identified in Chapter 6 (Table 6.5 and Table 6.6), with the maximum generation associated with each The levelised cost of electricity generation was calculated as be- zone. Five zones each were identified for solar PV and wind, low for the list of technologies analysed in the Masterplan. The respectively. Note that the technology parameters for different LCOE for natural gas technologies increases due to the escalated zones are defined differently reflecting the local methodologi- gas import price. For solar and wind technologies, the reduction cal conditions, and in Table 8.6, the numbers corresponding to in investment costs and O&M costs leads to a strong reduction in the representative zones (zone J for wind and zone C for solar the LCOE. The LCOE was calculated using the following formula: PV) are given. LCOE = {(Overnight capital costs × capital recovery factor + fixed O&M cost) /(8,760 × capacity factor)} + (fuel cost × heat rate) + variable O&M cost. TABLE 8.7 LEVELISED COST OF ELECTRICITY (USD/MWH), 2014 AND 2034 2014 2034 Diesel/Gasoline generator 240 240 Diesel/Gasoline 1 kW system (residential and commercial) 243 243 Diesel/Gasoline 1 kW system (rural) 521 521 Open-cycle gas turbine 441 592 Combined-cycle gas turbine 254 334 Coal 57 57 Hydro 55 55 Small hydro 64 64 Biomass (wood chip) 63 63 Bagasse (high efficiency) 65 65 Bagasse (regular efficiency) 60 60 Wind (zone J) 122 59 Solar PV (zone C) 167 59 Rooftop PV (urban and commercial) 196 83 Rooftop PV (rural) 353 115 Rooftop PV (rural, two-hour battery) 336 119 CSP 243 141 ENERGY MASTERPLAN 2034 89 Transmission and distribution of power Transmission costs are taken into account. Distribution is assumed to incur costs specific to the demand categories. The distributed generation does not require that the power be delivered through the network. TABLE 8.8 TECHNO-ECONOMIC PARAMETERS FOR TRANSMISSION AND DISTRIBUTION Transmission Distribution Industry/sugar/agriculture/ Urban Rural commercial Efficiency loss 5% 2% 10 % 15 % Investment cost 365 USD/kW 70 USD/kW 30 USD/kW 30 USD/kW

8.4 MODELLING OF ESWATINI ENERGY SYSTEM: Power generation mix KEY MODEL RESULTS The electricity supply mix for 2014–2034 under the Base Case

8.4.1 BASE CASE RESULTS: POWER SECTOR scenario is shown in Figure 8.5. Electricity demand is projected to increase from 1,270 GWh to 2,648 GWh over the planning The demand projections are a fundamental model input and a horizon. Currently electricity imports account for approximately major driver of scenario results. As introduced, the Base Case 70 % of the country’s electric energy requirement. Following outlines the least-cost path under business-as-usual policies. major investments in power generation from 2020, it is envis- The objective is to ensure that future energy supply options aged that the share of imports will fall to 65 % and further de- meet the prospective energy demand. The abbreviation “dist.” cline to 35 % by 2034. Two new sources of power generation, used in the figures refers to distributed generation capacity (for solar PV and wind, are projected to enter the generation mix, own consumption), while the rest is centrally located (feeds to together accounting for one-third of total generation by 2034. the national grid). There is also sizable biomass generation resulting in nearly four times the amount of centralised biomass compared to distrib- uted biomass in 2034. The increase in distributed biomass is driven mainly by an anticipated increase in the use of biomass for power generation for own-use (not grid-connected). 90 KINGDOM OF ESWATINI FIGURE 8.5 BASE CASE: ELECTRICITY GENERATION 100 % 90 % 80 % Electricity production (GWh) 70 % Dist. Solar PV 60 % Mini Hydro Dist. Biomass Dist. Coal 50 % Dist. Oil Net Imports 40 % Wind Solar Thermal 30 % Solar PV Biomass 20 % Hydro Nuclear 10 % Gas Oil 0% Coal 2014 2018 2022 2026 2030 2034 Generation capacity mix is expected to continue because importing electricity is more economical than constructing new power plants under the Base Figure 8.6 shows the results of a capacity build-up programme Case, which assumes an annual increase of 5 % (in nominal under the Base Case. With the additional investment in central- terms) in the import price. ised technologies of 585 MW and in decentralised technologies of 29 MW over the planning horizon, the total capacity (centralised By 2034 the total centralised capacity of 681 MW is dominated and decentralised) by 2034 is expected to reach 742 MW, an over by wind (165 MW) and solar PV (250 MW). The decentralised four-fold increase from the present installed capacity of 179 MW. capacity of 61 MW is dominated by biomass (59 MW), with a small amount of coal (2 MW). This is mainly because biomass is Under the Base Case assumptions, energy imports continue to the available resource in the country for co-generation; hence dominate throughout the planning horizon. New grid-connect- it is economically feasible for power generation. Wind and so- ed capacity is scheduled to be commissioned in 2020, namely lar PV technologies are the most economical renewable energy 12 MW of biomass, 14 MW of hydropower and 10 MW of solar sources due to the continuous cost decline in these technolo- PV. Additional significant investments in solar PV and oil-based gies as they become increasingly competitive with conventional capacity commence later in the planning period, in around sources in the country around 2030. From an operator’s per- 2025; biomass capacity is further expanded continuously from spective, studies will need to be undertaken to ensure that in- 2020 to 2034. Significant new wind capacity is added in 2032 troduction of large centralised solar PV and wind plants does (121 MW) and 2034 (44 MW). A degree of reliance on imports not compromise the integrity of the grid. ENERGY MASTERPLAN 2034 91 The total centralised capacity from oil amounts to 89 MW by Eswatini is considering developing a coal-fired power plant to 2034. This includes new capacity of 80 MW added between achieve self-sufficiency, even though coal is not a least-cost 2027 and 2034, which is triggered by the assumption that the option. The price of domestic coal makes it non-competitive price of imported electricity will rise later in the planning period as compared to the price of imported electricity. However, (including already expensive peak power). However, other al- whether domestic coal makes it into the capacity mix or not ternative measures including demand-side management inter- is highly sensitive to the assumptions made regarding the do- ventions and/or imports can be utilised to reduce or eliminate mestic coal price. the reliance on diesel generation for peaking purposes. FIGURE 8.6 BASE CASE: ELECTRICITY GENERATION CAPACITY BY TYPE 800 700 600 Installed Capacity (MW) 500 Dist. Solar PV Mini Hydro Dist. Biomass 400 Dist. Coal Dist. Oil Wind 300 Solar Thermal Solar PV Biomass 200 Hydro Nuclear 100 Gas Oil Coal 0 System peak 2014 2016 2018 2020 2022 2024 2026 2028 2030 2032 2034 Currently local power generation comprises mainly renewable total centralised power supply in 2014 falls to 27 %. With the energy sources (97 % of total domestic electricity production), projected decreasing share of imports, the share of renewa- and this remains at this high level throughout the planning pe- bles in centralised power generation is expected to increase riod. However, when accounting for imports in the total power to 55 % by 2034. supply, the share of renewable electricity production in the 92 KINGDOM OF ESWATINI Figure 8.7 shows the breakdown of the installed renewable zone E is deployed most, followed by zone C, zone A and zone energy capacity for centralised generation, including specific B. Further details on the constitution and location of the zones zones for new solar PV and wind capacity to be deployed un- is provided in Chapter 6. For biomass, all deployed capacity is der the Base Case. For wind, seven potential zones are incorpo- bagasse-based generation that uses coal and wood chips as rated, and the deployment will happen mainly in zone J and a supplementary fuels. No deployment of wood chip-only gen- small amount in zone I. For solar PV, five zones are evaluated; eration is envisaged under the Base Case. FIGURE 8.7 BASE CASE: CAPACITY EXPANSION OF RENEWABLE ENERGY 1,400 60 % System peak Hydro existing Hydro new 1,200 50 % Bagasse existing Bagasse new RE share of cent. supply (incl. imports) Electricity production (GWh) 1,000 Wood chips existing 40 % Wood chips new PV E 800 PV D 30 % PV C PV B 600 PV A 20 % Wind O 400 Wind M Wind K 10 % Wind J 200 Wind I Wind B 0 0% Wind A 2015 2017 2019 2021 2023 2025 2027 2029 2031 2033 ENERGY MASTERPLAN 2034 93 Fuel consumption for electricity generation biomass power plants. Because of limited consumption of fossil fuels, the CO₂ emissions from power generation remain at a low Bagasse and wood chips will remain the primary fuels for elec- level. In 2014 CO₂ emissions totalled 85 kilotonnes (Kt), due to tricity generation. Most bagasse is replaced by wood chips af- relatively large use of coal as a secondary fuel in current bio- ter 2018. Diesel consumption is very limited and used only for mass generation; emissions will fall to 39 Kt as the use of coal peaking power. Bagasse and wood chips, and small amounts decreases over time. of coal, will be used for generation in the projected 101 MW of FIGURE 8.8 BASE CASE: FUEL CONSUMPTION OF ELECTRICITY PRODUCTION 12,000 10,000 8,000 TJ 6,000 4,000 Diesel Wood chips 2,000 Natural Gas Coal 0 Bagasse 2014 2016 2018 2020 2022 2024 2026 2028 2030 2032 2034 94 KINGDOM OF ESWATINI Transmission and distribution needs 170 MW in 2014, has to more than double by 2034. Distribu- tion to the commercial sector has to grow to around 32 MW Figure 8.9 shows the investment needs for transmission and by 2034, and distribution to the urban sector is expected to distribution. As the investment in centralised power plants increase by about 70 %. Distribution to industry is expected to starts to grow, the needs for investment in transmission grow remain constant. accordingly. The transmission capacity, estimated at roughly FIGURE 8.9 BASE CASE: TRANSMISSION AND DISTRIBUTION NEEDS 800 700 600 500 MW 400 300 200 Distibution to rural Distribution to urban 100 Distibution to commercial Distibution to industy 0 Transmission 2014 2016 2018 2020 2022 2024 2026 2028 2030 2032 2034 ENERGY MASTERPLAN 2034 95 Total system costs imports are a cheaper alternative until solar PV and wind be- come competitive. From 2020 onwards, centralised generation The development of the total system costs is shown in Fig- investment costs will occupy a larger portion of the system ure 8.10. Investment costs are levelised over the life time of costs. The fuel costs primarily for bagasse, wood chips and oil the plants. The need for new capacity development to achieve start to increase as these technologies come into the system. sufficient national electricity supply will result in a continuous Overall average electricity supply costs (i. e., total system costs increase in system costs over the years. Import costs account divided by total electricity sold) increase from USD 28 per MWh for the bulk of the overall system costs. This is mainly because in 2014 to USD 74 per MWh in 2034. FIGURE 8.10 BASE CASE: SYSTEM COST SUMMARY 250 80 70 200 60 System Costs (million 2013 USD) Avg. unit cost (USD/MWh) 50 150 Exports 40 Imports Dist. Gen. O&M 100 30 Dist. Gen Investment Cent. Gen O&M 20 Cent. Gen Investment 50 Transmission and Distribution 10 CO2 Fuel 0 0 Avg. unit cost 2014 2016 2018 2020 2022 2024 2026 2028 2030 2032 2034 96 KINGDOM OF ESWATINI With regard to the financial implications of investment, the Forced Coal scenarios (300 MW and 600 MW) cumulative total investment needs over the planning horizon amount to around USD 888 million, of which 89 % is invest- The scenarios illustrate that large-scale coal-fired generation ments in centralised generation, 3 % investments in decentral- could be a part of the solution only when it is forced to be. ised generation and 8 % investments in transmission and dis- Under the Forced Coal 1 scenario, a 300 MW coal capacity is tribution. committed in 2025; under the Forced Coal 2 scenario, 600 MW is committed in 2025.

8.4.2 ALTERNATIVE CASE RESULTS: POWER SECTOR The respective electricity supply mix under the two coal sce- The alternative cases are policy options or “policy alternatives” narios is shown in Figure 8.11. Under the Forced Coal 1 scenario, for different pathways that could be pursued to achieve energy after 2025 the bulk of local electrical energy requirements are policy objectives (see Section 8.2 for existing energy policy ob- met through energy generated from the coal power plant, with jectives). some production from solar PV (after 2020), hydro, biomass and a small share of oil. Solar PV starts producing from 2020 at 17 GWh, which increases to 240 GWh in 2034. In the Forced Coal 1 scenario, Eswatini moves to being a net exporter from 2025 on- wards. However, the introduction of the coal power plant limits the local exploitation of renewable energy resources. FIGURE 8.11 FORCED COAL SCENARIOS: ELECTRICITY GENERATION Forced Coal 1 Forced Coal 2 6,000 5,000 4,000 Electricity Production (GWh) 3,000 2,000 1,000 0 2014 2018 2022 2026 2030 2034 2014 2018 2022 2026 2030 2034 -1,000 -2,000 -3,000 Dist. Solar PV Dist. Coal Wind Biomass Gas Mini Hydro Dist. Oil Solar Thermal Hydro Oil Dom. System dmd Dist. Biomass Net Imports Solar PV Nuclear Coal ENERGY MASTERPLAN 2034 97 FIGURE 8.12 FORCED COAL SCENARIOS: CAPACITY MIX Forced Coal 1 Forced Coal 2 900 800 700 Installed Capacity (MW) 600 500 400 300 200 100 0 2014 2018 2022 2026 2030 2034 2014 2018 2022 2026 2030 2034 Dist. Solar PV Dist. Coal Wind Biomass Gas Mini Hydro Dist. Oil Solar Thermal Hydro Oil System peak Dist. Biomass Net Imports Solar PV Nuclear Coal As shown in Figure 8.12, both Forced Coal scenarios exhibit a coal power plant will make the country a net exporter, the rev- similar pattern by eliminating wind power; however, the Forced enues do not cover the cost of the higher capacity. Coal 2 scenario also greatly reduces solar PV. In this scenario, domestic production is increased significantly through coal ca- Domestic Resources scenarios (Limited Import 1, Limited pacity, which eliminates almost all other technology options Import 2 with forced hydro and Limited Import 3 without coal/ and leads to higher exports. Besides coal, the energy mix is lim- with renewables only) ited to the existing biomass and hydropower plants. This puts a severe limitation on the national objective of diversifying the Improving self-sufficiency in power generation is one of the electricity generation mix. most important policy goals in Eswatini. Three scenarios were developed to explore the impacts of limiting imports from Apart from constraining diversification of the country’s gen- South Africa and making maximum use of domestically en- eration mix, both scenarios require significant amounts of dowed resources, with a wide range of technology options. The capital investment. The overall investment requirements for first scenario (Limited Import 1) includes all of the generation the Forced Coal 1 scenario and the Forced Coal 2 scenario are options as optional (i. e., not forced), while the second scenario USD 1,301 million and USD 1,879 million, respectively, whereas (Limited Import 2) imposes deployment of 120 MW of hydro- the Base Case investment amounted to USD 888 million. These power plants in 2025. The third scenario (Limited Import 3) ex- high investment costs result in increased average unit costs. plores supply met only with renewable energy sources. Under The average unit costs in 2034 for the Forced Coal 1 scenario all three scenarios, the share of imported electricity is reduced is USD 73.6 per MWh, which is comparable to the Base Case. to meet up to 35 % of total domestic electricity demand by The average unit cost in 2034 for the Forced Coal 2 scenario is 2025, and 25 % by 2030. roughly USD 88 per MWh. Even though building the 600 MW 98 KINGDOM OF ESWATINI In contrast to the Base Case, under the Limited Import 1 diversified with import needs reduced by 36 % over the plan- scenario coal generation enters into the solution, contribut- ning horizon, from 858 GWh in 2014, to a peak of 1,129 GWh in ing about 14 % of total generation by 2034. This is a result of 2019, to 513 GWh in 2034. the commissioning of 36 MW of coal generation in 2025 and an additional 9 MW in 2026 and 8 MW in 2027 under Limited Under the second scenario, Limited Import 2, imports are limit- Import 1. Before the coal power plant is installed, biomass- ed and the deployment of 120 MW of hydropower is forced; this based generation is increased significantly from 2020, adding reduces coal to 15 MW in 2025, as the 120 MW of hydro added 105 MW of capacity in the first half of the 2020s. Additional in 2025 together with other renewable options are sufficient to solar PV starts to get introduced in 2020, and by 2034 it con- meet the self-sufficiency target. By 2034 the generation mix is tributes about 12 % of the country’s electricity needs. Wind well diversified with hydro, biomass, solar PV and wind genera- power starts producing from 2030 with 91 GWh and increases tion, thus reducing imports by over 50 %, to 315 GWh, over the to over 440 GWh by 2034. By 2034 the generation mix has planning horizon (Figure 8.13). FIGURE 8.13 LIMITED IMPORT 1 AND 2 SCENARIOS: ELECTRICITY GENERATION Limited Import 1 Limited Import 2 3,500 3,000 2,500 Electricity Production (GWh) 2,000 1,500 1,000 500 0 2014 2018 2022 2026 2030 2034 2014 2018 2022 2026 2030 2034 Dist. Solar PV Dist. Coal Wind Biomass Gas Mini Hydro Dist. Oil Solar Thermal Hydro Oil Dom. System dmd Dist. Biomass Net Imports Solar PV Nuclear Coal ENERGY MASTERPLAN 2034 99 Compared with the Base Case, where solar PV and wind ac- Base Case. Solar PV installation varies, from 182 MW by 2034 count for as much as 60 % of the total installed capacity in under the Limited Import 1 scenario to 167 MW by 2034 under 2034, the capacity mix under both Domestic Resources scenar- the Limited Import 2 scenario with hydro. Under the Limited Im- ios is more diverse, as shown in Figure 8.14. The share of solar port 1 scenario, a larger share of oil and coal installed capacity PV and wind in the 2034 capacity mix is 46 % under the Limited is deployed to meet the peak, whereas under the Limited Im- Import 1 scenario and 44 % under the Limited Import 2 scenario port 2 scenario with hydro, the peak demand is met sufficiently with hydro. In 2034 installed capacity for wind is 165 MW un- with hydropower, and thus no investment is needed in the oil der both Domestic Resources scenarios, as well as under the installed capacity. FIGURE 8.14 LIMITED IMPORT 1 AND 2 SCENARIOS: CAPACITY MIX Limited Import 1 Limited Import 2 1,000 900 800 700 Installed Capacity (MW) 600 500 400 300 200 100 0 2014 2018 2022 2026 2030 2034 2014 2018 2022 2026 2030 2034 Dist. Solar PV Dist. Coal Wind Biomass Gas Mini Hydro Dist. Oil Solar Thermal Hydro Oil System peak Dist. Biomass Net Imports Solar PV Nuclear Coal 100 KINGDOM OF ESWATINI Figure 8.15 shows the results of the Limited Import 3 scenario, coal, wind is introduced in an earlier year (2025) compared with where only renewables are allowed to be introduced and the the Limited Import 1 scenario (wind introduced in 2030). Wind entry of coal- and oil-fired generation is limited. This scenario power and solar PV are introduced at about the same amount, depicts a 100 % renewable path. In this scenario, in place of reaching a capacity of around 180 MW each. FIGURE 8.15 LIMITED IMPORT 3 SCENARIO: ELECTRICITY GENERATION AND CAPACITY MIX Limited Import 3: Electricity Generation Limited Import 3: Capacity mix 3,500 800 700 3,000 600 2,500 Electricity Production (GWh) Installed Capacity (MW) 500 2,000 400 1,500 300 1,000 200 500 100 0 0 2014 2018 2022 2026 2030 2034 2014 2018 2022 2026 2030 2034 Dist. Solar PV Dist. Coal Wind Biomass Gas Dist. Oil Solar Thermal (1) Dom. System dmd Mini Hydro Hydro Oil (2) System peak Dist. Biomass Net Imports Solar PV Nuclear Coal The financial implication for the three Domestic Resources the Limited Import 1 scenario are similar to the Base Case, at scenarios is that they require substantially higher investment around USD 74 per MWh by 2034. However, under the Lim- compared with the Base Case. While the Base Case requires ited Import 2 scenario, the average supply costs are higher, at USD 888 million, the Limited Import 1 scenario and the Lim- around USD 90 per MWh by 2034. Under the Limited Import 3 ited Import 2 scenario would require USD 1,113 million and 1,731 scenario, the investment requirement is USD 1 027 million and million of investment, respectively. The average supply costs in average unit costs around 76.88 per MWh by 2034. ENERGY MASTERPLAN 2034 101 No Import Enhancement scenarios (High Demand with no im- Figure 8.16 shows the resulting supply mix under both scenari- port enhancement, Base Case with no import enhancement) os. As can be seen, the additional restriction on no import en- hancement has little impact on the Base Case results. The “No Import Enhancement” scenarios assume that the im- port capacity is not expanded. Under the High Demand with no Under the High Demand scenario, demand will reach 3,399 import enhancement scenario, electricity demand in the mining GWh by 2034, compared with 2,648 GWh under the Base Case. sector is expected to increase in the planning horizon, resulting The increased demand will be met with coal-based genera- in challenges in meeting the country’s electricity demand. This tion of 160 GWh in 2025, rising to 785 GWh in 2034. Imports scenario depicts a pathway for how electricity demand can be are stretched to the maximum degree. Coal-fired power plants met even as electricity imports are assumed to not exceed the come into play gradually from 2025 with 22 MW capacity, which 185 MW capacity limit outlined in the current bilateral agree- increases to an overall capacity of 111 MW by 2034. Solar PV ment with South Africa. The Base Case scenario was run with capacity increases by 289 MW, biomass capacity by 76 MW and a 185 MW capacity limit on imports, as a point for comparison. hydro capacity by 14 MW. FIGURE 8.16 NO IMPORT ENHANCEMENT SCENARIOS: ELECTRICITY GENERATION High Demand Base Case (with no import enhancement) 4,000 3,500 3,000 Electricity Production (GWh) 2,500 2,000 1,500 1,000 500 0 2014 2018 2022 2026 2030 2034 2014 2018 2022 2026 2030 2034 Dist. Solar PV Dist. Coal Wind Biomass Gas Mini Hydro Dist. Oil Solar Thermal Hydro Oil Dom. System dmd Dist. Biomass Net Imports Solar PV Nuclear Coal 102 KINGDOM OF ESWATINI Figure 8.17 illustrates that in order to meet the high demand, 799 MW between 2014 and 2034, of which 71 % comes from re- 927 MW of installed capacity (centralised and decentralised) by newable energy technologies. This includes 289 MW of solar PV, 2034 is needed, compared with 742 MW under the Base Case. 165 MW of wind, 76 MW of biomass and 14 MW of hydro. This would require additional investment in new capacity of FIGURE 8.17 NO IMPORT ENHANCEMENT SCENARIOS: CAPACITY MIX High Demand Base Case (with no import enhancement) 1,000 900 800 700 Installed Capacity (MW) 600 500 400 300 200 100 0 2014 2018 2022 2026 2030 2034 2014 2018 2022 2026 2030 2034 Dist. Solar PV Dist. Coal Wind Biomass Gas Mini Hydro Dist. Oil Solar Thermal Hydro Oil System Peak Dist. Biomass Net Imports Solar PV Nuclear Coal ENERGY MASTERPLAN 2034 103 To achieve this scenario, a cumulative investment total of equivalent to around 2,361 GWh). The average generation cost USD 1,310 million is required in the planning horizon, as com- per unit of electricity is around SZL 0.9 per kWh (USD 64 per pared to USD 888 million required in the Base Case. MWh) by 2034. The scenario represents a pathway with relatively low electricity costs but a high degree of dependency on imports. Low Import Price scenario Figure 8.18 shows the resulting supply mix under the scenario. Under this scenario, there is no limit on imports of electricity. The domestic system demand is expected to be 2,648 GWh by The scenario assumes a low import tariff, with no general cost 2034, similar to that under the Base Case. The mix of domestic escalation and an annual import tariff increase of 3 %. As a result, generation in centralised generation in 2034 consists of hydro the price of imported electricity is expected to be low over the (222 GWh), biomass (52 GWh), solar PV (42 GWh) and a very planning horizon (USD 57.1 per MWh by 2034), and imports will minor share of oil. In addition, domestic supplies come from dominate the supply of electricity (89 % of total demand by 2034, distributed biomass and distributed coal. FIGURE 8.18 LOW IMPORT PRICE SCENARIO: ELECTRICITY GENERATION 3,500 Dist. Solar PV Mini Hydro 3,000 Dist. Biomass Dist. Coal Electricity production (GWh) 2,500 Dist. Oil Net Imports Wind 2,000 Solar Thermal Solar PV 1,500 Biomass Hydro 1,000 Nuclear Gas 500 Oil Coal 0 Dom. System dmd 2014 2016 2018 2020 2022 2024 2026 2028 2030 2032 2034 104 KINGDOM OF ESWATINI Figure 8.19 shows the capacity mix under the Low Import Price To achieve this scenario a cumulative investment total of scenario. The only domestic capacity additions in the planning USD 204 million is required in the planning horizon, as com- period are new capacities of 14 MW of hydro in 2020, 11 MW of pared to the USD 888 million required in the Base Case. A large biomass in 2030, 10 MW of solar PV in 2020 and 14 MW of solar PV share (37 %) of the investment is required for enhancing do- in 2032, and overall new capacity in distributed biomass of 29 MW. mestic transmission and distribution. FIGURE 8.19 LOW IMPORT PRICE SCENARIO: CAPACITY MIX 450 Dist. Solar PV Mini Hydro 400 Dist. Biomass 350 Dist. Coal Dist. Oil Installed Capacity (MW) 300 Net Imports Wind 250 Solar Thermal Solar PV 200 Biomass 150 Hydro Nuclear 100 Gas Oil 50 Coal 0 Dom. System dmd 2014 2016 2018 2020 2022 2024 2026 2028 2030 2032 2034 ENERGY MASTERPLAN 2034 105 Natural Gas scenario The generation mix under the Natural Gas scenario is almost identical to that under the Base Case, with the difference of us- No natural gas options feature in the Base Case or in any of the ing gas instead of oil as a peaking plant. The non-competitive- alternative scenarios. In the Natural Gas scenario, these plants ness of gas-based generation is due to the fact that natural gas only feature where gas is explicitly enforced. As shown in Fig- combustion turbines are more expensive to operate than other ure 8.20, 150 MW of gas power plants is forced to be installed types of power plants, in particular hydropower plants, and are in 2021, but the electricity supply mix shows that no electricity also costlier than imports. Furthermore, the natural gas will be is generated from this capacity; hence this plant can be used as imported, and the infrastructure needs are capital intensive. In reserve capacity. this scenario, generation from solar PV is expected to increase from 17 GWh in 2020 to 476 GWh in 2034, and wind is intro- duced in 2032, similar to under the Base Case. FIGURE 8.20 NATURAL GAS SCENARIO: ELECTRICITY SUPPLY MIX AND CAPACITY MIX 3,500 900 800 3,000 Electricity Production (GWh) 700 Installed Capacity (MW) 2,500 600 Dom. System dmd 2,000 500 1,500 400 System peak 300 1,000 200 500 100 0 0 2014 2018 2022 2026 2030 2034 2014 2018 2022 2026 2030 2034 Dist. Solar PV Wind Nuclear Mini Hydro Solar Thermal Gas Dist. Biomass Solar PV Oil Dist. Coal Biomass Coal Dist. Oil Hydro In terms of financial implications, the investment cost require- cost is slightly higher than in the Base Case due to additional ment under the Natural Gas scenario is about USD 960 million, investment in gas that is not fully made use of. which is more than the Base Case scenario. The average supply 106 KINGDOM OF ESWATINI 8.4.3 COMPARATIVE ASSESSMENT OF SCENARIOS: penetration of coal-fired power plants in each scenario. Un- POWER SECTOR der the Forced Coal 2 scenario, the emissions increase to 4.5 metric tonnes after 2024, and under the Forced Coal 1 sce- Carbon dioxide emissions nario they remain at 2.2 metric tonnes; in the remainder of the scenarios emissions remain below 1 metric tonne. Figure 8.21 Across the scenarios, the annual CO₂ emissions from the shows a comparison of CO₂ emissions for all scenarios from power sector remain negligible until 2024. The amount of CO₂ 2014 to 2034. emissions from the power sector corresponds directly to the FIGURE 8.21 CUMULATIVE CARBON DIOXIDE LEVELS FROM 2014 TO 2034 FOR ALL SCENARIOS 5 Base Case Base Case (no import enhancement) 4 Forced Coal 1 (300 MW) Forced Coal 2 (600 MW) 3 MtCO2 Low Import 1 Limited Import 2 (with hydro) 2 Limited Import 3 (with renewables) 1 Natural Gas High Demand 0 Low Import Price 2014 2018 2022 2026 2030 2034 Share of renewable energy in total demand enhancement (66 %) and Natural Gas (67 %) scenarios, or more significantly, as in the Limited Import 1 (71 %) and Limited Im- Figure 8.22 shows the share of renewable energy in total en- port 2 (89 %) scenarios. The Limited Import 3 scenario depicts a ergy demand for all scenarios. In the Base Case the share grows pathway where 100 % of domestic electricity generation is met from 44 % in 2014 to 66 % by 2034. In the Forced Coal 2 (19 %), by renewables, although the overall share of renewables in total Low Import Price (20 %) and Forced Coal 1 (37 %) scenarios the demand is limited to 80 % since 20 % of domestic demand is share of renewable energy is declining over the planning pe- met by imports. The increased share of renewable energy can riod. In all other scenarios, the share of renewable energy is bring considerable economic benefits, mainly through reduced increasing between 2014 and 2034, either moderately, as is in reliance on and spending for fossil-based fuels. the case of the High Demand (53 %), Base Case with no import ENERGY MASTERPLAN 2034 107 FIGURE 8.22 COMPARATIVE SHARE OF RENEWABLE ENERGY FOR ALL SCENARIOS 100 % Base Case 80 % Base Case (no import enhancement) Forced Coal 1 (300 MW) 60 % Forced Coal 2 (600 MW) Low Import 1 40 % Limited Import 2 (with hydro) Limited Import 3 (with renewables) 20 % Natural Gas High Demand 0% Low Import Price 2014 2018 2022 2026 2030 2034 Import dependency 50 % of total electricity production. Apart from the Low Import Price scenario, which results in 90 % import dependency by Electricity trade is crucial to harness the rich mix of energy 2034, the share of imports gradually decreases throughout the resources in the country, while excessive reliance on imports planning horizon, to below 50 % from 2019 onwards, while in hampers energy security. The scenarios were developed using the Forced Coal scenarios the excess electricity supply poses different assumptions about trade. Across all scenarios, imports opportunities for exports from 2025 (Figure 8.23). are crucial for the country until 2019, where they remain above FIGURE 8.23 SHARE OF IMPORT ELECTRICITY IN TOTAL ELECTRICITY PRODUCTION FOR ALL SCENARIOS 100 % Base Case 50 % Base Case (no import enhancement) Forced Coal 1 (300 MW) 0% Forced Coal 2 (600 MW) 2014 2018 2022 2026 2030 2034 Low Import 1 Limited Import 2 (with hydro) -50 % Limited Import 3 (with renewables) Natural Gas -100 % High Demand Low Import Price -150 % 108 KINGDOM OF ESWATINI Investment requirements Investments under the Forced Coal 1 (300 MW) scenario (USD 1 301 million) and the High Demand with no import en- Figure 8.24 shows the total cumulative investment costs be- hancement scenario (USD 1,310 million), as well as those under tween 2014 and 2034 in each scenario, and their breakdown the Limited Import 1 scenario (USD 1,113 million), Limited Im- into investments related to centralised systems, decentralised port 3 scenario (USD 1,027 million) and Natural Gas scenario systems, and transmission and distribution. Total investment (USD 960 million) constitute the middle ground. The Low Im- needs for the Forced Coal 2 (600 MW) scenario (USD 1,880 port Price scenario requires the least amount of investment at million) and the Limited Import 2 scenario (USD 1,713 million) around USD 204 million, as it requires limited investment in are the highest among all scenarios at around, about twice the building domestic generation infrastructure. investment needs of the Base Case scenario (USD 888 million). FIGURE 8.24 UNDISCOUNTED CUMULATIVE LUMP-SUM INVESTMENT COSTS BY TYPE OVER THE PERIOD 2014 TO 2034 2,000 Centralised gen. Decentralised gen. Dom T&D 1,800 1,600 1,400 1,200 Million USD 1,000 800 600 400 200 0 Base Case Base Case Forced Forced Limited Limited Limited Natural High Low Import (no import Coal 1 Coal 2 Import 1 Import 2 Import 3 Gas Demand Price enhancement) (300 MW) (600 MW) (with hydro) (with renewables) ENERGY MASTERPLAN 2034 109 Average generation costs exception of the Forced Coal 2 (600 MW) scenario and the Limited Import 2 with hydro scenario, where the costs in- Figure 8.25 compares the average generation costs per kilo- crease drastically in 2025 and then decline slowly thereafter. watt-hour across all scenarios. The unit cost increases gradu- The costs remain below SZL 1.00 per kWh (USD 71 per MWh) ally over the time horizon for all of the scenarios, with the until around 2029. FIGURE 8.25 AVERAGE GENERATION COSTS FOR ALL SCENARIOS 1.3 Base Case 1.1 Base Case (no import enhancement) Forced Coal 1 (300 MW) SZL/kWh

0.9 Forced Coal 2 (600 MW) Low Import 1

0.7 Limited Import 2 (with hydro) Limited Import 3 (with renewables) Natural Gas 0.5 High Demand Low Import Price 0.3 2014 2018 2022 2026 2030 2034 110 KINGDOM OF ESWATINI

8.4.4 PRIMARY ENERGY REQUIREMENT whereas bagasse requirement follows an opposite trend, grow- ing from 8.4 PJ in 2014 to 14.8 PJ in 2034. The High Demand Figure 8.26 shows the primary energy requirements for all sce- with no import enhancement scenario overall assumes a higher narios. 9 out of the 10 scenarios assume the demand as under primary energy requirement. The primary energy requirement the Base Case, featuring a growth in primary energy require- first reduces until 2022, due to the constraints of the scenario, ment from 39.4 PJ in 2014 to 48.4 PJ in 2034. Growth is particu- but then gradually grows for the remainder of the planning larly strong in electricity requirement (more than doubles, from period. The difference in requirement between the two cases

4.6 PJ to 9.5 PJ). Demand for oil products (14.1 PJ by 2034) and stems from electricity requirement, which grows to 12.2 PJ in coal (2.6 PJ by 2034) increase moderately. Demand for wood 2034 under the High Demand scenario. based supply reduces from 14.5 PJ in 2014 to 6.4 PJ in 2034, FIGURE 8.26 PRIMARY ENERGY REQUIREMENTS FOR ALL SCENARIOS Base Case High Demand (No Import Enhancement scenario) 60,000 50,000 40,000 Final Demand (TJ) 30,000 20,000 10,000 0 2014 2018 2022 2026 2030 2034 2014 2018 2022 2026 2030 2034 Fuelwood Coal Bagasse Wood chips Oil products Electricity ENERGY MASTERPLAN 2034 111 Implications for the sugarcane sector ments by sugar companies to increase their power generation capacity from 106 MW to 160 MW during the planning horizon Bagasse is the most viable biomass fuel for power generation, (see Section 6.1). Ethanol production would increase to meet and is used together with wood chips for co-generation. To the blending ratio of 10 % ethanol (E10) in unleaded petrol produce the additional 76 MW from biomass by 2034 in the (see Section 7.3); the current production is not sufficient to Base Case there is a need to expand the domestic sugar sector achieve this blending ratio. Figure 8.27 shows the primary re- as well as the timber industry for wood chip production. This quirements of bagasse, molasses and ethanol throughout the would require additional ambition on top of the announce- planning horizon. FIGURE 8.27 PRIMARY REQUIREMENTS FOR ETHANOL, MOLASSES AND BAGASSE PRODUCTION 30 25 20 Molasses PJ 15 Bagasse Ethanol 10 5 0 2014 2018 2022 2026 2030 2034 112 KINGDOM OF ESWATINI 9 VISION FOR 2034 The energy landscape in 2034 will be fundamentally different population of Eswatini is expected to increase slightly in the from that of today. Given the long lifespan of energy infrastruc- planning time horizon (2014–2034); economic growth increases ture, proper investment decisions made today can enhance the modestly at an annual average rate of 1.8 %–3 % (2014 to 2020) energy supply system for decades, and wrong decisions now to 3 % (2020 to 2034) under the business-as-usual trajectory. can adversely affect the energy system for decades. For this Under all scenarios, 100 % electrification is assumed to be reason, today’s energy policy will decisively determine the fu- achieved within the planning time horizon. Various goals un- ture, by setting out clear pathways to improve the role of en- der the SE4ALL initiative – such as the promotion of alterna- ergy in the economic development of Eswatini. Long-term plan- tive modern fuels for cooking by replacement of wood; solar ning is crucial to support cost-effective development of energy water heating; and energy efficiency improvements including infrastructure, including integrating larger shares of renewable increased use of efficient lighting – are also taken into account. energy into the generation mix. While the demand for wood-based supply is expected to be re- This document provides a forward-looking plan to inform the duced significantly, demand for oil products and electricity are ex- direction of national energy policy. It provides a detailed evalu- pected to grow under the Reference scenario for economic growth, ation of trends in the energy supply and demand of the coun- from 9.4 TJ to 14.1 TJ for oil, and from 4.6 TJ to 9.5 TJ for electric- try, drivers and resource prospects for various energy sources, ity. Under the High Growth scenario (1.8 % to 3 % GDP growth per costs and benefits related to the deployment of renewable en- year until 2020, and over 3 % growth after 2020) the demand for ergy technologies, and implications of pursuing different path- oil is expected to reach 14.7 TJ and the demand for electricity is ways for policy and investment decision making. expected to reach 12.2 TJ, while under the Low Growth scenario (0.9 % to 1.8 % GDP growth until 2020 and 1.8 % growth thereafter The analysis presented in this Masterplan was presented to until 2034), they reach 12.0 TJ and 8.2 TJ, respectively. In all three stakeholders, and the concrete quantitative assessment of dif- economic growth scenarios, demand for wood-based fuel is re- ferent pathway options outlined in the plan proved to be an duced from the current level of 11.9 TJ to 1.6 TJ. effective basis for discussing the future of the national energy system. The process of developing this Masterplan highlighted To accommodate the modernisation of the energy supply the importance of defining an integrated and comprehensive though implementation of energy efficiency measures, a major long-term vision to guide effective policy making for a sustain- shift in the use of energy-saving devices, such as solar water able energy future. heaters and LED lighting, is required – supported by a clear government policy to such efforts. At the same time, the en-

9.1 KEY INSIGHTS FROM THE SCENARIO ergy supply structure in Eswatini needs to expand; oil imports ANALYSIS are expected to grow by 30 % to 60 % within the planning ho- rizon (depending on the scenarios), while electricity production At the outset of this Masterplan, the key long-term goal of needs to grow by 90 % to 180 %. energy sector development was defined as achieving a well- diversified energy mix that will 1) provide sufficient energy sup- With regard to power sector development, for adequate expan- ply; 2) provide access to clean energy technologies; 3) foster sion of the system the total system cost of providing electric- industrialisation; and 4) ensure security of energy supply. The ity is expected to increase about four- to five-fold, from the following sections assess results from the scenarios analysed in current level of around USD 40 million per year to the range this Masterplan in light of national policy goals. of USD 175 million to USD 200 million by 2034. The increase in investments will be driven primarily by the need to develop Provision of sufficient energy supply domestic power generation infrastructure with the objective of gradually reducing dependency on imports. The development Part II of this report provides an assessment of the long-term of generation infrastructure must be complemented by invest- energy needs to meet the growth ambition of the country as ments in transmission and distribution necessary to wheel en- stipulated in the National Development Strategy. This includes ergy to load centres and to support the transmission and dis- an assessment of trends in demographic changes as well as tribution network, including to enable the system to integrate in overall economic growth. According to the assessment the higher shares of renewable energy supplies. ENERGY MASTERPLAN 2034 113 Fostering industrialisation Energy access The Government of the Kingdom of Eswatini acknowledges Eswatini’s short-term goal is to accelerate the rural electrifica- energy as one of the key drivers for economic development. tion programme that seeks to provide electricity/clean energy Achievement of the country’s macroeconomic development to rural households. Household electricity access stood at 74 % goals depends on clear, measurable and practical policy di- in 2017; Eswatini aims to achieve universal access to electricity rections and actions for energy in the short, medium and long by 2022, as part of the National Development Strategy and also terms. The systematic uncertainty about the long-term supply supported by the country’s Sustainable Energy for All goals. Sub- or “just enough power” situations with high degree of uncer- stantial progress has been made over the past decade to improve tainty hampers the industrial investment appetite. This Master- household access to modern energy services in the country. plan is expected to be the primary point of reference in policy formulation, providing the foundation for clear policy goals. Yet, nearly 70 % of households still rely on inefficient fuelwood for cooking and heating; traditional biomass use creates social Currently energy supply relies on imports from South Africa. and environmental costs as a result of the negative impacts on Not only does the significant energy import undermine the air quality and health as well as having a high opportunity cost trade balance, the price for peak import is higher than the do- of time used for wood collection. Thus, besides expanding en- mestic selling price, which puts some pressure on the future ergy access, the country needs to also gradually replace the trajectory for the tariff in order to fully cover the cost of supply. use of traditional biomass. Efforts are underway to increase In this Masterplan, various future paths that have higher lev- access to efficient cook stoves and modern cooking fuels, in els of autonomy have been investigated. Under the Base Case particular LPG. scenario, import dependency is expected to be reduced from 66 % in 2014 to 38 % by 2034. Under the three Limited Import According to the demand assessment discussed in Chapter 4, scenarios, import dependency is reduced to a range of 10 % to it is envisaged that the use of fuelwood will decrease by 90 % 24 % by 2034. This is achieved at the expense of higher genera- in the residential sector, while use of LPG will increase almost tion costs, but the magnitude of the cost increase is relatively three-fold by 2034, relative to current levels. By 2030 the por- marginal at 1 %–5 %, except for the Limited Import 2 scenario tion of the population relying on fuelwood for cooking and where the additional investment in hydro leads to an increase heating could fall from the current 70 % to 25 % if clean and of 22 % in average generation costs. sustainable cooking alternatives are incentivised through poli- cy-based interventions such as rebate programmes to encour- Furthermore, the Limited Import scenarios would require build- age households to use modern energy cook stoves. ing domestic generation capacity, which would at the same time create jobs. In particular, the Limited Import 3 scenario With the envisaged reduction in the use of fuelwood, forest with renewables would foster creating a new industry around degradation is expected to slow, yielding important climate solar and wind and further strengthen the existing biomass sec- benefits. It is estimated that deforestation and forest degrada- tor. Under this scenario over 180 MW of new capacity would be tion globally contribute about 15 % to 20 % of all greenhouse required for solar PV and wind each, and 140 MW of additional gas emissions. To reduce forest loss, Eswatini is poised to boost capacity would be required for bagasse-fuelled generation. carbon sequestration in forests. Another set of scenarios explored new industry development Furthermore, there is potential for the adoption and roll-out of around coal resources. While new coal industry will also cre- decentralised/distributed generation based on household-level ate jobs, the coal industry suffers from negative externalities systems, mainly from solar PV. In addition, solar water heaters – notably mining-related death and health impacts expected could potentially be deployed by 50 % of households, thereby as a result of pollution from coal-based generation. In addition, replacing the most inefficient wood-based water heating de- global decarbonisation ambition makes it harder for the coal vices. At a larger scale, off-grid stand-alone systems have a industry to be attractive. Many financiers are valuing the risk of potential to be deployed to tap into locally available energy coal investments becoming stranded assets. To be compatible resources, such as bioenergy and solar PV, to electrify remote with the Government’s commitment to the Paris Agreement, areas and meet the growing energy demand. These measures industrialisation around clean energy appears to be more desir- form an important part in achieving SDG 7, and particularly its able than betting on new coal-based capacity. sub-target of reaching universal access to affordable, reliable and modern energy services by 2030. 114 KINGDOM OF ESWATINI Energy security (solar PV and wind) to a maximum of 70 % of generation at any point in time. It may be required that the country introduces Energy security has three key components: the affordability, state-of-the-art grid control technologies and operation rules reliability and environmental sustainability of energy supply, to accommodate the high share of variable renewable energy. known as the energy trilemma. Addressing all three objectives simultaneously may in the near term require decision makers Under the two Forced Coal scenarios, domestic energy produc- to strike a delicate balance between short-term (such as im- tion is dominated by coal-powered generation. These scenarios mediate economic growth gains) and long-term policy goals; are not consistent with the government policy of diversifying the long-term objective is to design energy policy and markets the energy mix. All other scenarios show a balanced mix of dif- in a way that aligns all three components. The main finding ferent energy sources including hydro, solar, biomass and wind, across scenarios is that renewable energy sources can play an and improve the diversification of energy sources compared increasingly important role in providing reliable, affordable and to the 2014 power system structure; the exception is the Low environmentally sound energy, while enhancing energy access Import Price scenario, where imports dominate the supply of including through decentralised solutions. electricity (89 % by 2034). Reliability of supply As for the transport fuel mix, the assessment shows that do- mestic production of ethanol from molasses can meet the 10 % Total national electricity demand is expected to rise by 113 % blending requirement for petrol to be used in the transport sec- between 2014 and 2034. Indigenous energy resources can pro- tor. This has the potential to moderately contribute to further vide a cost-effective and secure supply that underpins sustaina- diversification of the energy mix. ble economic growth, and the balanced deployment of diversi- fied resources is key to ensuring the secure supply of electricity. Affordability These resources encompass both fossil fuels and renewable energy options, such as hydro, solar, wind and bioenergy. Cur- The average generation cost per unit of electricity is expected rently the production of electricity is based mainly on biomass to increase in all of the 10 scenarios assessed. This is mainly and hydropower, and the share of domestic production in the driven by increases in the import price as well as the cost of total energy supply is below 35 %. new investment, as most of the current installations have al- ready been ‘paid off’. Two scenarios, namely the Forced Coal The 10 scenarios discussed in Chapter 8 explore different mixes 2 (600 MW) scenario and the Limited Import 2 with hydro sce- of energy supply options. The shares of domestic production nario, are at the high end of the generation costs – at around by 2034 range between 10 % and 100 % under the different SZL 1.3 per kWh (USD 93 per MWh). All other scenarios are scenarios. The 10 % domestic production corresponds to a Low around SZL 1.05 per kWh (USD 75 per MWh); the exception is Import Price scenario, whereas 100 % domestic production cor- the sensitivity scenario that assumed lower import price (Low responds to the two Forced Coal scenarios in which the country Import Price scenario), which is at the low end of the generation expands domestic coal capacity and becomes a net exporter. costs, at around SZL 0.9 per kWh (USD 64 per MWh). This indi- All other scenarios suggest domestic production shares in the cates that higher investment costs compared to the Base Case range of 62 % to 90 %, thus suggesting significant improve- in some scenarios (particularly, the Limited Import 1 scenario, ments towards reduced import dependency. the Limited Import 3 scenario and the Natural Gas scenario) are dampened by reduction of other energy system costs, such as Notably, in some scenarios, renewable energy sources – includ- fuel costs and import costs. ing solar PV and wind – account for the largest share of produc- tion. The scenarios show the feasibility of a renewable pathway It is notable that the rapid cost declines for renewable energy in the power sector without compromising other objectives, technologies will reinforce the business case for scaling up the including supply adequacy. The supply adequacy is ensured by deployment of renewable energy options, mainly solar PV and imposing a reserve margin of 10 % (firm capacity) and using a wind. This can serve as an encouraging signal for IPPs and in- constraint that limits the share of the intermittent renewables crease the appetite for investment in these technologies. ENERGY MASTERPLAN 2034 115 Environmental sustainability Comparison of scenarios and implications The need to diversify the national energy mix with a view to There are various pathways to achieve enhanced energy secu- achieving energy security has been emphasised in a number of rity and energy access through 2034, as illustrated in the sce- national energy policies, also with regard to the environmen- nario analysis. Each scenario emphasises different priorities and tal sustainability of energy supply. Currently the country relies sets different constraints on the future energy system so as to mainly on hydropower, so there is no issue of pollution. The test-run the use of specific energy carriers or measures. key environmental issues are presented by use of fuelwood, resulting in degradation of land and in health problems linked The Base Case defines a business-as-usual baseline through to indoor air pollution. If adequately deployed, renewables can 2034. It projects an increased share of renewable energy and mitigate negative impacts of energy use on the environment. a reduced import dependency. The country’s first utility-scale solar PV power project is expected to come online in 2019, In the power sector, the key environmental concern is the emis- with a capacity of 10 MW. Additional hydropower capacity of sion of CO₂ from fossil fuel-based generation. The introduction 14 MW is expected in 2020. The new deployment of hydro- of domestic coal power generation would greatly increase car- power is limited to this capacity of currently committed pro- bon emissions from the country’s power sector. The Base Case jects, since hydropower is observed to be increasingly less scenario, No Important Enhancement with High Demand sce- competitive as compared to solar and wind technologies for nario, Limited Import 3 with renewables scenario, Natural Gas power generation. Co-generation continues to be used to en- scenario and Low Import Price scenario all present a trajectory sure reliable heat and electricity supply for the sugar indus- for the country to achieve clean energy policy goals, while other try; co-generation is expected to account for 11 MW of new scenarios, particularly the two Forced Coal (300 MW, 600 MW) installed capacity in 2020. In the Base Case, the share of re- scenarios, would be inconsistent with these goals. newables in the country’s electricity mix is expected to reach 30 % by 2030, which is 20 percent short of the government Energy end-use sectors, such as transport, industry, households target of 50 % for that year. and agriculture, play a key role in reducing emissions in line with the Paris Agreement and the greenhouse gas reduction The alternative scenarios emphasise the use of particular en- targets set in Eswatini’s NDC. Eswatini is exploring opportuni- ergy carriers, including coal (under the Forced Coal 300 MW ties for scaling up renewable energy in end-use sectors beyond and 600 MW scenarios), natural gas (under the Natural Gas the power sector. For example, the domestic production of bio- scenario) and hydro (under the Limited Import 2 with hydro fuels has significant growth potential to diversify the transport scenario). All of these scenarios represent future options that sector away from the dominance of imported oil products (e. g., are technically feasible. petroleum) and towards a more sustainable fuel mix. Eswati- ni already has a biofuel policy in place that includes blending Table 9.1 provides a summary overview of the results for key mandates of biofuels in transport. This policy is particularly im- parameters, across the 10 scenarios developed for the power pactful in the country’s major ethanol-producing sector, where sector. Out of the 10 scenarios, the two No Import Enhance- a blending mandate for up to 10 % ethanol in petrol is currently ment scenarios and Low Import Price scenario are considered under development. as sensitivity scenarios, meaning that they are not established to explore alternative policy paths but rather to explore sen- Besides renewable energy deployment, the Natural Gas sce- sitivities to assumptions on the external factors. That leaves nario envisions that natural gas partly displaces the consump- seven scenarios for comparison of alternative policy paths. tion of carbon-intensive energy carriers, such as coal in power generation as well as liquid fuels in transport and the residen- As discussed above, those scenarios that meet the goal of re- tial sector. With regard to the power sector, assessment of the duced import dependency and further diversification of power Natural Gas scenario revealed that due to high fuel costs, there generation sources are the Base Case scenario, the three Lim- is no economic advantage to introducing gas-fired generation ited Import scenarios and the Natural Gas scenario. From an within the planning horizon. Since natural gas lacks competi- affordability point of view, the Forced Coal 2 (600 MW) sce- tiveness with other energy resources, the country has to formu- nario and the Limited Import 2 with hydro scenario are the late goals to take this opportunity and create favourable condi- least favourable. From the CO₂ emission perspective, the two tions that can lead the growth of natural gas use in the country. Forced Coal scenarios and the Limited Import 1 scenario are the least favourable. 116 KINGDOM OF ESWATINI TABLE 9.1 SUMMARY COMPARISON TABLE FOR THE SCENARIOS Value for Value for 2034 2014 Limited Import 3 Low Import Price Limited Import 2 Limited Import 1 Base Case High Demand Forced Coal 1 Forced Coal 2 Base Case Natural Gas (no import (no import enhancement) (300 MW) (600 MW) with hydro (with renewables) enhancement) CO₂ emissions (kt) 85 39 39 2,258 4,489 425 136 19 29 871 19 Share of RE in total demand 40 % 66 % 66 % 37 % 19 % 71 % 89 % 79 % 67 % 53 % 20 % Import dependency 66 % 38 % 38 % −11 % −70 % 18 % 10 % 24 % 37 % 28 % 89 % Average generation costs (SZL/kWh) 0.38 1.03 1.04 1.03 1.23 1.04 1.26 1.08 1.07 1.10 0.92 Shannon’s diversity index* 117 % 170 % 170 % 108 % 48 % 196 % 177 % 176 % 170 % 187 % 73 % * Index = 0 indicates zero diversity. The higher the value is, the more diversified the production mix is. Based on this assessment, among those scenarios that broad- In this context, the coal options were opposed based on the ly meet the overall policy goals, the Limited Import 3 with adverse impacts of coal-fired power generation; rather, more renewables scenario seems to possess the most balanced ambitious renewable energy deployment was called for as combination of the desirable policy features. Detailed results of the most appropriate means of power generation. Some par- this scenario are given in Section 8.4.2. ticipants supported the introduction of solar and wind earlier than the time frame applied in most scenarios (major uptake

9.2 RESULTS OF THE STAKEHOLDER mid-2020). As was pointed out, with the reduction in cost for CONSULTATIONS renewable energy technology options (both utility-scale and decentralised) power generation from renewables has become The development of this Masterplan document featured an increasingly cost competitive with fossil fuels, even more so extensive process of stakeholder consultation and review. At when factoring in external costs/benefits. Yet, there remains a country meetings with key national energy sector stakeholders, need to address the intermittent behavior of the resource. participants indicated a clear preference for embracing a re- newable energy pathway and for gradually phasing out the use of fossil fuels and/or rejecting new fossil fuel capacity options such as those assessed under the two Forced Coal scenarios. Among several national stakeholder meetings, a meeting or- ganised by the Ministry of Energy in November 2017 convened 40 stakeholders and was crucial in receiving feedback on the Base Case assumptions and on the draft set of scenarios. At this meeting, participants voiced their support for a strong na- tional commitment to clean energy including in the context of the Paris Agreement and Eswatini’s NDC, and urged the imple- mentation of national climate policies as required under inter- national commitments. ENERGY MASTERPLAN 2034 117 Cost-competitive renewable energy technologies are only Policy measures need to incentivise, or be complemented by, part of the equation; determined government action is also public and private investments in domestic power generation required to unlock the energy sector’s full potential. Stake- infrastructure, to gradually reduce dependency on imports. The holders pointed out the need for clear policy support schemes rapid cost declines that have been witnessed for solar PV and for solar and wind deployment in order to address the exist- wind technologies, and that are expected to continue, strength- ing constraints, and also stressed the social value that can be en the business case for embracing renewable energy options derived from inclusive development of renewable energy. The and provide an encouraging signal for IPPs. For signing con- current policy gaps and the suggested way forward are out- tracts with IPPs, however, careful evaluation of contract condi- lined in the next section. tions (e. g., take-or-pay contracts) needs to be performed, and a tool such as SPLAT-SW is useful for that purpose. Addressing the energy security issue, stakeholders pointed out that biomass could provide sufficient feedstock for up to about The zoning information (see Chapter 6 of this report) provides 150 MW of baseload power from the timber and sugar indus- guidance on particularly suitable project sites in Eswatini. The tries. Natural gas and biogas also should be considered as vi- development of a standardised PPA could increase the bank- able options in the future energy mix. ability of renewable energy power projects in the country and help to attract scalable investment and limit elaborate negotia-

9.3 WAY FORWARD: KEY POLICY GAPS tions. Solar systems in particular present an opportunity for Es- watini to further increase electricity access and improve energy Under the business-as-usual pathway to 2034, Eswatini’s de- efficiency. An increased awareness of local financial institutions pendency on fuel and electricity imports remains high, while regarding their commercial viability, as well as adequate finan- domestic renewable energy resources remain under-utilised. cial support (such as dedicated credit lines or guarantees), are Long-term oriented government action is required to unlock needed to make improved use of this opportunity. the full potential of Eswatini’s energy sector. With regard to bioenergy, the sugar industry needs to be fur- The global policy landscape shows that new mechanisms such ther encouraged to invest in energy-efficient equipment to as auctions/competitive bidding schemes are increasingly boost bagasse-based electricity production. This could be done supplementing traditional instruments such as feed-in tariff by extending existing tax credits and allowances to the sugar schemes to drive renewable energy growth in the power sec- industry as well as through a generation-based incentive (e. g., tor. A key challenge is the intermittency of certain renewable founded on the avoided costs of power generation (or import) sources such as wind and solar. Some operational practices to the utility). The ethanol blending mandate (E10) needs to be are known to support the integration of variable renewable en- operationalised, together with guidelines for price-setting and ergy, and introduction of these practices requires institutional control mechanisms. adaptation of power system operation rules. A comprehensive grid code was developed, and it iterates the enabling condi- Generation capacity investments need to be matched by invest- tions and clear market rules for integrating renewable energy ments in transmission and distribution infrastructure, including into the grid, while setting appropriate standards for the quality improvements to enable the grid network to integrate higher of electricity supply and services. Complementary technology shares of renewable energy supplies. Demand-side investment options and policy options are evolving to adapt to changing is expected to make up almost a quarter of total energy sector conditions relevant to solar PV, wind and hydropower, among investments to 2030, much of which will be focused on energy- others. These include smart grids and storage. Government ac- efficient technologies. These measures have been further de- tion is needed to increase access to clean energy for lighting, tailed in the National Energy Policy with a long-term perspec- cooking and heating, especially solar water heating. tive. The NEP has placed emphasis on implementing energy efficiency and conservation measures that can be applied to reduce energy consumption in all economic sectors, including through the development and enforcement of minimum energy performance standards for household and industrial equip- ment, through solar water heaters and through renewable en- ergy mini-grid/off-grid technologies. 118 KINGDOM OF ESWATINI Based on the opportunity posed by the natural gas discoveries consistency of the existing datasets as well as co-operation be- in Mozambique, there is a need to define areas of co-opera- tween key national stakeholders as well as with international or- tion and the role that natural gas can play in improving energy ganisations and experts will ensure the continued collection of security and encouraging economic development in Eswatini, meaningful data. Good energy policy is built on good data, and including in the context of the overarching objectives of the a sustained commitment to collecting adequate data is critical SADC industrialization Strategy Framework. The use of natural to meeting the country’s future energy needs. gas could help reduce heavy reliance on fuelwood as well as promote industrial development for addressing electricity se- 2. Institutional arrangement curity issues. The composition of the national working team reflected core

9.4 ENHANCING NATIONAL ENERGY PLANNING technical expertise on Eswatini’s energy sector that was com- CAPABILITY plementary and well balanced. The working team comprised experts from the Ministry of Natural Resources and Energy, This Masterplan document was drafted by a team of national Swaziland Electricity Company, Swaziland Energy Regulatory energy experts set up to carry out the national SPLAT model Authority, the Central Statistical Office and the University of development and scenario analysis, in close consultation with Swaziland. The team received training on energy statistics use the steering committee and through consultation with a wider in energy planning tools and on preparation of the Energy Mas- group of energy sector stakeholders. terplan. The training also focused on updating and maintaining the SPLAT-SW model as well as on expanding it to the entire Through the process, the working team collected the required energy system of the country (i.e., beyond the power sector). data and developed a database, created and calibrated a coun- The level of training received was adequate for the purpose, try model, developed scenarios, and analysed them. A dozen but continuous training and enhancement of expertise will be national energy experts were engaged throughout the process important. To ensure this, it is desirable that an energy planning and were equipped with the necessary skills for undertaking team is institutionalised within the government, to lead and or- robust energy planning (see Annex A for a list of the working ganise the master plan update process and further capacity en- team members, and of trainers and key contributors). Although hancement for policy analysis. this process benefited greatly from the support from IRENA and its international experts, it was truly a country-driven pro- 3. Updating/reviewing the Masterplan cess, and, as a result, ownership of this document is established fully by the Government. The Government acknowledges the The process of developing this Masterplan took over two years, importance of fully internalising the planning process and of starting in March 2016. In the final phase of this development further continuing the efforts to build and enhance the country- process, the national working team recognised that some as- owned planning process. sumptions made at the beginning of the process may require review or update. Once this Masterplan is adopted by the Cabi- This was the first such attempt for the government to develop net, the government should immediately start the process of an integral energy master plan driven by national experts, and updating the document, building on lessons learned from the the working team identified areas of possible enhancement to development process. The team recommends that the data be implemented for the next update of the Masterplan. management and model updates should be institutionalised and continuous in order to be able to deal with rapid changes in 1. Data collection the energy sector and to support decision makers with up-to- date information. It is suggested that the Energy Masterplan be The national working team developed a database of the current updated at a regular interval of, preferably, every two to three energy system and a catalogue of future scenario assumptions. years, but not at an interval of longer than five years. Although these are documented thoroughly in this document, more systematic management of the database would improve the transparency of the data. Publishing such a database, e. g., through a government web-portal, would improve the accessi- bility of data and scrutiny of the data quality. Maintenance and ENERGY MASTERPLAN 2034 119 4. Key areas for further analysis For the power sector analysis, further assessment of dispatch- ability under critical conditions (e. g., peak time) would be help- Given the importance of bagasse and molasses in the indicated ful to identify the required mitigation measures for operating energy mix, targeted consultation with the relevant industry the system with a high share of variable renewable energy. The stakeholders can be strengthened to further improve the qual- analysis should focus on a 100 % renewable energy trajectory ity of this analysis. resulting from tapping the renewable resource potential in bio- mass, solar PV, wind and hydro that is established in the Mas- The demand assessment in this Masterplan focused on the im- terplan. This analysis can guide expansion of the power sector pacts of different economic growth assumptions on energy using renewable energy technology options, systems integrity needs, but was limited in regard to exploring the impacts of a and infrastructure measures, as required to make the renewable fuller set of policy options. The potential of renewable energy energy path both technically and economically viable. deployment in the end-use sectors may require further analysis. In particular, the modelling of technology choices to assess path- Further, analysis is required of the use of untapped resources ways to achieve universal access to modern energy in the resi- such as biogas and natural gas in the residential and transport dential sector could be further strengthened by analysing effec- sectors, and for power generation in order to increase the use tive policy choices under heterogeneous economic conditions. of clean energy. The analysis can identify different pathways Policy analysis is needed related to the collection of fuelwood for crucial to most effectively directing investments in this area. cooking and heating and how the challenges related to the adop- tion of modern fuels can adversely affect households, especially During the next phase of the master plan development, more in terms of affordability since fuelwood is a “free” energy source attention could be given to the statistics on decentralised sys- whereas modern energy sources can be costly. tems. Despite vast resource potential, the deployment potential of renewable-based decentralised systems including solar PV and solar water heaters was not thoroughly evaluated. To tackle part of the further analysis required, the government will co-operate with the International Atomic Energy Agency (IAEA) for energy planning training and technical support, building on the analysis done for this Energy Masterplan. The IAEA has a long-standing track record in capacity building for energy planning in Africa, and can provide support in areas complementary to the support provided by IRENA. In particu- lar, this collaboration will serve to improve the existing energy planning models and encourage applications of an optimisation framework to a reference energy system. 120 KINGDOM OF ESWATINI ANNEX ANNEX A MEMBERS OF THE NATIONAL WORKING TEAM; TRAINERS AND KEY CONTRIBUTORS MEMBERS OF THE NATIONAL WORKING TEAM Thabile Nkosi Ministry of Natural Resources and Energy Mandla Vilakati Ministry of Natural Resources and Energy Thandazile Dlamini Ministry of Natural Resources and Energy Mzwandile Thwala Ministry of Natural Resources and Energy Thembinkosi Ndzimandze Ministry of Natural Resources and Energy Nolwazi Khumalo Ministry of Natural Resources and Energy Gugu Vilakati Ministry of Natural Resources and Energy Khulile Fakudze Swaziland Electricity Company Joseph Ncwane Swaziland Electricity Company Mzwandile Msibi Swaziland Energy Regulatory Authority Tsembayena Phakathi Central Statistical Office Thokozani Khumalo University of Swaziland TRAINERS AND KEY CONTRIBUTORS Asami Miketa IRENA Lennart Kuntze IRENA Bruno Merven IRENA consultant/University of Cape Town Dalius Tarvydas IRENA consultant/Lithuanian Energy Institute ENERGY MASTERPLAN 2034 121 ANNEX B DETAILED RESULTS OF THE DEMAND ASSESSMENT TABLE A1 ENERGY USE BY FUEL AND BY SECTOR IN 2014 (PJ) Commerce and Industry Agriculture Fuels Transport Residential Sugar Total Govt. Services Other Other Electricity – 1.2 0.4 0.8 0.9 1.3 4.6 Kerosene – 0.2 – – – – 0.2 LPG – 0.2 0.2 – – – 0.4 Oil 8.9 – – 0.1 0.4 0.0 9.4 Bagasse – – – – – 8.4 8.4 Wood – 11.9 – – – 2.7 14.5 Coal Unspecified – – 0.6 – – 1.2 1.9 Total 8.9 13.5 1.2 0.8 1.3 13.7 39.4 TABLE A2 DEMAND PROJECTION UNDER THE REFERENCE SCENARIO (PJ), 2014–2034 Fuels 2014 2020 2025 2030 2034 Electricity 4.6 5.7 7.0 8.5 9.5 Kerosene 0.2 0.1 0.1 0.0 0.0 LPG 0.4 0.6 0.7 0.9 1.0 Oil 9.4 10.5 11.7 13.0 14.1 Wood 14.5 12.1 9.5 6.0 6.4 Bagasse 8.4 9.8 11.3 13.1 14.8 Coal Unspecified 1.9 2.0 2.2 2.4 2.6 Total 39.4 40.8 42.5 43.9 48.4 TABLE A3 ENERGY USE BY FUEL AND BY SECTOR IN 2034 UNDER THE REFERENCE SCENARIO (PJ) Commerce and Industry Agriculture Fuels Transport Residential Sugar Total Govt. Services Other Other Electricity – 3.7 0.6 1.4 1.5 2.3 9.5 Kerosene – 0.0 – – – – 0.0 LPG – 0.6 0.4 – – – 1.0 Oil 13.3 – – 0.1 0.7 0.1 14.1 Wood – 1.7 – – – 4.7 6.4 Bagasse – – – – – 14.8 14.8 Coal Unspecified – – 0.5 – – 2.1 2.6 Total 13.3 6.0 1.5 1.5 2.2 24.0 48.4 122 KINGDOM OF ESWATINI TABLE A4 DEMAND PROJECTION UNDER THE HIGH GROWTH SCENARIO (PJ), 2014–2034 Fuels 2014 2020 2025 2030 2034 Electricity 4.6 6.0 8.0 10.1 12.2 Kerosene 0.2 0.1 0.0 0.0 0.0 LPG 0.4 0.6 0.8 0.9 1.0 Oil 9.4 10.5 11.8 13.3 14.7 Wood 14.5 8.0 5.7 6.2 6.7 Bagasse 8.4 9.8 11.5 13.6 15.7 Coal Unspecified 1.9 2.0 2.2 2.5 2.7 Total 39.4 37.0 39.9 46.7 53.0 TABLE A5 ENERGY USE BY FUEL AND BY SECTOR IN 2034 UNDER THE HIGH GROWTH SCENARIO (PJ) Commerce and Industry Agriculture Fuels Transport Residential Mining Sugar Total Govt. Services Other Other Electricity – 3.7 0.6 2.3 1.5 1.6 2.4 12.2 Kerosene – 0.0 – – – – – 0.0 LPG – 0.6 0.4 – – – – 1.0 Oil 13.8 – – – 0.1 0.7 0.1 14.7 Wood – 1.7 – – – – 5.0 6.7 Bagasse – – – – – – 15.7 15.7 Coal Unspecified – – 0.5 – – – 2.3 2.7 Total 13.8 6.0 1.5 2.3 1.6 2.3 25.4 53.0 ENERGY MASTERPLAN 2034 123 TABLE A6 DEMAND PROJECTION UNDER THE LOW GROWTH SCENARIO (PJ), 2014–2034 Fuels 2014 2020 2025 2030 2034 Electricity 4.6 5.4 6.3 7.3 8.2 Kerosene 0.2 0.2 0.1 0.1 0.0 LPG 0.4 0.5 0.7 0.8 0.9 Oil 9.4 10.0 10.7 11.4 12.0 Wood 14.5 12.9 11.1 8.6 6.1 Bagasse 8.4 14.9 10.0 11.0 11.8 Coal Unspecified 1.9 1.9 2.0 2.0 2.1 Total 39.4 40.1 40.8 41.1 41.1 TABLE A7 ENERGY USE BY FUEL AND BY SECTOR IN 2034 UNDER THE LOW GROWTH SCENARIO (PJ) Commerce and Industry Agriculture Fuels Transport Residential Sugar Total Govt. Services Other Other Electricity – 3.5 0.5 1.1 1.2 1.8 8.2 Kerosene – 0.0 – – – – 0.0 LPG – 0.6 0.3 – – – 0.9 Oil 11.3 – – 0.1 0.6 0.0 12.0 Wood – 2.4 – – – 3.7 6.1 Bagasse – – – – – 11.8 11.8 Coal Unspecified – – 0.4 – – 1.7 2.1 Total 11.3 6.5 1.2 1.2 1.8 19.1 41.1 124 KINGDOM OF ESWATINI ENERGY MASTERPLAN 2034 125 Ministry of Natural Resources and Energy Income Tax Building, Fourth Floor P.O. Box 57 Mbabane, Eswatini Tel: +268 2404 2644 Fax: +268 2404 4851 Website: www.gov.sz