SPECIAL FEATURE SEAR ENERGY ACCESS AND ELECTRICITY PLANNING Mark Howells, Royal Institute of Technology; Hans Holger Rogner, International Institute for Applied Systems Analysis; Dimitris Mentis, Royal Institute of Technology; and Oliver Broad, University College London b    S TAT E O F E N E R GY ACCES S R EPO RT  |  2 0 1 7 Copyright © 2017 International Bank for Reconstruction and Development / THE WORLD BANK Washington DC 20433 Telephone: +1-202-473-1000 Internet: www.worldbank.org This work is a product of the staff of the World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work and accept no responsibility for any consequence of their use. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for non- commercial purposes as long as full attribution to this work is given. Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: +1-202-522-2625; e-mail: pubrights@worldbank.org. Furthermore, the ESMAP Program Manager would appreciate receiving a copy of the publication that uses this publication for its source sent in care of the address above, or to esmap@worldbank.org Cover photo: © Malcolm Cosgrove-Davies | World Bank ENERGY ACCESS AND ELECTRICITY PLANNING Mark Howells , Royal Institute of Technology; Hans Holger Rogner, International Institute for Applied Systems Analysis; Dimitris Mentis, Royal Institute of Technology; and Oliver Broad, University College London INTRODUCTION A s developing countries look for ways to achieve the existing capacity stock is scheduled for retirement by sustainable energy services, which is essential to 2040 (IEA, 2015)—and the timely integration of new and lift people out of poverty, the big challenge cen- renewable energy sources into existing infrastructures. But ters around providing access for all while avoiding past investment decisions are clouded by demand uncertainty pitfalls without creating new ones. The reality is that this due to ongoing efficiency improvements in end-use sec- can only occur if there is a fundamental transformation of tors, the emergence of smart grids, and potential develop- energy systems along the entire set of resource to energy ments of new electricity markets (such as electric vehicles). service chains—and that will necessitate greater energy Competitive and private sector dominated energy markets efficiency and a bigger role for renewables in the global rely on clear and consistent government energy-environ- energy mix energy. Moreover it must occur at a time when ment policies to align their investment decisions with sus- projected global electricity demand calls for installing tainable development objectives. After all, energy system some 6.7 Terawatt (TW) of new electricity generating transformation is largely a capital-intensive affair that can capacity worth an aggregate investment of $20 trillion conflict with short-term profit maximization. from 2015 to 2040 (IEA, 2015). The beneficial role of access to energy for socio-eco- Clearly, this is a tall order, especially given that modern nomic development has been acknowledged since the energy systems are highly complex and capital intensive, onset of industrialization more than two centuries ago. But constantly interacting with many other sectors like the there are no roses without thorns—the thorns of energy environment, natural resource systems, and infrastructure. access are numerous: (i) scarred landscapes caused by This means that countries will have to undertake compre- mining activities; (ii) land-use change from fuel wood pro- hensive and systematic analyses and planning to identify duction; (iii) pollution emissions from fossil fuel combus- and avoid (or at least minimize) expensive stop-gap mea- tion that are chiefly responsible for adverse impacts on sures and long-term “lock-in” into inadequate and unsus- human health, environmental degradation, and climate tainable infrastructures. In many instances, short-term instability; and (iv) energy security concerns and interna- pressure for immediate action will take precedence over tional conflicts about the very issue of access to energy. In long-term consideration for sustainability. essence, sustainable energy avoids or minimizes these In practice, comprehensive energy planning at the adverse side effects of energy access. Past energy transi- national, regional, or local levels is further complicated tions from wood to coal to oil were often meant to miti- because there is no one size fits all energy system and gate some of these consequences, only to cause new and priorities vary sharply. In developing countries, access to potentially worse impacts aggravated by a seemingly affordable energy services is primarily a priority for rural ever-increasing demand for energy fueling economic areas to combat energy poverty, but increasingly also in growth—the blue print for unsustainability. the large metropolitan areas as urbanization accelerates. This paper tries to shed light on how developing coun- These countries have 2 billion people without electricity tries can carry out energy planning by reviewing the avail- and nearly 3 billion people relying on dirty fuels (such as able methodologies and tools, including their potential to firewood and animal dung) for cooking and heating. At integrate rural energy access and encourage the uptake of the regional level, nearly 90 percent of people suffering renewable energy technologies. It also probes how invest- from energy poverty reside in South Asia and Sub-Saharan ment needs and cost-effectiveness are reflected in differ- Africa (Bazilian, 2015). And recent projections for Sub-Sa- ent analytic and planning tools—with a case study on haran Africa indicate an increase in energy demand of 80 Ethiopia. And it examines the interaction of energy plan- percent and a fourfold expansion of electricity generating ning and scenario development and how these are capacity by 2040 to improve the socio-economic welfare applied to informed policy making. The findings suggest of a population twice as large as today’s (IEA, 2014). that energy planning is essential and feasible. However, In contrast, the developed countries of Europe, North support is required to improve data collection and access, America, and Asia struggle with the replacement of aging develop open accessible modelling tools, and build sus- plant and equipment—for electricity, some 40 percent of tainable national capacity to undertake planning.   1  2    S TAT E O F E L E C T RI CI TY ACCES S R EPO RT  |  2 0 1 7 PLANNING FOR ELECTRICITY ACCESS functional institutional framework to ensure the availability of funding, the timely readiness of the many pieces What exactly do we mean by electricity planning? It is the needed for energy infrastructure investments, and a act of assessing the ability of a regional system to provide mechanism to oversee progress and control quality. In this dependable energy services under constantly changing context one should note that: conditions—which involves variables such as the cost of materials and fuels, investment costs in technologies, Sound project economics mobilizes the necessary demand levels, and distribution. Drawing on the field of finance. This is particularly the case for large infrastructure operations research, planning applies advanced analytical investments (Goodman and Hastak, 2006). Costs and cash- methods and tools to make better decisions when faced flow streams must be established and mapped to national with complex decisions. This activity is inherently iterative budgetary, extra-budgetary, and external funding. The due to the fast, and potentially drastic, transformations matching provides insights for estimates of investment that can take place over very limited periods of time (IAEA requirements and operating and maintenance costs over 1984). Developed as a way of mitigating the impacts of the project’s life-cycle. It specifies projected costs to con- external events on the ability of a system to provide its sumers, expected revenues, and subsidies. It quantifies specific service, this process typically identifies the most potential implementation barriers resulting from budget- cost effective way of delivering energy to the final con- ary or financial constraints. And, depending on condition- sumer over time (Wilson, R. & Biewald, B., 2013). alities, it can help leverage funding from foreign and Of course, this process takes on different meanings in international sources (Onyeji et al., 2012). Examples of different parts of the world—especially in developing such funding include foreign direct investment (FDI), the countries’ poor rural areas and burgeoning megacities, Clean Development Mechanism (CDM), and financing where electricity access is a challenge in itself. Fortunately, from international development banks. quantitative energy modeling (using mathematically coded images of current and future energy needs), which Physical deployment of infrastructure needs to match is increasingly being used by industrial countries offers a schedule logistics. The introduction of a large hydro- promising tool (see box). The main barriers for developing power plant serving initially a small market may exceed countries are the lack of adequate data and a shortage of current electricity demand. Thus it would have to gener- skilled human resources to perform the analysis. Thus, ate the needed stream of revenue to cover generating investment decisions are often based on ill-informed pol- costs for many years, making it difficult to pay dividends icy targets and the need for ad-hoc stop-gap measures. and service debt. Then, again, shortages result when This means that targets or measures tend to focus on electricity demand grows faster than supply, prompting cheap and quick to build (often popular) technologies or stop-gap measures and delaying economic develop- emergency fuel purchases, resulting in high operating and ment. Moreover, connecting millions of new customers environmental costs and expensive end-use services. presents a formidable logistic challenge that requires Given that such actions serve the supply shortfalls of coordinating the timely and consistently staged deploy- already connected consumers, increasing access is rarely ment of generation, transmission, and distribution, along part of the strategy. with developing complementary chains for system main- Moreover, energy planning is not an end in itself and tenance and consumer services. Failing to map these involves more than mastering energy modeling tools. dynamically evolving factors may lead to severe demand Planning without subsequent implementation is an inef- and supply mismatches, thereby impeding the energy fective use of resources. Plus, implementation needs a system’s effective expansion. BOX 1 Challenges for Energy Access in a Warmer World Quantified models of complex systems help decision tal institutions can structure tariffs to minimize con- makers in numerous ways. From a technical perspec- sumer electricity bills. It also ensures that scenarios of tive, they enable analysts to compare different sys- future energy system developments are internally tem configurations without incurring the upfront cost consistent. Such scenarios can serve as effective com- of actually building them, which helps to mitigate munication tools for non-partisan political commit- uncertainty. From a practical perspective, they facili- ment—which will help both to garner and mobilize tate the design of systems in a way that accounts for private sector support and to solicit agreement and local resources, demands, and constraints that are feedback from society at large. And for developing placed upon real life electricity systems. This ensures countries, even minor system improvements often that generation meets demand in the most cost have disproportionally high positive economic and effective way and that public utilities and governmen- environmental returns. ENERGY ACCESS AND ELECT RICIT Y P L A NNING  3  Functional institutions are required to support the good energy statistics and well educated energy analysts. implementation and operation of an expanding energy As a result, these tools are, for the most part, ill adapted system. These include market regulators, system opera- to applications in developing countries. Adapting models tors, vendors, environmental protection agencies, line min- to the data situation in developing countries is not only istries, educational institutions, and public-private sector aggravated by the tools’ data intensities but also by a lack partnerships (Bazilian et al., 2011). Jointly these institutions of options for simplified yet meaningful reduced-form are tasked with developing milestones for implementation model setups. Proprietary codes, “closed-source” fea- (IAEA, 2009) and inter-ministerial coordination tures, and often poor documentation make their simplifi- cation difficult (Howells et al., 2011). Recent simplified open source models—such as OSeMOSYS (Howells et al, TOOLS AND METHODOLOGIES 2011), Temoa (Hunter et al, 2013), and SWITCH (i4e, n.d.) have a stripped down code base allowing analysts to add The key to such an enormous undertaking as energy plan- analytical features. This reduces barriers to entry for new ning is the availability of energy planning toolkits, which, at analysts and developing country practitioners (DeCarolis least in their contemporary form, date back to the oil crises et al., 2012 and Howells et al., 2011), although it limits of the 1970s, when governments and industries were their off-the-shelf analytical functionality. caught by surprise by an unforeseen and unprecedented Given the evolving nature of energy systems, a contin- curtailment of global oil supplies. Initially these models uous advancement of the modeling tools is inevitable. focused on sector specific issues such as ensuring a suffi- Delineating energy access or poverty in mathematical cient supply of oil or expanding electricity generating terms (Nussbaumer et al., 2013), costing energy access capacity. But they were later expanded to also account for (Fuso Nerini et al., 2014), and developing open access energy-economy-environment interlinkages or externali- energy planning tools (Bazilian et al., 2012) are new and ties (see figure 1)—leading to a vastly improved and expanding fields. The challenge is to align new model fea- increasingly inclusive energy systems planning process tures such as energy access with the moving “goal posts” (Bhattacharyya, C., and Timilsina, G.R., 2010). of energy systems that become more dynamic (Bazilian et Models and toolkits were primarily developed by, and al., 2013) and integrated well beyond the traditional for use in, the industrialized countries of the OECD and energy system boundaries (Howells, M., and Rogner, the former CMEA (or Comecon), which had relatively H.-H., 2014). FIGURE 1  A framework for comprehensive energy planning Social and Economic Perspective Assessment of Identification of natural resources Assessment of technology choices energy needs Exogenous Regional trade of assumptions electricity and fuels Analysis of energy supply options Environmental Financial and burdens and other resource mitigation requirements Sustainable supply strategy Source: Rogner, 2011 4    S TAT E O F E L E C T RI CI TY ACCES S R EPO RT  |  2 0 1 7 Energy systems models energy-related greenhouse gas (GHG) emissions and Electricity systems models are tools used by electricity ana- land-use changes, along with their impacts on climate lysts (such as engineers, economists, and planners) to man- change (and vice versa). Integrated Assessment Models age and plan the electricity system, trade electricity, and (IAMs)1 account for elements that cross boundaries of expand generation capacity (Foley et al. 2010). Here, we different domains (especially between energy, atmo- focus on three types. sphere, oceans, and land-use). • Macroeconomic or “top-down” models. These include Focusing on bottom-up models, which are most relevant econometric, input-output, and computable general for identifying investment requirements in developing equilibrium approaches (Welsch 2013). They are driven countries, they are chiefly characterized by their temporal by projected developments of major economic indica- scope (figure 2)—which can range from maintaining volt- tors, using prices to balance demand and supply within ages (stability analysis) to the operation and dispatch of the energy sector as well as the rest of the economy. electricity (load flow and market power analyses) to invest- They provide insights on broad relationships between ment requirements (long-term energy system analysis). economic development and the associated energy de- Models focusing on AC or DC load flow analysis may serve mand and supply. And they may or may not (but usually to investigate various grid configurations. Such models do not) include details on technology. may cover timeframes of hours or years. Based on the derived transmission capacities, steady-state, transient, or • “Bottom-up models. These are largely technology dynamic stability analyses may serve to assess distur- driven—accounting for the physical configuration of bances in power systems. Stability analyses usually cover the energy system’s technologies and infrastructures, timeframes of up to several minutes. They may provide their vintage situation, energy efficiency, and econom- important insights on the design of the components of the ic and environmental performance. They are driven by transmission and distribution system. the usually hard-wired requirement that supply has to meet demand, which is externally determined (possi- Geographic information system models bly by a macroeconomic model). The most popular Cost-effective electricity supply systems serving rural ones are optimization models, which identify optimal households and businesses are diverse and site specific, pathways for meeting demand; optimal can be least meaning that the cost-optimal technology choice cost, highest level of energy security, or fastest access depends on several parameters. These can be geophysi- to energy services. cal, technical, economic, or social—such as local popula- • Hybrid models. These incorporate aspects of both bot- tion density, distance to the grid, fuel costs, and electricity tom-up and top-down models, and are either integrat- usage, respectively—and many are strongly spatial in ed or “soft-linked” (data is explicitly transferred be- nature (like wind regimes, potential micro-hydropower tween two stand-alone tools in an iterative manner). sites, settlement positions, and grid expansion). Develop- Linking and integrating models from different scientific ing a clear transparent approach to capturing these disciplines has become a necessity for understanding parameters and translating them into potential technol- FIGURE 2  The temporal scope of power sector models Assessed year(s) Now 2020 2030 2050 2100 Stability analysis (secs–days) Short-term e.g. for various faults In fo Design Time frame covered within model rm at Load flow analysis io n for various iterations of grid configuration flo w (days–years) Mid-term Power market analysis considering bidding areas and interconnectors Srtategy Long-term (decades) Long-term energy system analysis for multiple connected aggregated systems Source: Welsch, 2013 ENERGY ACCESS AND ELECT RICIT Y P L A NNING  5  ogy suites to meet energy access goals, is crucial to DEMAND FORECASTING FOR POOR informing effective policy. COUNTRIES Geographical Information System (GIS) models respond to this need by enabling the analyst to assess the cost of Of course, a critical element in these models is how much electricity provision at each specific location in a given energy will be demanded. Typically, electrification efforts area. By combining detailed geo-referenced layers of data that focus on connecting new households are driven by for each relevant parameter, site specific investment needs policy, while for commercial users, the driver is economics and energy cost implications of competing technological (Gaunt, 2003). systems can be compared in space and time. Starting at the household level, where there are rela- The use of GIS-based analyses has increased since the tively high connection and distribution costs, demand pro- mid-1990s with a clear focus on using levelized energy jections for electrification efforts are typically engineered cost2 (that is, the breakeven cost) for choosing the appro- as a function of a limited number of key parameters (such priate technology. The value of the geo-referenced ap- as population density, location, and governmental targets proach in these situations lies in its ability to combine com- for energy access). For example, the Network Planner Tool prehensive information relating to site specific technologi- uses parameters like energy intensity per rural and urban cal information to in depth regional resource availability household, projected population growth, and economic data thus assessing an “integration of all the possibilities” demand elasticity to derive geo-spatial demands. Similar for electrification (Amador and Domínguez, 2005). end use accounting (Bhattacharyya and Timilsina, 2009) Further, the application of such tools to remote areas, techniques to derive rural and urban household demand where information is scarce, enables and supports analy- are employed by the popular LEAP (Heaps, 2014) and ses that could otherwise not take place. The use of remote MAED (IAEA, 2006) models (which do not provide any sensing data and technologies, combined with the inter- geo-spatial information). polation capabilities of GIS models can, when applied to At the country level, a variety of methods are used to macro-economic and statistical data for a given area, project demand for total electricity sales (which often “answer some of the key questions” relating to energy include little or no explicit geo-spatial consideration). (Bhat- planning and rural electrification (Szabó et al., 2011). Take tacharyya and Timilsina, 2009) group these approaches into the following examples: end use accounting and econometric. An end use account- In some Sub-Saharan African countries, the Network ing approach may begin with exogenous and detailed Planner model is used to compare the implications of economic projections that are delineated by economic either extending the national grid, rolling out solar PV subsector, with energy use split between thermal and household systems supplemented by diesel generators other requirement, and assumptions made about how for productive uses, or opting for low voltage diesel-based that energy intensity will change over time. These are then mini-grid systems (Kemausuor et al., 2014). In the case combined with potential fuel substitution for thermal where grid extension is the cost optimal solution for a requirements to make projections by sector and fuel. Sim- given location, the tool assigns each settlement an eco- ilar, but more flexible approaches are available in LEAP. nomic radius that the grid would reach from such a start- Less complex approaches include simple econometric ing point. Using a modified version of Kruskal’s minimum regressions with population and economic growth. Other spanning tree algorithm, it then connects the locations approaches account for feedback between the cost and using the least additional kilometers of additional grid configuration of the energy sector with the demand for its (Parshall et al., 2009). fuels. For example, fuel price elasticities by sector may be In Nigeria and Ethiopia, the ONSSET electrification accounted for directly (Loulou and Lavigne, 1996) or via tool is used to develop a cost model for comparing the broader macroeconomic feedback (Howells et al., 2010; levelized cost of electricity generation of grid extension Winkler et al., 2007). with mini-grid and off-grid diesel-based and renewable In addition, metrics have been developed to support options (Fuso Nerini et al., 2015). It generates a set of governmental goals for energy access and inform demand boundary conditions that inform a GIS related algorithm projections. The World Bank’s Multi-Tier approach (World assessing grid compatibility of all non-connected settle- Bank, 2015) determines tiers of energy intensity per house- ments and, in case of negative outcomes, selecting the hold, with each tier associated with different levels of elec- most cost effective mini-grid or household level solution tricity use—ranging from (at the lowest level) lighting and (Mentis et al., 2015) cooking to (at the highest level) services that provide com- In northern Brazil, GIS analysis is used to answer ques- fort (such as air-conditioning). Other metrics account for a tions of renewable energy management in semi-arid rural broader range of parameters involving other aspects of areas (Tiba et al., 2010). By crossing a variety of data banks energy poverty and development, such as the Multi-dimen- relating to (i) raw data for infrastructure, resource, and sional Energy Poverty Index (MEPI) (Nussbaumer et al., socio-economic parameters and (ii) technological data for 2013) and IEA Energy Development Index EDI (IEA, n.d.). solar power, water pumping requirements and other renewable energy systems, the study generates a repre- CASE STUDY: ETHIOPIA sentation of “the best localities for inclusion of a deter- mined renewable energy technology.” To better understand how these tools work in practice, we explore what would need to happen in Ethiopia to provide 6    S TAT E O F E L E C T RI CI TY ACCES S R EPO RT  |  2 0 1 7 better electricity access and services in a cost-effective lower targets, unit costs are higher. Note that costs near manner. We use two tools: (i) the ONSSET –GIS-based tool the grid in urban areas remain unchanged, following their for rural electrification to determine the cost optimal way constant electrification target. of providing high levels of electricity access; and (ii) the What would happen if electricity costs increase where OSeMOSYS tool to determine the cost optimal way of there is no systematic deployment of solar and mini-grids? expanding grid-based bulk generation. The combination As figure 5 (panel A) shows, if the grid is not extended and of these two tools forms a consistent approach to minimiz- users only have access to diesel generators, electricity costs ing the cost of electrification (Bekker et al., 2008) while are high. But if the PV market becomes more fluid, or the concurrently meeting the economics of supplying bulk government helps facilitate investment, the cost of rural quantities of low cost, reliable electricity. Current per cap- electrification drops significantly (Figure 5, panel B). This ita electricity consumption in Ethiopia is as low as about occurs because the deployment of PV stand-alone solu- 50kWh—compared to 13,200kWh in the United States and tions decreases the levelized cost of electricity in some set- 1,750kWh in neighboring Egypt (World Bank, 2014). tlements as compared to just diesel stand-alone options. PV stand-alone technology would be more viable than die- Providing high levels of electricity access sel stand alone for 22,624,921 people (or 32 percent of the We begin by considering the least cost configuration of population that needs to be electrified). If grid extension grid, micro-grid, and stand-alone technologies to meet and mini-grid technologies were to contribute to the elec- two rural (50 and 150 kWh/capita/year) and one urban trification mix of the country, only 656,767 people would be electrification target (300 kWh/capita/year). As figure 3 electrified by stand-alone systems (diesel, PV).) shows, a higher target results in the deployment of grid Thus, an optimal deployment strategy would include and mini-grid systems, with remote and low density pop- extra grid extension and the deployment of micro-grids— ulations relying on stand-alone electrification. The change information that could be used to support better poli- in technology from high to low is indicated in table 1, with cy-making. And knowing the cost optimal deployment a noticeably large shift to stand alone systems. characteristics could be used to develop specific poli- Underlying the shift in technology is how the cost of cies—ranging from state-led deployment to facilitation of electricity. Figure 4 indicates how the levelized cost of market development. At this point, Ethiopia is undergoing supply on a geo-spatial basis changes in response to the rapid expansion in its generation capacity. Consistent with higher and lower supply targets. With higher levels of pro- the most recent eastern African power pool development vision, the cost per unit is reduced in rural areas. With plan (EAPP/EAC, 2011), the power system grew by 20 per- FIGURE 3 Optimal electrification mix in Ethiopia A. Higher target B. Lower target Source: Author’s calculation based on Mentis et al 2016 b. TABLE 1  Optimal split for new connections (Population-based for different rural electrification targets) SPLIT POPULATION (150/300) POPULATION (50/300) CHANGE Grid 65,431,650 62,270,395 ↘ –4.8% Mini Grid 3,958,695 245,825 ↘ –93.8% Stand Alone 656,767 7,530,892 ↗ 1046.7% Source: Authors’ calculations ENERGY ACCESS AND ELECT RICIT Y P L A NNING  7  FIGURE 4 Higher levels of provision mean lower per unit rural area costs Spatial levelized cost of electricity A. Higher levels of provision B. Lower levels of provision Source: Author’s calculation based on Mentis et al 2016 b. FIGURE 5 A case for more grids and PV solar (Spatial levelized cost of electricity for the electricity access targets 150–300 kWh/capita/year A. Grid and stand alone diesel B. Grid, stand alone diesel and solar PV Source: Author’s calculation based on Mentis et al 2016 b. Note: Left panel: Population already connected to the grid is grid connected and the rest are electrified by stand-alone diesel. Right panel: Population already connected to the grid is grid connected and the rest are electrified by stand-alone diesel and PV solar. cent between 2013 and 2016, increasing by over 4.7GW. to form the foundation for Ethiopia’s electricity system One baseline projection (WB) of electricity growth is (Taliotis et al 2016), although recent analysis (IRENA, around 5 percent per year. 2014) also indicates relatively high potentials of non- hydropower renewables available. Plus there are limited Pinpointing the lowest cost route for grid reserves of crude oil and larger quantities of natural gas. expansion The model assumes that newly electrified households To determine the lowest cost expansion of the grid-based meet their demand target of 150kWh per capita in rural electricity system, we use the Open Source energy Model- areas and 300kWh per capita in urban areas. ing System (OSeMOSYS)—which is driven by demand for Our results show that generation investment is dom- “grid” electricity resulting from the ONSSET analysis, as inated by hydropower (Figure 6 panel A), with large well as a national projection of other (bulk) demand growth quantities used for export—although there are signifi- (based on GDP projections). It captures potential candidate cant new investments in capacity required for electrifica- power plants, fuel costs, and resource availability (fossil and tion (indicated hashed lines in figure 6 panel B). But if renewable) to calibrate the model cost and performance trade in Africa is to reach its cost optimal potential, Ethi- data relating to existing power plants and their retirement opia will need to join a number of countries that gener- schedule. A cost optimal system is then calculated (Howells ate significant quantities of electricity for export by 2030 et al 2011). On the resource front, hydropower is expected (figure 6 panel C) (Taliotis et al., 2016). 8    S TAT E O F E L E C T RI CI TY ACCES S R EPO RT  |  2 0 1 7 FIGURE 6 Hydropower will dominate in Ethiopia Generation mix Total capacity 80 16 14 60 12 Capacity (GW) 40 10 Generation (Twh) 8 20 6 0 4 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2 –20 0 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 –40 Generation mix per country, 2030 (%) 100 80 60 40 20 0 Central African Rep. Gambia Djibouti Eritrea Algeria Benin Burkina Faso Burundi Chad Congo Guinea Kenya Liberia Rwanda Somalia South Africa Sudan Swaziland Togo Zambia Zimbabwe Morocco Guinea-Bissau Libya Uganda Gabon Malawi Nigeria Ghana Equatorial Guinea Mauritania –20 Côte d’Ivoire Egypt Tunisia Niger Cameroon Sierra Leone Mali Tanzania, United Rep. of Senegal Mozambique –40 Ethiopia Congo, the Dem. Rep. of the Angola –60 Botswana –80 Namibia Lesotho Coal HFO Wind Hydro Nuclear Dist. diesel Net imports Diesel Gas Solar Geothermal Biomass Dist. solar Electrification Source: Taliotis et al., 2016 and author’s calculation based on Mentis et al 2016 b. INVESTMENT NEEDS AND SOURCES OF Once investment needs are clearly delineated, a variety of FUNDING options are available to raise and channel funds—at the international, bilateral, and national levels, involving the Between 2010 and 2030, an investment of $12 billion to public and private sectors and nongovernmental organiza- 279 billion per year will be needed to provide energy tions (NGOs). access for all (Bazilian et. al. 2014). These values vary sig- National and local governments play a key role in cre- nificantly as a function of electricity demand per capita. ating an enabling environment for private energy invest- Locally, the scale of investment depends on a variety of ments based on clear legislation, plans, policy, and parameters, such as geography, resource endowment, and effective institutions. The extent to which these will be investment risk. Recent studies estimate that the cost of called into play can be derived from planning, indicating electrification can range from less than $100 per house- the market size, scale of investment, and operation. More hold for basic access solutions, to about $7,000 per house- direct intervention is, in no small part, dependent on the hold for high targets of access (Fuso Nerini et al., 2014) . type of technology and investment that needs to be made. ENERGY ACCESS AND ELECT RICIT Y P L A NNING  9  For example, if solar panels are to play a key role, they it may crowd out funding needed for other purposes. In might be sourced from a producer country—which might the longer run, high upfront investments may result in low supply soft loans or other subsidies to secure the sale and running, and therefore low electricity prices. Low prices promote development. may reduce the cost of production factors and boost eco- At the international level, financing might be forthcom- nomic growth. The energy sector is often the most ing if expanding the system results in better environmental important contributor to GHG emissions. Yet environ- performance compared to “business as usual.” Perhaps ment ministries may be called on to communicate GHG the expanded system reduces carbon dioxide, by replacing projections and mitigation targets. fossil fuels. If so, it may be eligible for financing support The need for integration becomes particularly clear through global facilities such as the Global Environmental where there are physical links among the critical resource Forum (GEF) or the Nationally Appropriate Mitigation systems. The term nexus3 is increasingly being used to Action (NAMA) facility. Or perhaps it will reduce household describe the interlinked nature of resource systems. The woodfuel consumption in homes and thus demonstrably importance of interlinkages in the supply chains that pro- reduce deforestation. If so, REDD (Reducing Emissions vide water, energy, and food has been raised by the IAEA from Deforestation and forest Degradation) funding might (2009), among others, emphasising prudent integrated be available. Should transparent development progress be management of climate, land-, energy- and water-use demonstrated (which the planning will help establish), a (CLEW) strategies. As gains may be had with increased inte- variety of tied or untied aid and trade potential might be gration in these supply chains, it is argued that there is a unlocked through bilateral, regional, or global agreements. clear need to develop quantitative frameworks to support The public sector can directly support electrification future sustainability policies (Howells and Rogner (2014). and power system expansion via several routes. Common ones include: (i) fund raising through increased tariffs, cross Links to climate, water, land, and other resource subsidies, or taxes; (ii) providing tax credits; (iii) providing planning secure power or fuel purchase agreements; (iv) providing Planning approaches have been developed to study and grants or direct payments; (v) disseminating information develop policy for resource management. However, these and building capacity; and (vi) supporting the develop- approaches can be lacking—especially when different ment of supply chains. Again the type of measure will be a resource systems are tightly interwoven (UN, 2014). Exist- function of the technology choice. For example, a hydro- ing approaches typically examine future development sce- power plant may be supported by a power purchase narios of one sector, with little account of consistent and agreement. But a coal fired power plant may need facilita- concurrent scenarios of other sectors. Often termed “inte- tion with respect to granting mining rights, a fuel supply grated,”4 such planning processes make inter-sector link- agreement, and a power purchase agreement. The power ages explicit, but they do not necessarily look beyond purchase agreements may also vary. If climate change is those. Resource planning approaches typically assume expected to affect water flows, hydropower sales may that the related sectors are static, or that their develop- need to tailor the agreements to mitigate associated risks ment is not fundamentally changed by the primary sector (which can be uncovered in extended energy modeling being considered. This can result in important feedbacks exercises) (WB, 2015). being ignored or overlooked (M. Howells et al., 2013a). The right enabling environment can be used to secure For example, a drying climate change may drive up energy funding through local NGOs, with less “hands on” policy prices at a time when energy needs become amplified. intervention—or even no intervention at all, except for Unless considered concurrently under the same scenario information dissemination. For instance, cooperatives drivers, such a negative and reinforcing situation may go have been used to set up community owned and operated undetected. Efforts to overcome these methodological local energy systems. Additionally, local NGOs can sup- shortfalls are beginning to be made at a policy level, nota- port local electrification projects and help create enabling bly EU strategic environmental assessments (EU, n.d.). environments for energy access at the community level.  While in their infancy, nexus studies have started to zoom in on exploring different geographical scales: from global (United Nations, 2014) to regional (Smajgl and POWER PLANNING AND POLICY Ward, 2013; UNECE, 2014)) and national (Hermann et al., COHERENCE 2012; M. Howells et al., 2013b; Macknick et al., 2012; Sat- Given that energy needs to be both socially acceptable tler et al., 2012). At the sub-national level, Bartos and Ches- and environmentally compliant, it is vital that energy pol- ter (2014) illustrate missed opportunities in the United icy is integrated into broader development strategies. States from the lack of formal integration of the water and Consider, for example, expanding education to the rural energy service infrastructure in Arizona. In Sub-Saharan poor in a least developed country. This will require not Africa, a major World Bank study (WB, 2015) that covers only building schools but also supplying electricity. But over 40 countries combines agricultural, hydrological, cli- expanding the energy sector can be capital intensive; the mate, and energy modelling to assess the interference and demands for its fuels may be inelastic and taxable; and climate vulnerability of each. This allowed for not only a there might be a need for significant imports or produce resource consistent approach but also a regionally consis- exports. As a result, fiscal and economic policy would tent analysis. And it was delivered by a small team using need to be consistent with the needs of the energy sec- open tools5 over a relatively short period—which bodes tor. If expenditure on investments is too high, for instance, well for making such approaches easily available. 10    S TAT E O F E L E CTR I CI TY ACCES S R EPO RT  |  2 0 17 Role in economic, social, and regional development This is particularly the case with changes to capital strategies intensive energy investments or radical policy reforms, Energy provision to the poor is vital. Consider the typical which will likely require more capacity development to least developed country, which is highly dependent on obtain the needed skills to manage the new market situa- agriculture. Electricity is needed to support enabling basic tion. Take the case of reforms coupled with multinational activities such as lighting and powering ICT devices, and operations within a larger regional market that also involve mechanical power and heat are needed to farm and treat the entry of private players (AfDB, 2013). Such skills would crops. In addition, there is a strong correlation between include: (i) developing energy balances and projections; electricity use per capita and human development, with (ii) configuring energy supply and trade scenarios; (iii) esti- countries with higher per capita use ranking higher on the mating financing as well as the institutional support; and human development index. Studies in rural states of India (iv) understanding broader metrics for sustainable devel- have pointed out the potentials for increased literacy opment and cross-sector impacts ((IAEA, n.d.)), (Howells relating to electricity access (Kanagawa and Nakata, and Roehrl, 2012)). 2008). Others have theorized that such relations are best Capacity building should be seen as a long-term exer- described as saturation phenomena where we observe a cise—in effect, an investment project with limited imme- steep rise in human development relative to energy diate, higher long-term pay-offs. According to the African demand for energy-poor nations, a moderate rise for tran- Development Bank (AfDB, 2013), “in order to ensure sus- sitioning nations, and essentially no rise in human devel- tainability, capacity building should be migrated over opment for energy-advantaged nations (Martínez and time to the Centres of Excellence, tertiary institutions, Ebenhack, 2008). and the utility affiliated academies of learning.” Ideally At the regional level, extending modeling to look this should build and regional networks of experts, train- across borders has become important to assess integra- ers and trainees. tion, trade, and security issues. At the global level, extend- ing models to analyze the potential development of power pools is also occurring. A power pool coordination CONCLUSION program may use a computer generated power pool Thus, energy planning is feasible and essential. There is a “master plan” to help understand how best to mobilize strong imperative to calculate quantitative and internally resources (SAPP, 2009) or the role of technology deploy- consistent scenarios of a country’s energy sector develop- ment where resource rich countries may supply others ment to understand the needed institutional, incentive, (IRENA, Forthcoming). Other examples include continen- technological, and financial requirements. This information tal efforts, such as the Program for Infrastructure Develop- can also be used to: (i) help harmonize policies, (ii) plan ment for Africa (SOFRECO, 2011). across different resources systems, and (iii) inform energy and technology trade. Scenarios are developed with mod- els, and given the strategic nature of the energy sector, CAPACITY BUILDING FOR PEOPLE AND models and the capacity to run them are required for both INSTITUTIONS FOR PLANNING regions and countries. Without in-depth national energy planning capacity, the At present there exist a useful, but limited set of poorest will always be at the mercy of big industry and accessible cases studies and open toolkits. This means relying on the goodwill of the international and donor that support efforts should focus on assisting data collec- community (Rogner, 2011)—especially given that no one tion, contributing to open modeling development, and size fits all with respect to the energy system. Local condi- building human capacity to analyze the energy sector. tions are unique, physically, politically, economically, and Such capacity building should include, but go beyond, socially. Thus local capacity to develop, run, analyze, and technocrats. It should also include the establishment of interpret model results is crucial. Without the capacity to centers of excellence, tertiary education, and networks of undertake this cornerstone of energy policy development, experts—the latter will be needed to ensure that the national development strategies may be ill informed with planning process is sustainable and will continue after unwanted consequences. short-term assistance ends. NOTES 1. Notable integrated assessment models include: DICE (Dynamic Integrated Climate-Economy), RICE (Regional DICE), MERGE (Model for Estimating the Regional and Global Effects of greenhouse gas reductions), MESSAGE-MACRO, IMAGE (Integrated Model to Assess the Greenhouse Effect), IMAGE/TIMER (Targets IMage Energy Regional), MiniCAM (Mini Climate Assessment Model), GCAM (Global Change Assessment Model), WITCH (a World Induced Technical Change Hybrid System), DNE21 (Dynamic New Earth 21), MIND, ReMIND (Regional Model of Investments and Development), AIM/CGE (Asian Pacific Integrated Model). (Després et al., 2015) 2. The levelized cost is the total costs (including capital investments, operating costs, and financing costs) divided by the total energy output over the lifetime of the system. 3. Nexus refers to the interplay and interconnections among different societal or natural systems or resources. Most commonly, it covers water, energy, and food, but it can also involve security, eco-systems, climate, sanitation, health, and gender (see for instance Beck and Walker, 2013; UNECE, 2014) 4. Examples include Integrated Water Resource Management (IWRM), Integrated Energy Planning (IEP), Integrated Land-use Assessment (ILUA), etc. 5. For the electricity modelling OSeMOSYS was employed. ENERGY ACCESS AND ELECT RICIT Y PLANNING  11  REFERENCES AfDB, A.D.B., 2013. Energy Sector Capacity Building Després, J., Hadjsaid, N., Criqui, P., Noirot, I., 2015. Diagnostic & Needs Assessment Study. Modelling the impacts of variable renewable sources on the power sector: Reconsidering the typology of energy Amador, J., Domínguez, J., Application of geographical modelling tools. Energy 80, 486–495. doi:10.1016/j. information systems to rural electrification with renewable energy.2014.12.005 energy sources, Renewable Energy, Volume 30, Issue 12, October 2005, Pages 1897-1912, ISSN 0960-1481, http:// Després, J., Hadjsaid, N., Criqui, P., Noirot, I., Modelling the dx.doi.org/10.1016/j.renene.2004.12.007. impacts of variable renewable sources on the power sector: Reconsidering the typology of energy modelling Angel, S., J. Parent, D. L. Civco & A. M. Blei, 2010. Atlas of tools, Energy, Volume 80, 1 February 2015, Pages Urban Expansion. Available at http://www.alnap.org/ 486-495, ISSN 0360-5442, http://dx.doi.org/10.1016/j. resource/6808.aspx energy.2014.12.005. Bartos, M.D., Chester, M.V., 2014. The Conservation Nexus: Energyplan, 2015. Advanced energy system analysis Valuing Interdependent Water and Energy Savings in computer model [Online]. Available at: http://www. Arizona. Environ. Sci. Technol. 48, 2139–2149. energyplan.eu/markaltimes/ [Accessed May 12 2015] doi:10.1021/es4033343 EU, E.U., n.d. EU Strategic Environmental Assessment Baser, H., 2011. Special Report: Perspectives on Capacity Directive [WWW Document]. Eur-Lex—Rech. Simple. URL Development in Fragile Situations. Ottawa, Canada. http://eur-lex.europa.eu/LexUriServ/LexUriServ. Available at http://www.oecd.org/dac/governance-peace/ do?uri=CELEX:32001L0042:EN:NOT (accessed 2.25.14). governance/docs/48311487.pdf FAO. “Energy-Smart Food For People And Climate.” Issue Bazilian, M., 2015. Power to the Poor. Foreign Affairs [Online] paper, Food and Agricultural Organization, 2011. March/April 2015 Issue, Available at: https://www. foreignaffairs.com/articles/africa/2015-02-16/power-poor Foley, A.M., Ó Gallachóir, B.P., Hur, J., Baldick, R., McKeogh, [Accessed] 2015 05 07 E.J., A strategic review of electricity systems models, Energy, Volume 35, Issue 12, December 2010, Pages Bazilian, M., Cordes, L., Nussbaumer, P., Yager, A., 2011. 4522-4530, ISSN 0360-5442, http://dx.doi.org/10.1016/j. Partnerships for access to modern cooking fuels and energy.2010.03.057. technologies. Curr. Opin. Environ. Sustain., Energy Systems 3, 254–259. doi:10.1016/j.cosust.2011.05.004 Fuso Nerini, F., Dargaville, R., Howells, M., Bazilian, M., 2014. Estimating the cost of energy access: The case of the Bazilian, M., Miller, M., Detchon, R., Liebreich, M., Blyth, W., village of Suro Craic in Timor Leste. Energy. Futch, M., Modi, V., Jones, L., Barkett, B., Howells, M., MacGill, I., Kammen, D.M., Mai, T., Wittenstein, M., Fuso Nerini, F.; Broad, O.; Mentis, D.; Bazilian, M.; Howells, Aggarwal, S., O’Malley, M., Carvallo, J.P., Welsch, M., M.; A Cost Comparison Of Technology Approaches for Pugh, G., Weston, R., Arent, D.J., 2013. Accelerating the Improving Access to Electricity Services. Elesevier’s Global Transformation to 21st Century Power Systems. Energy, 2015 Electr. J. 26, 39–51. doi:10.1016/j.tej.2013.06.005 Goodman, A.S., Hastak, M., 2006. Infrastructure Planning Bazilian, M., Rice, A., Rotich, J., Howells, M., DeCarolis, Handbook: Planning, Engineering, and Economics. ASCE. Joseph, Macmillan, S., Brooks, C., Bauer, F., Liebreich, M., Hafele, W., 1975. Objective Functions (Working Paper No. 2012. Open source software and crowdsourcing for WP-75-025). International Institute for Applied Systems energy analysis. Energy Policy. Analysis, Laxenburg, Austria. Bazilian, M.; Economy, R.; Nussbaumer, P.; Haites, E.; Hermann, S., Welsch, M., Segerstrom, R.E., Howells, M.I., Yumkella, K.; Howells, M.; Takada, M.; Rothman, D.; Levi, Young, C., Alfstad, T., Rogner, Hans-Holger, Steduto, P., M; Behond Basic Access, The Scale of Investment 2012. Climate, land, energy and water (CLEW) interlink- Required For Universal Energy Access, Book Chapter, ages in Burkina Faso: An analysis of agricultural intensifi- Energy Poverty Global Challenges and Local Solutions, cation and bioenergy production. Nat. Resour. Forum 36, 2014 245–262. doi:10.1111/j.1477-8947.2012.01463.x Beck, M.B., Walker, R.V., 2013. On water security, sustainabil- Howells, M., Hermann, S., Welsch, M., Bazilian, M., ity, and the water-food-energy-climate nexus. Front. Segerström, R., Alfstad, T., Gielen, D., Rogner, H., Fischer, Environ. Sci. Eng. 7, 626–639. doi:10.1007/s11783-013- G., van Velthuizen, H., Wiberg, D., Young, C., Roehrl, R.A., 0548-6 Mueller, A., Steduto, P., Ramma, I., 2013a. Integrated Bekker, B., Eberhard, A., Gaunt, T., Marquard, A., 2008. South analysis of climate change, land-use, energy and water Africa’s rapid electrification programme: Policy, institu- strategies. Nat. Clim. Change 3, 621–626. doi:10.1038/ tional, planning, financing and technical innovations. nclimate1789 Energy Policy 36, 3125–3137. doi:10.1016/j. Howells, M., Hermann, S., Welsch, M., Bazilian, M., enpol.2008.04.014 Segerström, R., Alfstad, T., Gielen, D., Rogner, H., Fischer, Bhattacharyya, C., and Timilsina, G.R., A Review or Energy G., van Velthuizen, H., Wiberg, D., Young, C., Roehrl, R.A., System Models, INternational Journal of Energy Manage- Mueller, A., Steduto, P., Ramma, I., 2013b. Integrated ment, Vol. 4, No. 4, 2010, pp. 494-518, DOI analysis of climate change, land-use, energy and water 10.1108/17506221011092742 strategies. Nat. Clim. Change 3, 621–626. doi:10.1038/ nclimate1789 DeCarolis, J.F., Hunter, K., Sreepathi, S., 2012. The case for repeatable analysis with energy economy optimization Howells, M., Roehrl, R.A., 2012. Perspectives on Sustainable models. Energy Economics Energy for the 21st Century., Sustainable Development in 12    S TAT E O F E L E CTR I CI TY ACCES S R EPO RT  |  2 0 17 the 21st Century (SD21) project. United Nations Depart- Generation Expansion Planning Computer Code (No. ment of Economic and Social Affairs, Division for ORNL-4945). Oak Ridge National Lab., Tenn. (USA). Sustainable Development., New York. Kaijuka, E., GIS and rural electricity planning in Uganda, Howells, M., Rogner, H., Strachan, N., Heaps, Charles, Journal of Cleaner Production, Volume 15, Issue 2, 2007, Huntington, H., Kypreos, S., Hughes, A., Silveira, S., Pages 203-217, ISSN 0959-6526, http://dx.doi. DeCarolis, Joe, Bazillian, M., 2011. OSeMOSYS: the open org/10.1016/j.jclepro.2005.11.057. source energy modeling system: an introduction to its Kanagawa, M., Nakata, T., Assessment of access to electricity ethos, structure and development. Energy Policy 39, and the socio-economic impacts in rural areas of 5850–5870. developing countries, Energy Policy, Volume 36, Issue 6, Howells, M., Rogner, H.-H., 2014. Water-energy nexus: June 2008, Pages 2016-2029, ISSN 0301-4215, http:// Assessing integrated systems. Nat. Clim. Change 4, dx.doi.org/10.1016/j.enpol.2008.01.041. 246–247. doi:10.1038/nclimate2180 Kemausuor, F., Adkins, E., Adu-Poku, I., Brew-Hammond, A., Howells, M., Rogner, H.-Holger, 2014. Water-energy nexus: Modi, V., Electrification planning using Network Planner Assessing integrated systems. Nat. Clim. Change 4, tool: The case of Ghana, Energy for Sustainable Develop- 246–247. doi:10.1038/nclimate2180 ment, Volume 19, April 2014, Pages 92-101, ISSN 0973-0826, http://dx.doi.org/10.1016/j.esd.2013.12.009. Hunter, K., Sreepathi, S., DeCarolis, JS. 2013. Modeling for Insight Using Tools for Energy Model Optimization and Loulou, R., Goldstein, G., Noble, K., 2004. Documentation Analysis (Temoa), Energy Economics, 40, 339-349, ISSN for the MARKAL family of Models. 0140-9883 Macknick, J., Sattler, S., Averyt, K., Clemmer, S., Rogers, J., i4e, i4 energy, n.d. The SWITCH modelling project. [WWW 2012. The water implications of generating electricity: Document]. URL http://i4energy.org/index.php?op- water use across the United States based on different tion=com_content&view=arti- electricity pathways through 2050. Environ. Res. Lett. 7, cle&Itemid=206&id=473:switch 045803. doi:10.1088/1748-9326/7/4/045803 IAEA, 2009. Annex VI: seeking sustainable climate land Martínez, D.M., Ebenhack, B.W., Understanding the role of energy and water (CLEW) strategies—Nuclear Technology energy consumption in human development through the Review 2009 Supplement. Presented at the 53rd IAEA use of saturation phenomena, Energy Policy, Volume 36, General Conference. Issue 4, April 2008, Pages 1430-1435, ISSN 0301-4215, http://dx.doi.org/10.1016/j.enpol.2007.12.016. IAEA, I.A.E.A., n.d. Planning & Economic Studies Section (PESS): Capacity Building for Sustainable Energy Martinot, E., Energy efficiency and renewable energy in Development [WWW Document]. URL http://www.iaea. Russia: Transaction barriers, market intermediation, and org/OurWork/ST/NE/Pess/capacitybuilding.html capacity building, Energy Policy, Volume 26, Issue 11, September 1998, Pages 905-915, ISSN 0301-4215, IIASA, MESSAGE–A modeling framework for medium -to http://dx.doi.org/10.1016/S0301-4215(98)00022-6. long—term energy system planning, energy policy analysis, and scenario development, International Institute Mentis, D., Welsch, M., Fuso Nerini, F., Broad, O., Howells, of Applied Systems Analysis, Available at: http://www. M., Bazilian, M., Rogner, H., 2015. A GIS-based approach iiasa.ac.at/web/home/research/modelsData/MESSAGE/ for electrification planning—A case study on Nigeria. MESSAGE.en.html [Accessed May 12, 2015] Energy Sustain. Dev. 29, 142–150. doi:10.1016/j. esd.2015.09.007 International Atomic Energy Agency (IAEA), 1984. Expansion Planning for Electrical Generating Systems, A Guidebook. Mentis, D., Andersson, M., Howells, M., Rogner, H., Siyal, S., Broad, O., Korkovelos, A., Bazilian, M., 2016.a. The International Atomic Energy Agency (IAEA), 2009. IAEA tools benefits of geospatial planning in energy access— and methodologies for Energy Systems Planning and A case study on Ethiopia. Appl. Geogr. 72, 1–13. Nuclear Energy. doi:10.1016/j.apgeog.2016.04.009 International Energy Agency (IEA), 2009. IEA Energy Training Mentis, D., Andersson, M., Howells, M., Rogner, H., Siyal, S., Capacity-building Programme. [Online] Available at: Broad, O., Korkovelos, A., Bazilian, M., 2016.b. The https://www.iea.org/publications/freepublications/ benefits of geospatial planning in energy access— Accessed May 12 2015] A case study on Ethiopia. Appl. Geogr. 72, 1–13. International Energy Agency (IEA), 2014. Africa Energy doi:10.1016/j.apgeog.2016.04.009 Outlook: A Focus on Energy Prospects in Sub-Saharan Niet, T., Pitt, L., Rowe, A., Wild, P., Storage and the Shift Africa. Organisation for Economic Co-operation and to Low Carbon Energy. Journal of Clean Energy Development, Paris. Technologies vol. 4, no. 1, pp. 26-31, 2016. International Energy Agency (IEA), 2015. World Energy Nerini, F.F., Dargaville, R., Howells, M., Bazilian, M., 2014. Outlook. Organisation for Economic Co-operation and Estimating the cost of energy access: The case of the Development, Paris. village of Suro Craic in Timor Leste. Energy. International Renewable Energy Agency (IRENA) 2012, Nussbaumer, P., Fuso Nerini, F., Onyeji, I., Howells, M., 2013. Capacity Building Strategic Framework for IRENA Global Insights Based on the Multidimensional Energy (2012-2015). IRENA, United Arab Emirates, pp. 25 Poverty Index (MEPI). Sustainability 5, 2060–2076. IRENA, I.R.E.A., 2014. Estimating the Renewable Energy doi:10.3390/su5052060 Potential In Africa. Bonn. Ohiorhenuan, John F.E.; Wunker, Stephen M.; UNDP; UNEP. Jenkins, R.T., Joy, D.S., 1974. Wein Automatic System 1995. Capacity building requirements for global environ- Planning Package (WASP): An Electric Utility Optimal mental protection. Global Environment facility working paper series ; no. 12. Washington, D.C. : The World Bank. ENERGY ACCESS AND ELECT RICIT Y PLANNING  13  Onyeji, I., Bazilian, M., Nussbaumer, P., 2012. Contextualizing Tiba, C., Candeias, A.L.B., Fraidenraich, N., Barbosa, E.M. de electricity access in sub-Saharan Africa. Energy Sustain. S., de Carvalho Neto, P.B., de Melo Filho, J.B., 2010. A Dev. 16, 520–527. doi:10.1016/j.esd.2012.08.007 GIS-based decision support tool for renewable energy management and planning in semi-arid rural environ- Parshall, L., Pillai, D., Mohan, S., Sanoh, A., Modi, V., National ments of northeast of Brazil. Renew. Energy 35, 2921– electricity planning in settings with low pre-existing grid 2932. doi:10.1016/j.renene.2010.05.009 coverage: Development of a spatial model and case study of Kenya, Energy Policy, Volume 37, Issue 6, June 2009, Timmerman, J., Vandevelde, L., Van Eetvelde, G., Towards Pages 2395-2410, ISSN 0301-4215, http://dx.doi. low carbon business park energy systems: Classification of org/10.1016/j.enpol.2009.01.021. techno-economic energy models, Energy, Volume 75, 1 October 2014, Pages 68-80, ISSN 0360-5442, http://dx. Ravita D. Prasad, R.C. Bansal, Atul Raturi, Multi-faceted doi.org/10.1016/j.energy.2014.05.092. energy planning: A review, Renewable and Sustainable Energy Reviews, Volume 38, October 2014, Pages UN, U.N., 2014. Prototype Global Sustainable Development 686-699, ISSN 1364-0321, http://dx.doi.org/10.1016/j. Report . Online unedited edition. rser.2014.07.021. UNECE, W.C. secretariat, 2014. Alazani/Ganikh RiverBasin Rogner, H.H., 2011. Energy Planning and Application to Water-Food-Energy-Ecosystems Nexus assessment—First Nuclear Power. [Online] Available at: http://www.iaea.org/ draft for comments. NuclearPower/Downloadable/Meet- United Nations, 2014. Prototype Global Sustainable ings/2011/2011-May-Africa/EnergyPlanningApplica- Development Report. United Nations Department of tions-H.Rogner-IAEA.pdf Economic and Social Affairs, Division for Sustainable SAPP, S.A.P.P., 2009. SAPP Regional Generation and Development, New York. Transmission Expansion Plan Study. WB, World Bank, 2015. Enhancing the Climate Resilience Sattler, S., Macknick, J., Yates, D., Flores-Lopez, F., Lopez, A., of Africa’s Infrastructure. Washington. Rogers, J., 2012. Linking electricity and water models to Welsch, M., 2013. Enhancing the Treatment of Systems assess electricity choices at water-relevant scales. Environ. Integration in Long-term Energy Models. Ph.D. Stock- Res. Lett. 7, 045804. doi:10.1088/1748-9326/7/4/045804 holm: Royal Institute of Technology. Schrattenholzer, L., 1981. The energy supply model MES- Welsch, M., 2013. Enhancing the Treatment of Systems SAGE. IIASA Laxenburg, Austria. Integration in Long-term Energy Models. KTH—Royal Smajgl, A., Ward, J., 2013. The Water-Food-Energy Nexus in Institute of Technology, Sweden, Stockholm. the Mekong Region: Assessing Development Strategies Welsch, M., Deane, P., Howells, M., Gallachóir, B.Ó., Rogan, Considering Cross-Sectoral and Transboundary Impacts. F., Bazilian, M., Rogner, H.-H., 2014. Incorporating Springer Science & Business Media. flexibility requirements into long-term energy system SOFRECO, 2011. Africa Energy Outlook 2040: Study on models–A case study on high levels of renewable Programme for Infrastructure Development in Africa electricity penetration in Ireland. Appl. Energy 135, (PIDA) [WWW Document]. URL http://www.pidafrica.org/ 600–615. PIDA-Sector-speccific%20Terms%20of%20Reference%20 Wilson, R., Biewald, B., 2013. Best Practices in Electric Utility for%20Energy%20Sector.pdf Integrated Resource Planning, Examples of State Sustainable Energy for All (SE4ALL), 2015. Capacity Building Regulations and Recent Utility Plans. Synapse Energy Hub for Sustainable Energy. [Online] Available at: http:// Economics (June 2013) www.se4allcapacityhub.org/index.php?option=com_ World Bank (W.B.), 2010. Africa’s Infrastructure—A time for content&view=article&id=2&Itemid=160##vision Transformation. [Accessed May 12 2015] World Bank (W.B.), 2014. World Development Indicators Szabó, S., Bódis, K., Huld, T., Moner-Girona, M., 2011. Energy [WWW Document]. URL http://data.worldbank.org/ solutions in rural Africa: mapping electrification costs of indicator/EG.USE.ELEC.KH.PC (accessed 9.5.14). distributed solar and diesel generation versus grid extension. Environ. Res. Lett. 6, 034002. World Bank (W.B.). 2015. Africa. Planning Africa’s Infrastruc- doi:10.1088/1748-9326/6/3/034002 ture in An Uncertain Climate Future. Available at: http://www.worldbank.org/en/region/afr/publication/ Taliotis, C., Shivakumar, A., Ramos, E., Howells, M., Mentis, planning-africas-infrastructure-uncertain-climate-future D., Sridharan, V., Broad, O., Mofor, L., 2016. An indicative [Accessed May 12 2015] analysis of investment opportunities in the African electric- ity supply sector—Using TEMBA (The Electricity Model Base for Africa). Energy Sustain. Dev. 31, 50–66. doi:10.1016/j.esd.2015.12.001 SPECIAL FEATURES To download the State of Electricity Access Report, overview, and Special Features, visit: http://esmap.org/SEAR