© Clean Cooking Alliance Designing Responsible End-User Subsidies for Energy Access: A toolkit featuring case studies for the design of subsidies in off-grid solar and clean cooking Copyright © July 2024 | International Bank for Reconstruction and Development The World Bank 1818 H Street NW, Washington, DC 20433 Telephone: 202-473- 1000; Internet: www.worldbank.org 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 noncommercial purposes if full attribution to this work is given. Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; Fax: +1-202-522-2625; e-mail: pubrights@worldbank.org. Lighting Global/ESMAP. 2024. Designing Responsible End-User Subsidies for Energy Access. Washington DC: World Bank. The World Bank does not necessarily own each component of the content contained within the work and does not warrant that the use of any third-party owned individual component or part contained in the work will not infringe on the rights of those third parties. If you wish to reuse a component of the work, it is your responsibility to determine whether permission is needed for that reuse and to obtain permission from the copyright owner. Examples of components can include, but are not limited to, tables, figures, or images. Disclaimer The conclusions and judgments contained in this report should not be attributed to, and do not necessarily represent the views of the World Bank or its Executive Directors, or the countries they represent. The World Bank does not guarantee the accuracy of the data in this publication and accepts no responsibility for any consequences of their use. The information contained in this publication is derived from carefully selected sources that are believed to be reasonable. Any opinions expressed reflect the current judgment of the authors of the relevant article or features and do not necessarily reflect the opinion of the World Bank. The opinions presented are subject to change without notice. The World Bank accepts no responsibility for any liability arising from the use of this document or its contents. Lighting Global is the World Bank’s initiative to rapidly increase access to off-grid solar energy for the hundreds of millions of people living without electricity world-wide. Managed by the Energy Sector Management Assistance Program (ESMAP), we work with governments, the private sector, development partners, and end-users, continually innovating to unlock key market barriers and enable access and affordability to those that would otherwise be left behind. Our support has expanded to technologies that go far beyond lighting, including stand-alone solar systems to power the needs of households, farms, businesses, schools, health centers, and more. We operate with funding gratefully acknowledged from ESMAP and their donors. The Energy Sector Management Assistance Program (ESMAP) is a partnership between the World Bank and over 20 partners to help low- and middle-income countries reduce poverty and boost growth through sustainable energy solutions. ESMAP’s analytical and advisory services are fully integrated within the World Bank’s country financing and policy dialogue in the energy sector. Through the World Bank (WB), ESMAP works to accelerate the energy transition required to achieve Sustainable Development Goal 7 (SDG7) to ensure access to affordable, reliable, sustainable, and modern energy for all. It helps to share WB strategies and programs to achieve the WBG Climate Change Action Plan targets. Learn more at https://esmap.org. The End-User Subsidy Lab is a platform to promote the uptake of responsibly-designed and well- informed end-user subsidies that are effective in reaching the poor, use public resources efficiently and minimize market distortion The End-User Subsidy Lab crowds in knowledge, resources and expertise from all stakeholders interested in participating; offers a platform for exchange, dialogue and extensive consultation among stakeholders; shares lessons learned, tools, and information; and tests prototype end-user subsidy designs. The lab is coordinated by ESMAP, EnDev, GOGLA and the Clean Cooking Alliance. For more information, visit: www.gogla.org/what-we-do/policy-regulations/end-user-subsidy-lab. Energising Development (EnDev) is an international flagship programme for providing energy access. EnDev works in more than 20 countries across Africa, Asia and Latin America. The driving force behind EnDev is the partnership of Germany, the Netherlands, Norway and Switzerland: donors who are committed to accelerating energy access. EnDev’s Demand-Side Subsidies (DSS) component is funded by the Directorate-General for International Cooperation (DGIS) of the Netherlands Ministry of Foreign Affairs. DSS enables low-income and displaced populations to sustainably improve their lives and livelihoods in Liberia, Malawi, Niger, and Uganda. GOGLA is the global association for the off-grid solar energy industry. Their almost 200 members provide millions of low-income and climate-vulnerable people with affordable, high-quality products and services; rapidly increasing customers’ productivity, connectivity, and resilience. To enable sustainable businesses and accelerate energy access, they provide market insights, standards and best practice, and advocate for catalytic policies, programmes and investment. To find out more, visit www.gogla.org. The Clean Cooking Alliance (CCA) works with a global network of partners to build an inclusive industry that can make clean cooking accessible to all. Established in 2010, CCA is driving consumer demand, mobilizing investment, and supporting policies that allow the clean cooking sector to thrive. To find out more, visit cleancooking.org. Open Capital Advisors (OCA) is a management consulting and financial advisory firm that drives growth, enables investment, and builds markets across Africa. We help businesses, investors, development partners, and the public sector to identify opportunities and deliver unique, impactful solutions. Our mission is to advance African economies and build future generations of business leaders. Since 2010, we have completed over 1600 engagements across over 30 countries in Sub-Saharan Africa and raised over $1.5 billion in capital for impactful businesses across the continent. Our locally based team of over 170 full-time staff brings experience from the world’s top consultancies, private equity firms, investment banks, and development organizations. For more information, please visit www.opencapital.com ACKNOWLEDGEMENTS The World Bank’s Lighting Global program, with funds gratefully acknowledged from the Shell Foundation commissioned this End-User Subsidy Toolkit. We would like to acknowledge the following authors and contributors (from Open Capital Advisors): David Ng’ethe, Davis Luboyera, Duda Slawek, Jonathan Maraka, Laurens Coeveld, (from the World Bank): Besnik Hyseni, Charlie Miller, Federico Hinrichs, Jennifer Lee Tracy, Raihan Elahi, (End-User Subsidy Lab partners): Christian Borchard, Collin Gumbu, Colm Fay, Myrte van der Spek, Olivia de Vesci, Pamela Lee, Patrick Tonui, Sarah Leitner. We would also like to thank those who reviewed the report for their valuable comments and contributions: Andrea Arricale, Andrew Lee Kent, Esmeralda Cecile Sindou, Ifunanya Genevieve Nwandu, Joern Huenteler, Jennifer Samantha Lynch, Johanna Christine Galan, Karl Skare, Kevin Johnstone, Mary Wilcox, Michelle Carvalho Metanias Hallack, Thomas Flochel, Tobi Ogunlesi, Yuko Okamura. iv End-User Subsidies for Energy Access: A toolkit End-User Subsidies for Energy Access: A toolkit i Table of Contents Key definitions iv Executive Summary 1 Introduction 9 1.1 Role of decentralized energy in universal energy access 9 1.2 Challenges and interventions for closing the energy access gap 10 1.3 End-user subsidies as a solution to close the affordability gap 13 1.4 The need to design effective EUS programs 14 Design framework for responsible end-user subsidies 16 2.1 Introduction 16 2.2 Pre-design elements 16 2.3 Beneficiaries and Targeting Mechanisms 20 2.4 Eligible Products and Subsidy Level 31 2.5 Delivery of Subsidy 40 2.6. Verification 56 2.7 Exit Strategy 59 2.8 Communication and Feedback 64 2.9 Conclusion 68 The design framework in practice 70 3.1 Introduction 70 3.2 Designing an EUS pilot targeting the poorest in Malawi 70 3.3 Designing an EUS pilot targeting refugees and host communities in Uganda 73 3.4 Analyzing EUS of Rwanda REF Window 5 76 3.5 Analyzing EUS of Nigeria’s OBF for SHS under NEP 79 Appendix 83 ©GIZ EnDev ii End-User Subsidies for Energy Access: A toolkit List of figures Figure 1: Affordability of off-grid solar technologies for unelectrified households 10 Figure 2: Energy access market types with a focus on the access and affordability gaps 11 Figure 3: Interventions to close the affordability gap 12 Figure 4: Total funding required to provide universal access to energy by 2030, 13 Figure 5: Launch dates for selected OGS and clean cooking end-user subsidy programs 14 Figure 6: Design framework for EUS programs 16 Figure 7: Pre-design elements 16 Figure 8: Target population, eligible households, and actual beneficiaries 21 Figure 9: Illustrative decision options based on the level of targeting 21 Figure 10: Summary of targeting mechanisms 23 Figure 11: Mapping poor villages using machine learning-enhanced satellite imagery 27 Figure 12: Image showing results of analyzed mobile phone metadata for target locations 27 Figure 13: Vulnerability Access Index in Tanzania 28 Figure 14: Summary of product selection considerations 34 Figure 16: Subsidy level options 38 Figure 17: Delivery channel choices 41 Figure 18: Company selection choices 47 Figure 19: Payment schedule choices 50 Figure 20: Fund management choices 53 Figure 21: Verification options 57 Figure 22: Exit strategy options 60 Figure 23: Continuous Monitoring, Evaluation, and Adaptation of subsidy programs 67 Figure 24: Wood-fired cook stove eligible for subsidy (left), and various Tier 1 OGS products (right) 71 Figure 25: Profile of OGS products sold, and customers reached, by project phase 85 List of tables Table 1: Questions to help determine the targeting approach 22 Table 2: Difference in subsidies for product purchase versus subsidizing electricity service or fuels (duration of the subsidy) 32 Table 3: Questions to help determine the delivery approach across the five elements 41 Table 4: Factors to help determine the delivery channel approach 42 Table 5: Questions to help determine company selection approach 48 Table 6: Questions to help determine the selection of a subsidy disbursement approach 51 Table 7: Questions to help inform the verification approach 58 Table 8: Questions to help determine exit strategy 61 Table 9: Summary of design element selected in an EUS pilot in Malawi 70 Table 10: Summary of design elements selected in an EUS pilot in Uganda 73 Table 11: Example of subsidy setting mechanism 74 Table 12: Summary of design elements of the subsidy under Rwanda REF 76 Table 13: Summary of design elements in Nigeria’s end-user subsidy 79 End-User Subsidies for Energy Access: A toolkit iii List of case studies Case study 1: Self-targeting in the Energy Access Scale Up Project (EASP) 25 Case study 2: Geographic and economic targeting in Togo’s Novissi program 26 Case study 3: Geographic targeting in EnDev’s Tanzania RBF 28 Case study 4: Demographic targeting in Nepal 29 Case study 5: Economic targeting in Rwanda Renewable Energy Fund (REF) Window 5 30 Case study 6: SHS subsidies provided by the Ngwee Ngwee Ngwee Fund under the Malawi Electricity Access Project (MEAP) 34 Case study 7: Combined percentage and fixed subsidy level in Rwanda’s REF Window 5 39 Case study 8: Subsidy delivery through private companies in refugee settings – Mercy Corps’ AMPERE project in Uganda 43 Case study 9: Direct delivery of subsidies to beneficiaries in Bangladesh’s BEAM Fund 46 Case study 10: Open company participation to all in Uganda’s Energy Access Scale-up project 48 Case study 11: Multiple recurring disbursements in Togo’s CIZO program 52 Case study 12: Mixed approach to fund administration in EnDev’s Pro-Poor RBF in Rwanda 55 Case study 13: Manual verification in Bangladesh’s IDCOL SHS Program 58 Case study 14: Decreased need for subsidies in Bangladesh 61 Case study 15: Long term subsidies in South Africa 63 iv End-User Subsidies for Energy Access: A toolkit Key definitions Terms Definitions Affordability Difference between the market price of a product and a consumer’s ability to pay. gap Clean cooking Cooking technologies that either reduce, eliminate, or support the transition away solutions from the use of biomass, such as charcoal or firewood, or other polluting fuels, such as coal and kerosene. These technologies range from Tier 4 to Tier 5 of the Multi-Tier Framework for clean cooking (refer to definition below), which are the technologies providing significant health benefits. Connections Households that have access to electricity through either a connection to an electricity distribution network or off-grid electricity providing Tier 1 or above access, as defined in the Multi-Tier Framework for electricity access. Refer to the definition below. Disposable The amount of money that an individual or household must spend or save after income income taxes, if any, have been deducted. End-user Final consumer of an energy product or service (either off-grid solar or clean cooking solution). In this report, it typically refers to an individual or a household. The terms end user and consumer are used interchangeably in this report. End-user Subsidies are provided to directly reduce the price of a service or product for the subsidies (EUS) end user, primarily aimed at bridging the affordability gap. Also known as demand-side subsidies (DSS), consumer subsidies, or price subsidies. Energy as a Energy as a service (EaaS), also referred to as fee-for-service, is a model in which Service customers pay periodically for a service, instead of purchasing a product or device. The product or device (e.g., a solar home system) in this case is owned by the company or service supplier, who sells electricity to the customer. Hard-to-reach Communities that are challenging to serve via commercial means due to being communities remote and/or because a large portion of the population is poor. Improved Cookstoves are commonly called “improved” if they are more efficient, produce Cookstoves lower emissions, or are safer than the traditional cook stoves or three-stone fires. Tier (ICS) 2 and 3 of the of the Multi-Tier Framework for clean cooking are also referred to as Improved Cookstoves. Leakage Situation where beneficiaries who could otherwise afford a product receive it through a subsidy. Market A situation in which the prices of goods and services on the market are influenced distortion by anything other than the principles of supply and demand, or in which competition amongst companies is distorted. It is viewed as any interference that significantly affects prices or market behavior. Subsidies, regulations, taxes and tariffs could represent sources of market distortion. In the case of end-user subsidies, they may distort a consumer’s perception of price, thus decreasing willingness to pay. Market maturity Degree of market development for off-grid solar products and clean cooking solutions in a given country or region. This report uses the classification of market maturity proposed in the Off-grid Solar Market Trends Report 2022: State of the Sector. Market maturity is classified in four categories, based primarily on a combination of cumulative sales penetration and recent sales growth rates: nascent, emerging, mature, and peaked. End-User Subsidies for Energy Access: A toolkit v Multi-Tier The Multi-Tier Framework (MTF) for clean cooking measures the dimensions of Framework for access to modern energy cooking services in various levels (tiers). clean cooking The MTF for clean cooking includes six attributes: (i) exposure, (ii) efficiency, (iii) convenience, (iv) safety, (v) affordability, and (vi) fuel availability. To measure progress, each attribute has six Tiers, ranging from 0 to 5. Multi-Tier The Multi-Tier Framework (MTF) for electricity access measures electricity access on Framework a tiered spectrum, from Tier 0 (no access) to Tier 5 (the highest level of access) for electricity access It provides a multi-dimensional definition for electricity access as ‘the ability to avail energy that is adequate, available when needed, reliable, of good quality, convenient, affordable, legal, healthy and safe for all required energy services.’ Off-grid solar Solar powered energy products, such as solar lanterns, multi-light kits, and solar (OGS) solutions home systems (SHS), including solar-powered appliances which are energy efficient. Pay-as-you-go PAYGo is a form of consumer financing that allows users to pay for their products (PAYGo) in small installments. It is commonly associated to selling OGS or clean cooking products through rent-to-own or lease-to-own contracts, i.e. the customer makes a down payment, followed by regular payments for a term ranging from six months to eight years, after which they own the product. The PAYGo business model is often technology-enabled, with payments usually made via mobile money (although other approaches exist i.e., payments done through scratch cards and cash) and a mechanism for the product to be “locked” in case payments are not made. PAYGo technology can also enable energy-as-a-service (EaaS) or fee-for-service models. EaaS is not as widespread as rent-to-own for OGS and clean cooking. For that reason, in this report PAYGo is used as a synonym for rent-to-own, and EaaS as an alternative business model. Results-based Financing instrument that provides financing (typically grants) based on achieving financing (RBF) specific and pre-agreed milestones. Supply-side Financial incentives for companies to reduce risks or costs of operations, typically subsidies (SSS) used to incentivize them to serve a market segment that is not commercially attractive. Vulnerable In this report, vulnerable refers primarily to people experiencing poverty or extreme groups poverty and other forms of marginalization often correlated with poverty, such as displacement or refugee status, gender, climate impact vulnerability, and so on. Willingness to The price that a person is willing to pay for a good or service. Pay (WTP) vi End-User Subsidies for Energy Access: A toolkit List of abbreviations AMPERE Accessing Markets through Private Sector MTF Multi-Tier Framework Enterprises for Refugees Energy MSME micro, small and medium enterprises ATP ability to pay MTR Off-Grid Solar Market Trends Report BEAM Bangladesh Energy Access to NASSP National Social Safety Net Project (Nigeria) Modernisation Fund NGO non-governmental organization BRD Development Bank of Rwanda NNNF Ngwee Ngwee Ngwee Fund (Malawi) CRM customer relation management NEP Nigeria Electrification Project or National DSS demand-side subsidies Electrification Plan (Rwanda) EASP Energy Access Scale Up Project (Uganda) OGS off-grid solar EDCL Energy Development Corporation Limited PAYGo pay as you go EEAS energy as a service PUE productive use of energy EnDev Energising Development PV photovoltaic (solar technology) ESMAP Energy Sector Management Assistance RBF results-based financing Program REF Renewable Energy Fund (Rwanda) EUS end-user subsidies REG Rwanda Energy Group EUSL  end-user subsidy lab RREA Rural and Renewable Energy Agency FBAE  Free Basic Alternative Energy Program (South Africa) SDG Sustainable Development Goal (United Nations initiative) FBE Free Basic Electricity Program (South Africa) SHS solar home systems ICS improved cookstoves SNV Netherlands Development Organisation IDCOL Infrastructure Development Company Limited SSS supply-side subsidies IEC  International Electrotechnical Commission UBR Unified Beneficiary Registry IVA  independent verification agent UECCC Uganda Energy Credit Capitalization Company LCY local currency UNHCR United Nations High Commissioner for LPG liquefied petroleum gas Refugees GOGLA Global Off-Grid Lighting Association VAI vulnerablilty access index GRM grievance redress mechanism Wh watt hour LODA  Local Administrative Entities Development Wp watt peak Agency (Rwanda) WTP willingness to pay MEAP Malawi Electricity Access Project MINALOC Minstry of Local Government (Rwanda) MMP Mwangaza Mashinani Program (Kenya) All currency is in United States dollars ($, US$, or USD) unless otherwise indicated End-User Subsidies for Energy Access: A toolkit 1 Executive Summary The Need for End-User Subsidies now and 2030, of which $12 billion are needed to address End-user subsidies, alongside other public financing the affordability gap alone.8 EUS bridges the affordability gap mechanisms, have a key role to play in accelerating by directly lowering the price of a service or product for the global energy access in line with SDG 7. end-user. They are also known as demand-side subsidies (DSS), consumer subsidies, or price subsidies. Other public Off-grid solar (OGS) and clean cooking solutions have financing mechanisms include supply-side subsidies, lines played a significant role in accelerating universal energy of credit for companies to finance working capital and access. However, based on current projections, by 2030, receivables and guarantees. Coordination between end- 660 million people will be without access to electricity, and user subsidies and other public financing mechanisms can 1.8 billion people without access to clean cooking, mostly in support the development of markets for clean energy, while Sub-Saharan Africa (SSA).1 To ensure no one is left behind, addressing affordability challenges simultaneously. households must be reached with OGS and clean cooking solutions at an accelerated rate.2,3 End-user subsidies are an essential instrument to close the affordability gap and have been implemented at a Affordability remains a key barrier in preventing more greater scale in recent years. This report builds on the people from accessing energy solutions.4 This has been experience of ten countries having deployed or adjusted exacerbated by inflation and local currency depreciation EUS programs within the last decade. Notable examples in several key markets. In addition, the majority of people include Rwanda, where the government deployed targeted without energy access live in hard-to-reach areas, where subsidies for Solar Home Systems (SHS) reaching 330,000 markets for off-grid solar products and clean cooking households (more than 10 percent of the population) whom solutions are only nascent or emerging. These affected otherwise would be unable to afford those products, populations might not have access to clean energy products, thus significantly contributing to the country’s National even if they could afford them.5,6 Electrification Plan.9 In the case of Nigeria, the introduction of an end-user subsidy for SHS was critical to the acceleration An integrated approach, leveraging end-user subsidies of growth in an emerging market despite widespread (EUS) alongside other public financing mechanisms7, is affordability challenges, reaching over one million needed to overcome these challenges. Over $60 billion will households within one year of implementation.10 be required for OGS and clean cooking solutions between 1 International Energy Agency, 2023. SDG7: Data and Projections. Projections under Stated Policies Scenario. 2 Lighting Global/ESMAP, GOGLA, Efficiency For Access, Open Capital Advisors (2022), Off- Grid Solar Market Trends Reports 2022: Outlook. Washington, DC: World Bank. 3 International Energy Agency 2023, A Vision for Clean Cooking Access for All. 4 Affordability refers to the difference between the market price of a product and a consumer’s ability to pay. 5 Emerging OGS markets are characterized by both high penetration of sales and a large remaining electricity access gap. 6 Nascent markets refer to markets that have low adoption rates for OGS solutions, with OGS sales being less than 10 percent of market potential. 7 The linkages between end-user subsidies and other public finance mechanisms are explored in this toolkit and presented in more detail in ESMAP publication ‘Designing Public Funding Mechanisms in the Off-Grid Solar Sector’ 8 OCA Analysis. The financing gap related to affordability for OGS is taken from [Lighting Global/ESMAP, GOGLA, Efficiency For Access, Open Capital Advisors (2022), Off- Grid Solar Market Trends Reports 2022: Outlook. Washington, DC: World Bank report] and is 20% of the total financing gap. We applied the same estimate (20%) for clean cooking using the overall financing need from [International Energy Agency 2023, A Vision for Clean Cooking Access for All]. 9 Subsidy window under Rwanda’s Renewable Energy Fund (REF), funded by the World Bank and managed by the Development Bank of Rwanda (BRD). A case study of this subsidy is available in section 3.4. 10 Output-Based Fund (OBF) for SHS under Nigeria Electrification Project (NEP), funded by the World Bank and managed by the Rural Electrification Agency (REA). A case study of this subsidy is available in section 3.5. 2 End-User Subsidies for Energy Access: A toolkit End-user subsidies must be designed ‘responsibly’ The EUS Design Framework to be effective in reaching the poor, to use public resources efficiently and to avoid market distortion. As Design framework end-user subsidies directly influence pricing, they carry a risk of distorting market dynamics, as well as political Communication challenges. For example, subsidies may affect consumers’ value perception of OGS and clean cooking products and services, reducing their willingness to pay unsubsidized prices. A subsidy program may favor certain companies 3. Delivery selected for the implementation, leaving companies that do not participate in the subsidy program at a significant disadvantage. Companies participating may also become 2. Subsidy level 4. Verification reliant on subsidy revenue and find it difficult to operate after subsidies are removed. Another important risk is the political challenge of removing or restructuring subsidies, even when this is economically justified. These risks can be mitigated 1. Targeting 5. Exit strategy by designing responsible programs, with learning and adaptive mechanisms embedded in them, which this toolkit sets out to support. Pre-design elements This toolkit provides a framework to design responsible subsidies, building on lessons learned, considering Feedback different contexts and objectives, and balancing trade- offs. The way a subsidy is designed will have an impact on the cost to the government, the speed of rollout, the This paper presents a toolkit for designing EUS programs number of people reached and its scalability, as well as the for off-grid solar and clean cooking solutions across market-distortion and political risks discussed above. The different markets to address the affordability gap. toolkit provides recommendations on how to inform the subsidy design, options to set specific parameters (targeting, This design framework consists of pre-design subsidy level, delivery, verification, exit or adjustment), as elements, five design components, communication, well as guidelines for communication about subsidies. It also and feedback. The pre-design elements are grouped into provides recommendations on monitoring, evaluation, and two main categories: 1) conducting market assessments adaptation mechanisms. and 2) determining program goals and resources. Market assessments aim to better understand the beneficiaries’ This toolkit is primarily focused on subsidies for the context, ongoing initiatives, and the maturity of the clean purchase of off-grid solar products and clean cookstoves; energy market. The second category looks at defining it is limited in its application to fuel or electricity the program goals, funding amount and the type of pilot subsidies. Off-grid solar electrification is making the most required, if any. An example of a program goal may be for progress through the selling of devices to end users (on the EUS to focus on providing first-time access to basic cash or credit through mechanisms such as Pay-As-You- energy solutions for the ‘poorest’ of the population. A Go). This toolkit draws primarily from the experience of different program goal may be to support the growth of the subsidizing such sales and is therefore most applicable for clean energy market by addressing widespread affordability the design of subsidies for product purchase. The report challenges. Collectively, the pre-design elements set however acknowledges the importance of emerging Fee-for- important program and market-specific contexts to support Service or Electricity-as-a-Service models, and most of the the design of effective EUS programs. They help inform recommendations in this toolkit are also suitable for these the design of the program more broadly, as well as the models. However, they are not cited as prominently. Similarly, prioritization of resources, and ensure that the EUS program this toolkit is most applicable to subsidies for the purchase of is complementary to other energy access initiatives in the improved and clean cooking devices. Less so for subsidies market. for fuels, such as liquefied petroleum gas (LPG), or the electricity consumed by electric cooking devices. End-User Subsidies for Energy Access: A toolkit 3 Targeting Targeting Untargeted Self-targeting Geographic Demographic Economic Subsidies are open Subsidy is accessible Leverages existing Leverages a range Targets households to all, and eligible to all, but designed borders e.g., regions, of demographic or individuals individuals are not to attract the target or districts, where data (e.g., age, whose income restricted based on population (applying the population face gender, ethnic levels cannot unique identifiers it to products they similar affordability group, etc.) to afford OGS or clean such as income prefer) challenges target end users cooking products levels Less targeted More targeted (less precise) (more precise) Targeting identifies which individuals qualify for was untargeted, with the main objective of addressing subsidies. While there is not a single perfect design widespread affordability issues. Self-targeted subsidies refer option, designers may opt for one or more of the five to subsidies accessible to all but designed to attract the options, ranging from less targeted to more targeted. voluntary participation of the target population. Limiting the subsidy to specific product tiers is one technique for self- More targeted approaches enable designers to reach targeting. specific target populations more effectively, but require more data, and can be more costly and complex to The decision on the targeting approach will depend on implement. Assuming the target population is defined the objective of the subsidy, on context factors, as well as the people that cannot afford clean energy products as on trade-offs related to complexity of implementation, at unsubsidized prices, the most precise way of targeting cost, scalability, and market distortion and political risks. subsidies to them is to use data on income or expenditure More targeted approaches are suitable when the target at the household level (economic targeting). Such was the population is a relatively small portion of the total population, case of the end-user subsidy for SHS deployed in Rwanda, who need to be accurately identified. Accurate targeting which leveraged the country’s “Ubudehe” categorization will help minimize the leakage of subsidies to people of household socio-economic status. However, this data that do not need them, as well as mitigate risks of market is often unavailable, inaccurate, or outdated. As such, distortion or political challenges. On the other hand, targeted designers may consider other targeting approaches, such as approaches can be costly and complex to implement, using demographic characteristics or geographic location, requiring extensive monitoring and verification mechanisms. provided such characteristics have some correlation to Such complexity may also affect companies participating in income. Programs may also refine their targeting approach the subsidy program (who may need to develop mechanisms by leveraging additional data sources that can complement to make sure they sell subsidized products only to eligible one of the targeting approaches, such as mobile money individuals), and end-users (who may need to fulfill more usage or satellite imagery, or by using data from other burdensome administrative requirements to demonstrate programs or organizations serving similar beneficiaries. eligibility). Untargeted (or self-targeted) subsidies have the potential to reach a broader population, they are easier and Alternatively, designers may opt for less targeted faster to roll out, but they also carry the risk of allocating approaches, particularly where the majority of the support to people who do not need it, leading to resource population is facing affordability constraints. For example, inefficiencies and market distortion. the end-user subsidy for SHS introduced in Nigeria 4 End-User Subsidies for Energy Access: A toolkit Subsidy determination Product selection Subsidy level determination Subsidy level principles Actual subsidy levels Limit to Open to all specific product types Vary by Vary by meeting minimum Fixed / absolute Proportional to product or Tier Invariable target group product Tier Combination standards amount price or size The subsidy Subsidy level Subsidy level Subsidy Subsidy level Subsidy Eligibility is All products is defined as can adapt to can vary for level is set is set as a level is restricted to offered by a unique an affordability of different as a fixed $ percentage of determined as a narrow set companies invariable each group, product amount per price or fixed a combination of products participating in amount for whether categories / product and $ per unit of of percentage meeting certain the program, all eligible socio- Tiers customer capacity ($/Wh of price with specifications or and meeting products and economic, or $/Wp) for a maximum from a certain Tier. minimum customers geographic or the selected fixed amount For example, Tier-1 eligibility demographic. product or price SHS providing basic criteria, can service energy access only benefit from the subsidy. Subsidy determination involves selecting the products Overall, it is important to set the subsidy levels not and services eligible for a subsidy and determining the too low and not too high to ensure the most efficient subsidy amount for each end-user. When determining and effective use of limited resources. Over-subsidizing the subsidy amount, it is important for designers to set a products can lead to market distortion and depletion of reasonably simple subsidy structure that will be easy to program resources. On the other hand, setting a subsidy communicate, understand, and manage. too low may result in limited adoption, as beneficiaries may still be unable to pay. Beyond these subsidy level options, An important design choice is to determine which designers can further differentiate the subsidy level, for products and services are eligible for the subsidy. example, by varying the subsidy level by product tiers or by Designers can restrict eligibility to specific products or target group. tiers, or open eligibility to all products meeting minimum specifications, quality standards, and after-sales service.11 The decision on eligible products and subsidy levels is These choices are driven by what the target population typically linked to the targeting approach. As a general needs, what products are available in the market, and guideline, subsidy levels are higher for more targeted what type of products and services the program wants to subsidies, and lower for less targeted subsidies. When incentivize. subsidies are targeted to the poorest, they need to be high to fill a wider affordability gap. When subsidies are Designers also need to define the subsidy level, which untargeted, they are usually kept lower, recognizing that a is the process of defining how much subsidy each significant portion of subsidies will benefit people who do beneficiary will receive. This choice depends on the price not need them. Making subsidies eligible for only certain of the product and the amount the beneficiary can pay. The product categories (e.g., basic features, entry level) better difference between the two represents the affordability suited to the poor is an alternative to keeping untargeted gap. Ideally, the subsidy level matches the affordability gap. subsidies pro-poor (referred to as ‘self-targeting’ under However, this can be difficult in practice, as affordability will targeting approaches). vary by beneficiary and product. Therefore, designers need to decide on a structure to set the subsidy level, which can be based on a fixed amount, a percentage of the price, or a combination of both methods. As an alternative to program designers setting the subsidy level, this could be done by companies through a reverse-auction process. 11 Energy Sector Management Assistance Program, Multi-Tier Framework End-User Subsidies for Energy Access: A toolkit 5 Delivery Subsidy delivery Delivery channel Company selection Through companies Direct to beneficiaries Open to all companies Restrict through competitive meeting minimum criteria process Subsidy amount Subsidy amount All companies are eligible Designers issue a is given to the is given to eligible for participation provided competitive call for participating beneficiaries through they meet minimum criteria proposals and select a company, which in either vouchers or cash on quality, after-sales limited number of companies turn sells the product transfers (conditional or service, and environmental that are best placed to at a discounted price unconditional). and social safeguards deliver the subsidy (e.g., that to eligible consumers. Cash transfers Vouchers can deliver at scale) (conditional or (paper or unconditional) digital) A well-designed subsidy delivery mechanism aims to cash or vouchers, making this approach preferable. promote transparency and accountability, encourage stakeholder participation, and maximize outcomes Company selection is usually done either through an for end-users. Designers need to balance ease of open call to all companies meeting certain minimum access, robustness of approach, and cost-effectiveness eligibility requirements or through a competitive process when selecting an approach across the five delivery aimed at selecting only a few companies. The most components: 1) delivery channel, 2) company selection, 3) appropriate approach depends on the program goals, subsidy disbursement, 4) fund management, and 5) claim targeting approach and size of the target population, maturity management. of the market, and product selection. Opening participation to all companies is effective for growing local markets and Subsidies can be delivered directly to beneficiaries or encouraging competition, which helps keep prices low. On through a company. The choice of approach will depend the other hand, designers may want to restrict participation on factors such as the size of the target population and to a limited number of companies when the target population the targeting approach, availability of quality products, is relatively small (e.g., if targeting remote areas where the market maturity and structure, and beneficiary high costs would make it difficult for multiple companies in characteristics. Delivery through companies can be simpler competition to profitably serve them), or when they want to for the government and has lower administrative costs, minimize transaction costs by working with only a few firms since subsidies are provided in bulk. It is usually much that can achieve economies of scale. easier and cheaper to channel a subsidy to a small number of companies rather than channeling a cash transfer, or a voucher, to potentially tens or hundreds of thousands of households. Because of these advantages, delivery through companies is the most widely used approach, often in the form of results-based financing (RBF). A clear challenge with delivering through companies is that beneficiaries may not be aware that the prices are subsidized or © Clean Cooking Alliance carry a temporary discount. However, direct delivery to end users may work well when the target population is more narrowly defined and relatively small, and allows for direct communication, better data collection on end users, and opens an opportunity to provide additional support if needed. For example, in some situations such as refugee camps, mechanisms may already be in place to distribute 6 End-User Subsidies for Energy Access: A toolkit Subsidy delivery Subsidy disbursement One off (after Multiple milestone- Recurring automatic Advance payment verification) based installments top-ups (before sale) Companies are Subsidy payments Subsidy payments are made Portion of subsidy reimbursed in full, one off, are made in regular to match the repayment is paid before sale for all subsidized products intervals determined pattern of the beneficiaries and verification. The sold within a certain by completion and remaining subsidy is period, after such sales verification of pre-defined disbursed following one have been verified by the service conditions of the prior approaches. program When working with companies to deliver subsidies, Fund management can be done by government agencies, there are various approaches for disbursement, which third-party administrators, or a combination of both. guide the payment schedule (or cadence) of subsidies. Strong administration structures consist of defined roles and Selecting the disbursement approach depends on the responsibilities for the players involved in the implementation mode of sale of the subsidized products (e.g., cash, and clear processes and procedures for claiming and credit, leasing, rental, or service fees), the level of need to disbursing funds. Deciding whether to involve a government incentivize companies to provide after-sales services, and agency is often dependent on whether the relevant the operational capacity of the participating companies. A government agency has adequate institutional capacity one-off payment approach minimizes both transaction costs and infrastructure to manage the subsidy program. Other and administration. Disbursement through multiple payments approaches designers may opt for include engaging a third- allows implementors to ensure that companies provide party administrator or taking a mixed approach that involves after-sales services to beneficiaries. Recurring top ups are a third-party working alongside the relevant government a suitable method for PAYGo payments and for monitoring agency. Third-party administrators can often bring best companies’ after-sales services. Finally, providing some practices that allow for efficient program delivery while advance payment may improve cashflows for companies to building local capacity, but this can also lead to increased purchase stock in advance, when other forms of finance are administrative costs. Regardless of the approach adopted, it limited. is advisable to engage independent auditors to review the management of funds periodically. Subsidy delivery Claim management involves coordinating with the participating companies to submit information about Fund management the sale of a product. A claim form is typically submitted electronically, by email, or uploading it onto an online IT platform. It is important for program designers to keep the Government Mixed phased claim management procedure simple, set realistic timelines, Third party administrators approach and document all requirements for companies to submit claims. Subsidy program Program Third-party is managed by designers hire administrators Verification agencies or an external firm jointly manage ministries directly with specialized the program The verification process allows to confirm that subsidies related to the technical alongside were used as planned (i.e., that products claimed are real, subsidy program knowledge government and that they were sold at the agreed subsidized price, to agency eligible end users). Verification in EUS programs informs the subsidy disbursement. End-User Subsidies for Energy Access: A toolkit 7 Verification (complementary options) Exit and adjustment strategies A program-level exit strategy addresses what the next Traditional / manual Automated / technology- step is once the specific program comes to an end. It is verification methods enabled verification critical to define this next step from the program’s start methods to ensure program objectives, timing, and funding are Involves combined aligned with what will follow. techniques of desk-based Leverages digital verification (paper trail), technology to automate Exit and adjustment strategies phone surveys, and field verification (e.g., remote verification monitoring via GSM or GPS, customer payment information through Subsidy no New subsidy Long-term companies’ CRM systems longer needed program subsidies or mobile money providers, automatic SMS/ When a Transitioning to Continuous Whatsapp surveys, etc) government has a new subsidy end-user fully achieved program, with subsidy facility There are multiple options and tools to carry out its targets and restructuring that consumers verification, ranging from more traditional or manual a subsidy is no of subsidies if and companies forms of verification to more automated verification longer needed, needed participate in, leveraging data and technology. Verification is typically an unsubsidized can be funded done by independent verification agents (IVAs). The market can take from the more traditional or manual approach typically involves over national treasury the IVA investigating companies’ subsidy claims through or sector level documentation checks and conducting a combination funds of phone and field surveys of a sample of customers. The automated verification refers to leveraging data and technology, such as using information on the use of products When a government has fully achieved its targets, and (if they are equipped with remote-monitoring technology) a subsidy is no longer needed, a non-subsidized market and customer payment information drawn from companies’ can take over. In this scenario, the exit strategy should customer relation management (CRM) systems or mobile ensure program goals are met and that there is a successful money operators (if such payments are primarily done transition to the non-subsidized market for products or through mobile money). Manual verification can be used services after a subsidy is withdrawn. Market monitoring is across all types of products and regions as it does not critical to understand changes in market dynamics that may require any form of remote connectivity. However, manual impact the subsidy scheme and to identify when subsidies verification processes are labor-intensive, costly, and more are no longer needed. prone to human error. Automated technology-enabled verification can provide quick results and lower transaction The need for end-user subsidies will, at some level, costs. But it may not work in all situations today, for example, continue until market developments (such as economies projects delivered in remote areas where connectivity is an of scale, technology advances driving cost reduction, issue and projects involving lower-tier products that are not and economic development) reduce the affordability equipped with remote monitoring technologies. Often, a gap to the point where access is achievable on a purely system needs to be tailored to each program’s operations commercial basis. When designing a subsidy, it is important and unique needs, and this comes at a high cost that only to consider what recurring subsidies, if any, may be makes financial sense for programs of a large scale. needed to sustain universal energy access. Where long- term subsidies are envisioned, programs should consider There is a trend for subsidy programs to integrate how these can be sustainably funded and operated. This automated verification features where this is possible, may mean transitioning into new programs, or a long-term to improve the speed and accuracy of the verification facility funded by the government or development partners. process. Where possible, program designers should try to use technology as much as possible but leverage manual methods as a complement. 8 End-User Subsidies for Energy Access: A toolkit Other funding options may include integration into a social audiences with the goal of achieving transparency. It is very protection program, an energy sector fund operating cross important to ensure that intended beneficiaries are well subsidies, or carbon finance. informed about the subsidy, with campaigns tailored to them, on the right channels and in the culturally appropriate media. Based on the identified exit or adjustment strategy, Beyond communication plans for intended beneficiaries, it designers also need to consider how to stop subsidies may be important to develop clear messaging for non-target within the specific program. This can either be outright populations to help them understand why they are excluded. or through a gradual phase-out process, for example, by reducing the subsidy level over time or incrementally Designers must continuously monitor the program’s narrowing the target group. Ideally, the exit strategy is performance against its set goals, remain aware of thought through from the program’s start and actively current market contexts and dynamics, and create a communicated throughout implementation to both framework for program adaptation. Creating feedback households and participating companies. loops, periodical program reviews and subsidy redesign when necessary is crucial for subsidy programs to remain successful over time. Subsidies may need revision if uptake Communications and Feedback is too low or too high. Macroeconomic factors like currency fluctuations, inflation, and global supply chain costs can Communications involve developing a plan to guide significantly influence program implementation. In addition, stakeholder interactions before and during program it is helpful to continuously engage end-users, companies, implementation. Feedback refers to the monitorting and governments, and development partners through channels evaluation of the program’s results and impacts, used such as round tables and one-on-one check-ins to gather for learning and adaptation of the program’s design as feedback on how the program can improve. This may needed. include changes in the targeting approach, increasing or decreasing subsidy levels, engaging additional companies, The framing of the subsidy is important to set the streamlining procedures, etc. For a subsidy to remain right expectations for all stakeholders involved. Key effective and efficient, it needs to be able to adapt. stakeholders include end users, participating companies, and government entities in the program areas. Communication ought to be tailored to suit the context of intended A call to action As we draw closer to 2030, urgent action is required to ensure the most vulnerable populations are not left behind in the journey to energy access. End-user subsidies are an essential instrument to close the affordability gap. More well-designed EUS, tailored to varying market contexts, are needed to address the affordability gap. Governments, development partners, and the private sector must come together to pool resources and leverage knowledge to scale these programs and ensure no household is left behind. This toolkit provides guidance on the design of responsible end-user subsidies, building on lessons learned in recent EUS programs. It also provides examples to illustrate how the design framework for EUS programs can be used in practice. End-User Subsidies for Energy Access: A toolkit 9 01. Introduction This report targets stakeholders active in the energy sector, 2030.15 Meanwhile, clean cooking stoves and clean fuels such as governments, development partners, companies, reduce dependency on biomass such as firewood and and industry experts. The report’s primary objective is to charcoal. In line with SDG 7, improved cookstove solutions serve as a toolkit, offering insights into the design and (ICS) increase access to clean energy and decrease fuel implementation of end-user subsidy programs for off-grid requirements by 20 to 75 percent, reducing harmful smoke solar and clean cooking solutions. The document is centered and emissions. Higher-tier clean cooking technologies around a design framework that has been developed based provide a large decrease in hazardous pollutants and deliver on learnings from end-user subsidy (EUS) programs to date significant health benefits.16 and uses examples and case studies where relevant. The toolkit is not meant to be prescriptive; it is designed to Progress in expanding energy access has slowed down provide the design options available to support program in recent years, and the world is at serious risk of not designers. achieving universal energy access by 2030. As of 2022, an estimated 760 million people globally lack access to electricity, while 2.3 billion people do not have access 1.1 Role of decentralized energy in universal to clean cooking solutions17. Affordability, impacted by energy access inflation and local currency depreciation in several key markets, excludes people from access to clean energy Off-grid solar (OGS) and clean cooking solutions solutions. Under current projections, by 2030, there will (collectively clean energy solutions) have already played be approximately 660 million people without access to a significant role in accelerating global energy access in electricity and 1.8 billion people without access to clean line with SDG 7. Between 2010 and 2022, OGS solutions cooking, mostly in Sub-Saharan Africa.18 To reverse this (Tier 1 and above) provided energy access to 493 million trend, households must be reached at an accelerated people, and 700 million people accessed clean cooking rate. solutions (Tier 3 and above).12,13 Traditionally, governments have provided access to electricity via the national grid. Progress in expanding energy access However, grid connections are predominantly found in has slowed down in recent years, and the urban and peri-urban areas, and grid extension is often a world is at serious risk of not achieving costly solution for remote and sparsely populated regions.14 universal energy access by 2030. As of Decentralized systems such as OGS are, therefore, essential 2022, an estimated 760 million people tools for providing first-time access to electricity, and they globally lack access to electricity, while can also serve as backups for unreliable grid electricity. Based on geospatial least-cost analysis, OGS is expected 2.3 billion people do not have access to to account for 41 percent of all new global connections by clean cooking solutions. 12 IEA, IRENA, UNSD, World Bank, WHO, 2023 Tracking SDG 7: The Energy Access Report, 13 International Energy Agency 2023, A Vision for Clean Cooking Access for All. 14 Africa Clean Energy (ACE) Technical Assistance Facility (TAF) and Open Capita 2020, Demand-Side Subsidies in Off-Grid Solar: A tool for achieving universal energy access and sustainable markets. 15 Lighting Global/ESMAP, GOGLA, Efficiency For Access, Open Capital Advisors 2022, Off- Grid Solar Market Trends Reports 2022: State of the Sector. Washington, DC: World Bank. 16 International Energy Agency 2023, A Vision for Clean Cooking Access for All. 17 International Energy Agency, 2023. SDG7: Data and Projections. 18 International Energy Agency, 2023. SDG7: Data and Projections. Projections under Stated Policies Scenario. 10 End-User Subsidies for Energy Access: A toolkit Under current projections, by 2030, there Zooming in on the affordability challenge, even will be approximately 660 million people assuming universal availability, a significant portion of without access to electricity and 1.8 the population would not be able to afford clean energy billion people without access to clean products. The uptake of the PAYGo business model, which cooking, mostly in Sub-Saharan Africa. allows customers to pay with an upfront down payment and To reverse this trend, households must be installments, has been crucial in reducing the affordability reached at an accelerated rate. gap, but a significant gap remains. For illustration (see Figure 1 below), only 62 percent of unelectrified households could 1.2 Challenges and interventions for closing afford a multi-light and mobile charging system providing the energy access gap basic Tier-1 access to electricity, and only 30 percent could afford a SHS under current prices and over typical PAYGo The majority of people without energy access live in repayment periods. Higher-tier products are unaffordable for underdeveloped markets. For OGS solutions, over 80 most unelectrified households. percent of the unserved today are in markets that may be categorized as “nascent” and “emerging,” as defined in the Figure 1: Affordability of off-grid solar technologies for Off-Grid Solar Market Trends Report (MTR) 2022.19 Most of the bottom of the pyramid23 these markets are in Sub-Saharan Africa, which accounts for 77 percent of the electricity access gap. In addition, around Affordability (2021) half of unconnected households live in fragile and conflict- Cash PayGo affected countries.20 For clean cooking solutions, the gap 2% also occurs more strongly in challenging markets. This is illustrated by 29 countries in Sub-Saharan Africa with access 26% to clean cooking below 20 percent.21 30% 45% Populations across these markets remain unserved for one of two reasons22: 58% 62% 1. The “availability or access gap” refers to the situation 57% 74% 98% where individuals have no practical ways of obtaining 95% OGS or clean cooking products despite some having 55% the ability and willingness to pay. These individuals 29% are usually in remote and hard-to-reach areas where 34% the market for decentralized clean energy solutions has not taken off. This may be due to remoteness, 13% 13% lack of infrastructure, low population density, and 4% 4% Lantern(<1.5 Wp) Light & Charger (1.5-3 Wp) Multilight & charging (3- 11Wp) SHS(11-21 Wp) SHS(21-50 Wp) SHS(50-100 Wp) SHS(100+ Wp) perceived affordability issues, making some markets not commercially viable for companies. Some individuals may also lack access due to experiencing different forms of marginalization, for example, due to gender and disabilities. Tier 0 Tier 1 Partial Tier 2 Tier 1 2. The “affordability gap” refers to the situation of individuals who are unable to afford clean energy products, regardless of whether or not they have access Affordable at a stretch Affordable Unaffordable to purchase them. Affordability is a factor of both low Note: graph shows the percentage of the unelectrified households that can afford consumer ability to pay, especially among the most different off-grid solar products based on their income and market prices (cash prices for vulnerable households, and high product prices coupled Tier-0 products, and PAYGo prices for products of tier-1 and above). For PAYGo products, ‘affordable’ is defined as a household spending up to 5%of monthly income on PAYGo with the lack of consumer financing options. repayments. Affordable at a stretch as spending up to 10% of monthly household income. 19 Lighting Global/ESMAP, GOGLA, Efficiency For Access, Open Capital Advisors (2022), Off- Grid Solar Market Trends Reports 2022: Outlook. Washington, DC: World Bank. 20 IEA (2021), Tracking SDG 7: The Energy Progress Report 2021. 21 International Energy Agency 2023, A Vision for Clean Cooking Access for All. 22 Africa Clean Energy (ACE) Technical Assistance Facility (TAF) and Open Capital 2020, Demand-Side Subsidies in Off-Grid Solar: A tool for achieving universal energy access and sustainable markets. 23 Lighting Global/ESMAP, GOGLA, Efficiency For Access, Open Capital Advisors 2022, Off- Grid Solar Market Trends Reports 2022: State of the Sector. Washington, DC: World Bank. The methodology for this analysis is presented in detail in Annex 4 of the Market Trends report. End-User Subsidies for Energy Access: A toolkit 11 Considering the access and affordability gaps, the unserved population can be divided into four segments (see Figure 2). Even in commercial markets (top left segment), there will be people without access to energy, but these can be served by the market. Market-building interventions, including supply-side, demand-side, and enabling environment interventions, are needed for the other three segments to help consumers and companies alike. Figure 2: Energy access market types with a focus on the access and affordability gaps24 Able to Commercial Market: Consumers able Logically Challenged Market: Consumers can pay for afford to pay for products and in commercially products but not within commercial serviceable areas products serviceable areas (Access gap) Financially Challenged Market: Consumers Non-Commercial Market: Consumers are Unable are unable to pay but are in commercially unable to afford and not in commercially to afford serviceable areas serviceable areas products (Affordability gap) (Access and affordability gap) Within commercial geographic reach Not within commercial geographic reach Governments and development partners have a range of A holistic approach, leveraging multiple instruments, instruments available to address these gaps. Supply-side is typically needed to overcome the affordability gap. instruments aim to incentivize companies to serve more Supply-side instruments can indirectly bring down market people in a particular market, thus addressing the access prices by supporting technology and business model gap. These typically take the form of financial incentives for innovation, fostering competition, and lowering costs companies to enter a market or expand their consumer base, through scaling. They could provide support or incentives to such as grants, results-based financing, credit lines, and risk- companies that sell products on a PAYGo basis. Alternative sharing instruments. Demand-side instruments are designed models may also be incentivized, such as ‘Energy-as-a- to support the consumer and are often aimed at addressing Service’ (EaaS), in which the company maintains ownership the affordability gap. These may include end-user subsidies, of the system, which helps reduce the price levels for concessional consumer financing, and public procurement of consumers and enhances the quality of the system and energy access solutions on behalf of end-users.25 In addition service levels provided. Increased consumer awareness may to these financing mechanisms, governments may further further increase customers’ willingness to pay. In addition enable the market by providing an enabling environment to other demand-side instruments, end-user subsidies may (adopting favorable policies such as tax exemptions or be used to further bridge the affordability gap by directly setting up national energy agencies), raising consumer reducing the price of a product. awareness, and building infrastructure (for example, roads and connectivity), among other solutions. 24 GOGLA 2018, Providing Energy Access through Off-Grid Solar: Guidance for Governments. 25 Energy Sector Management Assistance Program (ESMAP) 2022. Designing Public Funding Mechanisms in the Off-Grid Solar Sector. 12 End-User Subsidies for Energy Access: A toolkit ©GIZ EnDev When both supply-side and demand-side instruments projects cited in this report are implementing supply-side are needed, they can be deployed sequentially or in interventions coupled with end-user subsidies; such as the parallel, depending on the needs of the market. In an Renewable Energy Fund (REF) in Rwanda, the Output-Based ideal scenario, an enabling environment is first developed, Fund for SHS under the Nigeria Electrification Project (NEP), followed by supply-side interventions to address the the Electricity Access Scale-up Project (EASP) in Uganda, accessibility of products and reduce costs (see Figure and the Ngwee Ngwee Ngwee Fund under the Malawi 3 below), with EUS saved as a last option to bridge the Electricity Access Project (MEAP). Using these and other remaining affordability gap. This ensures the efficient use examples, this toolkit includes specific guidance to assess of public funds and reduces the risk of market distortion.26 in what cases both supply-side and end-user subsidies are With only six years left to achieve SDG 7, some countries needed in simultaneously, and how these interventions can and stakeholders are thinking through how to responsibly be designed to effectively complement each other while implement interventions in parallel. Several subsidy avoiding over-subsidization. Figure 3: Interventions to close the affordability gap Interventions to improve supply and demand End-user interventions Initial market Technology and Production scale Increased Low cost of doing price design innovation and distribution competition business and efficiency among companies improved enabling environment Reduced Gap Vulnerable Initial Gap Early adopters and first Growing customer households Mass Market movers base End-user subsidies to bridge the Initial remaining willingness Increased Aspirational Savings or income Consumer affordability gap to pay awareness and product and generating financing demand consumer value potential 26 Energy Sector Management Assistance Program (ESMAP) 2022. Designing Public Funding Mechanisms in the Off-Grid Solar Sector. End-User Subsidies for Energy Access: A toolkit 13 Across instruments, significant funding will be required customers. For OGS, end-user subsidies can help drive to address the access and affordability gaps and ensure electricity access among the 760 million unserved people, that hundreds of millions of people, especially in Africa, 280 million of whom cannot afford a Tier 1 system even on are not left behind. Over $60 billion is required for OGS PAYGo.31 EUS are an effective tool to support the segment and clean cooking solutions between now and 2030 to of the population that simply cannot afford these products, reach SDG 7, of which $12 billion is needed to address and they have proven to drive significant impact. A notable the affordability gap alone.27 Approximately 90 percent of example is Rwanda’s REF Window 5, which has deployed this funding is required in emerging and nascent markets, targeted subsidies for SHS reaching 330,000 households especially in fragile and conflict-affected regions.28 (more than 10 percent of the population) otherwise unable to afford those products. In the case of Nigeria’s NEP, the introduction of an end-user subsidy for SHS was critical Figure 4: Total funding required to provide universal access to accelerate growth in an emerging market despite to energy by 203029,30, widespread affordability challenges, realizing over one million SHS sales within one year. Both case studies are presented in detail in this toolkit32, and used to illustrate USD, Billion subsidy design decisions and trade-offs. End-user subsidies for OGS and clean cooking products Clean have been implemented at a greater scale in recent Cooking years. Since 2017, there has been an acceleration in the development of programs providing subsidies for off-grid OGS solar and clean cooking devices, the majority of which are in Sub-Saharan Africa (see Figure 5 below for select EUS 0 5 10 15 20 25 30 35 40 programs). While the reach of these programs has varied across markets, development partners and governments Affordability gap financing Access gap financing increasingly recognize the potential for EUS to work alongside other supply- and demand-side initiatives to ensure no one is left behind.33 1.3 End-user subsidies as a solution to close the affordability gap End-user subsidies bridge the End-user subsidies are an essential instrument to close affordability gap by directly the affordability gap. End-user subsidies bridge the affordability gap by directly reducing the price of clean reducing the price of clean energy energy products below commercial rates for eligible products below commercial rates. 27 OCA Analysis. The financing gap related to affordability for OGS is taken from [Lighting Global/ESMAP, GOGLA, Efficiency For Access, Open Capital Advisors (2022), Off- Grid Solar Market Trends Reports 2022: Outlook. Washington, DC: World Bank report] and is 20% of the total financing gap. We applied the same estimate (20%) for clean cooking using the overall financing need from [International Energy Agency 2023, A Vision for Clean Cooking Access for All]. 28 Lighting Global/ESMAP, GOGLA, Efficiency For Access, Open Capital Advisors (2022), Off- Grid Solar Market Trends Reports 2022: Outlook. Washington, DC: World Bank. 29 Lighting Global/ESMAP, GOGLA, Efficiency For Access, Open Capital Advisors 2022, Off- Grid Solar Market Trends Reports 2022: Outlook. For OGS, funding needs are estimated on the basis of reaching 1.1 billion people with Tier 1 and above OGS products by 2030. This includes current users, new primary connections, and ‘new weak’ grid connections. To achieve SDG 7, OGS electrification needs to be complemented with new grid and mini-grid connections. 30 International Energy Agency 2023, A Vision for Clean Cooking Access for All 31 Lighting Global/ESMAP, GOGLA, Efficiency For Access, Open Capital Advisors 2022, Off- Grid Solar Market Trends Reports 2022: State of the Sector. Washington, DC: World Bank. 32 Refer to section 3.4 for a detailed case study on Rwanda REF Window 5 and to section 3.5 for a detailed case study on Nigeria’s output-based fund for SHS under the Nigeria Electrification Project (NEP). 33 Open Capital Advisors Consultations 14 End-User Subsidies for Energy Access: A toolkit Figure 5: Launch dates for selected OGS and clean cooking end-user subsidy programs34 EnDev/RREA/WB DSS Nepal’s Rural Energy Policy. (Launched in Yemen Rwanda REF Liberia 2006 and updated in 2009, 2013 and 2016) Window 5 EEAP EnDev DSS Malawi Rwanda Clean Rwanda Pro- Cooking Fund EnDev DSS Uganda Bangladesh IDCOL Togo CIZO Poor RBF 2002 2006 2016 2017 2018 2019 2020 2021 2022 2023 2024 South Africa Free Kenya Energy and Cash Malawi NNNF Basic Electricity Plus initiative I Accessing Markets through Private Enterprises for Kenya Energy and Uganda EASP Refugees’ Energy access (AMPERE) Cash Plus initiative II Pilot Nigeria Electrification Bangladesh Energy Access to Full program implementation Project (NEP) Modernization (BEAM) Fund 1.4 The need to design responsible EUS programs While end-user subsidies are a critical tool to address may be tempted to increase their prices ahead of a affordability, it is important to recognize their potential subsidy program, to increase margins or to compensate risks, such as market distortions, political challenges, and for the costs associated with the subsidy management issues of fairness and transparency. Given that subsidies (administrative costs, delays in reimbursements, etc.). directly influence pricing, they may lead to market distortion, Companies participating may also become reliant on whereby supply and demand for energy solutions are subsidy revenue and find it difficult to operate after negatively affected and hinder the ability of companies to subsidies are removed. sustainably serve customers in the long run. Some of the issues that can arise are35: Another important risk associated with subsidies is the political challenge of removing them. Many developing • On the demand side: subsidies may affect consumers’ countries have decades of experience subsidizing bottled value perception of OGS and clean cooking products and gas for cleaner cooking, for example. When trying to services, reducing their willingness to pay unsubsidized restructure or remove these subsidies, governments have prices. Subsidies may also change consumer behaviors, faced significant opposition.36 altering preferences for products and services depending on whether they are subsidized or not. With regards to perceived fairness and transparency, subsidies that are targeted to certain populations (and thus • On the supply side: companies and products that do not exclude others) or provide different subsidy amounts to participate in the subsidy program will be at a significant different population groups may be economically justified but disadvantage. Companies intending to enter the market, a very sensitive issue from a social perspective. These issues may see a subsidy program as a market entry barrier need to be considered in the design and the communication if they cannot participate in it. In addition, companies of subsidies. 34 End-user subsidies for LPG, which have been deployed in multiple countries in Latin America and Asia over the last decades, have been excluded from this figure. Refer to limitations of this report in section 2.1. 35 These issues are well documented. For example, issues of price increases or companies struggling to self-sustain once subsidies were withdrawn are reported in IIED’s discussion paper Energy for all: Better use of subsidies to achieve impact (chapter 3, demand-side subsidies in Nepal). Similar issues were reported after the subsidy under Nigeria’s output-based fund for SHS was removed (case study in section 3.5). 36 World Bank, ESMAP, 2022. Reforming Subsidies for Bottled Gas: Recent Experience in Developing Countries End-User Subsidies for Energy Access: A toolkit 15 While the risks of subsidies cannot be fully eliminated, of a subsidy program will never be perfect, and learning EUS programs can be designed in a responsible way that and adapting through its implementation is critical. Subsidy minimize and mitigate them. While sometimes challenging levels may need to change in response to changes in market to implement, if designers are mindful of the risks and ‘smart’ dynamics, or the targeting strategy improved if an evaluation in mitigating them, then effective EUS programs can have a reveals that most of the subsidy is being captured by wealthy strong impact and make a major contribution to SDG 7. This individuals. Continuous monitoring, regular evaluations, toolkit provides guidance on the design of ‘responsible’ and a framework for the adaptation of the subsidy need end-user subsidies, i.e. subsidies that are (i) effective in to be built into the design. This toolkit also contains reaching the poor, (ii) efficient in terms of use of limited recommendations for monitoring, evaluation and adaptive public resources, and (iii) that minimize market distortion management. and political challenges. The tools in the report can be used to target subsidies effectively, set the right subsidy This toolkit also acknowledges the trade-offs when levels, and establish suitable monitoring and evaluation designing end-user subsidies. Very targeted subsidies mechanisms to adapt the subsidy design when necessary. designed to be most effective in reaching the poor and efficient in their use of limited public resources may become End-user subsidies are complex and thus may require very complex to implement, increasing administrative more capacity to design and implement than other costs and limiting the speed of rollout and scalability. support mechanisms. Designs need to be well informed Depending on the context and subsidy program objectives, and frequently adapted to the context. Some concepts effectiveness and efficiency need to be balanced with may need to be piloted before they can be taken to simplicity and management costs. The toolkit builds on scale. And linkages with other market-building activities lessons learned from a wide range of EUS programs having need to be carefully considered. Building capacity and considered these trade-offs to make design choices. establishing partnerships is essential for subsidy projects to be successful. It is helpful to establish partnerships with End-user subsidies are a vital tool for achieving SDG 7, government institutions across ministries, departments which the global community has pledged to do by 2030. and agencies (covering energy, financial inclusion, social EUS programs help overcome one of the key barriers protection, and digital development), development partners to electrification and clean cooking – affordability. The with strong presence on the ground, and financiers such as affordability gap is significant and puts hundreds of millions the World Bank. of people, mostly in sub-Saharan Africa, at risk of being left behind. Governments and companies are increasingly It is critical for a feedback mechanism to be embedded in deploying EUS programs to bridge this gap. More well- the design of subsidy programs and to have the capacity designed EUS programs will be needed to achieve SDG 7. to adapt the subsidy design when needed. The design 16 End-User Subsidies for Energy Access: A toolkit 02. Design framework for responsible end-user subsidies 2.1 Introduction This design framework consists of pre-design elements however acknowledges the importance of emerging Fee-for- and design components. It builds on existing literature Service or Electricity-as-a-Service models, and most of the on designing EUS programs, the combined experience of recommendations in this toolkit are also suitable for these End-User Subsidy Lab (EUSL) member organizations37, and models. They are however not cited as prominently. Similarly, knowledge from other stakeholders currently designing and this toolkit is most applicable to subsidies for the purchase of implementing EUS. The toolkit includes considerations of improved and clean cooking devices. Less so for subsidies each of the pre-design elements, design components, and for fuels, such as liquefied petroleum gas (LPG), or the communication, and feedback (see Figure 6 below) that may electricity consumed by electric cooking devices. be used for applying the framework across different markets. Figure 6: Design framework for EUS programs38 2.2 Pre-design elements Design framework The pre-design elements inform the detailed design of an Communication end-user subsidy program. They set important program and market-specific context to support the design of responsible EUS programs. The pre-design elements are grouped into 3. two main categories: conducting market assessments and Delivery determining program goals and resources, as depicted 2. Subsidy 4. Verification below: level 1. 5. Exit Figure 7: Pre-design elements Targeting strategy Pre-design Market assessment elements • Understand the beneficiaries’ context Feedback • Evaluate ongoing government initiatives and national electrification plans This toolkit is primarily focused on subsidies for the • Assess the maturity of the private sector purchase of off-grid solar products and clean cookstoves; • Determine the availability of data it is limited in its application to fuel or electricity Program goals and resources subsidies. Off-grid solar electrification is making the most progress through the selling of devices to end users (on • Define EUS goals cash or credit through mechanisms such as Pay-As-You- • Determing funding amount Go). This toolkit draws primarily from the experience of • Establish the need for and type of pilot necessary subsidizing such sales and is therefore most applicable for • Establish long-term goals and plans for continuity the design of subsidies for product purchase. The report 37 The End-user Subsidy Lab is integrated by ESMAP/Lighting Global, EnDev, GOGLA and the Clean Cooking Alliance. 38 End-user Subsidy Lab and Open Capital Analysis End-User Subsidies for Energy Access: A toolkit 17 2.2.1 Market assessment • Awareness, perception and interest of the potential and current customers of OGS and clean cooking A good market assessment is a pre-requisite for the products, including regarding different product features design of a subsidy. It should include both demand-side and quality of such products. and supply-side assessments, as well as research on the enabling environment. The market assessment should • Access to consumer finance for OGS and clean cooking provide information on (i) how many people can afford products, whether provided by clean energy companies what product at what price; (ii) what products, services are (through mechanisms such as PAYGo or Energy-as-a- available in the market and at what prices; (iii) government Service) or financial institutions. and development-partner plans and initiatives for energy access; and (iv) other context factors, such as access to • Factors which may constrain subsidies’ reach to target consumer finance and digital infrastructure. groups, for example, having an official identification document, a fixed address or a mobile phone. Subsidies Understand the beneficiaries’ context39 can be designed to minimize the impacts of these factors, and also to set realistic expectations of how many are A key starting point of the program design is ascertaining likely to be reached through a subsidy program40. the access and affordability gaps among the intended beneficiaries. Understanding the current energy landscape For this part of the market assessment, designers involves an assessment of intended beneficiaries’ specific may leverage available knowledge in the market and needs, preferences, and constraints around energy access. complement this with targeted research. Available sources Beyond this, it is important to assess any social, economic, may include socio-economic censuses, poverty mapping, cultural, and environmental factors that drive the affordability MTF country-specific surveys where available41, and other gap and may impact the program design. A good research conducted by the government or development understanding of beneficiaries’ context enables designers partners. Complementary research may include household to develop a cost-effective, accurately targeted program that surveys for a sample representative of the intended helps minimize risks of market distortion. beneficiaries, or other mechanisms to estimate socio- economic indicators described in section 2.4.2. Important aspects to cover in the market assessment regarding the beneficiaries’ context are: Market assessments require additional and proactive efforts to reach marginalized populations living in remote • The number of households that could benefit from areas or settings of fragility, conflict and violence (FCV). OGS and clean cooking products, by providing These are populations with extremely low affordability, estimates of market size (existing and potential) and who are likely to face multiple challenges to benefit from a mapping areas that could be economically appropriate. subsidy program. For example, in refugee settings people This assessment can be guided by government plans and may not have a permanent address, have limited access least-cost electrification analysis. to government ID, limited access to products and services, and further challenges. In Mercy Corps’ AMPERE program • Segmentation of that potential market from a in Uganda42, substantial market research was conducted geographical, demographical, socio-economical and prior to the design of a subsidy, to understand not only behavioral point of view. affordability barriers but all other constraints to accessing clean energy products (such as product appropriateness and • Ability and willingness to pay for energy products and availability, consumer awareness, market linkages and last services, for each of these segments. Guidance on mile distribution infrastructure). The AMPERE case study is methodologies to estimate the ability and willingness to described in more detail in section 2.5.1. pay is provided in this toolkit in section 2.4.2. 39 WFP 2022, Clean and Modern Energy for Cooking: A Path to Food Security and Sustainable Development. 40 Practical Action, 2023. Can market mechanisms enable energy access for people living in extreme poverty? 41 Multi-Tier Framework (MTF) Surveys 42 AMPERE: Accessing Markets through Private Sector Enterprises for Refugees Energy. The results of their market analysis can be found here. 18 End-User Subsidies for Energy Access: A toolkit Understand the maturity of the private sector Subsidy programs are advised to align with national electrification strategies or national electrification plans Programs should make sure they have a clear (NEPs). Governments develop such plans to guide their understanding of the private sector market within the approach to achieving universal energy access. As of 2017, targeted regions. Program implementors may benefit from 77 countries in regions with significant energy access gaps understanding which players are active in the country, their had outlined the role of OGS in their NEPs. Where these products and services, pricing, and their current traction. It is plans are available, end-user subsidies need to be designed important to leverage the knowledge, capacity, and interests to feed into these strategies.43 For example, in the case of of active and interested companies during program design. Rwanda’s Renewable Energy Fund (REF), a subsidy for OGS This will enable the subsidy program to reach those currently products was designed specifically for areas identified for not served by these companies in a responsible manner. off-grid electrification in the National Electrification Plan. The Rwanda example is presented in detail in section 3.4. Information to collect on the state of the market includes: In addition to energy-sector initiatives, governments • Characterization of OGS and clean cooking companies. implement various relevant programs across ministries, This will include the analysis of international and local departments and agencies, such as poverty safety nets, companies and their market shares, business models, climate vulnerability support, and many others. Energy financing strategies, manufacturing and import strategies, subsidies could be paired with these types of programs etc. As part of this analysis, assess leading international to better complement the support. For example, Kenya’s companies present in the region but not in the country, Mwangaza Mashinani Program (MMP), an end-user subsidy and their interest and capacity to enter the country. scheme targeting the poorest households, was integrated in Kenya’s National Safety Net Program, a cash transfer system • Products available in the market, including quality- implemented by the government and supported by various verified and non-quality-verified products. funders and partners. This kind of integration with other government-led activities in the targeted area increases the • Companies’ existing distribution channels and chances of long-term success.44 geographical coverage, to understand to what extent clean energy products are available for the population, including in hard-to-reach areas. Understand the availability of data for targeted subsidies • Cost structure and pricing of OGS and clean cooking products and services, for sales made on cash, PAYGo, It is important to understand what data is available to or energy-as-a-service packages. inform the EUS program design. Program designers need data to identify the target population and set subsidy levels, facilitate companies to identify eligible customers, and Understand ongoing government initiatives and verify sales. Designers may leverage data available from national electrification plans government repositories or related programs. If not available, designers may need to conduct data gathering or consider It is important to align the subsidy program with existing alternative data approaches, such as using proxy data to government initiatives. The government or other players assess poverty levels. in the market may be implementing several market-building initiatives to promote the sector’s development, such as Data may be available (or generated) at the household supply-side financing or adopting favorable policies. An EUS level. Household-level data from censuses or social- program will be more effective when it complements other protection programs may be useful for economic existing initiatives in the market. targeting of subsidies, provided it is regularly updated and comprehensive. Such is the case of Rwanda’s Ubudehe program, used for subsidies designed under REF45. 43 World Bank 2022, Designing Public Funding Mechanisms in the off-grid Solar Sector 44 End-user Subsidy Lab, Case study: Mwangaza Mashinani Program 45 A detailed case study on Rwanda REF Windows 5 is available in section 3.4. End-User Subsidies for Energy Access: A toolkit 19 Providing higher-tier energy products to a target In Malawi, EnDev used the Unified Beneficiary Registry (UBR) population – a government database that classifies households into five income categories – to inform the targeting approach of the Providing energy access to a population in a poor or EUS pilot.46 vulnerable context (for example, extreme poverty or refugee setting) Where adequate household-level data is not available, geographic and demographic data can be a good Beyond the primary goal, designers may have secondary alternative. Poverty maps, for example, may have been goals for their EUS program. Additional goals may include developed by social protection programs to identify the supporting specific technologies, reducing the ongoing costs specific communities in which poor people live. For example, of energy consumption, increasing gender or other social in Nigeria, only 1.6 percent of households are enrolled in inclusion criteria, or other considerations. the National Social Safety Net Project (NASSP), the country’s flagship social protection program, and the most recent census dates back to 200647. To provide up-to-date and Determine funding amount granular poverty estimates for Nigeria, a high-resolution poverty map has been constructed on the basis of geo- The funds available for the subsidy program depend located household surveys, satellite imagery and other on government budgets or donor funding. The available sources of geospatial data48. funding influences all design components, including which beneficiaries can be targeted, the product selection, the Where no comprehensive social protection systems or subsidy levels, and how verification is conducted. population data sources are available, an energy subsidy program has different alternatives for its targeting Program designers ought to identify funding needs, strategy. These are covered under targeting mechanisms, in sources, and potential funders. Each of these elements section 2.3. may be refined when the components are fully designed. However, it is important for program goals to be set and communication with key stakeholders to be done with 2.2.2 Program goals and resources the available funding in mind to avoid setting the wrong expectations. Programs should avoid running out of money Define EUS goals unexpectedly, which can send beneficiaries back into energy poverty, reduce trust in future programs, in government, Setting clear and specific goals complementary to other energy enterprises, and more, thus reducing future program initiatives within each market is necessary. EUS goals are efficacy. If funding is expected to be raised in tranches, important as they signal the government’s commitment to the programs can be designed in phases. program, serve as guiding principles throughout the project, and inform all decisions. These goals may facilitate first-time connections or support the ongoing adoption of OGS or Establish the need for and type of pilot necessary clean cooking solutions. In the case of clean cooking, for instance, facilitating first-time access to liquefied petroleum A pilot project will be helpful to test the design and inform gas (LPG) requires ensuring that customers have a gas the full-scale rollout of the subsidy scheme. A pilot helps cylinder, an appropriate stove, or a burner. On the other designers test their design choices and assess the program’s hand, facilitating ongoing consumption involves supporting impact on the market, both for consumers and companies. households to refill LPG at a lower cost.49 A pilot may also help ensure alignment between key stakeholders, such as government and private companies, Examples of subsidy program goals include: and allow the refinement of key processes, such as subsidy disbursement and verification. Feedback gathered during Providing first-time access to a target population 46 EnDev 2023, Demand-side subsidy pilot – Malawi: Concept note. This case study is discussed in more detail in section 3.2. 47 At the time of writing, the 2023 census of Nigeria had not been completed. 48 World Bank blogs, 2021. Using Big Data and machine learning to locate the poor in Nigeria. 49 Sustainable Energy for All (SEforALL) 2020, Energy safety nets: A guide for policy makers 20 End-User Subsidies for Energy Access: A toolkit this pilot stage may inform the full-scale project rollout and In the pre-design phase, designers help convince funders that the program is efficient and should ask themselves: What can responsible in its design. For example, in Rwanda, EnDev we accomplish now? What would a piloted the Pro-Poor RBF subsidy scheme50, which informed a subsequent government-implemented national roll out of transition to a next phase have to look the program under the REF51. like? And how can we prepare today to ensure the next phase comes on stream without delay that could cause Establish long-term goals and plans for continuity gaps? When introducing complex and risky interventions such as end-user subsidies, it is important to plan beyond a 2.3 Eligible Households and Targeting single project. This is key in ensuring the government can Mechanisms ultimately reach its goal, but also in ensuring participating companies know what to expect. A big risk to the market is a situation where a subsidy abruptly stops as there is no Focus of adequate plan to bridge subsidy from one project cycle this section 3. Delivery to the next. In the pre-design phase, designers should ask themselves: What can we accomplish now? What would a transition to a next phase have to look like? And how can we 2. Subsidy level 4. Verification prepare today to ensure the next phase comes on stream without delay that could cause gaps? 5. Exit strategy 1. Targeting ©GIZ EnDev 50 End-User Subsidy Lab 2022, Case study: EnDev’s Pro-poor Results Based Financing in Rwanda 51 A detailed case study of the REF Window 5 subsidy is available in section 3.4 of this report. End-User Subsidies for Energy Access: A toolkit 21 The first step in the design framework is to select the Figure 8: Target population, eligible households, and actual targeting approach. This is the process of identifying which beneficiaries Eligible households individuals qualify for subsidies. For EUS programs, the target (potential beneficiaries) Total population populations are those who cannot otherwise afford the As defined by eligibility All households product or service without a subsidy. However, identifying criteria. For example, without electricity or households in certain this population can be difficult in practice. Targeting is the clean cooking hard-to-reach geographic process of directing subsidies to specific population groups, areas, if subsidy is targeted Target population by defining criteria that make them eligible. Targeting may geographically (intended beneficiaries) be more or less accurate depending on how closely eligible People who cannot households match the target population (see Figure 8): afford the product or service Actual beneficiaries People that buy or • Targeting is said to be accurate if all or most of the subscribe to the subsidies reach those who need it (i.e., the target subsidized product or Exclusion error population) People that are part of the service, thus capturing target population but are the subsidy • Targeting is not accurate if a significant portion of the unable to meet eligibility Inclusion error subsidies do not reach the target population, or there is criteria. For example, not able People that captured the to produce a proof of address subsidy that are not part a significant loss or leakage of subsidies to other groups. that may be required of the target population Having people who don’t need subsidies but do receive them is also known as inclusion error. • It is also important to highlight that targeting mechanisms Designers may use one of several targeting approaches may result in exclusion errors (people who need subsidies (untargeted, self-targeting, geographic targeting, but don’t receive them), due to not being able to meet demographic targeting, and economic targeting). These eligibility criteria. range from less targeted to more targeted, as shown in Figure 9. Figure 9: Illustrative decision options based on the level of targeting52 Untargeted Self-targeting Geographic Demographic Economic Less targeted More targeted (less precise) (more precise) Does not define beneficiaries based on specific data points but Utilizes income or expenditure or proxy rather rely on implicit characteristics (e.g assumed product data with strong correlation to ability to pay preference of individuals who are unable to afford) or other to identify individuals who are unable to assumptions to identify individuals who are unable to afford afford energy product or services. energy product or services. Untargeted: No specific criteria are required for selecting beneficiaries. As such, anyone can theoretically benefit from the subsidy program. Self-targeting: The subsidy criteria naturally favor a certain segment of the population that correlates with the intended beneficiaries. For example, a subsidy may be linked to a low-priced energy product popular among the most vulnerable people in a society. If more affluent populations do not widely use the product, they may “self-select” out of the subsidy. 52 Demographic targeting is not always more precise than geographic targeting but depends on the degree of correlation of the geographic or demographic group to the intended beneficiaries 22 End-User Subsidies for Energy Access: A toolkit Geographic: Beneficiaries are selected based on their location in a geographic area that program designers believe predominantly consists of the targeted population. This area is defined using an administrative boundary (a district, county, village, or region), a climatic zone, a settlement type, or the distance from regions well-covered by energy solutions.53 Demographic: In some cases, the intended beneficiaries may share some demographic characteristics that enable designers to allocate subsidies based on these. The categories typically used are age, civil status, and gender. Female-headed households, family size, veteran status, ethnicity, and refugee status may also be employed, but this list is not exhaustive.54,55 Economic: This approach defines the target beneficiaries by income level, occupation, energy expenditure, or energy consumption levels. It is often considered the method that most closely correlates with affordability.56 The more targeted an approach is, the more accurately Table 1: Questions to help determine the targeting approach designers can use it to identify the program’s target What proportion of the market is made up of the target population (that is, those who cannot afford an energy population? product without subsidies). The most precise way of targeting is to use data on income or expenditure amount • The target population for end-user subsidy programs are at the household level. In practice, most EUS programs households who might remain unelectrified or without access have taken a geographic approach to targeting. Lack of to clean cooking without a subsidy. household-level expenditure data has rendered more • A low proportion of the target population in a given country targeted approaches unfeasible. Geographic targeting is or region favors a more targeted approach, whereas sometimes combined with self-selection through higher a widespread affordability gap will favor less targeted subsidies for smaller products, and sometimes combined approaches. with untargeted tax exemptions. What data is available and what is its quality and relevance? What mechanisms can be leveraged to complement this There is not one perfect targeting approach. Designers data? may explore what works best, considering the market context, available resources, administrative costs, and the • Designers may retrieve data through national social registries, program objectives. In addition, designers are invited to online demographic data tools, or similar databases. In the absence of this data, designers may leverage data from other consider each approach’s feasibility, accuracy, and cost- programs or mechanisms (e.g., social safety program). effectiveness. Targeting can also exclude some of the very people intended as targets, because they lack the means • In some cases, designers may choose to collect new data (e.g., documentation) needed to demonstrate eligibility or for a program, which can be a lengthy and costly process. because demonstrating eligibility carries costs (monetary, Designers need to weigh the trade-offs of gathering this time). information as they determine the most appropriate targeting approach. It is also possible to use more than one approach. For • When data needed for targeting is not available or unfeasible instance, designers may use geographic targeting to define to generate, designers may opt for less targeted mechanisms. the administrative region where the subsidy will be provided What limitations does the target population have to access and then leverage economic targeting through proxy data targeted subsidies? (for example, household consumption) to identify eligible households within that region. • More targeted subsidies may require individuals to produce documentation (e.g., ID, proof of permanent address, proof of refugee status or female-headed household) to prove eligibility. If a large portion of the target population cannot do this, then less targeted approaches may be warranted. 53 Sustainable Energy for All (SEforALL), 2020, Energy safety nets: A guide for policymakers 54 Sustainable Energy for All (SEforALL), 2020, Energy safety nets: A guide for policymakers 55 World Bank, 2022, Full Report: A New Look at Old Dilemmas: Revisiting Targeting in Social Assistance 56 Sustainable Energy for All (SEforALL), 2020, Energy safety nets: A guide for policymakers End-User Subsidies for Energy Access: A toolkit 23 The diagram in figure Figure 10 provides a summary of the context factors or considerations, pros and cons, of the different targeting mechanisms. Each of these is explained in more detail and with examples in the following sub-sections. Figure 10: Summary of targeting mechanisms Targeting Untargeted Self-targeting Geographic Demographic Economic Subsidies are open Subsidy is accessible Leverages existing Leverages a range Targets households to all, and eligible to all, but designed borders e.g., regions, of demographic data or individuals whose individuals are not to attract the target or districts, where (e.g., age, gender, income levels cannot Description restricted based on population (applying the population face ethnic group, etc.) to afford OGS or clean unique identifiers such it to products they similar affordability target end users cooking products as income levels prefer) challenges • Target population • Same as untargeted • Marked energy • Marked energy • Target population is a large portion of • Subsidy program access and poverty access and poverty is a small portion of the population (e.g., intended to be pro- differences across differences across the population (e.g., Considerations low electrification poor geographies demographic high electrification, rates, widespread • Existing poverty groups low poverty) poverty) maps or data to • Poverty and access • Adequate social • No robust data sets construct them data across different registries / demographic protection systems groups that can be leveraged • Other adequate income / proxy data • Easiest to • Easy and quick to • Simplest and least- • Can promote access • Most accurate administer deploy costly targeted across marginalized targeting method • Quickest to • Lower risk of method to roll out groups, such as (provided suitable Pros deploy and reach leakage and to a high number of refugees or female- targeting systems high number of market distortion beneficiaries headed HH are in place) beneficiaries than untargeted • Compared to • Compared to • Minimizes leakage subsidies untargeted untargeted and market subsidies: more subsidies: more distortion risks accurate and lower accurate and lower risk of market risk of market distortion distortion 24 End-User Subsidies for Energy Access: A toolkit • Highest risk of • More complex • Prone to inclusion • May create social • Highest complexity subsidy leakage subsidy structure errors if eligible tension among and administrative • Population may than untargeted geographic areas demographic costs Cons be unaware of subsidies are large and groups • Slow rollout existence of • Compared to diverse • Prone to exclusion • Prone to exclusion subsidies more targeted • Prone to complexity errors if individuals error if systems like • Highest risk of approaches: higher in eligibility criteria cannot demonstrate national IDs, civil long-term market risks of leakage, and verification, being eligible (e.g., registry, and social distortion market distortion, and exclusion refugee status, or protection are not and political errors, if geographic female-headed HH) comprehensive • Political challenge of pressure to keep areas are narrowly • Complexity results ending untargeted subsidies defined in higher cost and subsidies • Complexity results longer timelines in higher cost and longer timelines 2.3.1 Less targeted approaches Designers may consider less targeted approaches, such needing to provide proof that the product is being used by as untargeted and self-targeted subsidies, when quality the intended beneficiaries. For example, the untargeted data is unavailable or unfeasible to capture or when there is subsidy for SHS provided under the Nigeria Electrification limited correlation between geographic and demographic Project (NEP) was very successful in scaling the off-grid characteristics and the ability to pay. Less targeted solar market, reaching one million households within one approaches may work in nascent markets where a large year. However, the majority of subsidy beneficiaries were in portion of the population has affordability challenges, and the top wealth quintile of the population. Details of the NEP the risks of market distortion are lower. For mature markets, subsidy are provided in section 3.5. self-targeting or untargeted subsidies aimed at the lower-tier solutions may also work to reach those groups of people left Another important risk of untargeted subsidies is the behind who truly cannot afford the available products. political challenge of removing them. For example, many developing countries have provided untargeted subsidies Untargeted subsidies have the potential to reach a for bottled gas for decades to promote clean cooking. broader population, but they also carry the risk of Advances in digital technology and evidence of market allocating support to people who do not need it, which distortions caused by the untargeted subsidies have led can lead to resource inefficiencies and market distortion. some governments to instead use targeted cash transfers, Designers should be careful in selecting situations where slashing illegal diversion and fiscal costs. But the recent an untargeted subsidy is considered. In these cases, rise in fuel prices has put pressure on governments to the risks of market distortion may be outweighed by the reinstate universal price subsidies. A review by ESMAP of lower administrative and verification costs of less targeted the experience of five countries providing subsidies for subsidies.57 This may allow programs to channel these cooking LPG58 reveals difficulties of ending price subsidies costs into additional subsidies, reaching more people. In once started, and the high risk of a return to universal less targeted approaches, it may be enough to verify that a price subsidies despite their well-documented and widely sale has been made and the product is in use rather than acknowledged shortcomings. 57 Africa Clean Energy (ACE) Technical Assistance Facility (TAF) and Open Capital 2020, Demand-Side Subsidies in Off-Grid Solar: A tool for achieving universal energy access and sustainable markets. 58 World Bank, ESMAP, 2022. Reforming Subsidies for Bottled Gas: Recent Experience in Developing Countries. End-User Subsidies for Energy Access: A toolkit 25 One way to minimize the risk of untargeted subsidies is CASE STUDY 1 through self-targeting, where the subsidy is accessible to all but designed to attract the voluntary participation of the Self-targeting in the Energy Access Scale Up Project target population. Limiting the subsidy to specific product (EASP) tiers and varying the level of the subsidy across product tiers is one technique for self-targeting. For instance, a program Program objectives: To increase access to energy for with the goal of providing first-time access to low-income households, commercial enterprises, and public institutions. households in a region with a flourishing off-grid market might opt to limit subsidies to Tier 1 OGS products.59 The Overview: With funding from the World Bank, the Uganda EASP program follows this logic, providing different government of Uganda set up the Energy Access Scale subsidy amounts based on the product tier, with lower- Up Project (EASP) to increase energy access. The financial tier products attracting higher subsidies (see Case study component of the program is implemented by Uganda 1). Higher-income households who can afford an energy Energy Credit Capitalization Company (UECCC), and its goals product are likely to have one already and, therefore, less include the provision of end-user financing for solar home likely to take advantage of the subsidy.60 In this scenario, the and commercial systems, as well as results-based grants to targeting approach and the eligible products (lower-tier) are facilitate access to Tier 1 solar lanterns, clean cooking, and complementary. productive use of energy (PUE).61 Another way to minimize market distortion when using Targeting approach: Self-targeting less targeted approaches is to provide relatively low The consumer subsidy is applied to all customers in all subsidy amounts. One approach is to begin with a lower communities in Uganda equally. The customer pays the subsidy amount, even if the affordability gap is high. This balance of the cost, either in cash or through a credit PAYGo provides a window to monitor how the market reacts mechanism, depending on the terms of the customer’s and to explore other options for making products more preferred energy supply company (ESCO). The lower-tier affordable (for example, tax exemptions). Over time, the SHS and clean cooking equipment benefit from higher subsidy amount may be gradually increased based on percentage contributions from the subsidy compared to feedback from companies or if the intended impact (that is, higher-tier products. an increase in OGS or cleaning cooking access rates) has not been achieved. Taking this approach ensures longer- The program is available to all customers equally, but the term sustainability of commercial markets as it minimizes differing percentage contributions per tier incentivize the over-subsidizing products and sets a more realistic value most vulnerable households more than those purchasing perception for customers. This has been the approach higher-tier products. chosen for the Ngwee Ngwee Ngwee Fund under the Malawi Electricity Access Project (MEAP), which is described Key insight on untargeted approaches: Self- in sub-section 2.4.1. targeting, coupled with appropriate communication, may be an effective way for designers to take Finally, in less targeted approaches, ensuring an advantage of the greater simplicity of less targeted appropriate communication strategy is critical. When approaches while also managing the risk of market beneficiaries are not aware of the extent of the subsidy, distortion. they may undervalue the system and be unwilling to pay the full cost in the future, even if they can afford it. This can be particularly challenging when the subsidy is delivered 2.3.2 More targeted approaches through a company because end-users do not directly Designers may consider more targeted approaches, such engage with the subsidy and therefore, may not know the as geographic, demographic, and economic targeting, product’s actual market price. in scenarios where good data is available and can be correlated to the ability to pay. 59 Given that the principle of self-targeting is based on assumed preferences of the target population, it is important to validate such assumptions. Subsidized products can be labelled as undesirable by more affluent families, so self-selection might work well in these contexts. These families don’t want to be seen with ‘subsidized’ products, which may have a distinct color, branding, etc. On the other hand, this might also affect the desirability for target households, as community perceptions and norms can be very important drivers of behavior. Targeted individuals or families may not want to display ‘subsidized’ products for fear that others will judge their status. 60 Open Capital Advisors analysis 61 Uganda Energy Credit Capitalization Company, 2023, Results Based Finance Manual (Part C of EASP Project Operations Manual) 26 End-User Subsidies for Energy Access: A toolkit Targeted approaches also work in markets with a mix of area is pre-established, this method is relatively easy to individuals who can and cannot afford electricity access implement. Geographical targeting is beneficial when a or access to clean cooking, as these may be more prone large share of the population in the target geography to leakage (i.e., significant inclusion errors). More targeted requires subsidies to afford products. For example, in approaches aim to provide subsidies to a select the World Bank’s KOSAP program, fourteen significantly population segment based on a shared set of factors. underserved counties were identified in Kenya’s north and Subsidies can be directed toward groups by considering northeastern regions, where nearly 70 percent of residents factors such as location, demographics, and income level, lived in poverty and had poor access to basic services.64,65 among others.62 In cases where there is a need for more narrow Given the complexity and higher administrative costs targeting, geographic targeting may be refined using of more targeted mechanisms (whether geographic, various methods. Poverty maps may have been produced demographic or economic targeting), some important by governments to estimate poverty of small areas or considerations include: administrative boundaries, based on household surveys and censuses. More recently, such maps may be produced • Collaboration with social protection programs which through satellite imagery and machine learning algorithms. allow energy subsidy programs to leverage existing With the increased availability of big data through satellite data and delivery mechanisms. This has been the case imagery, mobile phones, or digital content (and appropriate in Rwanda’s REF Windows 5, Endev’s demand-side data protection mechanisms), the applicability of geographic subsidy pilot in Malawi, and Kenya’s Mwangaza Mashinani targeting has increased dramatically. Poverty maps can Program63. be developed based on satellite images by focusing on night lighting or housing characteristics (for example, the • Developing suitable eligibility tools and streamlined type of roof), coupled with geo-located household surveys verification mechanisms will simplify the identification of to confirm patterns of poverty. The Togo Novissi program eligible households and reduce verification costs, such as demonstrates the use of satellite images to map areas of explained in the Rwanda and Malawi case studies. interest (Case study 2).66 Mapping exercises like that of Togo’s Novissi program, layered with electrification data, may • Eligibility and verification requirements need to be be used to target energy subsidies. taken into consideration to minimize possible exclusion errors, if a significant portion of the target population is unable to meet them. CASE STUDY 2 • Speed of rollout and number of beneficiaries reached Geographic and economic targeting in Togo’s Novissi with a given budget will be lower than for untargeted program subsidies, necessitating clear political will for pro-poor subsidy approaches. Overview: In 2020, the government of Togo sought to provide emergency assistance to the most vulnerable households. Collaborating with a team of researchers from the Center for Effective Global Action at UC Berkeley and Innovations for Poverty Action, the government initiated the “Novissi” program using a contactless, digital system to Geographic targeting utilizes existing borders such as transfer a monthly stipend of $15 directly to beneficiaries’ regions, districts, or neighborhoods to target beneficiaries mobile phones.67 with similar affordability challenges. Since the geographic 62 Sustainable Energy for All (SEforALL) 2020, Energy safety nets: A guide for policy makers 63 More information on these projects is available in sections 3.4, 3.2, and the End-User Subsidy Lab website, respectively. 64 Lighting Global 2018, Kenya Off-Grid Solar Access Project for Underserved Counties 65 Norken International 2017, Kenya Off-grid Solar Access Project (KOSAP) for 14 underserved counties: Social assessment report 66 World Bank, 2022, Full Report: A New Look at Old Dilemmas: Revisiting Targeting in Social Assistance 67 Innovations for Poverty Action, Using Mobile Phone and Satellite Data to Target Emergency Cash Transfers in Togo End-User Subsidies for Energy Access: A toolkit 27 Targeting approach: Geographic and economic (with phone usage as proxy for income) Accurate targeting of these households posed a significant challenge. To address this, the government engaged a team of researchers, who leveraged Togo’s recently concluded representative household survey, satellite imagery, and mobile phone data. This research team used satellite imagery to map out the geographic locations of the most impoverished villages and leveraged results from the survey to provide ground truth and machine learning to inform expenditure partners. Figure 11: Mapping poor villages using machine learning-enhanced satellite imagery Estimates of wealth were determined for each 2.4km grid cell by applying deep learning to satellite images (left) with estimates of population density (center) to arrive at the 100 poorest cantons (right). Second, in order to target the poorest individuals within the selected geographical areas, the team used mobile phone metadata to identify the individuals within these villages with the greatest need, analyzing their mobile phone usage as a proxy for income68. Figure 12: Image showing results of analyzed mobile phone metadata for target locations The team then trained machine learning algorithms to predict consumption and ‘wealth’ of each mobile phone subscriber based on mobile phone data and surveys. Within the 100 identified cantons (red distribution in the right), those consuming less than $1.25 a day (dashed vertical line) were prioritized for Novissi. 68 For this program, one important concern is exclusions that occur for people who don’t have mobile phones. However, the only way the government could quickly distribute cash en masse during the pandemic was by using mobile money. Based on the research team's analysis of recent nationally-representative household survey data, roughly 90% of households in Togo have at least one mobile phone, which might limit the scope of such exclusions. 28 End-User Subsidies for Energy Access: A toolkit Where such mapping tools are not available, an energy additional subsidy. This led to 128,000 solar product sales to subsidy program may choose to create its own using 570,000 individuals, the most vulnerable within six regions available data. For example, as part of the supply-side in the Lake Zone in Tanzania (see Case study 3).69Together subsidy program “RBF for Rural Market Development with the Rural and Renewable Energy Agency (RREA) and of Off-Grid Solar in Tanzania,” (RBF) EnDev developed a the World Bank, EnDev is also piloting end-user subsidies vulnerability access index (VAI) to determine regions with the in Liberia using locally available data and a similar VAI poorest populations, and sales in these regions received an methodology. CASE STUDY 3 Geographic targeting in EnDev’s Tanzania RBF Program objective: To incentivize OGS suppliers of certified OGS products to serve the most vulnerable households. Overview: The goal of the RBF project in Tanzania was to provide quality off-grid solar products for low-income, rural households. During the second phase of the RBF, SNV developed a vulnerability access index (VAI) to incentivize the distribution of solar products to the most vulnerable regions in Tanzania. It covered six regions of the Lake Zone and three regions of the Central Zone. Approach: Geographic targeting This index combined socioeconomic data (such as population density, electrification rates, energy access, biomass usage, gender equity, child and maternal health, and access to water supply and sanitation) from Tanzania’s National Bureau of Statistics with market performance data (including historical sales) gathered from solar companies. The program scored the regions on a scale of 1 to 5, with the higher-scored regions being the most vulnerable and attracting higher subsidies. Introduction of the VAI led to a shift in distribution patterns to the more vulnerable regions. A total of 128,000 solar products were sold to these regions between January 2019 and September 2020, impacting 570,000 people. Amongst these sales, more vulnerable regions realized a 24% increase in share of total sales compared to periods before the VAI was implemented. Figure 13: Vulnerability Access Index in Tanzania 69 EnDev Nov 2021, The Vulnerability Access Index (VAI) A Pro-Poor Approach to Develop Solar Markets in Rural and Vulnerable Areas of Tanzania End-User Subsidies for Energy Access: A toolkit 29 Key insight on geographic targeting: Geographic CASE STUDY 4 targeting may be the simplest and least costly targeted method to serve a high number of beneficiaries. Demographic targeting in Nepal72,73 Designers may benefit from developing mapping tools that use demographic data and market information from companies participating in the subsidy Overview: The government of Nepal has provided subsidies program, enabling greater accuracy when identifying to accelerate access to renewable energy technologies underserved regions and vulnerable households.70 since the 1970s. In 2006, the government developed the Rural Energy Policy. It updated this policy as Renewable Geographic targeting may come with operational Energy Subsidy Policy in 2009, 2013, 2016 and 2021. The challenges depending on how geographic areas are policy provided guidance to accelerate access to renewable defined. Geographic targeting is prone to significant energy technologies for the most remote and marginalized inclusion errors if eligible geographic areas are large and communities. Over time, the government refined the diverse in terms of household wealth. On the other hand, targeting approach, combining demographic, geographic, narrowly defined geographic areas bring complexity in and economic targeting to better identify and reach implementation; for companies, for the target population, and intended beneficiaries. in terms of verification. The latter is explained in more detail in the verification section (section 2.6). Approach: Demographic, geographic, and economic targeting Demographic targeting uses existing demographic data to The updates in 2013, 2016, and 2021 included additional target end-users. This approach is beneficial when people subsidy amounts linked to target beneficiary groups with affordability challenges share specific demographic (demographic targeting) and remoteness (geographic characteristics, for example, women-led households or targeting). The range of eligible products encompasses refugee status. Similar to geographic targeting, program solar PV systems, solar pumping systems, micro hydro designers need to verify, through a third party, the accuracy operated lift irrigation, biogas systems, and improved cook of the demographic data of targeted beneficiaries. In stoves of various varieties. Up to 90% of the product cost is Uganda’s EASP program, implementors used additional subsidized for the targeted population in marginalized and data from the UNHCR to authenticate beneficiaries residing hard-to-reach communities. in refugee settlements.71 With demographic targeting, there is also a risk of social tensions owing to allegations The latest subsidy policy, from 2021, defined target of discrimination on racial, ethnic, or religious grounds. It is beneficiaries as “women-led households with dependent crucial to (i) consider whether this approach is feasible at all, children, earthquake victims from earthquake-affected and (ii) if it is, develop a communication strategy that explains districts, endangered indigenous communities identified to stakeholders the rationale for selecting the demographic by the government, and Dalit.” In addition, the program set group to target. Lastly, leakage of subsidies may be mitigated geographical parameters of remoteness defined across by combining demographic targeting with other approaches, three levels: very remote, remote, and accessible. such as geographic targeting. Nepal’s Renewable Energy Subsidy Policy has used a mix of demographic targeting Lastly, the policy used local government registries to identify (which identified women-led households, earthquake victims, the most vulnerable people (that is, economic targeting). and endangered indigenous communities) with geographic The progressive changes to targeting helped increase targeting based on level of remoteness (see Case study 4). overall access to energy while helping to reduce market distortion. By 2023, 98% of households in Nepal had access to electricity (95% connected to the grid and 3% to decentralized renewable energy systems). 70 Project 90 by 2030, 2021 ,Have you heard about Free Basic Electricity? 71 UECC, 2023, Result Based Financing Manual 72 Nipunika Perera, Kevin Johnstone, Ben Garside, 2020. Energy for all: Better use of subsidies to achieve impact. Chapter 3: Demand-side subsidies: Lessons from Nepal. 73 End-user subsidy lab webinar: Targeting: Designing subsidies to prioritize the most vulnerable, 12 September 2023. 30 End-User Subsidies for Energy Access: A toolkit beneficiaries receive the product, which can be costly. Key insight on demographic targeting: Demographic Therefore, it is important to ensure that available data can targeting can offer a more targeted approach when a correlation between demographic characteristics be independently verified in an efficient manner. In addition, and affordability is present and when relevant data is without appropriate communication strategies, potential readily available. In addition, demographic targeting customers may find it confusing to determine their eligibility. can complement geographic targeting for a more precise identification of beneficiaries. CASE STUDY 5 Economic targeting in Rwanda Renewable Energy Fund (REF) Window 5 Program objective: To increase electricity access in Rwanda through SHS. Economic targeting identifies beneficiary households or individuals who cannot afford off-grid energy solutions Overview: In 2020, the Rwanda Energy Access and Quality based on income or expenditure data. Due to the direct Improvement Project (EAQIP) together with the Renewable correlation between these data points and affordability, Energy Fund (REF) launched a $30 million RBF subsidy called economic targeting is often considered the most accurate REF Window 5 with a goal to connect 370,000 households. form of targeting, if data is accurate and recent.74,75 In many REF Window 5 builds on learnings from EnDev’s Pro-Poor cases, designers leverage existing economic data from RBF pilot, which looked to accelerate access to off-grid other government social assistance programs. For example, solar products to the poorest households through a subsidy the Rwanda’s Renewable Energy Fund (REF) Window 5 mechanism. 78,79 leverages the government’s “Ubudehe” classifications, which categorize households by socioeconomic factors (see Case Approach: Economic targeting study 5).76,77 This case study is presented in detail in section REF Window 5 aims to benefit the poorest households, 3.4. specifically those in areas where the government has not yet prioritized grid extension. The economic targeting approach However, economic data is often not available and/or previously used was based on the socioeconomic categories difficult to gather, making use of this approach difficult known as “Ubudehe”. in practice. When available, designers can leverage proxy data sources, or data sources which closely correlate Through various survey data on disposable income and with income or expenditure to mitigate this challenge. energy expenditure of households in the target areas, the For example, mobile money usage or current lighting and program estimated end-users’ ability to pay and calculates cooking expenditure, obtained from telecom operators or the affordability gap that informs the subsidy level for each energy companies. However, even where such data exists, Ubudehe category80. companies are often reluctant to share or restricted in doing so by privacy laws. Building on the success of EnDev’s pilot, which provided access to 22,000 low-income households, REF Window 5 Economic targeting also requires extensive monitoring achieved 330,000 connections by August 2023. and verification methods to ensure that the right 74 Africa Clean Energy (ACE) Technical Assistance Facility (TAF) and Open Capital 2020, Demand-Side Subsidies in Off-Grid Solar: A tool for achieving universal energy access and sustainable markets. (Page 24) 75 World Bank, 2022, Full Report: A New Look at Old Dilemmas: Revisiting Targeting in Social Assistance 76 World Bank and Development Bank of Rwanda, 2021, Window 5 Operations Manual 77 End-User Subsidy Lab, Case study: EnDev’s Pro-poor Results Based Financing in Rwanda 78 End-User Subsidies Lab, 2022, End-User Subsidies Lab Official Launch Session: Rwanda End-User Subsidy 79 Gogla, Africa Clean Energy, World Bank, 2022, End-User Subsidies lab Official Launch: Session 1 80 Note: The use of the Ubudehe classification has been suspended in the program, and a new approach is currently in development End-User Subsidies for Energy Access: A toolkit 31 amount and adjusting the subsidy level over time. This Key insight on economic targeting: Economic enables companies to take the next step in reaching slightly targeting may be the most accurate form of targeting to ensure that intended beneficiaries who otherwise lower-income and more rural households since companies cannot afford a product receive a subsidy. However, are unlikely to reach the poorest households from the designers need to assess the availability, reliability, and start. While balancing these complexities, it is important for accuracy of economic data and the ease of verification designers to set a simple subsidy structure that will be easy to determine how feasible it will be to implement this to communicate, understand, and manage. approach. As indicated, the subsidy level is related to the affordability gap, which is a function of the product price and the ability to pay, as shown below. 2.4 Eligible Products and Subsidy Level Price of - Ability = Affordability = > Subsidy Focus of Product to pay gap Level this section 2.4.1 Selection of eligible products and services 3. Delivery An important design choice is to determine which products and services are eligible for the subsidy. To 2. Subsidy level 4. Verification specify eligible products, designers can leverage the tiered frameworks available for both OGS products and clean cooking solutions, which identify different levels of functionality, ranging from Tier 0 to Tier 5.81 Additionally, 1. Targeting 5. Exit strategy product selection may be based on quality standards such as certification by VeraSol for solar home systems and Burn design lab, Centre for Integrated Research and Community Development Uganda, Centre for Research in Energy and Energy Conservation for clean cooking stoves, among Subsidy determination involves selecting the products others.82,83 Beyond products, designers may opt to limit eligible for a subsidy and determining the subsidy level. eligibility to certain business models, for example, around This is the process of defining how much subsidy each the payment model (cash versus PAYGo) or the mode of beneficiary will receive. This choice depends on the price of delivery (ownership, rental, or Energy-as-a-Service). These the product and the amount the beneficiary is able to pay. choices are driven by what the target beneficiaries need, The difference between the two represents the affordability what products are available in the market, and what type gap. Ideally, for EUS programs, the subsidy matches the of products and services the program wants to incentivize. affordability gap. However, this can be difficult in practice, PAYGo products are typically more expensive than products as affordability will vary by beneficiary and product, which bought in cash, given the additional costs associated with means that designers may need to make several choices this model (e.g., PAYGo technology, additional customer regarding subsidy levels. services, payment collection, cost of capital of providing consumer finance) and the risk that a customer may not It is important not to set the subsidy too high to avoid fulfill all payments. However, a program may nevertheless over-subsidizing the selected products, which can lead choose to favor the PAYGo model as it is more affordable for to market distortion and depletion of program resources. end-users by spreading payments over time and allows for Setting a subsidy too low may result in limited adoption, as remote monitoring. beneficiaries may still be unable to pay. A good practice is to gradually phase in subsidies, starting with a lower subsidy 81 ESMAP 2022, Multi-Tier Framework 82 Verasol, Certification process 83 Clean cooking alliance, Regional Testing and Knowledge Centers 32 End-User Subsidies for Energy Access: A toolkit Table 2: Difference in subsidies for product purchase Many of the methodologies in this toolkit may apply to versus subsidizing electricity service or fuels (duration of both retail sales and service models. The principles of targeting, subsidy levels, delivery and verification can be the subsidy) used for both with some adjustments. The crucial difference is in the duration of the subsidy and the exit strategy. Given Product subsidies (linked to retail sales / purchase of service contracts are long term, subsidies should also be clean energy products): Product subsidies are intended to designed in a way that enables a long-term affordable energy reduce the price for customers to purchase and own a product. supply, factoring the cost of replacing devices at the end of For SHS or clean cooking solutions bought in cash, subsidies their lifespan. The exit strategy section (section 2.7) discusses are often provided as a one-off amount. In the case of lease- recurring subsidies in more detail. to-own sales via PAYGo, subsidies are typically provided periodically (with the subsidy reducing both the downpayment For product selection, program designers have a few and recurring payments). Such periodic subsidy payments end choices, summarized in Figure 15 and described in more when the product is fully paid, typically after a period ranging detail below. from 6 months to three years. Notable examples of this are Nigeria’s NEP and Rwanda REF programs, which provided Restricting eligibility to certain products or product subsidies for product purchase through both cash and PAYGo tiers. Under this approach, subsidies are only provided to payment models.84 beneficiaries buying a product with certain specifications Subsidies for electricity or fuel (linked to recurring (e.g., a solar lantern with a specific wattage) or a product service provision / consumption of energy): In contrast, from a particular tier. Often, eligibility is linked to tiers subsidies for electricity (from SHS under a fee-for-service or instead of specific products to allow consumer choice and energy-as-a-service model) or clean cooking fuels involve the market to develop. For example, depending on the ongoing support based on usage and are regularly paid out, program goals, designers may subsidize only lower-tier determined by the beneficiaries’ usage patterns. Subsidies products to reach a larger number of beneficiaries with for clean cooking fuels are aimed at lowering the ongoing the available funding. On the other hand, the project could costs to beneficiaries. The duration of such subsidies may be instead focus on subsidizing higher-tier products to enable unlimited, for as long as customers meet eligibility criteria. For greater energy-level access. For instance, in Malawi’s example, with Indonesia’s LPG subsidy program, beneficiaries Ngwee Ngwee Ngwee Fund, only Tier 1 OGS solutions receive a subsidy each time they refill.85In electricity, examples are eligible for the subsidy. This is inscribed in the logic of include South Africa’s Free Basic Electricity program86, self-selection targeting presented in the previous section. providing discounts to eligible fee-for-service SHS customers. Limiting subsidies to Tier-1 SHS makes the subsidy pro-poor. This report does not provide specific recommendations See Case study 6 below. regarding the suitability of retail sales vs service models. The general recommendation is to adapt the design to the Allowing all products and product tiers to be eligible. context and available models in the market where the subsidy Under this approach, subsidies are provided on all products will be introduced. In markets where both models co-exist offered by companies participating in the program (for and/or are desirable, both can be made eligible for subsidies. further detail, see the section on Subsidy Delivery and Retail sales of SHS and clean cooking solutions through cash Company Selection therein). Allowing all products to be or lease-to-own PAYGo are more widespread, and thus more eligible has the benefit of giving customers the freedom referenced in this toolkit. Electricity supply via fee-for-service to choose their preferred product. It also encourages / energy-as-a-service (EaaS) is however more prevalent competition. This was the approach preferred under under rural electrification public-private partnerships found Nigeria’s NEP for SHS, which is described in detail in section in francophone West Africa, for example. The EaaS model is 3.5. Program designers may still differentiate the subsidy also gaining prominence as a model well-suited to reach the amount by product tier or target group. More on this is poorest87, promoted by companies like Solar Aid in Malawi discussed in the subsidy level section below. and Zambia, and Moon in Senegal and Togo88. 84 Details of these projects are provided in sections 3.4 and 3.5. 85 International Institute for Sustainable Development, Global Subsidies Initiative 2021, LPG Subsidy Reform in Indonesia: Lessons learned from international experience 86 A case study of South Africa’s FBE program is available in section 2.7 Exit Strategy. 87 The rationale for this is: (i) Retail models, even when subsidized, ask the poorest people to make an investment decision. Investment decisions inherently involve risk which act as a barrier to access. (ii) All products have a lifespan, whereas energy service contracts can be long-term, with the replacement of equipment when needed. 88 Solar Aid: Light A Village. Moon – public private partnership to provide electricity as a service for last-mile rural households. See p. 120 of IEA’s Financing Clean Energy in Africa report (2023). End-User Subsidies for Energy Access: A toolkit 33 In addition to the customer choice and subsidy cost- requiring product verification by Burn Design Lab.89,90,91 effectiveness considerations above, it is also important to Beyond using these standards, or if they cannot be used, acknowledge that subsidies can entice companies to offer programs may put their own requirements on products or different products or services. For example, under Nigeria’s perform more stringent due diligence on companies and NEP, making subsidies available for entry-level products, their products, as part of the selection process. attracted several solar energy companies to this market segment. In this sense, subsidies have ‘market shaping’ Service standards: Service level standards can define power. a certain warranty period, consumer protection plans, or require accessibility of after-sales services in the In all cases, it is important for subsidy programs to set program’s geographic location. Minimum standards minimum standards for product quality and after-sales can be based on local government standards, industry service. standards, or both. Program designers should implement a Grievance Redress Mechanism (GRM) where customers Product quality standards: Setting quality standards can report instances where companies have failed to can help to minimize repairs or replacement of the offer required services within a stipulated period. subsidized product. Examples of leveraging standards include requiring OGS products to meet the International Electrotechnical Commission (IEC) standards and requiring product verification by VeraSol or requiring clean cooking products to meet ISO 19867 standard and ©GIZ EnDev 89 End-User Subsidy Lab, Case study: EnDev’s Pro-poor Results Based Financing in Rwanda 90 End-User Subsidy Lab, Case study: Togo CIZO Check Program 91 International Standards Organization, ISO 19867-1:2018 Clean cookstoves and clean cooking solutions 34 End-User Subsidies for Energy Access: A toolkit Figure 14: Summary of product selection considerations Product selection Limit to specific product or Tier Open to all product types Eligibility is restricted to a narrow set of products meeting All products offered by companies participating in the Considerations Description certain specifications or from a certain tier. For example, program, and meeting minimum eligibility criteria, can Tier-1 SHS providing basic energy access only. benefit from the subsidy. • When limiting to entry-level products (eg, Tier-1 SHS or • Customers have more freedom to choose preferred Tier-2 and Tier-3 ICS), subsidy can be considered pro- product, and encourages market development poor (self-selection) and help conserve limited resources • If subsidy is untargeted, opening to the whole product (more people reached for a given budget) range (low and high tiers) can lead to more leakage • Compatible with self-targeting / pro-poor subsidies if subsidy levels are higher for entry-level products • For all products and services, it is important for subsidy programs to set minimum standards for product quality and after-sales service. • All types of delivery models can be made eligible: product sales on cash or lease-to-own PAYGo, rental, or Energy-as-a- Service, depending on the context CASE STUDY 6 SHS subsidies provided by the Ngwee Ngwee Ngwee Fund under the Malawi Electricity Access Project (MEAP) Program objectives: To increase electricity access among the off-grid population in Malawi through SHS. Overview: Malawi’s electricity access rate in 2023 was estimated at 19% (13% through connections to the grid, and 6% through SHS) with severe disparities between urban (42%) and rural areas (5%). The government set the goal to achieve universal energy access by 2030, with SHS playing a major role92. Under the World-Bank funded Malawi Electricity Access Project (MEAP), the Ngwee Ngwee Ngwee Fund (NNNF) was set up to support the scaling of the OGS market in Malawi’s rural areas. It has a $20 million financing window that includes working capital financing to companies and RBF grants as end-user subsidies, with a goal of providing access to 200,000 households.93 The OGS market in Malawi is emerging, with a few international and local companies with well-developed distribution networks and expanding their operations. However, with widespread poverty in Malawi (71% of the population living under the poverty line94 ), the market for SHS at unsubsidized prices is likely to saturate quickly, justifying the end-user subsidies provided by the NNNF. Unsubsidized prices of Tier-1 SHS are unaffordable for about 80% of the off-grid population, based on assessment of their ability to pay (see Figure 15). For this reason, subsidies under the NNNF are mainly untargeted, but with some geographic and self-targeting elements (only rural households are eligible, only Tier-1 SHS are eligible for subsidies). Product selection: Limited to specific Tiers. As mentioned above, the program is intended to remain pro-poor and is therefore limited to basic Tier-1 SHS (as defined under the multi-tier framework). This is inscribed in the logic of self-targeting of subsidies described in section 2.3. It also ensures cost-effectiveness for the limited funding available. 92 World Bank, 2023. Malawi Economic Monitor - Powering Malawi's Growth: Rapidly and Sustainably Increasing Energy Access. 93 IDCOL, Off-Grid fund launched in Malawi with support from IDCOL 94 The proportion of people living on less than US$2.15 Purchasing Power Parity (PPP) per capita a day was 71.3% in 2022. World Bank, 2023. Malawi Economic Monitor - Powering Malawi’s Growth: Rapidly and Sustainably Increasing Energy Access. End-User Subsidies for Energy Access: A toolkit 35 Affordability gap and subsidy level: an analysis of Key insights on product selection:96 Choose high- ability to pay of the off-grid population suggests that only quality products that are delivered with good after-sale the wealthiest 20% (quintile 5 in Figure 15) can afford service and certified by an independent party, such as unsubsidized prices. This is a market segment quickly VeraSol for OGS or Burn Design lab for clean cooking saturating. In order for the market to continue to expand, stoves.97 a subsidy of $20 per unit was estimated to suffice for PAYGo pricing to be reduced to what the next quintile of the population can afford. This subsidy level is relatively 2.4.2 Ability to pay and affordability gap low, representing about 15% of the PAYGo price (sum of all payments) of a Tier-1 SHS in Malawi. The next element in determining the subsidy level is to estimate the affordability gap, based on the price of Figure 15: Affordability of SHS for the off-grid eligible products and how much beneficiaries can pay for population in Malawi, by quintile95 them. Average monthly household The product price used in this calculation is typically the Weighted average price of lighting/ energy eligible PAYGo package for sales-volume-weighted average price of eligible products spend per Tier - 1 SHS ($5.60/month) quintile ($/month) in the market. This can apply to products sold on cash, on 10.6 PAYGo, or fee-for-service packages. Ideally, designers will 4.5 3.7 2.8 1.5 not only look at the cheapest product available but also include higher-priced products and those most popular in the market. For nascent and emerging markets, it is 1.1 1.9 2.8 4.1 also important to compare prices in the local market with 1 2 3 4 5 HH Income quintiles international benchmarks and understand the pricing structure (including manufacturing, shipping, importation, Results: as of April 2024, more than 50,000 households distribution, consumer finance, profit, after-sales service, have benefitted from subsidized SHS, allowing the market etc.). Local prices significantly higher than the benchmark to grow despite deteriorating economic conditions in the may signal the need for supply-side incentives (such as country. supply-side RBF or credit lines) and enabling-environment interventions. See Box 1 in the next subsection for more Piloting targeted subsidies: for products to be affordable details on rolling out supply-side and demand-side subsidies for the poorest among the off-grid population (quintile 1), simultaneously. the amount of the subsidy per unit would have to increase by a factor of 3 to 4. As SHS sales increase over the coming The ability to pay is driven by the amount of household years and the market expands into poorer segments of income that is available for energy expenditures. Another the population, it is expected that subsidies will need to point of reference is ‘willingness to pay’, which is the be increased to overcome the affordability gap and realize maximum price a household is willing to pay for a product or universal electrification targets. Such increases may be service.98 Willingness to pay is more subjective and varies accompanied by more narrow targeting. In anticipation of more among target beneficiaries, typically influenced by the this need, Endev is piloting targeted subsidies in Malawi, value perception of the products involved. making use of Malawi’s social registry, the Unified Beneficiary Registry (UBR). This pilot project is described in detail in To estimate the ability or willingness to pay for energy section 3.2. access solutions, designers can use data from existing surveys on household income and expenditure or conduct their own. Existing household surveys may come from 95 ECA, MARGE, 2019. Off-grid solar market assessment in Malawi. 96 End-User Subsidy Lab learnings from partner organizations 97 Clean cooking alliance, Regional Testing and Knowledge Centers 98 Harvard Business School 2020, Willingness to Pay: What it is & How to calculate 36 End-User Subsidies for Energy Access: A toolkit government entities, NGOs, or other organizations aiming choices in purchasing energy and anticipates a slight to determine disposable income levels in the target region, increase in spending for a higher-tier service. Revealed often to inform social protection programs. Such surveys willingness to pay can be more effective than stated may be explicit about energy expenditure. If this is not the willingness to pay since actual consumer behavior does not case, assumptions can be made about the proportion of their always align with what people say they will do. global expenditure they can afford to spend on energy.99 If data is missing, incomplete, or outdated, designers may The affordability gap is calculated based on the product conduct their own surveys on ability or willingness to pay. price and ability to pay. This calculation will always be an For example, for the Malawi case study presented above, estimate due to data constraints and may change depending ability to pay was estimated based on household energy on macroeconomic factors affecting households’ disposable expenditure reported in integrated household surveys income (e.g., depreciation of local currency, food price conducted by the government, combined with surveys increases). Other factors may also affect the affordability specifically conducted for the design of the project. In gap, such as price changes due to technological changes designing a pilot EUS project for Uganda’s refugee settings and sector maturity. As such, the affordability gap requires (see details in section 3.3), EnDev conducted a “willingness regular monitoring throughout implementation. If this leads to pay” field survey in three settlements to complement to revisions of subsidy levels, these need to be carefully other available data points.100 EnDev typically uses ‘revealed’ planned and communicated in alignment with the companies willingness to pay, which is based on consumers’ past involved. © Netherlands Enterprise Agency 99 For OGS products, the OGS market trends report proposes an affordability estimation which assumes that households can allocate 5% of their monthly expenditure to OGS products (between 5% and 10% is considered affordability at a stretch). This 5% of monthly household expenditure is also in line with the Multi-Tier Framework definition of affordability for electricity access and clean cooking. 100 EnDev, 2023, Demand-side subsidy pilot – Malawi End-User Subsidies for Energy Access: A toolkit 37 2.4.3 Subsidy level identified, such as mentioned above for Rwanda REF Subsidy levels are set to match the affordability gap of the using Ubudehe categories or EnDev’s pilot using the target population, they therefore depend on the targeting wealth categories of the Unified Beneficiary Registry. approach. In general, the more targeted the subsidy is Other options include higher subsidies for marginalized towards poor populations, the higher the subsidy level. groups, such as women-led households or refugees. In the case of untargeted subsidies, like the Malawi subsidy presented above (see case study 6) or the subsidy under By product range (Tier): This option may be used Nigeria’s NEP (see section 3.5), end-user subsidies were set with less targeted approaches. Following the logic of relatively low (to about 15-20 percent of price) with a logic of self-targeting, higher subsidies can be given to lower- allowing markets to continue to expand despite widespread tier products, thus encouraging higher uptake by the affordability challenges. Keeping untargeted subsidies low beneficiaries with the lowest ability to pay. This subsidy helps reduce leakage, the risk of market distortion, and the level differentiation is relatively simple to communicate risk of political challenges when removing such subsidies. and manage, as product categories are clearly defined. On the other hand, targeted pro-poor subsidies like This option was chosen by Uganda EASP, varying Rwanda’s REF or EnDev’s pilot in Malawi (see sections 3.4 subsidy levels by tier and technology for both OGS and and 3.2 respectively) subsidy levels were set high (ranging clean cooking products101. from about 45 percent to 90 percent of price) to meet the affordability challenges of much more narrowly defined By payment model or mode of delivery: This option target populations. may be used to incentivize certain payment models or compensate for price differences. Designers may vary Designers can further vary the subsidy level by target the subsidy by payment model (such as cash versus group, product range, or payment model – or they can PAYGo) or by mode of delivery of the subsidized product choose to keep the subsidy level consistent. The different (such as ownership, rental, or Energy-as-a-Service). For options for varying subsidy levels are: example, designers may set a higher subsidy level for products sold through PAYGo to incentivize this model By target group: This may be used with a more targeted because it increases affordability and allows for remote approach in which different target groups can be easily monitoring. Box 1: Rolling out end-user subsidies simultaneously with supply-side incentives or enabling-environment initiatives102,103 Another factor to consider when setting subsidy levels is whether the end-user subsidy is being rolled out in parallel to supply- side incentives and enabling environment initiatives. This is especially relevant for nascent and emerging markets, where prices are high (in comparison with those of mature markets), distribution networks are not as well developed (poorer and remote areas not yet covered), and the policy and regulatory framework is not yet favorable for clean energy products and services. Projects like Nigeria NEP and Uganda EASP rolled out supply-side RBF simultaneously with end-user subsides, so that companies would have an incentive to expand their distribution networks. Projects like Malawi MEAP provided a line of credit to companies, potentially reducing the cost of capital for them. Enabling-environment measures like removing taxes or import duties for clean energy products may also translate into price reductions which need to be considered when setting the subsidy level. As a general rule, supply-side incentives and enabling environment initiatives (which unlike end-user subsidies do not mandate price reductions) should be used to (i) help reduce general prices to levels reasonably similar to those of mature markets, and (ii) help products be accessible throughout the country or region, thus removing the access gap (refer to discussion in Chapter 1). End-user subsidies should be used to address the affordability gap in relation to the reasonable price level. For example: Uganda is considered a mature clean energy market, but with a significant access gap in hard-to-reach areas, such as refugee-hosting and surrounding rural districts. In addition to an end-user subsidy, Uganda EASP provides a supply-side RBF 101 UECCC, 2023. EASP Project Operations Manual. 102 Refer back to discussion in Chapter 1 103 World Bank, 2022. Designing Public Funding Mechanisms in the Off-Grid Solar Sector. 38 End-User Subsidies for Energy Access: A toolkit to incentivize companies to distribute and service eligible OGS and clean cooking products in such areas104. The amount of the supply- side RBF is calibrated to offset the additional cost companies incur to serve hard-to-reach areas, so that products can be sold at the same price as in the rest of the country. The end-user subsidy amount is calibrated to address the affordability gap in relation to that reference price. For additional guidance on the complementary use of end-user subsidies and other forms of incentives, the World Bank has produced a toolkit on “Designing Public Funding Mechanisms in the Off-Grid Solar Sector”. 105 Once the principles of setting subsidy levels are defined, the actual subsidy level can be defined as a percentage of the product price, a fixed amount, or a combination of both.106 These options are summarized in the diagram of Figure 16 and explained in more detail below. Figure 16: Subsidy level options Subsidy level determination Subsidy level principles Actual subsidy levels Invariable Vary by target Vary by product Fixed / absolute Proportional to Combination group Tier amount price or size The subsidy Subsidy level Subsidy level can Subsidy level is set Subsidy set as a Subsidy level is is defined as can adapt to vary for different as a fixed $ amount percentage of price determined as a a unique an affordability of product categories per product and or fixed $ per unit combination of invariable amount each group, / Tiers customer of capacity ($/Wh percentage of price Description for all eligible whether socio- or $/Wp) for the with a maximum products and economic, selected product or fixed amount price customers geographic or service demographic. • Well-suited to • Applicable • Well-suited to • Well-suited to • More subsidy • Helps level untargeted / to targeted self-targeted untargeted / is captured by subsidized self-targeted subsidies only. subsidies, if self-targeted more expensive pricing across subsidies subsidies are subsidies products products Considerations • Subsidy levels • Well-suited to can be set high higher for lower- • Well-suited to • Perverse • Avoids over- relatively low for the poorest tier products relatively low incentive for subsidizing subsidy levels • Much more • Defining tier/ subsidy levels companies to cheaper • Simple to complex to category is • Simple to inflate price products implement implement complex for implement • Not commonly • More complex to clean cooking used without implement • Less prone • Options not mutually exclusive to subsidy subsidy cap • Better suited to • Subsidy can also vary by payment manipulation high, targeted model or mode of delivery (e.g., subsidies cash, lease-to-own PAYGo, EaaS) • Price references can be the sales-volume-weighted average price of eligible products or services in the market. Understanding price structure and international price benchmarks are also an important reference. • Affordability, based on ATP or WTP, can be estimated through economic data, average spend on supplementary products, income data, and surveys 104 UECCC, 2023. Electricity Access Scale Up Project (EASP) documents. 105 World Bank, 2022. Designing Public Funding Mechanisms in the Off-Grid Solar Sector. 106 Africa Clean Energy (ACE) Technical Assistance Facility (TAF) and Open Capita 2020, Demand-Side Subsidies in Off-Grid Solar: A tool for achieving universal energy access and sustainable markets. End-User Subsidies for Energy Access: A toolkit 39 of the product price, with an absolute cap of $50). This way, there is a ceiling to the actual subsidy. Furthermore, the lower of the two amounts is always used, which avoids over-subsidizing larger products and certain companies. The Percentage of price: The subsidy is set as a percentage complexity associated with this method is justifiable when of the product price during the program’s implementation subsidy levels are high, like Rwanda REF or Uganda EASP. (for example, 70 percent of the product price on all eligible products). This would mean a product that costs $100 receives a $70 subsidy, and a higher-tier system that costs CASE STUDY 7 $500 receives a $350 subsidy. Such an approach can result in a large quantity of funding going to higher-priced systems. Combined percentage and fixed subsidy level in Therefore, a percentage subsidy level is almost always Rwanda’s REF Window 5107 accompanied by a cap amount, which is the ‘combined’ subsidy level option described below, or limited to specific Rwanda’s REF Window 5 end-user subsidy provided targeted tiers (e.g., only Tier 1 products). A significant risk associated subsidies to the poorest families living within identified off-grid with this approach is the potential for companies to inflate electrification areas, as described in section 3.4. Subsidy levels product prices to maximize the subsidies they are eligible for. were high to match the affordability gap of the target population, To mitigate this, program designers would need to monitor justifying a more complex calculation of subsidy amounts. country and regional price trends before and during the program. The subsidy amount for each SHS product is set as a percentage of the retail price, but the subsidy is capped at a maximum absolute level. With the Ubudehe categorization, the relative subsidies were 90% , 70%, and 45% for Tier 1, 2, and 3 OGS solutions, respectively. A maximum absolute subsidy was set at Rwandan Francs (FRW)100,000, FRW 80,000, and FRW 50,000 for the three respective categories. The subsidy was then Fixed amount (absolute value): The subsidy is a fixed dollar disbursed in three installments for PAYGo customers and two or local currency (LCY) amount (for example, $50) to cover installments for cash customers. The subsidy categorizations are a portion of the product price. Keeping the amount fixed however currently under review to reflect market dynamics. across all products results in the lowest-priced products attracting the highest relative subsidy levels. This approach Category108 Percentage coverage Maximum subsidy can be considered a self-selection targeting approach, (of final price) (in FRW) as the subsidy is relatively more valuable for lower-priced Ubudehe 1 95% 120,000 products. This approach has been used in Malawi’s NNNF Ubudehe 2 80% 95,000 (one single subsidy level) and Nigeria’s NEP (amount is fixed for each product category), both of which offer relatively low Ubudehe 3 70% 80,000 end-user subsidy levels. If subsidy levels were high, setting fixed amounts may result in over-subsidizing the cheaper By combining an absolute and a relative maximum subsidy products. amount, the program design safeguards against the risk of over- subsidization and price inflation by participating companies. The subsidy levels will be reviewed regularly to reflect changes in market conditions; hence, the subsidy amount may vary over the program’s lifetime. Combination of percentage and absolute value: The Key insight on subsidy level calculations: Subsidy subsidy level is set to both an absolute amount and a structures need to be kept simple where possible. percentage of the product price (for example, 70 percent Complexity may be justifiable for higher subsidy levels, where risks of market distortion increase. 107 Development Bank of Rwanda, 2021. Window 5 Operations Manual, Version 04/23.06.2021. 108 Note: The use of the Ubudehe classification has been suspended in the program, and a new approach is currently in development 40 End-User Subsidies for Energy Access: A toolkit The more complex the subsidy structure, the more difficult it is to communicate and administer. Beneficiaries may find it difficult to understand why certain groups or products attract a higher subsidy level. Subsidy administrative processes such as verification and disbursement are also more complex with differentiated subsidy levels. Designers need to balance the trade-off between being specific and keeping subsidy delivery simple. Box 2: Setting subsidy levels via reverse auction? As an alternative to program designers setting the subsidy level as described in this section, they can ask companies to set the subsidy through a reverse-auction process. In this approach, program designers issue a tender to companies operating in the region with broad guidelines for the target beneficiaries, types of products to be subsidized, and total subsidy funding available. Companies place bids on the subsidy they require for selling a certain volume of OGS or clean cooking products, at a defined price, in a particular area. The winning bidder(s) are the ones needing the lowest subsidies. The main benefit of this approach is the potentially reduced subsidies109 However, reverse action has potential drawbacks, including giving an unfair advantage to larger companies.110 This approach calls for active monitoring during implementation, as companies may set subsidy levels that are not fully aligned with market realities. There is ample experience of reverse auctions to set supply-side RBFs, but not as much for end-user subsidies. The KOSAP project used a reverse auction for their solar RBF component111. The BRILHO programme in Mozambique uses a reverse auction for their clean cooking RBF.112 RBFs provided by the Beyond the Grid Fund for Africa (BGFA) are also set based on reverse auction113. None of these projects are end-user subsidies in the sense that no price reductions are mandated. More research is needed in this area to draw recommendations for auctions specifically for end-user subsidies. 2.5 Delivery of Subsidy Focus of In this section, the toolkit covers how the subsidy is delivered to the this section beneficiary, including the delivery channel, company selection, subsidy disbursement, fund management, and claim management. A well-designed delivery mechanism aims to promote transparency and accountability, encourage stakeholder participation, and maximize customer outcomes.114 The five delivery elements are defined as 3. Delivery follows: Delivery channel – how a subsidy is given to a beneficiary, for 2. Subsidy 4. Verification example, directly through cash or vouchers, or indirectly through a level company Company selection – the process of selecting eligible companies 1. Targeting to participate in the subsidy program 5. Exit strategy Subsidy disbursement – the process by which payments are made, for example to companies selected to deliver a subsidized product or service (when working through companies) 109 MECS and ENERGY 4 IMPACT, 2021. Clean cooking: results-based financing for modern energy cooking solutions: Analysis of a potential scale-up tool for the sector 110 MECS and ENERGY 4 IMPACT, 2021. Clean cooking: results-based financing for modern energy cooking solutions: Analysis of a potential scale-up tool for the sector 111 Kenya Off-grid Solar Access Project. Project website here. 112 MECS and ENERGY 4 IMPACT, 2021. Clean cooking: results-based financing for modern energy cooking solutions: Analysis of a potential scale-up tool for the sector 113 Beyond the Grid Fund for Africa (BGFA). Project website. 114 GOGLA (2021), How End-User Subsidies Can Help Achieve Universal Energy Access: Views From the Off-Grid Solar Industry End-User Subsidies for Energy Access: A toolkit 41 Fund management – the form of administration of the • Time of deployment and scalability: How does program the selected delivery approach impact the time of deployment of the subsidy? Can the approach be scaled Claim management – the process by which companies in line with subsidy program objectives? submit documentation notifying the EUS program of a sale of the subsidized product and requesting reimbursement where relevant 2.5.1 Delivery channels Table 3: Questions to help determine the delivery approach Subsidies can be delivered directly to beneficiaries or across the five elements115 indirectly through a company. Direct delivery of subsidies takes the form of cash or vouchers to offset the cost of • Ease of access: Can the target population easily the product. Beneficiaries may receive cash up front with access the subsidy through the selected delivery the intent that they use it to purchase the product, or they approach? may receive periodic payments in a bank or mobile money account as they make recurring payments for an eligible • Robustness of approach: Is the approach reliable product or service. Alternatively, consumers may receive a (provided effectively and without delays), and can it voucher (electronic or physical) that they can provide to a guarantee delivery of quality products or services? Is company to purchase the product at a reduced price. In this the governance sound, and is there sufficient security to approach, the company needs to apply for reimbursement guarantee the subsidies will not be misused? of the voucher amount to the subsidy program. Finally, • Cost-effectiveness: Do the benefits of the selected the subsidy may be given directly to the company. In this delivery approach justify its complexity and scenario, the company provides a reduced price to eligible administrative cost? consumers and typically receives the subsidy after a sale is made (see the subsidy disbursement section below).116 Figure 17: Delivery channel choices Delivery channel Through companies Direct to beneficiary Subsidy amount is given to the participating company, which Subsidy amount is given to eligible beneficiaries through either in turn sells the product at a discounted price to eligible vouchers or cash transfers (conditional or unconditional). consumers. Description Unconditional cash Conditional cash transfers and transfers (less common in vouchers (more common in energy access) energy access) • Suitable for less targeted subsidies / large eligible • Suitable for more targeted subsidies, ideally with existing population channels to identify beneficiaries and disburse subsidy (e.g., a social protection program) Considerations • Low availability of quality products (need to incentivize and control quality) • Risk of consumers purchasing substandard products is low • Suitable for customers with low access to technology and (most products in the market are high quality) information (customers don’t engage with subsidy directly) • Administrative requirements are compatible with target • Important to support companies deliver to hard-to-reach population customers • Raising consumer awareness is a key aspect of the program 115 World Bank, 2018, Social Protection Payments in the Directorate of Social Protection and Solidarity – Sao Tome and Principe 116 Africa Clean Energy (ACE) Technical Assistance Facility (TAF) and Open Capita 2020, Demand-Side Subsidies in Off-Grid Solar: A tool for achieving universal energy access and sustainable markets. 42 End-User Subsidies for Energy Access: A toolkit • Has lower administrative costs for program • Makes users aware of existence of subsidy and true cost of • Leverages existing distribution and marketing channels product Pros • Supports the delivery of quality products • Provides additional channel for feedback from intended beneficiaries • Can be effectively implemented even in untargeted approaches • Allows EUS designers to reach target population directly (minimizes inclusion error) • Most widely used delivery approach, and thus more substantial experience • Requires awareness of products for beneficiaries to seek • Risk of diverting subsidies (if unconditional subsidies) subsidies • Increased administration costs (for example logistics), Cons • May lead to market distortion as the consumer may not especially with vouchers fully understand subsidization • Requires beneficiaries to be aware of eligible products and • Companies may increase prices to compensate for subsidy companies program management costs • Risk of exclusion error due to administrative requirements • Risk of selling subsidized products to ineligible customers the target population may not be able to meet and claiming for subsidizes Table 4: Factors to help determine delivery channel Scenarios where delivery through a company is approach appropriate: • Targeting approach and size of target population: This Designers may consider delivery through a company based relates to the ease and cost of subsidy delivery. The larger on the following factors: the target population (and the less targeted the subsidy), the more difficult and costly it is to deliver subsidies Targeting approach and size of target population: directly to beneficiaries. • Less targeted subsidies (untargeted, self-targeted, • Availability and consumer awareness of quality products: some delivery approaches allow designers to and, to some extent, geographic) will make a large restrict product quality standards (e.g. delivery through part of the population eligible for the subsidy. The eligible companies or conditional cash transfers and government may lack mechanisms to provide vouchers), while others (e.g., unconditional cash transfers) subsidies directly to such a large population do not. Therefore, the availability and consumers’ (comprehensive population databases or capacity awareness of quality products in the market influences to deal with individual subsidy applications). On the which delivery approach to prioritize to minimize risk for other hand, companies may integrate the subsidy end-users purchasing substandard products. transaction to their existing sales’ processes and claim subsidies to the government in bulk. • Awareness of subsidy: when delivering subsidies through companies (notably untargeted subsidies), consumers • On the other hand, if subsidies are more targeted may not be aware products are subsidized, affecting value and the government already has direct access to the perception, and contributing to market distortions. target population - for example, via a social safety net program - direct subsidies may be considered a • Administrative requirements placed on consumers: suitable option. Levels of literacy (reading, writing, digital literacy, ownership of phone, ownership of bank account, etc.) may Limited availability of quality products: influence the delivery approach. • When penetration of low-quality energy access • Administrative requirements for governments: handling cash transfers and vouchers requires higher products in a given market is high, designers may opt investment from the subsidy administrator. to deliver the subsidy indirectly through companies pre-vetted for quality. This will incentivize companies End-User Subsidies for Energy Access: A toolkit 43 (existing and new entrants) to provide quality companies with significant support. For example, Mercy products. This eliminates the risk that beneficiaries Corps recently partnered with two private companies to will use their cash or voucher to purchase low-quality deliver quality SHS to refugees in Uganda (see case study products. It is however important to highlight that such below). a risk can also be addressed by making cash transfers and vouchers conditional to the purchase of quality products. CASE STUDY 8 Beneficiary characteristics: Subsidy delivery through private companies in refugee settings – Mercy Corps’ AMPERE project in • Delivery through private companies may be more Uganda117,118,119,120,121 appropriate for consumers with poor access to information and technology, because it does not Project objective: To increase access to energy for refugees require customers to handle the subsidy themselves. and their host communities by facilitating the introduction of private sector energy providers. Delivery through companies leverages existing sales’ networks and processes, lowering administrative costs for Overview: Affordability of OGS solutions remains a major the program. It may also support the delivery of quality challenge for refugee populations in Uganda, whose ability products and the transparency of implementation. With to pay for these products is three to four times lower than delivery through companies, subsidies can be delivered in the national average. The Accessing Markets through Private bulk to a few partner companies – as opposed to thousands Sector Enterprises for Refugees Energy (AMPERE) project of individual beneficiaries – requiring lower administrative was designed to incentivize the entry of energy companies costs for logistics and facilitation. RBF is the most common into refugee settlements and to better understand both the delivery tool when using this approach because it market and the ability of potential consumers to pay for the allows program implementors to verify that the intended relevant products. The project was implemented at pilot beneficiaries have received the subsidy. Additionally, RBF scale over one year (from July 2019 to June 2020), in the Bidi schemes allow implementers to steer market development Bidi refugee settlement in West Nile, Uganda. Initial market in the right direction by offering incentives for specific research highlighted key constraints, including affordability achievements, such as program transparency and product and consumer finance, appropriateness of products, quality. consumer awareness, and last-mile distribution infrastructure. The project sought to address these constraints through Because of these advantages, delivery through producing market intelligence, engaging private sector companies is one of the most widely used delivery players through an RBF scheme, leading product awareness approaches. This approach was taken in Rwanda during campaigns, and training sales agents. Endev’s Pro Poor subsidy and REF Window 5, in Uganda during the World Bank’s EASP, in Nigeria during the NEP SHS Subsidy design: the subsidy was targeted to refugees and output-based fund, and during various other World Bank and their host community, made eligible to all residents of the Bidi EnDev programs, among many other examples worldwide. Bidi Refugee Settlement and its host community. The subsidy level was high, at about 50-60% of the price of eligible Delivery through companies may affect how effective products (quality-verified solar lanterns and SHS), to adapt to a subsidy is in reaching poorer and excluded users, so the limited ability to pay of the target population. supporting companies to adopt inclusive modes may be necessary. Box 1 described how supply-side incentives can Delivery channel: Through private companies. Given be deployed simultaneously with subsidies to reach poor the (i) target population was large (all residents of target and remote populations. Some designers have opted for area were eligible) and (ii) that the objective was a scalable subsidy delivery though companies even in challenging approach to developing markets for high-quality OGS contexts such as remote refugee settlements, by providing 117 End-User Subsidies Lab, Case Study: The role of end-user subsidies in strengthening solar markets for refugees in Uganda 118 Response Innovation Lab, D.LIGHT AND VILLAGE POWER: Supporting private sector energy actors to enter refugee markets 119 UOMA, 2020. Reaching unserved refugee markets in Uganda 120 Mercy Corps, 2022, Ensuring Access to Affordable, Reliable, Sustainable and Modern Energy for All 121 USAID Power Africa, , Assessment of Market-Driven Solutions for Energy Access in Refugee Settlements in Sub-Saharan Africa, Master Card Foundation 44 End-User Subsidies for Energy Access: A toolkit price reduction and reasons for the subsidization (e.g., due products and PAYGo consumer finance in refugee settings, to health benefits or government goals related to energy private solar companies were selected to partner with access and climate change). Companies could also be held Mercy Corps to implement the project. The selection was accountable by being obliged to provide this information competitive, and two companies (d.light and Village Power) to beneficiaries through, for example, program-branded were tasked with selling the lanterns SHS at subsidized subsidy certificates delivered together with the product. prices, targeting potential customers through marketing events. A key goal of selecting these companies was to Another challenge of working through companies is that facilitate increased proximity to the beneficiaries. they may not always entirely pass the subsidy on to their customers. Experience of delivering subsidies through Results and lessons learned: By the end of the pilot in companies has shown companies increase retail prices to June 2020, a total of 4,000 products were sold which compensate for the complex or hard to-administer subsidy equates to roughly 9% of refugee households in the Bidi Bidi delivery systems, subsidy program transaction costs, long settlement. The pilot demonstrated that supporting a market procedures and verification delays incurred by them in the systems approach could be successful, even in remote subsidy release process. Therefore, end-users do not access and underserved refugee markets. However, such support clean energy products at the least-cost.123 This challenge can needs to be scaled and regularly adapted to the context to be mitigated through balancing more efficient procedures be sustainable. A two-year follow-up survey continued to on the one hand, and monitoring of end-user prices on the monitor the state of the market. It was determined that as other. a result of COVID, sales stopped, the companies reduced their personnel capacity and system repairs had a tendency to go unmet. Being that AMPERE was a limited pilot without Scenarios where direct delivery to beneficiaries is continued scaling momentum, the cost of serving and appropriate: servicing these markets increased. Two years on, there was a considerably reduced company footprint in the settlement Designers may consider direct delivery to beneficiaries especially for PAYGo systems. based on the following factors: The AMPERE pilot made it apparent that though pilot projects Size of target population: can provide valuable data (such as evidence that refugees can pay for energy services), they will not result in systemic • When there is a smaller target population or a clear change. A long-term outlook with strong partnerships and targeting mechanism that enables easy identification of consistent momentum is essential to long-term gains and beneficiaries, the subsidy can be efficiently distributed a sustainable market approach. Though the subsidy was directly to end-users. Direct delivery requires meant, in part, to reduce the risk to the private sector, solar precise targeting approaches because the program distributors still see significant risks in these markets, and implementor needs to know the exact households to this requires appropriate support to enable them to operate serve. profitably in this market. Availability of products: • When there is a low risk of consumers purchasing A challenge with delivering through companies is that substandard products. beneficiaries may not be aware that the prices they are paying are subsidized.122 This creates a misconception • When there is a wide range of possible products that the product is much cheaper than it is and thus affects available on the markets, subsidies directly to future willingness to pay for the product at commercial rates. beneficiaries provide additional flexibility to potential To mitigate this challenge, it is crucial that designers and beneficiaries to select the product most relevant to implementors adopt a communication plan to ensure the them. target communities are aware of the temporary nature of the Administrative requirements: 122 Africa Clean Energy (ACE) Technical Assistance Facility (TAF) and Open Capita 2020, Demand-Side Subsidies in Off-Grid Solar: A tool for achieving universal energy access and sustainable markets. 123 IIED, 2020. Energy for all: Better use of subsidies to achieve impact. End-User Subsidies for Energy Access: A toolkit 45 • When the administrative requirements are suitable for the target population (e.g., in terms of technology Direct delivery to beneficiaries and literacy), given they will need to understand and requires a clear targeting approach complete the required verification processes, which to identify the beneficiaries to frequently include filling out subsidy request forms or receive the subsidy. sharing feedback. Consumer awareness: and are available in both electronic and physical formats. An advantage of vouchers is that they can incorporate • When a key goal of the program is to ensure advanced security measures, such as watermarks or awareness among beneficiaries of the existence multifactor authentication, to prevent duplication or of subsidies, delivery directly to beneficiaries helps unauthorized resale. awareness development. Cash transfers may be conditional or unconditional. • Can empower the customer by teaching them about Most direct subsidy programs linked to OGS and clean new energy products and their benefits. Some subsidy cooking solutions are delivered as either conditional cash programs explicitly include this kind of consumer transfers or vouchers. Conditional cash transfers imply education as a goal.124 subsidy recipients are subject to meeting certain desirable behaviors. In the case of energy access, such a condition Data collection may be to use the subsidy to pay for clean and high-quality energy products or services (the program’s eligible products • Program designers can opt to deliver subsidies directly and services). Unconditional cash transfers would remove to beneficiaries to collect additional data on the target such a condition, giving the beneficiary the freedom to beneficiaries that can be leveraged for other goals. choose how to spend it. Unconditional cash transfers are i.e., provision of capacity building services in these not common in energy access programs, where there are communities in areas like agriculture or access to often specific objectives linked to beneficiaries acquiring concessional third-party financing for productive use of a certain tier and quality of access to electricity or clean energy (PUE) equipment to bring further benefits. cooking solutions. A conditional cash transfer program needs to be vigilant in announcing it (so that all parties Direct delivery to beneficiaries requires a clear targeting involved understand the rules), monitoring it, and enforcing approach to identify the beneficiaries to receive the it. Vouchers are an alternate format for delivering conditional subsidy. It is important for program designers to accurately cash transfers. They are useful to maintain control of the identify the beneficiaries and provide them with a cash quality of products acquired by linking voucher redemption transfer or a voucher. This is contingent upon the targeting to only qualified suppliers. approach and the quality of data available to allow for accurate targeting. Delivery of subsidies directly to beneficiaries provides a straight communication channel with beneficiaries and Direct delivery to beneficiaries can be done through ensures reach. This is because direct delivery provides an cash transfers or vouchers. Cash transfers may take additional channel for feedback from intended beneficiaries various forms, including physical cash, mobile money, or during the subsidy delivery process. Beneficiaries are also bank transfers. They can be comprised of either a single kept aware of the true price of systems and the existence upfront payment during the initial purchase or smaller of subsidies because they engage directly with product installments paid out alongside verified customer payments, costs. Additionally, direct delivery to beneficiaries allows for example, through a PAYGo model. Vouchers offer an EUS designers to reach the target population directly, thus alternative method for delivering subsidies to beneficiaries minimizing risk of leakage or inclusion error. 124 Stephen Nash and Jo Khinmaung-Moore 2020, Designing Sustainable Subsidies To Accelerate Universal Energy Access A briefing paper on key principles for the design of pro- poor subsidies to meet the goal of sustainable energy for all. 46 End-User Subsidies for Energy Access: A toolkit The effectiveness of direct delivery to consumers may be details on the procedure followed in Bangladesh to select limited by administrative processes that require literacy the partner for delivering the subsidy. and technology access, among other requirements. This is because both cash and vouchers require customers to CASE STUDY 9 engage administratively with the subsidy through digital or written channels. This may carry risks of exclusion error if Direct delivery of subsidies to beneficiaries in customers are unable to do so. For example, beneficiaries Bangladesh’s BEAM Fund might have to submit applications for subsidies or meet routine monitoring requirements, among other administrative requirements. Transfers via mobile money or digital Program Objectives: To test the effectiveness of conditional e-voucher systems may risk leaving out people that do not end-user subsidies and enhance the effectiveness of financing to vulnerable market segments. have a phone/mobile money account. Overview: The clean cooking market in Bangladesh has Cash transfers may limit the risk of price distortion as witnessed significant developments driven by various customers directly interact with the full price of a system, initiatives and partnerships. The Bangladesh Energy Access sometimes even paying the full price upfront and receiving to Modernisation Fund (BEAM Fund) aimed to reach 10,000 cash payments after the fact. Customers are keenly aware people in 2,000 households with clean cooking access of the possible alternative uses of cash, which may lead to and provide a total of €50,000 in end-user subsidies to consumers. This intervention was aligned with the them placing greater value on subsidized energy products. Government of Bangladesh’s priority to promote the practice Without proper control mechanisms, however, there is a of stove stacking to achieve universal access to clean greater risk that households will use the funds for other cooking by 2030.128,129,130 products or services, or low-quality products not intended to be eligible for the program. To limit this risk, designers often Delivery channel: Direct to beneficiary through prefer to deliver cash in smaller amounts alongside recurring conditional cash transfer verified customer payments for the product.125 The subsidy was delivered through a post-purchase cash transfer to beneficiaries through a mobile banking Before opting for cash transfers, designers may want to payment system. The cash transfers were conditional assess the possibility of leveraging existing disbursement on customers acquiring clean cooking equipment. The channels. These include mobile money services or program encouraged suppliers to provide PAYGo options for microfinance institutions.126 Using existing channels is often disbursements such that the upfront cost was significantly preferable because setting up a new delivery channel reduced for potential beneficiaries. that is efficient and robust is administratively costly and, in most cases, impractical. For example, Kenya’s Mwangaza SNV, the implementation partner, conducted a rigorous Mashinani Program leveraged the subsidy disbursement procurement process and registered with Bkash, a mobile channels of the National Safety Net Programme (NSNP)127. financial services provider, to ensure the transparent When existing channels are available, extensive due distribution of subsidies to the intended beneficiaries. Since diligence is recommended to ensure all intended some beneficiaries did not have Bkash accounts, SNV also beneficiaries can access the subsidy. Other factors to supported their registration of accounts to facilitate the consider include the availability of security measures to transfer of the subsidies.131 avoid misuse of the funds and the efficiency of systems to ensure timely disbursements. See Case study 9 below for 125 End-User Subsidy Lab, Case study: Mwangaza Mashinani Program 126 Africa Clean Energy (ACE) Technical Assistance Facility (TAF) and Open Capital 2020, Demand-Side Subsidies in Off-Grid Solar: A tool for achieving universal energy access and sustainable markets. 127 More information on this project is available the End-User Subsidy Lab website. 128 Note: Stove stacking involves using multiple stoves concurrently. The choices in stove and fuel use are influenced by factors such as accessibility and availability of clean cookstoves and fuels, cultural considerations, and environmental influences. These choices are primarily shaped by household income and the costs associated with acquiring and using stoves and fuels, including both initial and recurring expenses. Shankar, A. V., et. al, 2020. Everybody Stacks: Lessons from household energy case studies to inform design principles for clean energy transitions. 129 EnDev, 2023, Bangladesh Energy Access to Modernization, BEAM Fund: Final Project Report 130 SNV, Bangladesh Energy Access to Modernization Fund (BEAM Fund) 131 EnDev, 2022, Bangladesh Energy Access to Modernization, BEAM Fund Final project report End-User Subsidies for Energy Access: A toolkit 47 The voucher approach minimizes the risk of funds being companies participating in the subsidy program in order to used incorrectly since they are designated for the specific redeem their vouchers. As such, communication and raising purpose outlined by the subsidy program. In addition, awareness about the participating companies to the end- when delivering the voucher, the program may utilize the users is critical for the success of an EUS program using opportunity to gather feedback about the program from this approach. Additionally, the voucher system means that the beneficiary. The level of mobile penetration or other companies must make claims to receive reimbursement for relevant technological or financial inclusion indicators among the subsidy. the target beneficiaries plays a role in deciding between electronic and physical voucher delivery methods.132 2.5.2 Company selection However, creating and managing the vouchers requires additional administrative capacity. This administrative When delivering subsidies through companies, the process may result in voucher shortages, under or over- company selection process is an important step for budgeting, and other operational complexities. The voucher designers. The subsidy program can allow all companies approach requires that customers are made aware of the to participate, after screening for certain minimum eligibility requirements, or restrict participation to a limited number of competitively selected companies. While both options The voucher approach minimizes set minimum eligibility criteria, a key difference is that the the risk of funds being used second option is more restrictive in nature and assigns incorrectly since they are ‘lots’ of the subsidy to a few players at the start of the EUS program. The most appropriate approach depends on the designated for the specific purpose program goals, targeting approach, market maturity, and outlined by the subsidy program. administrative cost as key factors. Figure 18: Company selection choices Company selection Open to all companies meeting minimum criteria Restrict through competitive process All companies are eligible for participation provided they Designers issue a competitive call for proposals and select a Description meet minimum criteria on quality, after-sales service, and limited number of companies that are best placed to deliver environmental and social safeguards the subsidy (e.g., that can deliver at scale) • Suitable for large and more mature markets • Suitable for small and nascent markets, where attracting • Effective for growing local markets and encouraging new market entrants is important competition, which helps keep prices low • Can encourages participation of specific company Considerations • Increases transaction costs and delivery costs of multiple segments (e.g., local companies) companies • Leverages economies of scale and minimizes transaction • Frequency of call for proposals: typically on a rolling costs, and delivery risks associated with multiple basis companies • Subsidy allocation: more likely on first-come-first-serve • Increases the risk of market distortion because of limiting basis, as companies make eligible sales company participation and consumer choice • Frequency of call for proposals: typically limited, based on program progress and market developments • Subsidy allocation: more likely to earmark subsidy based on companies’ projected sales and other considerations 132 Africa Clean Energy (ACE) Technical Assistance Facility (TAF) and Open Capital 2020, Demand-Side Subsidies in Off-Grid Solar: A tool for achieving universal energy access and sustainable markets. 48 End-User Subsidies for Energy Access: A toolkit Table 5: Questions to help determine company selection study below explains why Uganda’s Energy Access Scale approach Up Project (EASP) opened participation to all companies alongside an untargeted approach to beneficiary selection • How many and what eligible companies are active in the within a mature OGS market. market? • Is the goal to incentivize market entry and attract Opening participation to all companies may carry higher companies to serve hard-to-reach communities? administrative costs. For each company participating, the subsidy program will need conduct due diligence, issue • Is the development of local companies a priority? contracts, run monitoring and verification procedures, and • Does the program designer want to foster partnerships manage disbursements. between international and local companies? • What are the avenues for keeping the management costs CASE STUDY 10 low? Open company participation to all in Uganda’s Energy Access Scale-up project Open to all companies meeting minimum eligibility requirements Program objectives: To increase access to energy for households, commercial enterprises, and public institutions. In more mature markets, designers typically opt to open participation to all companies. That is because in these Overview: With funding from the World Bank, the markets, there will likely be enough companies that meet the Government of Uganda set up the Energy Access Scale qualification criteria to deliver the subsidy to the intended Up Project (EASP) to increase energy access. The financial beneficiaries, and open participation allows designers to component of the program is implemented by Uganda maintain the competitive dynamics between them. In these Energy Credit Capitalization Company (UECCC), and its goals cases, designers define eligibility criteria for companies to include the provision of results-based grants to facilitate participate in the program. Eligibility may be determined by access to SHS, clean cooking, and Productive Use of Energy requirements around company registration and compliance, (PUE) technologies.135 The RBF includes both an untargeted company structure and ownership, company business model end-user subsidy and a supply-side incentive for companies and operations, company capacity to adhere to minimum to expand operations into more remote and underserved consumer protection practices, company capacity to meet areas. consumer data collection and reporting requirements, the type of products and consumer finance offered, and Company selection approach: Open to all eligible meeting social and environmental safeguards.133 Program companies designers may also include a gender focus as part of the Uganda is classified as a mature OGS market.136 Participation eligibility criteria for participating companies. For example, is open to all energy service companies that initially and Malawi’s Ngwee Ngwee Ngwee Fund (NNNF) explicitly continually meet the eligibility criteria to ensure the market stated, “Among equally capable applicants, preference remains competitive. These criteria are as follows: adequate will be given to women-owned or women-led businesses ownership structure; possession of adequate funding; and companies that can demonstrate employing a gender possession of a bank account; a satisfactory end-user pricing diverse workforce.” 134 scheme; proof of the quality of operations; ability to provide a warranty and after-sales service; commitment to gender Opening participation to all companies is effective for quotas (at least 10% of the workforce should be women); growing local markets and encouraging competition, ability to maintain relevant data systems; and ability to pass a which helps keep prices low. A key benefit is that it gives due-diligence inspection. These criteria were set by UECCC. the end-user the widest product options, thus encouraging open market dynamics. It is also effective when program Launched in 2023, the program outcomes are yet to be designers are using less targeted approaches. The case established. 133 End-User Subsidy Lab learnings 134 Malawi Energy Access Project 2023, Request for Proposals, Market Catalyst Fund under Ngwee Ngwee Ngwee Fund (Off-Grid Market Development Fund). 135 Uganda Energy Credit Capitalization Company, 2023, Results Based Finance Manual (Part C of EASP Project Operations Manual) 136 Based on the Off-Grid Solar Market Trends Report 2022: State of the Sector, page 80. End-User Subsidies for Energy Access: A toolkit 49 Restrict participation to a limited number of companies and put in measures to ensure companies operate in a through competitive selection competitive environment where they price their products fairly and provide adequate after-sales services. Designers restrict participation in scenarios where the market has a limited number of companies or where they want to prioritize a specific type of company (e.g., local vs Other considerations: subsidy allocation to international). Restricting the participation of companies participating companies and frequency of onboarding can also be applied to minimize the transaction costs, new companies minimize delivery risks associated with multiple companies, and leverage economies of scale by larger established For companies participating in a subsidy program, companies to ensure public funds are utilized efficiently. subsidies can either be earmarked or allocated to them Restricting participation may also be used to incentivize in advance on the basis of their projected eligible sales, companies to enter ‘less commercially viable’ or ‘hard- or they can be allocated based on actual sales, on a first to-reach’ regions by guaranteeing companies a more come first serve basis. substantial portion of the subsidy. An example of restricted participation is the Togo CIZO program. Given OGS solutions First come, first serve: eligible companies prequalify for played a significant role in their national electrification a program and then claim the grant as they make sales strategy, but that the OGS market was only nascent, on a first come, first serve basis subject to any company the government used a competitive selection process caps. Transparency should be provided to all firms, to incentivize market entry for international companies, showing how much of the total grant has been paid out selecting five participating companies.137 and how much is available.139 Programs may also consider encouraging collaboration Earmarking subsidy amounts per company: In between international and domestic enterprises, especially some circumstances, it is also helpful to incorporate when promoting local businesses is a key objective. a maximum subsidy amount per company. This is to Governments may require that subsidies be restricted to avoid situations where a fast mover monopolizes the local companies, which are frequently smaller in scale and market and prevents other actors from participating. typically operate in rural, harder-to-reach areas. Generally, it Conversely, program implementors may also start with is advisable to work with more than one company to avoid small subsidy amounts for specific companies to test creating a monopoly.138 Program designers and funders are their ability to implement and avoid overcommitting on also advised to set up an impartial committee that evaluates non-performing contracts. Setting a maximum amount, the selection process results to ensure it is fair, transparent, which can be periodically reviewed, also helps to and in line with program goals. manage the risk that the program funds become ’over- subscribed’ or depleted too quickly. When restricting participation, minimizing the risk of distorting market competition is important. Program Similarly, a program can choose to issue one or several designers should be cautious not to limit consumer choice calls for proposals to onboard companies into the subsidy program, or to keep applications open on a rolling basis. When restricting participation, minimizing the risk of distorting market competition 2.5.3 Subsidy disbursement is important. Program designers should be cautious not to limit consumer When working through participating companies to choice and put in measures to ensure deliver subsidies, there are four subsidy disbursement approaches. Companies participating in the subsidy companies operate in a competitive program sell products and services at a subsidized price and environment where they price their then claim reimbursement. The following options guide the products fairly and provide adequate payment schedule (or cadence) of subsidy reimbursement after-sales services. to participating companies: 137 End-User Subsidy Lab, Case study: Togo CIZO Check Program 138 Africa Clean Energy (ACE) Technical Assistance Facility (TAF) and Open Capital 2020, Demand-Side Subsidies in Off-Grid Solar: A tool for achieving universal energy access and sustainable markets. 139 Open Capital consultations 50 End-User Subsidies for Energy Access: A toolkit One-off payment (after verification): Subsidy payments Recurring automatic top-ups: Subsidy payments are are reimbursed in full, one off, for all products sold within made to match the repayment pattern of the beneficiaries, a certain period, after such sales have been verified by for instance in PAYGo sales or EaaS contracts. the program. Upfront payment before a sale: A portion of the subsidy Multiple milestone-based installments: Subsidy payment is made before a sale has been made and payments are made at regular intervals determined verified, often to support companies with upfront capital by completion and verification of pre-defined service expenditure. The remaining subsidy can be disbursed conditions. following one of the above approaches. Figure 19: Payment schedule choices Subsidy delivery Subsidy disbursement One off (after Multiple milestone- Recurring automatic Advance payment verification) based installments top-ups (before sale) Companies are reimbursed Subsidy payments are Subsidy payments are made Portion of subsidy is paid in full, one off, for all made in regular intervals to match the repayment before sale and verification. subsidized products sold determined by completion pattern of the beneficiaries The remaining subsidy is Description within a certain period, and verification of pre- disbursed following one of after such sales have been defined service conditions the prior approaches. verified by the program • More suitable for relatively • More suitable for relatively • More suitable for PAYGo • Useful to reduce low subsidy amounts high subsidy amounts sales and EaaS contracts companies’ working • Lower transaction and • Higher transaction and (rather than cash sales) capital requirements Considerations verification costs verification costs • Similar considerations as • May increase transaction • Convenient for companies • Less convenient multiple milestone-based costs as it reduces working for companies as it instalments (in terms of • Suitable where access to capital requirements increases working capital cost and quality aspects) finance for companies is • Limits options to enforce requirements (which they limited after-sales service may pass on to consumers through price) • May limits accuracy of reporting (given once-off • Tool to enforce after-sales verification) service • Improves accuracy of reporting (given multiple verifications) End-User Subsidies for Energy Access: A toolkit 51 Table 6: Questions to help determine the selection of a risk, such as (i) working with companies with a proven subsidy disbursement approach track record in after-sales service and establishing an appropriate grievance redress mechanism, (ii) use one-off • How are eligible products and services sold (for example, payments only in program’s where subsidies are relatively cash, PAYGo, or fee-for-service)? low, leaving a substantial amount for the customer to pay • What mechanisms or incentives exist to guarantee over time. companies provide after sales services? The secondary impact of this is that the EUS program may • Do suppliers have sufficient working capital / financial have inaccurate reporting on households connected resources? to electricity or provided with clean cooking solutions. • Are there other supply-side incentives in place, such as Because of the single verification, there is a possibility credit lines? that a subsidy may be paid out to an OGS company, and the beneficiary household counted as having access, even when the product may stop working or households defaulted on their payments. This risk can also be One-off payment (after verification) is an efficient subsidy mitigated by (i) working with companies committed to disbursement approach because each product sold has a after-sales service, and (ii) establishing suitable monitoring single subsidy transaction and only needs to be verified systems, such as remote-monitoring platforms (more once. As a result, a one-off payment approach minimizes details in section 2.6 on verification). both transaction costs and administration. Nigeria’s NEP subsidy program has over 50 participating companies While the multiple milestone-based installments and uses this approach to reduce costs and expedite approach attracts higher transaction costs, it also allows disbursements (see section 3.5 for details). implementors to ensure that companies provide after- sales services to beneficiaries and that beneficiaries This approach is attractive to energy companies selling complete their payments in the case of PAYGo sales. This products on credit because they can access additional is because the program implementers stay in contact with working capital at no cost. This is because the full subsidy the company over a longer period. Criteria for payments payment is given after the customer has only made the first typically include the level of customer satisfaction and quality down payment. To illustrate this benefit, consider a SHS on of service, among others. For instance, Rwanda’s REF pays sale at a price of $100 under a PAYGo arrangement, whereby companies in three installments for PAYGo sales and two the customer needs to make a 20 percent down payment installments for cash sales. The final installment is made after upon purchase. Suppose this product has a 40 percent three years of service (details in section 3.4). subsidy. In that case, the energy company will receive $60 at the moment of purchase (the sum of the customer’s down The multiple milestone-based installments approach payment and the subsidy) instead of $20 (the down payment has two main disadvantages. First, the multiple payments under the PAYGo arrangement). and verification processes lead to higher costs and more cumbersome processes. Second, the program designers While the one-off payment approach minimizes may have to spend more time with participating companies transaction costs and is attractive to companies, it has a to align on milestones for the payment. few challenges. Recurring automatic top-ups are used when the The program implementor, for instance, has no avenue to beneficiary is required to pay fixed amounts over a regular verify that the OGS company is providing after-sales repayment schedule. The subsidy amount is disbursed to services to the subsidy beneficiary because verification the companies per this repayment schedule. For instance, if is carried out only once at purchase. As a result, the a PAYGo beneficiary is obligated to make 24 equal monthly beneficiary may face technical challenges with the payments on specific dates, then the program will disburse purchased OGS or clean cooking product and not receive parallel payments to their service provider following the support. It is important to highlight that for PAYGo sales, schedule. Like all other approaches discussed where the the company still has an incentive to provide after-sales program implementers stay in touch with the company over service, since future customer payments are dependent a long period, recurring automatic top-ups ensure that the on the customer having a functional system. There are energy company addresses technical challenges through also ways for a subsidy program to mitigate against this after-sales services. 52 End-User Subsidies for Energy Access: A toolkit This method is more appropriate for a fee-for-service subsidy or PAYGo programs with monthly payments. An additional subsidy disbursement option Togo’s CIZO program uses this approach to settle subsidy includes upfront payments, which can be used in amounts to participating companies (see case study below). combination with one of the approaches above. Advance payments are when the subsidy program CASE STUDY 11 pays out part of the subsidy at a verifiable milestone before the sale of a product. Typically, after the Multiple recurring disbursements in Togo’s CIZO upfront payment is made, companies must continue program submitting claims to the EUS program until their total sales make up the amount of the upfront payment. Policy Objectives: To achieve universal electricity access This approach is often combined with one of the for people living in rural Togo through increased access to other approaches to ensure that follow-on payments off-grid solar systems.140 are made. This could be in line with agreed-upon milestones and thus follow a multiple recurring Overview: In line with the government of Togo’s National milestone-based payments approach. The method of Electrification strategy to reach universal electrification by upfront payments is used in various RBF programs, 2030, the CIZO program was initiated to deploy SHS through for example, the Global LEAP RBF made payments public-private partnerships. The program was designed to directly to both manufacturers and distributors after target rural populations and electrify more than 555,000 orders had been placed. households with SHS of 20Wp, which is above the minimum Tier 1 access defined by the multi-tier framework. During the This payment option is especially helpful in contexts program, the government has sought to address affordability where it is challenging for companies to access using two means: consumer financing (provided by MFIs and financing because it provides extra cash flow for SHS companies through their PAYGo platforms) and an end- companies to purchase stock in advance. While user subsidy. this method might be attractive to participating companies, there is a risk of misuse of the funds, in Subsidy disbursement: Multiple recurring disbursements which companies use the upfront payment for other Upon identification of eligibility for the program, beneficiaries purposes. Program designers should, therefore, be acquire an eligible SHS and make a payment to activate an cautious in using this method. They should apply associated mobile money account. The telecom company rigorous screening of companies during selection then requests an eligibility confirmation from the government as well as close monitoring during implementation and, if it is successful, channels the subsidy to the SHS to assess whether the use of funds is in line with company. Given the PAYGo nature of the program, the contractual agreements. company collects regular payments from the customer, and in parallel, the telecom company transfers subsidy funding to the company monthly for 36 months. 2.5.4 Fund management Fund managers provide oversight of implementation and Like the multiple milestone-based installments approach, program administration. Strong administration structures the multiple recurring top-ups approach attracts higher consist of defined roles and responsibilities for the players transaction costs. In addition, maintaining the automatic involved in implementation, as well as clear processes and payments usually requires an IT system, which increases procedures for claiming and disbursing funds. This minimizes administrative costs, both for government and for companies. the risk of abuse of funds. There are three approaches to fund management: government agencies, third-party administrators, or a mixed approach.141 140 End-User Subsidies Lab, 2022, Case Study: Togo CIZO Cheque Program 141 Africa Clean Energy (ACE) Technical Assistance Facility (TAF) and Open Capita 2020, Demand-Side Subsidies in Off-Grid Solar: A tool for achieving universal energy access and sustainable markets. End-User Subsidies for Energy Access: A toolkit 53 Figure 20: Fund management choices Subsidy delivery Fund management Government administrators Third party Mixed phased approach Subsidy program is managed by Program designers hire an external Third-party administrators jointly Description agencies or ministries directly related firm with specialized technical manage the program alongside to the subsidy program knowledge government agency • Leverages capacity and • Encourages efficient use of • Build local capacity within infrastructure within governments resources governments • Provides buy-in from the • Creates opportunities for knowledge • Utilizes learnings from third party Pros government which could sharing and capacity building of administrators to inform decisions encourage participation from other local talent during subsidy delivery stakeholders • Boosts governments capacity to manage other future programs • Provides a long-term governance structure in the case of long-term facilities • Additional costs might be required • May not encourage government • May lead to inefficiencies and for sufficient oversight capacity-building overlap of responsibilities if not • Bureaucratic processes may cause • Procurement and contracting can be planned well Cons delays in contracting and subsidy a lengthy process • Introduces additional costs due disbursement • Introduces additional oversight cost to the use of both government • May require deliberate engagement for fund administrators agencies and employment of third- with, and reassurance to the private party administrators sector where the government agency has a weak reputation An important driver for this choice is whether the relevant implementation. While this is not a primary objective of EUS government agency has adequate institutional capacity programs, it is important for their sustainability. Another and infrastructure to manage the subsidy program. general consideration is to engage independent auditors to Designers may opt for a third-party administrator in instances periodically review the management of funds.142 where the government agency does not have the capacity, or other considerations may be at play, such as trying to Government administrators are usually situated within attract a fund manager into the market or aiming to run a agencies or ministries directly related to the subsidy pilot requiring short-term, technical expertise. Designers program. The use of government agencies allows programs may also opt for a mixed approach that involves a third to take advantage of pre-existing government infrastructure party working alongside the relevant government agency. and human resources. For example, the subsidy for SHS Regardless of the approach adopted, it is important to under World Bank-funded NEP in Nigeria is managed involve public institutions and build their capacity throughout through the Rural Electrification Authority. Similarly, the Rwanda REF Window 5 is managed by government-owned 142 Africa Clean Energy (ACE) Technical Assistance Facility (TAF) and Open Capita 2020, Demand-Side Subsidies in Off-Grid Solar: A tool for achieving universal energy access and sustainable markets. 54 End-User Subsidies for Energy Access: A toolkit EDCL and BRD. Working through governments is also a more and a good understanding of the local context. This is also natural fit for a long-term subsidy program since it ensures appropriate if program designers want to encourage the alignment with government priorities and proper governance third party to set up operations in the country or a particular accountability. region. Engaging a third-party administrator is typically more costly than fund management by the government. It Deciding whether to involve a government agency is is also a lengthy process, with competitive procurement often dependent on the following competencies within and contracting taking a significant amount of time, which agency:143 may delay program implementation. However, it may be a cost-effective solution for specialized projects, allowing for Knowledge of OGS and cleaning cooking leveraging of skills and best practices and for building local technologies: Knowledge about OGS or clean cooking capacity. markets differs significantly from country to country, and it is important to work with an agency that already When selecting a suitable third-party fund manager, understands these technologies well. Where the designers are advised to take the following steps: 145,146 institutional knowledge of the technologies is low, designers may consider allocating a budget towards Ensure that the role, expectations, and service training and oversight in the first few years. levels for the fund manager are well documented in the contract. Create explicit terms of reference, Past track record in executing similar programs: ideally accompanied by an operational manual It is advisable to engage with agencies that have delineating all the necessary steps, procedures, and demonstrated strong financial controls in the past without required documents. instances of misappropriation of funds. Assign fund managers activities that fall within Level of bureaucracy: Government agencies have their core business and capacities, such as varying levels of bureaucracy due to variances in executing transactions, overseeing clients, and governance structure and operational policies. conducting compliance-related tasks. Designers Government agencies may have to seek approval and implementors should seek to enhance these from their line ministries, a process that is often time- strengths and avoid urging fund managers to consuming and can impact contracting and fund assume responsibilities like technical verification disbursement. that are best suited for independent technical experts. Program designers may opt for a third-party administrator with specialized technical knowledge, such as a private Engaging a third-party sector entity or development partner. This is especially administrator is typically more appropriate when the relevant government agencies in costly than fund management the country have limited capacity to execute the project. For example, in Malawi, the Ngwee Ngwee Ngwee Fund by the government. It is also a (under the Malawi Electricity Access Project) selected a lengthy process, with competitive fund manager through a competitive procurement process. procurement and contracting Infrastructure Development Company Limited (IDCOL) taking a significant amount of was selected to manage the fund.144 Having demonstrated time, which may delay program experience managing similar funds, notably Bangladesh’s IDCOL Solar Home System program, IDCOL is well placed implementation. However, it may to support the Malawi government run the program, while be a cost-effective solution for in parallel help develop internal capacity of the Ministry of specialized projects, allowing for Energy. Although selecting a local fund manager helps boost leveraging of skills and best practices local capacity, this does not restrict program designers from and for building local capacity. hiring international firms with the required technical capacity 143 Africa Clean Energy (ACE) Technical Assistance Facility (TAF) and Open Capita 2020, Demand-Side Subsidies in Off-Grid Solar: A tool for achieving universal energy access and sustainable markets. 144 GET.Invest, 2023, Malawi Off-Grid Market Development Fund (OGMDF) - Debt Window. 145 Open Capital Advisors Consultations 146 EnDev 2021, Transforming energy access markets with Results-based Financing: Lessons from 7 years of implementation under EnDev’s RBF Facility financed by UK Aid End-User Subsidies for Energy Access: A toolkit 55 Consider the fund manager’s understanding of Fund management: Mixed approach the local market as well as their human capital From November 2019 to March 2021, EnDev, Rwanda resources and access to target beneficiaries in Energy Group (REG), and Urwego Bank, a local microfinance the form of their rural presence, agents, or branch institution, joined forces to execute the pilot program. networks. This collaboration utilized the unique strengths of each organization and fostered a sense of ownership among all Select an institution with a proven interest in the stakeholders. energy sector. Such a firm will be more likely to set up internal procedures and structures leading to the EnDev took the lead in project implementation and further development of energy-focused services management, development of IT monitoring tools, and beyond the subsidy program. provision of funds. Urwego Bank played a crucial role in the management of the fund, selection of eligible companies, Designers may opt for a mixed approach when the relevant contract management with participating companies, and government agency has the technical capacity to design overseeing the disbursement of funds. REG was co- and execute the program but face limitations in relation to implementing the program with EnDev Rwanda, focused on contracting, fund management, and/or disbursement. The communication campaigns, participating in the selection of program may be designed in such a way that a third-party eligible companies, and participating in periodic program administrator manages the program alongside the relevant reviews. The verification was carried out jointly by all the government agency’s personnel to capitalize on each three stakeholders prior to the disbursement of the subsidy. institution’s expertise. As seen in the case study below in Rwanda, it is important to have a clear division of roles and responsibilities. The third-party administrator may also train government agency personnel on the essential responsibilities for CASE STUDY 12 successful subsidy programs. Responsibilities can be progressively transitioned towards the government in phases Mixed approach to fund administration in EnDev’s or based on certain milestones. Pro-Poor RBF in Rwanda Program objective: To provide affordable off-grid solar 2.5.5 Claim management products through a subsidy mechanism targeted to the poorest and delivered through companies147 Claim management involves coordinating with the participating companies to submit information about the Overview: This pilot project aimed to accelerate access sale of a product. The claim process starts with companies to electricity for low-income households in off-grid areas submitting a form with information on the customer and through targeted incentives. While existing programs the product sale. Information about the customer may focused on access to finance for households and working include their name, identification number, address or GPS capital support to companies, affordability remained the key coordinates, and phone number. Product information may challenge to SHS uptake.148 The different subsidy levels were consist of product type and model, serial number, date informed by the ability to pay of various socio-economic of sale, price, and payment terms (for example, cash or categories, a government of Rwanda system called Ubudehe. PAYGo).149,150 A claim form is typically submitted electronically Those categorized as Ubudehe 1 are the most vulnerable by sending an email or uploading the form onto an online households and received the highest incentive. The concept IT platform. For example, in Rwanda’s Energy Fund Window of this pilot project was scaled under the government’s REF 5, companies are required to submit a claim form through Window 5 subsidy. email.151 In Nigeria’s NEP, companies submit information through an IT platform.152 The claim management process 147 End-User Subsidies Lab, 2022, Case Study: EnDev’s Pro-Poor Results Based Financing in Rwanda 148 USAID, 2022, Pro Poor Results-Based Financing: Increasing Off-Grid Access To Electricity In Rwanda 149 Development Bank of Rwanda, EDCL, The World Bank 2023, Window 5 Operations Manual. 150 End-User Subsidy Lab learnings 151 Development Bank of Rwanda, EDCL, The World Bank 2023, Window 5 Operations Manual. 152 NEP uses the Odyssey platform. Details provided in section 3.5. 56 End-User Subsidies for Energy Access: A toolkit may be further enhanced via a direct integration with 2.6 Verification a company’s Customer Relations Management (CRM) software. In this case, claims can be made automatically Focus of through synchronization with the company’s latest sales and this section repossessions. 3. Delivery It is important for program designers to keep the claim management procedure simple and set realistic timelines. 2. Subsidy level 4. Verification Multiple claims by multiple companies may be submitted in parallel, and typically, claims are submitted and handled in batches. For example, a batch may contain 500 sales. The processing of claims mostly involves verification (see next section) and, is ideally quick, accurate, and simple. 1. Targeting 5. Exit strategy The process, timelines, and documentation requirements should be well communicated to companies. Transparent communication with companies on the status of claims and any data compliance issues is also important. Digitizing records can significantly enhance efficiency. Using Verification is used to confirm that the subsidies are specialized IT platforms for claim management, such as implemented as designed, i.e. that eligible products Odyssey, Prospect, or others, can support improvement or services have been sold to eligible customers, at a in automating the claim management process, reducing subsidized price, and, to the extent possible, that these processing time and potentially reducing overall cost. Further products are being used as intended (by the eligible details are provided in the section on verification. customer, and in good operating condition). Designers should also consider how to handle specific Verification starts at the point of sale, supported by scenarios related to claims such as repossessions, eligibility tools where necessary. That is because vendors non-working systems, and fraud. Products sold on credit need to confirm that a customer meets the subsidy eligibility (PAYGo) may be repossessed by companies when customers criteria before selling them the subsidized product. The do not pay. Other scenarios include those in which products level of verification at the point of sale is related to the are not working during verification or cases of fraud by the targeting approach. With a less targeted approach, pre- customer or company. Processes for handling claims in these sale verification may be quite simple. At a minimum, the instances should be defined in advance in the operations program or company will want to verify that a beneficiary manual.153 For example, in the case of repossessions, it is has not already received a subsidy from the program. With recommended that these must be declared to the fund more targeted approaches, verification at the point of sale administrator, if not automatically made aware, e.g., in the may involve confirming the demographic, geographic, or case of recurring payments linked to PAYGo payments. economic characteristics of a beneficiary as selected in the There is a risk that companies may hide information about targeting approach. Companies typically do this verification repossessions, as it may impact their subsidies, which means supported by technology in the form of an online platform that fund administrators should apply clear processes, or database with data on eligible customers. For example, monitoring, and potential sanctions. In cases where the under Rwanda REF Window 5, an eligibility tool was provided subsidy for such products had been fully disbursed, the to companies (for more details, refer on eligibility tools, refer program may decide to deduct that subsidy amount on the to section 2.3 or targeting or section 3.4 for details about the next sale of a similar product. Program designers should Rwanda REF eligibility tool). The remainder of this section further capture processes and policies that handle other focuses on verification post-sale. issues, such as cases of theft, fraud, and on selling, among others. 153 Open Capital Advisors Consultations End-User Subsidies for Energy Access: A toolkit 57 The core focuses of verification in EUS programs are and field surveys of a sample of customers. The automated around product delivery and use, which inform subsidy verification refers to leveraging data and technology, such disbursement. For subsidies delivered through companies, as using information on the use of products (if they are verification confirms whether the product was delivered equipped with remote-monitoring technology) and customer to the intended recipient as “claimed” or reported by payment information drawn from companies’ customer the company. Verification is thus connected to the claim relation management (CRM) systems or mobile money management process (described in sub-section 2.5.5) and operators (if such payments are primarily done through helps inform whether a subsidy can be disbursed. Beyond mobile money). product delivery, the program needs to verify if the product is still operational and has not been resold by the beneficiary. There is a trend for subsidy programs to integrate automated verification features where this is possible, Verification is typically done by independent verification to improve the speed and accuracy of the verification agents (IVAs). Using a third party for verification helps process. Where possible, program designers should try to mitigate the risk of fraud and avoids potential disputes use technology as much as possible but leverage manual between companies and the fund manager. The IVA can be methods as a complement. In practice, a hybrid approach is contracted by the program designer or the fund manager the most often adopted as some level of manual verification with a mandate to verify product delivery in line with the is often still needed to complement results from the tech- subsidy design. enabled processes. Manual verification may also be needed in situations where tech-enabled verification will not work. There are multiple options and tools to carry out Examples of such situations include projects delivered in verification, ranging from more traditional or manual remote areas where connectivity is an issue and projects forms of verification to more automated verification involving lower-tier products that are not equipped with leveraging data and technology. The more traditional or remote monitoring technologies. For example, the more manual approach typically involves the fund manager and basic improved cook stoves (ICS) are typically not GSM or IVA investigating companies’ subsidy claims through paper GPS enabled. trail checks, and the IVA conducting a combination of phone Figure 21: Verification options Verification (complementary options) Traditional / manual verification methods Automated / technology-enabled verification methods Involves combined techniques of desk-based verification Leverages digital technology to automate verification (e.g., Description (paper trail), phone surveys, and field verification remote monitoring via GSM or GPS, customer payment information through companies’ CRM systems or mobile money providers, automatic SMS/Whatsapp surveys, etc) • Works for all types of products and regions as it does not • Allows for quicker results and avoids the risks of human require any form of remote connectivity errors • May involve lower upfront set-up costs compared to • Allows for quicker subsidy disbursement to the companies Pros automated tech-enabled verification • Lower running costs compared to manual approach, and • Flexibility in cases where verification is difficult, as potential for economies of scale in-person judgement calls can be made, with nuanced • Allows for data accessibility and transparency insights • Longer execution time than other methods • May not work for all types of sales (i.e. those not leaving a • Phone surveys may not work in areas with low mobile digital trace, such as cash sales of low-tier products) or in phone ownership or low coverage regions with low internet connectivity Cons • Labor-intensive and therefore costly in implementation • Relatively novel approach, and might require manual verification to supplement • More prone to human error or fraud • Set-up of the system requires high up-front costs 58 End-User Subsidies for Energy Access: A toolkit Table 7: Questions to help inform the verification approach It is important for program designers • Does automated / technology-enabled verification to create alternative verification increase the price and, therefore, affect the affordability approaches as needed to adapt to of products? beneficiary circumstances. • What data needs to be tracked during the verification process? Can all of the data be provided through customers do not know the specific name of their product technology, or does it also require manual processes? and are instead allowed to describe what the product looks • Are there existing databases or platforms that the program like for verification purposes. To increase the chances may leverage for verification? Or would the program need of reaching a beneficiary, phone verification calls can be to develop something new? conducted at different times of the day. • How do program costs (and timelines) compare for both Despite its drawbacks, manual verification has several methods? advantages. First, it is widely applicable because it does not require any form of remote connectivity. It can, therefore, • Is automated / technology-enabled verification feasible be used across all types of products and regions. In for all products in the program scope? Or is it limited to higher-tier products or products sold with PAYGo as a addition, this approach allows for more flexibility in cases payment model? where verification is difficult because alternative verification methods exist. Lastly, it allows for more nuanced insights • Does network coverage allow for automated / technology- because interactions are in person. enabled verification in all the program's target areas? CASE STUDY 13 2.6.1 Manual verification Manual verification in Bangladesh’s IDCOL SHS Program In manual verification, an IVA combines techniques of desk-based verification, phone surveys, and field Program objectives: Provide electrification through solar verification. These activities are labor-intensive and, home systems (SHS) therefore, costly to implement. They are more prone to human error or fraud (for example, false narratives from Overview: The IDCOL SHS Program was one of the world’s beneficiaries and other stakeholders) and may involve largest national off-grid electrification programs. The program significant logistical challenges, particularly in remote ran from 2003 to 2018 and provided energy access to about areas. Desk-based verification can include a paper trail 20 million people through grants and loans. The program was managed by Infrastructure Development Company and payment verification, including company visits to verify Limited (IDCOL), a government-owned organization created orders and receipts. Field visits and phone calls are typically to support the uptake of SHS. This program was implemented only done for a representative sample of customers or a through partnerships with various participating organizations given percentage of each company’s claim. IVAs may use that supplied the SHS.154 field officers to confirm in person that beneficiaries have received the subsidized products and that those products Verification: Manual are still operational. Each of these steps is labor intensive, Monitoring and verification were conducted in-house by and there is a notable risk that customers may not be IDCOL, which built the capacity to manage the process to minimize program costs. For this purpose, IDCOL set up three present or responsive to calls. Therefore, manual verification divisional and 12 regional inspection offices, employing 103 is a time-consuming process that can potentially delay the technical inspectors responsible for monthly checks. disbursement of subsidies to participating companies. The program’s cost-effectiveness was attributed to the It is important for program designers to create internal verification operations and the strong sense of alternative approaches as needed to adapt to beneficiary accountability among partnering organizations to deliver circumstances. For example, when beneficiaries do not and maintain high-quality and well-functioning SHS. This approach led to heightened consumer confidence and, own a phone, program implementors may allow them to consequently, a notable level of program success. provide alternative contact numbers from neighbors or family members for phone verification. Another example is when 154 End-User Subsidies Lab, How IDCOL Addressed the Affordability Gap: Lessons from Bangladesh’s Solar Home System Program End-User Subsidies for Energy Access: A toolkit 59 2.6.2 Automated verification Automated technology-enabled Automated verification leveraging digital technology verification may not work in all allows for quicker results and avoids the risk of human situations errors. In cases where verification is fully automated, the system can do the relevant verification checks and trigger 2.7 Exit or Adjustment Strategy the subsidy disbursement immediately when a new customer is registered. This means companies do not need to wait a long time for the disbursement of funds. Of course, the setup 3. Delivery of such a system, whether it leverages an existing solution or builds a new one, involves a significant upfront investment. However, once the system is operational, the running costs 2. Subsidy level 4. Verification are likely to be considerably lower than those related to manual verification. Automated technology-enabled verification also scales up well, as it may be applied in larger programs without the need for many additional staff. Another 1. Targeting 5. Exit strategy advantage of working with a centralized online platform is that it allows for data accessibility and transparency. The fund manager, participating companies, and the IVA have real- time access and a single “source of truth.” It is important to note that the system should operate in line with the relevant data privacy and security regulations. Focus of Examples of platforms used for automated verification this section include Prospect and Odyssey. Prospect is an open- access platform based on open-source code. Companies An exit strategy aims to ensure a sustainable market regularly upload their data (ideally via API to enable real- for products or services after a subsidy is withdrawn. time-monitoring), and the platform can process the data Traditionally, this occurs when consumers no longer need the to produce analytics. Personal data is protected through subsidy to purchase this product or service. Without a clear anonymization and there will soon be an option to choose exit strategy from program designers, it is more difficult for encryption. Companies own the rights to the processed companies to plan their approach to continuing operations in data. The fund manager and IVA get access to a selected an unsubsidized market.155 subset of the information to verify submitted data and monitor subsidy disbursement. It is also possible to make a Exit strategies are more complicated when it comes direct interface between the remote monitoring system of a to end-user subsidies. Since EUS aim to address the company and Prospect. That way, the system also has “live” affordability gap in energy access and ensure no one is data on the products in use. Similarly to Prospect, Odyssey left behind, the “need” for the subsidy will continue until offers this service through a license to a platform that can universal access is affordable on a purely commercial basis perform automated claim management and verification, (through a combination of economies of market scale, among other fund management functions. The Odyssey technology advances driving cost reduction, and economic platform was used in Nigeria’s NEP SHS output-based fund. development). It is clear that programs have limited Automation of verification through the platform played a resources and must operate within their realities. Therefore, role in its rapid expansion, as explained in the case study of considering a program-level exit strategy for the moment a section 3.5. program runs out of time and money is crucial to ensuring beneficiaries are not sent back into energy poverty and Automated technology-enabled verification may not work markets are not distorted. Supply-side subsidies may be in all situations today because, in some cases, there are simpler to ‘exit’ given their aim to expand access by reducing challenges in building and integrating the system across the upfront risks or costs to companies with an intrinsic participating organizations. Oftentimes, a system needs to expectation of building a commercial market. For example, be tailored to each program’s operations and unique needs, incentivizing companies to enter a new market by offering and this comes at a high cost that only makes sense for incentives for the first 1,000 systems sold, assuming that programs of a large scale. As a result, programs can adopt a companies will choose to continue operating in that market mix of both manual and automated verification. after the subsidies are removed. 155 GOGLA Discussion Paper: How End-User Subsidies Can Help Achieve Universal Energy Access 60 End-User Subsidies for Energy Access: A toolkit A program-level exit strategy, in the context of EUS, Exit strategies are more complicated means addressing the next step once the specific when it comes to end-user subsidies. program ends. In cases where subsidies are expected Since EUS aim to address the to continue through another subsidy program, subsidy affordability gap in energy access ‘adjustment’ may be more appropriate terminology than and ensure no one is left behind, the ‘exit’. It is critical to define this next step from the program’s start to ensure program objectives, timing, and funding are “need” for the subsidy will continue aligned with what will follow. At the same time, it is helpful until universal access is affordable on a to review subsidy schemes regularly to account for changes purely commercial basis in market dynamics that may impact the exit or adjustment strategy. There are three potential exit paths shown in Figure 22, including: Therefore, considering a program- i. Subsidies are no longer needed: Letting an unsubsidized level exit strategy for the moment a market take over because the subsidy is no longer program runs out of time and money needed. is crucial to ensuring beneficiaries are not sent back into energy ii. New subsidy program: Designing a new program for the poverty and markets are not next phase of the subsidy (in which case it is important to think through the transition and any gaps between the distorted. programs). iii. Long-term subsidies: Envisioning long-term subsidies through a sustainable long-term facility. Figure 22: Exit strategy options Exit and adjustment strategies Subsidy no longer needed New subsidy program Long-term subsidies When a government has fully achieved Transitioning to a new subsidy Continuous end-user subsidy facility Description its targets and a subsidy is no longer program, with restructuring of that consumers and companies needed, an unsubsidized market can subsidies if needed participate in, can be funded from the take over national treasury or sector level funds • Affordability is no longer a concern, • Affordability continues to be a • Affordability likely to remain a long- or can be addressed through concern to achieve and sustain term concern to achieve and sustain other mechanisms (e.g., consumer energy access energy access Considerations finance, a social protection cash • Programs are designed to operate • It is important to consider what transfers for the most marginalized) within specific timelines and budget recurring subsidies, may be needed • The exit should ensure alignment constraints, but can transition into to sustain universal energy access with program goals new programs where appropriate • Sustainability in funding and • Program should make certain that • A smooth transition from program operations is crucial where long- the market is not distorted to program is critical to ensure that term subsidies are envisioned • Market monitoring is critical to the expectations of beneficiaries are understand changes in market appropriately managed dynamics to identify when subsidies are no longer needed End-User Subsidies for Energy Access: A toolkit 61 Table 8: Questions to help determine exit strategy CASE STUDY 14 • What is driving the current need for subsidies? Are these Decreased need for subsidies in Bangladesh market dynamics expected to change? • What are the project’s targets? Are there enough funds to Program objectives: Provide electrification through solar fulfill this target? home systems. • What are the government’s targets (e.g., universal access, Overview: The IDCOL SHS Program was the largest national minimum tier), and what are the resources needed to off-grid electrification program in the world. The program ran achieve them? from 2003 to 2018, providing electricity to about 20 million people through grants and loans that facilitated access to • Can subsidies be phased out after the project’s target is SHS. The program was run and managed by Infrastructure achieved? After the government’s target is achieved? Development Company Limited (IDCOL), a government- owned organization that supported the uptake of SHS.158 • Will recurrent subsidies be needed? Exit: Subsidies no longer needed When the SHS Program was launched in 2003, there were 2.7.1 Subsidies no longer needed around 15 million unelectrified rural households in the When a government has fully achieved its targets, and country, and the rural electrification rate was under 27%. a subsidy is no longer needed, the market can take over At that time, the pace of grid electrification was slow. The number of unelectrified rural households declined slowly, on a purely commercial basis. Depending on the targets reaching about 13 million by 2013. IDCOL estimated at that set in the context of EUS, that could mean, for example, that time that the market for SHS was about 6 million households all households have access to at least Tier-1 electricity and or about 50% of unelectrified rural households. However, at least Tier-2 improved cookstoves. In addition, it would in the following years, the pace of grid electrification mean that customers who received a subsidy to purchase a accelerated. Among the remaining unelectrified households, product can now afford replacement parts or fuel costs, for an expectation of imminent connection to the grid made example, LPG.156 Alternatively, it may mean that affordability many reluctant to invest in SHS. Coupled with an increase in is no longer a major barrier to access, and beneficiaries can commercial SHS sales and the impact of other government afford basic lighting and cooking products on their own. This initiatives (e.g., social safety net programs), subsidized SHS sales in the IDCOL program began to decline in 2014, may be due to a rise in income or product innovations that following 11 years of growth. By 2018, more than 80 percent have driven down costs. Another alternative may be that of rural households had access to electricity from either the affordability is still a gap for a smaller group of people. This grid or SHS. could also be addressed by a social assistance program rather than a specific energy access EUS program. While At the same time, the technology for SHS changed, bringing all these scenarios are unlikely to happen on a large scale down their cost and altering where subsidies were needed. within the timeframe needed to achieve SDG 7, it may be the At the same time, rural households’ average incomes increased, and they began to seek additional services from case in specific regions. their SHS. The average grant amount per SHS dropped from 19% of the retail price in 2003 to under 5% in 2017. In this scenario, the exit strategy should ensure program goals are met and that there is a successful transition to Following a study to ascertain an appropriate exit strategy for a market functioning without subsidies. It is critical that IDCOL, the government opted to gradually reduce the grant there is a proper phase-out of the subsidy program to make component by narrowing down the systems that qualified certain that the market is not distorted (see the section below and reducing the amount of the subsidy itself. From 2012, only systems smaller than 30Wp were eligible for a subsidy. on phasing out subsidies). It is recommended that market The subsidy amount was set between $9 and $13, down from monitoring is conducted alongside the phase-out so that $55 in 2005 (at which point the subsidy could be used on a program implementors can inform participants of risks – and system of any size). help them manage these – as the program ends. These risks include business collapse, financial losses, and customers having no recourse to repair or replacement services.157 156 Private sector paper on “Recommendations for Demand-Side Subsidy Design for Accelerating SHS-Driven Energy Access” 157 Cabraal, Anil, William A. Ward, V. Susan Bogach and Amit Jain. 2021. Living in the Light: The Bangladesh Solar Home Systems Story. A World Bank Study. Washington, DC: World Bank. 158 End-User Subsidies Lab, How IDCOL Addressed the Affordability Gap: Lessons from Bangladesh’s Solar Home System Program 62 End-User Subsidies for Energy Access: A toolkit Key insight on exiting when subsidies are no Key insight on exiting through new subsidy longer needed: Market monitoring is critical to programs: Program designers are advised to plan well understand changes in market dynamics that may in advance for what could happen after the program impact the subsidy scheme and to identify when ends (i.e., when funds run out or the scheduled subsidies are no longer needed. timeframe is over) and incorporate this into the exit strategy of the program. 2.7.2 New subsidy program 2.7.3 Long-term subsidies Subsidy programs are designed to operate within specific timelines and budget constraints but can transition into When designing a subsidy, it is important to consider new programs where appropriate. For example, EUS what recurring subsidies, if any, may be needed to sustain programs often start with a pilot to test a given market’s universal energy access.159 Programs with fixed end dates need for subsidies. The narrow scope of a pilot helps avoid often fail to fully eliminate the affordability gap in a market market distortion as crucial information is gathered. Programs either because they run out of funds or because there is can then be scaled up by raising new funding or become continued need by marginalized communities for subsidies embedded within existing programs. Close involvement (for example, to buy, repair, and replace a product). In the of the scale-up players during implementation allows for case of clean cooking, the poorest consumers may need a smooth pilot transition into the scale-up phase. Beyond an ongoing subsidy to buy clean cooking fuel for years to pilots, more extensive programs may also transition into a come. The idea of a long-term facility is to have an end- newer phase, for example, when shifting towards a new user subsidy scheme that both consumers and companies region or target group. can tap into on a continuous basis. Companies may find participating in these long-term subsidy facilities appealing A smooth transition from program to program is critical because they give them long-term access to potential to ensure that the expectations of beneficiaries are customers who would otherwise be beyond their commercial appropriately managed. For instance, if the subsidy’s target reach.160 group or product category changes from one program to the next, this must be clearly communicated to avoid Where long-term subsidies are envisioned, programs confusion in the communities being served. It is important should consider how these can be sustainably funded to note that during the design phase, it may not be known if and operated. A long-term facility may be funded by the there will be a scale-up or follow-on program, but designers government or supported by donors. Donor funding may and implementers should make an effort to shape this next not always allow for long-term facilities. Governments may phase where possible. Government and potential funders decide to fund subsidies in other ways, for example, by of such new programs should be involved as early as introducing an electricity levy to subsidize off-grid solutions. possible to help design an effective ‘transition’ strategy. Energy markets in many developed countries have such In the EnDev pilot for Malawi, for example, the designers sustainable cross-subsidies in place, which are designed identified a government-led fund, supported by the World to make energy access universally affordable.161 This is also Bank, as a potential avenue for a scale-up and involved being explored in less developed markets, for example, by both stakeholders in the design and implementation of using income from electricity supplied through the grid to the pilot to ensure a smooth handover. This case study is help fund mini-grid development in off-grid settings. Ongoing further discussed in 3.2 Designing an EUS pilot targeting the discussions advocate for extending this pricing mechanism poorest in Malawi. to include mini-grids operated by private companies or cooperatives as well.162,163 Carbon finance associated to A smooth transition from program the use of clean energy products may also be considered as a source of funding for long-term subsidies. Besides the to program is critical to ensure that funding challenge, other potential obstacles for a long-term the expectations of beneficiaries facility may be the cost and time involved in the set-up and are appropriately managed. the government’s capacity to operate this. 159 Terafund Designing 2020, Designing Sustainable Subsidies To Accelerate Universal Energy Access: A briefing paper on key principles for the design of pro-poor subsidies to meet the goal of sustainable energy for all 160 Hamayun, Mansoor 2020. Op-Ed: This is how Africa can accelerate energy access. 161 Terafund Designing 2020, Designing Sustainable Subsidies To Accelerate Universal Energy Access: A briefing paper on key principles for the design of pro-poor subsidies to meet the goal of sustainable energy for all 162 Energy & Petroleum Regulatory Authority March 2023, Press Release : Retail Electricity Tariff Review For The 2022/23-2025/26 4th Tariff Control Period (TCP) Effective 1st April 2023 163 Climate Compatible Growth 2021, At what price? The Political Economy of mini-grid electricity development and deployment in Kenya End-User Subsidies for Energy Access: A toolkit 63 CASE STUDY 15 Key insight on long-term facilities: A future energy market with sustained universal access may Long term subsidies in South Africa include long-term support to the poorest and most marginalized people through recurring facilities. Program objectives: South Africa’s Free Basic Electricity (FBE) and Free Basic Alternative Energy (FBAE) programs provide a limited amount of electricity (or alternative source 2.7.4 Phasing out subsidies of energy for off-grid households, such as OGS solutions) for free to eligible poor households. Regardless of the longer-term exit or adjustment strategy Overview: Municipalities in South Africa are responsible chosen, designers need to consider how to ‘exit’ subsidies for providing electricity to their population. They are also within a specific program. Designers can plan to end a responsible to determine which households are eligible subsidy outright at a specified time, or they can choose a for subsidized electricity through the FBE program. South gradual phase-out process, which either reduces the subsidy Africa’s municipalities have registered 3.5 million indigent level over time or incrementally narrows the target group. It is households in 2017, about 20% of the population.164 About best to think through this phase-out during the design phase 2.1 million among them are receiving FBE through the electricity network.165 For households identified as ‘indigent’ to avoid ambiguity around funding amounts and timelines. that are not connected to the grid, the FBAE policy instructs municipalities to supply alternative sources of energy. As of When designers plan to end a subsidy completely, it is 2022, of South Africa’s 257 municipalities, 21 municipalities important that the end date is appropriately planned have supplied SHS to 150,000 households. This represents for and communicated to all stakeholders involved. about 4% of the 3.5 million indigent households, as identified Ideally, a subsidy will not end abruptly because funding has by municipalities.166 been depleted. Rather, it will run until a target date that is Adjustment / Exit: Setting up a long-term facility planned for in advance and that beneficiaries are aware of. In the case of Nigeria’s NEP program, in part due to an The FBE and FBAE were policies launched in 2003 and uptake that was higher than expected, the available funds 2006 respectively, establishing long-term institutional were depleted before the scheduled end date in 2023. arrangements to provide electricity to the poor. Funds are This sudden stop had many negative consequences, such disbursed from the National Treasury to municipalities. as sudden price increases for consumers and companies Municipalities then pay service providers (Eskom, municipal having to lay off people and stop distribution.168,169 This kind distribution companies and off-grid concessionaires) on of sudden end to the subsidy scheme should absolutely be behalf of FBE/FBAE recipients. avoided by careful design and ongoing monitoring. Households are required to register with municipalities to qualify for free basic services. A municipality’s role is to vet A gradual phase-out process allows for a smoother every application, selecting only those households that meet transition. Where subsidies are no longer needed, various criteria. Successful applicants are granted indigent companies can operate through phases of lower subsidy status. Municipalities often run awareness campaigns to amounts until the subsidy is completely removed, and ensure that households are aware of the application process. their own revenue streams can sustain them. Where a new Indigent status is not without end. The economic status of program or long-term facility is envisioned, a gradual phase- a family might improve over time, thus affecting its status. out creates a longer period for collecting feedback to inform To ensure that only the poorest families are catered for, the design of the follow-up program or facility. municipalities require registered households to reapply for indigent status on a regular basis, often once a year. • One option for such a phase-out is to progressively reduce the subsidy level, so that consumers and There are additional constraints. A municipality might not be companies can adjust to commercial rates and minimize able to service all indigent households that have successfully market distortion. registered, due to lack of funds or inadequate infrastructure. There is also criticism about administrative procedures at • Another alternative is to narrow the subsidy target municipalities making it difficult for eligible households to group, in which case the subsidy level could also be complete the application process.167 increased. This particularly applies to programs with a pro-poor strategy that aims to leave no one behind. 164 Government of South Africa, Cooperative Governance Traditional Affairs website. 165 ESI, 2021. The status of Free Basic Electricity in South Africa. 166 Department of Statistics of South Africa, 2022. Solar energy for the poor. 167 ESI, 2021. The status of Free Basic Electricity in South Africa. 168 World Bank, Nigeria Electrification Project 169 Energy for Growth, An update on the Nigerian Electrification Project: electrifying Nigeria’s most underserved 64 End-User Subsidies for Energy Access: A toolkit For example, Nepal’s energy subsidy started with a with stakeholders before and during program less targeted approach but progressively narrowed its targeting to reduce spillover risks and ensure efficient implementation. Communication is an important use of financial resources. In the last phase, the program aspect of a successful subsidy program, because targeted only the households experiencing the most multiple stakeholders in any program need to marginalization with highly subsidized products. have a clear understanding of the program’s setup and parameters. These stakeholders include both Active and transparent communication is crucial to any beneficiaries and those excluded by the program, as exit strategy. This is critical for households considering well as companies and local government entities. The using the subsidy and for participating companies who need framing of the subsidy – that is, how it is presented this information to plan their operations.170,171 Regardless, – is important to set the right expectations for all programs need to monitor and evaluate progress and market involved. It may be presented as a temporary discount, dynamics regularly and adjust the exit strategy if needed. promotion, or waiver, but the language should clearly convey that it applies only for a specified period unless the program has been designed as a long-term Key insight on phasing out: Program designers facility.172 should plan for a phase-out during the design phase and ensure this is planned for, monitored, and communicated well. Communication with beneficiaries and non-target populations 2.8 Communication and Feedback Effective communication is critical to beneficiaries Design framework since they are the subsidy recipients. A clear understanding on their part not only promotes subsidy Communication adoption but also helps mitigate the risks of market distortion and social tensions. The messaging can elaborate on the importance of subsidy programs for spurring commercial activity, for instance, and 3. Delivery explain why specific households are eligible while others are not. Ensuring that beneficiaries are well- 2. Subsidy informed about the subsidy allows them to appreciate 4. Verification level the product’s true value, and to understand that the price they pay is subsidized and does not reflect its market value.173 Some of the effective avenues of 5. Exit 1. Targeting communicating to end-users include promotional strategy campaigns and peer-to-peer networking. Public awareness campaigns may include door-to-door Pre-design elements campaigns, local events, and digital campaigns.174 Feedback Effective communication is critical to beneficiaries since they are the subsidy recipients. A clear 2.8.1 Communication understanding on their part not only Designers should prioritize developing a promotes subsidy adoption but also communication plan that will guide interactions helps mitigate the risks of market distortion and social tensions. 170 Open Capital Advisors Consultations 171 Africa Clean Energy (ACE) Technical Assistance Facility (TAF) and Open Capital 2020, Demand-Side Subsidies in Off-Grid Solar: A tool for achieving universal energy access and sustainable markets. 172 End-User Subsidy Lab learnings 173 SeforAll 2022, The Role Of End-User Subsidies In Closing the Affordability Gap. 174 Clean Cooking Alliance, Busara Center for Behavioural Economics 2023, Formative Literature Review: Clean Cookstove End- User Subsidy Research. End-User Subsidies for Energy Access: A toolkit 65 When designing communication plans for for non-target populations to ensure transparency. beneficiaries, designers may consider the following: This is mostly needed for people not targeted by the program who may encounter it, such as when the • Tailoring methods and messages to suit the program is running in their village or area. A clear specific context of beneficiaries: Tailoring the understanding of the selection criteria and the overall message to the unique characteristics of target design of the subsidy helps non-target populations beneficiaries is important in driving behavior change understand why they may be excluded. Additionally, in communities. This may be achieved through this understanding may provide them with insights highlighting how the product can be used.175 To on obtaining the same or similar products that are improve the power of subsidy messaging, program not subsidized. Communication can be done through designers may emphasize that the price of the community engagement activities and print media product is subsidized and their additional benefits. such as handbills. For example, communication to beneficiaries using clean cooking technologies could explain how these devices promote better health and benefit the Communication with companies environment.176 Lastly, the communication method needs to be aligned with the target population. For Companies are an important stakeholder and example, TV commercials may be ineffective in source of information during the design phase. some marginalized communities because they lack Companies provide an important perspective on televisions, making radio advertisements, flyers, and the market because they know their customers and handbills more effective options.177 potential subsidy beneficiaries, and their preferences around products and prices, among other types of • Explore partnering with trusted role models expertise. Companies may be involved in the design to deliver messages: Trusted role models may process through co-design workshops and in-person be influential figures within a community, such consultations, alongside consultations with national as teachers, spiritual leaders, or the elderly. For and local governments, communities and end-users. It example, SolarAid used schools to help promote is important to engage companies with experience in the adoption of its products.178 Messaging that supplying OGS and clean cooking solutions, ideally in originates from government entities should be the regions targeted by the program. treated carefully, as it may interfere with creating a free market. For example, end-users in some Designers need to clearly communicate the contexts may be accustomed to hearing about program design and requirements to companies. It public free goods from the government, so they must be clear to companies which people are eligible may have the impression that the energy products to participate, what products are eligible, and what are free.179 implementation will involve. It is beneficial to all parties if there is transparency on program elements, like how • Work with companies to deliver messages: When companies are expected to check customer eligibility, delivering through companies, hold companies which IT tools will be used, and how and when accountable to provide subsidy information to disbursements will be made. This allows companies to beneficiaries through, for example, program- consider participating in the program and to plan their branded grant certificates delivered together with operations once selected. It is also helpful to clarify the the product. role of companies in communication with beneficiaries. For instance, what companies are expected to Beyond communication plans for intended communicate about the subsidy, and whether there beneficiaries, it is important to develop messaging are any communication guidelines to adhere to. 175 Shields et al. 2020, Improving Smoke Alarm Self-Report Via a Prompted Questionnaire. 176 Clean Cooking Alliance, Busara Center for Behavioural Economics 2023, Formative Literature Review: Clean Cookstove End- User Subsidy Research. 177 Africa Clean Energy (ACE) Technical Assistance Facility (TAF) and Open Capital 2020, Demand-Side Subsidies in Off-Grid Solar: A tool for achieving universal energy access and sustainable markets. 178 SolarAid 2022, Annual Review. 179 End-User Subsidy Lab learnings 66 End-User Subsidies for Energy Access: A toolkit Regular engagement of companies during program to disseminate information about the program and, implementation helps monitor performance. in some cases, facilitate its implementation. For There should be direct lines of communication with example, EnDev’s Pro-Poor RBF in Rwanda used local each participating company individually, and it is governments to validate contracts between companies advisable for the program to appoint a focal point for and customers by signing and stamping them. This company communication in the team. This ensures was an additional step to mitigate the risk of leakage that companies always know who to talk to and that and helped with verification. Furthermore, local there is no miscommunication. Beyond individual governments may support the resolution of challenges communication, it may also be helpful to engage in implementing the program, for example, when participating companies and other key stakeholders in customers claim that systems were stolen or when group settings as well, particularly when more general there is a conflict between a customer and a company. feedback on the program is required. This helps to identify potential improvements and ensure continued Conversely, without alignment, local government alignment.180 may be a roadblock to implementation. Officials may inadvertently block the work of companies involved in the subsidy scheme if they are unaware of Communication with local governments the project and its goals, and thus do not know how the project will affect their constituents. Or they may Alignment with local governments improves the mistakenly misinform potential beneficiaries due to dissemination of program information and may a lack of knowledge. Overall, local governments are support select implementation components. Local charged with increasing access to basic services for governments have critical roles to play through their populations, so the program is advised to engage their offices and officials who are integrated into them to advance joint objectives. Designers may communities and, as such, are a key resource for create communication channels with local government engaging end-users and for the design process. through periodic check-ins, appointment of contact There is an opportunity to leverage their networks individuals, and email communication. © Netherlands Enterprise Agency 180 https://www.ace-taf.org/wp-content/uploads/2020/09/Demand-Side-Subsidies-in-Off-Grid-Solar-A-Tool-for-Achieving-Universal-Energy-Access-and-Sustainable-Markets.pdf End-User Subsidies for Energy Access: A toolkit 67 2.8.2 Feedback: Monitoring, Evaluation, and Subsidy programs should follow best practice around Monitoring and Evaluation (M&E) and Adaptation adaptive management181, from external standpoints Designers must continuously monitor the program’s (e.g. adjusting to market trends) and internal (e.g. performance against its set goals, remain aware of from an operational/implementation standpoint). It is current market contexts and dynamics, and create very important for subsidy programs to integrate a a framework for program adaptation. Subsidies are systematic feedback loop, by closely monitoring it and complex and their design needs to improve over time, taking an iterative approach to the design, adapting it based on learning. Creating feedback loops, periodical during implementation when needed. program reviews and subsidy redesign when necessary is crucial for subsidy programs to remain Monitoring: Program implementors must actively successful over time. Subsidies may need revision and continuously monitor the program’s performance if uptake is too low, or too high. Macroeconomic against its goals as well as continuously monitor the factors like currency fluctuations, inflation, and global market dynamics. Monitoring is about overseeing the supply chain costs can significantly influence program broader performance of the program and may inform implementation. In addition, it is helpful to continuously any necessary adjustments. A first step in effective engage end-users, companies, governments, and monitoring is determining what data are needed, development partners through channels such as includes the following steps: round tables and one-on-one check-ins to gather feedback on how the program can improve. This may Evolution of subsidized sales or service include changes in the targeting approach, increasing subscriptions over the duration of the program, or decreasing subsidy levels, engaging additional and broken down by the different segments of companies, streamlining procedures, etc. For a the population (socio-economic, geographic, subsidy to remain effective and efficient, it needs to be demographic) and by company, and data on able to adapt. whether customers continue to make payments and use subsidized products and services over time Figure 23: Continuous Monitoring, Evaluation, and Service provision by companies participating in the Adaptation of subsidy programs program and customer satisfaction Changes in the competitive environment and Design framework technology development, such as entry of new Implementation Communication companies into the market, or new technologies allowing for cost reductions Macroeconomic factors like currency fluctuations, 3. Delivery inflation, and global supply chain costs can significantly influence prices and affordability, thus 2. Subsidy impacting the subsidy program implementation. 4. Verification level Monitoring, evaluation, General feedback on the program, by keeping open lines of communication with participating companies aaptation 5. Exit and other stakeholders and make provisions for 1. Targeting strategy feedback from end-users Evaluation: using monitoring data, combined with Pre-design elements additional research (e.g., stakeholder interviews, surveys) and analysis, periodic evaluations (independent or self-initiated) can help program Feedback 181 A recent publication of IIED (IIIED, 2022. From what works to what will work: Integrating climate risks into sustainable development evaluation — a practical guide) provides insights and best practices for monitoring, evaluation, learning and adaptive management. 68 End-User Subsidies for Energy Access: A toolkit implementors assess what is (or is not) working, and and assumptions, must build in mechanisms to improve the possible reasons behind the results. Implementors the design as data and evidence are gathered, as well and funders may find it helpful to conduct such as to evolve as the context changes. evaluations regularly (for example, quarterly and annually) to assess any need for adaptation. Adjustments may be necessary to accommodate 2.9 Conclusion external influences on prices or other factors influencing the affordability gap.182 Well-designed end-user subsidies have the potential to generate significant impact and contribute Adaptation: Adaptive management is often to achieving SDG 7 by tackling a key barrier to represented as a cycle of planning, implementing, electrification and access to clean cooking – monitoring and learning. This process integrates the affordability. Although implementing end-user design, management, monitoring and evaluation subsidies poses challenges, careful consideration of the program in a framework that can be used to of risks allows for the design and implementation of test assumptions and adapt and learn as project responsible EUS programs, i.e. subsidies that are implementation unfolds. Each of the design effective in reaching the poor, efficient in terms of components of the program can be adapted based on use of limited public resources, and that limit market this process, for example: distortion. This is illustrated in the many case studies throughout this document. Eligible households and targeting approach can be defined more narrowly if a survey reveals that most It is important to reinforce the linkages between of the subsidy is being captured by urban wealthy subsidies and other market-building incentives households, whereas the program’s objective was (including policy and other forms of public to increase access among the rural poor financing). End-user subsidies are one of the many public finance mechanisms that can be implemented Eligible products and subsidy level can be revisited to improve the affordability of off-grid solar products if local currency has depreciated, pushing prices up and clean cooking solutions. Other mechanisms such for imported OGS products and deteriorating ability as supply-side subsidies, concessional consumer to pay finance, grant funding to research and development are needed for markets to grow. Beyond the energy Delivery of the subsidy: more companies can be access space, it is important to acknowledge the invited to participate into the program if the pace critical need for complementary market functions and of sales is lower than expected, or new companies enabling environment support for certain services and have entered the market products that are often unavailable and contribute socioeconomic development (such as business or Verification procedures may benefit from being smallholder farmer credit, business or farm inputs, streamlined if consultations with companies reveal advisory support, access to markets, aligning they are too slow or complex, and carry significant institutions and policies, etc.). costs and delays in disbursements In order to achieve and sustain SDG 7, more EUS Exit (or adjustment) strategy: may be modified if programs are needed at an accelerated pace, macroeconomic conditions change significantly, alongside other initiatives and innovations to increasing or decreasing the size of the population address affordability and overall socioeconomic that can afford clean energy development. We hope this toolkit supports designers and encourages collective learning to ensure no one It is important to understand that a subsidy program, is left behind. often designed on the basis of incomplete information 182 End-User Subsidy Lab learnings End-User Subsidies for Energy Access: A toolkit 69 © Clean Cooking Alliance 70 End-User Subsidies for Energy Access: A toolkit 03. The design framework in practice 3.1 Introduction This chapter aims to illustrate how the design framework Subsidy Delivery channel: Through companies via an for EUS programs can be used in practice. This is done delivery RBF by outlining two example projects that leveraged Company selection: Open to companies the framework in their design phase: EnDev EUS meeting minimum requirements through a call pilot projects in Uganda and Malawi. In addition, two for proposals projects having rolled out EUS for off-grid solar and clean cooking are analyzed from the perspective Subsidy disbursement: After verification of this toolkit: the Rwanda Electrification Fund and Fund management: A consortium of two third- Nigeria’s Electrification Project. Following from the party administrators previous chapter, which explained various design components and considerations, these examples Claim management: Through fund manager will show how such design choices can be applied Verification Led by an IVA that uses manual processes in specific market contexts. The goal here is not to prescribe solutions, but rather to highlight how various Exit Transition to a scale up program or gradual choices are made based on context factors, and how phase out at end they are interdependent. Communication Channels include mobile phone communications, radio, and community meetings 3.2 Designing an EUS pilot targeting the poorest in Malawi Key context factors for the design include:   EnDev has designed an EUS pilot project for Malawi, Malawi has one of the lowest electrification rates in the focusing on households designated as “poorer” and world. Only 18 percent of Malawians have electricity “poorest” by the Malawian government.183 This project access, with 11 percent connected to the grid and 7 was designed between Q3 2022 and Q2 2023, and percent utilizing OGS solutions. implementation is planned for 2024 and 2025. While several international and local OGS companies Table 9: Summary of design element selected in an EUS are operating in Malawi (including Yellow, Vitalite, pilot in Malawi SolarAid, Zuwa, among others), affordability of OGS systems poses a significant barrier to access, as 80 Design Selected option percent of the off-grid population is unable to afford Tier element 1 access at the prevailing market price. Targeting Geographic targeting Eligible Eligible products: Tier 1 OGS and locally Clean cooking market development has been limited products and manufactured Tier 1 ICS due to wider social economic contexts. Accessibility subsidy level and affordability of modern cooking solutions remains a Subsidy level: Percentage of product price with a maximum amount (cap) challenge. 98.8 percent of households rely on firewood and charcoal for cooking, leading to forest depletion and soil erosion. 183 Based on the Unified Beneficiary Registry (UBR) established by the Department of Poverty Reduction and Social protection of the Ministry of Economic Planning and Development. The UBR classifies households into five income categories ranging from poorest to rich. End-User Subsidies for Energy Access: A toolkit 71 The government of Malawi has put in place a strong who could support the government in scaling up. If a enabling environment for off-grid electrification, scale-up is not possible, the project will instead focus on including enhanced regulations and the removal of a smooth and well-communicated phase-out. various taxes and levies, as well as market development initiatives, such as the World Bank-supported Malawi The project is set up as a 3.5-year pilot with a strong Electricity Access Project (MEAP) which provides EUS focus on identifying a scalable model and developing for solar home systems. That said, commercialization a scale-up strategy with support from the government remains limited due to the low ability to pay and difficult of Malawi and the World Bank to reach ultra poor market conditions. households not benefitting from other government projects The Department of Poverty Reduction and Social protection of the Ministry of Economic Planning and Targeting – The project leverages an economic targeting Development has established a Unified Beneficiary approach to identify ultra-poor households. Economic Registry (UBR). The UBR provides a central, digital targeting was selected due to the presence of available repository for data on geographic, demographic, and data on the ultra-poor households targeted. Based on socioeconomic characteristics of households and was proxy means testing, the households registered in UBR are developed to support the implementation of various classified into five income categories, with “poorest” and social protection and other programs in Malawi. The “poorer” being the lowest two. 40 percent of the Malawi UBR classifies households into five income categories population is classified as “poorer” and 10 percent as ranging from poorest to rich. “poorest.” During implementation, the pilot will utilize KOBO Toolbox digital tool to confirm beneficiary eligibility, specify The above points clearly demonstrate the need for end- the requisite subsidy level per technology, and register sales.   user subsidies in Malawi, given the high proportion of the population without energy access and their limited Eligible products and subsidy level – Given the high ability to pay for relevant goods and services. Given poverty level of the target population, the project will only this context, the pilot aims to reach 20,000 ultra-poor support Tier 1 OGS products – defined under the multi-tier households with Tier 1 OGS and 190,000 ultra-poor framework – and locally-manufactured, wood-burning stoves. households with ICS. These stoves are an entry-level solution that saves more than 40 percent on fuel consumption compared to traditional World Bank-funded Ngwee Ngwee Ngwee Fund (NNNF) cooking methods. Their inclusion in the project is based provides end-user subsidies to an off-grid population on the consideration that a) charcoal that is often illegally that is close to being able to afford commercial prices. produced and at a high environmental cost; and b) there are Their subsidies are thus significantly lower and less already a handful of pellet-fueled stoves on the local market targeted than the ones proposed under this pilot. For with limited fuel distribution chains the cooking sector, the project aims to identify a partner Figure 24: Wood-fired cook stove eligible for subsidy (left), and various Tier 1 OGS products (right) 72 End-User Subsidies for Energy Access: A toolkit Two fixed subsidy levels have been established for both subsidized products are sold only to eligible beneficiaries, OGS and ICS, with a higher subsidy level for households that beneficiaries have not already received a subsidy, and classified as “poorest” in the UBR. EnDev calculated can reveal whether beneficiaries have resold products for a subsidies based on available data on prices, as well as profit. The fact that this is a pilot project, the choice of remote recent “ability to pay” surveys. The project sets a subsidy regions and lower tier products has led to an emphasis on level as a percentage of the product price for each manual verification processes (notably the cook stove is not category and household category up to a maximum (cap) “technology enabled”). amount, which is calibrated to the affordability gap. Based on the price point of locally manufactured wood-burning Exit strategy – If successful, the pilot may be scaled-up stoves, the indicative subsidy amount for an ICS solution through a government fund with additional World Bank is approximately $3 for the “poorest” households. The support. Lessons from this pilot may also enhance the indicative subsidy for an OGS solution is approximately $153 targeting of subsidies in the existing NNNF under the MEAP for the “poorest” households, which is close to 90 percent project, potentially allowing for broader implementation of the average price of Tier 1 OGS products (with PAYGo) in of that project. If a scale up is not possible, the project will Malawi. These preliminary calculations will be adjusted prior instead focus on a smooth and well-communicated phase- to the start of the project and may be readjusted again over out. the course of the pilot.   Subsidy communication and coordination – Given that Subsidy delivery – The pilot will use qualifying cook the project targets those most marginalized, the project will stove and OGS companies as subsidy delivery channels partner with government and community leaders to provide via an RBF approach. Delivery through companies was awareness-raising and logistical support. All communication chosen over direct subsidies to consumers given that a will be developed with consideration for the literacy rates of large proportion of the population is eligible, making direct beneficiaries. Key channels for communication will include delivery of consumers costly. The selection process for mobile messages (the UBR includes phone numbers) participating companies is a competitive call for proposals, provided that this does not violate data privacy rules, printed encouraging local company involvement. Designers opted materials, radio programs, and meetings. to select a limited number of companies given the small size of the project. Criteria for selection will include registration EnDev will collaborate with key stakeholders including the with the Malawi government, rural presence, and distribution Ministry of Energy, the government’s team responsible for experience. International companies will be required to the UBR, the World Bank on the MEAP project, and the World partner with local entities to ensure sustainability. Subsidy Bank’s Energy Sector Management Assistance Program disbursement will follow independent verification of eligible (ESMAP) team. EnDev plans to establish a sector-specific sales. OGS companies may request an advance payment working group for key players in OGS to coordinate and (up to 30 percent of the contract value) to use as working share knowledge. In the cooking domain, EnDev will utilize capital for acquiring products. Given the much lower prices existing national coordination groups for updates and and subsidy levels for the selected ICS solution, no advance feedback. payment will be offered for ICS suppliers. A consortium of two third-party fund managers will be hired by EnDev to oversee the pilot and coordinate with the Malawi Key take-aways for designers  government for a seamless transition in case of scaling up.  The program is able to adopt a highly targeted approach Verification – The eligibility tool (though KOBO Toolbox) focusing on the poorer and poorest households by works as a verification mechanism at the point of sale, leveraging data from the existing social protection scheme allowing companies to confirm a customer is eligible for a implemented by the Government of Malawi. given subsidy level. In this case, the tool will confirm the recipient’s socioeconomic category in the UBR, and they Given that Malawi is a nascent market, with low access and a have not already benefitted from the subsidy, among other high affordability gap, there is a strong need to complement criteria. In addition, participating companies will be required an end-user subsidy scheme with other market-building to submit sales reports for eligible products to the fund interventions, which the government has embarked on in the manager. An IVA will be recruited to confirm sales using form of several favorable policies (i.e. tax exemptions) and data and document verification, phone interviews, and field other programs such as the NNNF under MEAP. visits. This manual verification will double-check that the End-User Subsidies for Energy Access: A toolkit 73 For this pilot, the designers are using a digital tool (hosted Verification Led by an IVA who will utilize an online data by Kobo Toolbox) to serve as a beneficiary eligibility check platform complemented by manual processes through identification of ultra-poor households. Verification such as phone and field verification by the IVA will be manual. In case of a scale-up after Exit Transition to a scale up program or gradual this pilot, the designers may opt for more technology- phase out at end enabled verification, where possible, to lower the costs of verification. Communication Communication channels include mobile phone communications, radio, and community meetings There is potential pathway for scaling after this pilot, which forms part of a clear exit strategy. Key context factors for the design include:   3.3 Designing an EUS pilot targeting 94 percent of households in Uganda rely on biomass for refugees and host communities in cooking and, although electrification has been rising, 43 Uganda  percent of households do not have any form of energy access. This is often linked to affordability challenges, EnDev has designed an EUS pilot project for Uganda, particularly for rural/refugee hosting households and focusing on both refugees and host communities as well as refugees. rural poor. Refugee hosting districts have the highest poverty levels in the country per the Multidimensional Poverty Index The off-grid solar market in Uganda is considered (MPI) and are thus the most at risk of being left behind for mature, marked by healthy competition among SDG 7.184 The pilot was designed between Q3 2022 and Q2 companies and a high share of pay-as-you-go (PAYGo) 2023, and implementation is planned for 2024 and 2025. It sales. However, refugees and their host communities supports both off-grid solar and cooking solutions. are not considered commercially viable markets. Literature suggests that the ability to pay for these communities is three to four times lower than the Table 10: Summary of design elements selected in an EUS national average.185 pilot in Uganda Given the large population with limited ability to pay in Design element Selected option the target communities, the pilot aims to reach 20,000 Targeting Geographic targeting households with Tier 1 OGS and 210,000 households with improved cook stoves (ICS), which are Tier 1-2, Eligible products Eligible products: Tier 1 OGS and Tier 1-4 CCS and subsidy locally manufactured cook stoves that are more fuel Subsidy level: Percentage of product price efficient than traditional cooking methods. It also level with a maximum amount (cap) allocates budget to reach 6,000 households with more Subsidy delivery Delivery channel: Through companies via an expensive, higher-tier clean cooking products RBF approach The project is set up as a 3.5-year pilot with a strong Company selection: Open to companies focus on learning and adaptation. It has potential for meeting minimum requirements; international companies required to partner with local scaling by the Government of Uganda through the entities. World Bank-funded Electricity Access Scale-up Project (EASP). A key goal is to focus on both refugees and host Subsidy disbursement: After verification; communities as well as rural poor. option to obtain 30% advance for companies that have demonstrated working capital Targeting – The project has adopted a geographic constraints targeting approach by focusing on selected refugee-hosting Fund management (including claims): districts, as well as additional districts targeting the rural Through third-party administrator (poor) population. Refugees and host communities/rural poor in these districts are eligible. The choice of these districts 184 Multidimensional Poverty Index Report 2022 – Uganda 185 UOMA, 2020. Reaching unserved refugee markets in Uganda. 74 End-User Subsidies for Energy Access: A toolkit is informed by data related to poverty status as measured affordability gap for each product category (tier). The project by the Multidimensional Poverty Index (MPI) of Uganda, sets one subsidy level as a percentage of the product and by data available from other programs implemented by price for each category up to a maximum (cap) amount, government and development partners, such as UNHCR.186 which is calibrated to the affordability gap. The subsidy Given the correlation between these districts and poverty calculation for a Tier 1 solar lantern system is illustrated in the levels, geographic targeting of vulnerable areas was chosen table below. as an appropriate targeting approach. The pilot will employ a phased approach, considering expansion into additional communities after nine months of implementation. The Table 11: Example of subsidy setting mechanism risk of subsidized products being resold in commercial Equipment Price Estimated Estimated Subsidy markets adjacent to targeted counties will be mitigated by reference willingness affordability technology-enabled customer eligibility checks as described to pay gap below, as well as regular monitoring of market prices.   Tier 3 ICS $125 with $7.3 $52.4 42% of the on PAYGo installments monthly for price with Eligible products and subsidy levels – Linked to the project for 10 10 months (= price maximum of goals, support will be provided for Tier 1 OGS products, months minus $52.4 as well as cooking solutions ranging from Tier 1 to 4 as (totals Willingness defined by the respective multi-tier frameworks. Because ($21 deposit $73.4) to Pay) (to be and $10.5 refined of widespread poverty in the selected areas, the majority of monthly how the funding will be allocated to lower-tier solutions. payment) subsidy will reduce both Given that Uganda has a well-developed market of energy deposit and companies, the project leveraged price data from reference installments) products already available in the market for both cash and Tier 1 solar $34 (cash) $13.10 $21 62% of price PAYGo sales to estimate the affordability gap for selected lantern with max of products. EnDev used existing data on ability and willingness + phone (= price $21 to pay and complemented this by conducting additional charging minus WTP) “willingness to pay” surveys to calculate the indicative on cash ©GIZ EnDev 186 Uganda Bureau of Statistics 2022, Multidimensional Poverty Index Report End-User Subsidies for Energy Access: A toolkit 75 From previous experience, EnDev chose to equalize the periodically checking 100 percent of the data for consistency subsidy level between refugee and host communities to and conducting spot checks on the accuracy of relevant avoid social tensions. EnDev may revisit the subsidy levels paperwork. This will be complemented by additional checks when expanding to other locations (where affordability on a representative sample of customers, in the form of metrics may be different) as well as more broadly throughout phone verification (90 percent of the sample) and field implementation depending upon product uptake. verification (10 percent). Additionally, verification will be used to confirm the quality of access obtained by the households. Subsidy delivery Exit strategy Given the market maturity, the project will leverage existing cook stove and OGS companies via an RBF approach to If successful, the pilot may be scaled up through the existing deliver the subsidies. The identification and selection of EASP, which is managed by the Government of Uganda companies for the project will be based on a competitive through UECCC and supported by the World Bank. Close call for proposals, allowing many companies to participate in involvement of these two players during implementation will the pilot. Subsidy disbursement will be done upon verified allow for a smooth transition of the pilot into the scale-up sales, with an option for companies to receive a 30 percent phase. If a scale up is not possible, the project will aim for a advance payment. To qualify for this advance, companies will smooth and clearly communicated phase out. need to demonstrate that they do not have sufficient means to meet the working capital requirements. EnDev will hire a Subsidy communication and coordination third-party fund manager to manage the pilot, who will work closely with the Ugandan government to ensure a smooth EnDev will collaborate closely with key stakeholders in transition if later scaled up. the design, implementation, and exit strategy of the EUS pilot. A collaborative working group is envisioned for Considering the difficulties and higher costs of serving the knowledge sharing, harmonization of subsidy designs, and selected locations, the pilot project will partner with other communication strategies. The pilot will develop detailed programs and stakeholders to provide complementary communication activities to increase awareness about the supply- and demand-side support, such as awareness- program using a variety of channels. raising campaigns, logistical support, supply-side RBFs, and credit facilities. Despite Uganda being considered a mature market, its more remote and poor areas remain underserved. Key take-aways for designers Companies need incentives to enter the refugee-hosting districts, which are commercially less attractive. The program builds on the supply-side subsidies implemented by EnDev As multiple companies are operating in Uganda, the designers could leverage existing data and work with and other partners, including Mercy Corps, to support existing companies, as opposed to having to collect companies’ expansion into these underserved areas. significant data as part of the design process. Eligibility Tool The program builds on blended demand and supply-side subsidies implemented by EnDev and other partners, An eligibility tool is utilized to confirm customer eligibility including Mercy Corps, to support companies’ expansion into under the RBF. The tool (hosted by Kobo Toolbox) will also underserved areas. This is important because while Uganda inform companies about the subsidy level the customer is may be a mature market, its more remote and poor areas entitled to. Only if the tool confirms that the eligibility criteria remain underserved. are met, can the company then proceed with the sales and Given multiple demand- and supply-side interventions register the client in the tool. The tool will also register sales being implemented in Uganda, there is need for strong made under the RBF to update the list of eligible customers coordination of subsidy structures across programs, which and track progress to ensure that households only benefit the designers have catered for by involving a variety of once from a subsidized system. stakeholders in the design and by planning for a sector specific working group during implementation made up of Verification industry stakeholders, local associations, and private sector companies. Because the project will partner with many participating The potential pathway for scaling up after this pilot has been companies and in remote areas, an independent verification made clear during the design phase, which has informed agent (IVA) will be recruited to manually verify sales by communication plans for the project. 76 End-User Subsidies for Energy Access: A toolkit 3.4 Analyzing EUS of Rwanda REF Subsidy delivery Through companies via RBF Window 5 Open to all companies meeting eligibility criteria (until all funds are allocated to In 2020, Rwanda’s Renewable Energy Fund (REF), funded companies) by the World Bank and managed by the Development Bank of Rwanda (BRD), launched a $30 million RBF subsidy called Subsidy claims disbursed to companies in 2-3 REF Window 5 with a goal to connect 370,000 households instalments within a period of three years through SHS. Window 5 complements another four REF windows providing end-user financing and financing to Subsidy is managed by government institutions OGS companies in the form of debt. REF Window 5 builds Verification Verification of customer eligibility by on the learnings from EnDev’s Pro-Poor RBF pilot, which companies at point of sale via eligibility tool looked to accelerate access to OGS products for the poorest households.187,188 Verification of claims via IVA desk check, and phone calls and field visits for a sample of customers within claim In addition to the OGS RBF, funded via the same World- Bank project189, a clean cooking RBF window subsidizes Adjustment and Periodic adjustments of subsidy level and purchases of clean and efficient cooking solutions, following exit delivery based on progress and market the same design and implementation structure of REF conditions Window 5. The specifics of the clean cooking window are Subsidy eligibility may need to evolve to cover not covered in this case study. the replacement of SHS at end of life In August 2023, the project design described in this case study had to change due to modifications in the Ubudehe Key context factors for the design include: social protection system used to target subsidies. The new subsidy structure adopted by REF Window 5190 is not Rwanda’s National Electrification Plan (NEP) foresees explored in this case study. This case study focuses on the a major role for off-grid electrification—reaching 30 targeted subsidy provided until August 2023, which has percent of the population by 2024—as transitory been used to connect 330,000 households over a period of solution before the grid arrives. The government has three years. identified the villages that by 2024 are meant to be served with off-grid electrification solutions192. Table 12: Summary of design elements of the subsidy under Rwanda’s OGS market is classed as a mature market193, Rwanda REF191 with a high penetration of OGS products (~22 percent of households connected by off-grid solutions by January Design element Selected option 2024), many companies participating, a high share of Targeting Geographic and economic targeting, PAYGo sales, relatively strong quality assurance, and focused on off-grid areas and lower-income policy stability. households Eligible products Quality-verified SHS providing Tier-1 access At the time of the design of Window 5, off-grid access and subsidy and above. expansion through SHS had been slowing down despite level the provision of debt financing under REF and favorable Subsidy level varies with socio-economic policies, such as tax exemptions. Affordability of OGS category of customer (Ubudehe 1, 2 and 3). products of Tier 1 and above had emerged as the Levels are defined as a % of price of the SHS with an absolute cap in Rwandan Francs. key challenge. As the OGS market became saturated 187 End-User Subsidies Lab, 2022, End-User Subsidies Lab Official Launch Session: Rwanda End-User Subsidy 188 Gogla, Africa Clean Energy, World Bank, 2022, End-User Subsidies lab Official Launch: Session 1 189 Rwanda Energy Access and Quality Improvement Project 190 The project removes the economic targeting based on the Ubudehe classification, but continues to applicable to residents of SHS electrification areas only, which are primarily inhabited by households in Ubudehe 1-3 categories. 191 Information sourced from the Project Operation Manual for REF Window 5. This design does not reflect changes introduced since August 2023 due to the modification of the Ubudehe system. 192 Rwanda Energy Group, 2022. Rwanda Electricity Access Development Plan 2018-2024. 193 Lighting Global/ESMAP, GOGLA, Efficiency For Access, Open Capital Advisors 2022, Off- Grid Solar Market Trends Reports 2022: State of the Sector. Washington, DC: World Bank. End-User Subsidies for Energy Access: A toolkit 77 for higher-income households, companies started to Eligible products and subsidy level expand to customers with lower or irregular income, which led to slower sales and an increase in their The subsidy is applicable to products providing Tier- default rate. 1 access and above, that comply with the Ministerial Guidelines for Minimum Standards of SHS194, which details Rwanda has developed a system to classify the all quality standards, warranty requirements, and after-sales population into socio-economic categories. The care and system service level requirements for the Rwandan Ubudehe program, under the Ministry of Local market. Government (MINALOC) and managed by the Local Administrative Entities Development Agency (LODA), The subsidies are expected to be fully passed on to end classifies the population into 4 levels (Ubudehe 1 users to match the affordability gap. The subsidy level to 4, with Ubudehe 1 being the lowest), based on was estimated based on (i) the average PAYGo price of factors such as income, employment, profession, and quality-verified market-leading Tier-1 products and (ii) end- land ownership. These levels are assigned by local users’ ability to pay in each of the three eligible Ubudehe government. categories195. The subsidy level was determined as a percentage of the retail price, however, it was also capped A pilot project led by EnDev (Pro Poor RBF, 2019- at a maximum absolute subsidy level. Subsidy levels range 2021) had developed mechanisms to provide targeted from 45 percent of price for Ubudehe 3 customers (capped subsidies to lower income households within off-grid at FRW 50,000, or ~$55 at the exchange rate when the zones (leveraging information systems of LODA and project started in 2020) to 90 percent of price for Ubudehe 1 Rwanda’s utility company REG). customers (capped at FRW 100,000, or ~$110). By combining an absolute and a relative maximum subsidy amount, the Targeting program design safeguarded against the possibility of over- subsidization or of participating companies inflating the retail Given the government had identified specific areas for price. While this combined percentage and absolute cap off-grid electrification via the National Electrification Plan, method introduces significant complexity, this is reasonable and affordability of SHS had been identified as the main given the high level of the subsidy. barrier for lower-income households, a targeted subsidy was warranted. Eligibility criteria for the subsidy include Subsidy delivery being a resident of SHS zones as identified by National Electrification Plan, and pertaining to Ubudehe categories 1, The subsidy was delivered through companies via RBF. To 2 or 3. Economic and geographic targeting was possible participate in the project, companies need to apply providing given the availability of information systems to support evidence they meet eligibility criteria196 and a business plan both. The project relies on two government databases to including their proposed pricing scheme, expected volume determine the eligibility: (i) LODA’s Monitoring and Evaluation of sales eligible for RBF, and the total amount of grant they Information System (MEIS) to check whether the customer are requesting. meets the Ubudehe eligibility criteria, and (ii) OMIS, an online information system, which allows REG to track developments in grid and off-grid electrification. By combining an absolute and a relative maximum subsidy amount, the An eligibility tool (a web-based tool) was developed by the government to support the implementation of RBF. The tool program design safeguarded against is used by companies to do a subsidy eligibility check on the possibility of over-subsidization or potential customers at the point of sale, as well as determine of participating companies inflating the the amount of subsidy the customer is eligible for. For this retail price. purpose, the tool connects with the databases mentioned above. 194 The ministerial guidelines provide standards for plug-and-play SHS under 350 Wp (matching quality standards used by Verasol, contained in IEC TS 62257-9-8) and for component-based SHS. 195 The estimated end-users’ ability to pay was based on various survey data on disposable income and energy expenditure by Ubudehe categories. 196 Eligibility criteria include being licensed to operate, have adequate funding, satisfactory quality of products and operations, satisfactory pricing scheme, warranty and after-sales service, adequate accounting and information management systems, and gender quotas, among others. 78 End-User Subsidies for Energy Access: A toolkit Once accepted into the project, a grant agreement is signed, During the project, subsidy levels can be adjusted to and a certain amount of grant funding is earmarked to reflect market changes (such as rate of inflation, foreign them. This amount is based on their application and the currency fluctuation, overall cost of living, etc.) and to appraisal of their capacity by the project’s implementation ensure sustainability of the program. The review of market unit. This grant allocation can be modified over time based conditions takes place at least every six months or on an on the performance of companies (reduced or increased). as-needed basis. This review takes place every 6 months. Results The management of REF Window 5 is done by different institutions within the government. Energy Development The project has been successful in increasing off-grid Corporation Limited (EDCL), a subsidiary of REG, focuses on connections among the poorer populations and in attracting the implementation of RBF197, and the Development Bank OGS companies to the market. By August 2023, 24 OGS of Rwanda (BRD), focuses on the grant management and companies had signed agreements with BRD, with the full disbursements. $30m allocated among them. 330,000 off-grid households have been connected under PAYGo (31 percent of sales) With regards to disbursements, companies make quarterly and cash sales (69 percent of sales), with 75 percent of the subsidy claims to EDCL for eligible sales during the period. subsidy ($23m) committed to these sales. About 40 percent Considering the high proportion of the subsidy on the of this commitment has already been disbursed, mostly total price of product, the subsidy is disbursed in 2-3 against Milestone 1 (after installation) and some Milestone 2 instalments (milestones) spread over a period of three (after 12 months of service) verification. years to incentivize after-sales service from companies. The disbursement schedule depends on the price of the product, Reaching the remaining 8 percent of households allocated the Ubudehe category, and whether the sales are made for electrification via OGS (to reach the 30 percent objective) on cash or PAYGo. All disbursements are contingent on the will likely be challenging, with the remaining market being results of a verification process. The initial disbursement largely composed of households in hard-to-reach areas with takes place after the verification of the installation and the limited ability to pay. final disbursement after adequate customer service for three years. Key take-aways for designers Verification Strong electrification planning and coordination, and clear The verification of reaching the milestones for each policy guidance on the allocation of subsidies, play a very disbursement is conducted by EDCL through an IVA. The important role in the design of a good subsidy project. verification process includes a desk check of all customers submitted in a claim, and phone and field verification for a Availability of good-quality data on the population eligible for sample of customers. subsidies allows for effective targeting mechanisms. Adjustment and exit strategy Despite their complexity, well-designed targeted subsidies The project was designed to substantially contribute can be delivered through companies. to Rwanda’s objective of achieving universal access to electricity by 2024, reaching the off-grid electrification rate Digitizing eligibility checks and claims leads to efficient of 30 percent as a temporary measure, and connecting program management for both companies and households to the grid over the longer term. Depending administrators. With greater complexity, the need for on how long it takes to achieve universal grid access, the digitalization increases to be able to ensure process quality, effective monitoring and fast processing. Important factors project may need to be adjusted to be able to sustain off- to consider for digitalization: local and company level data grid connections, by replacing SHS at the end of their usage protection regulations and policies, compatibility with capacity. devices used by company agents, network coverage and reliability, and available local capacity for trouble shooting and improvements. 197 Manages eligibility tool, OMIS, onboards companies, receives claims and checks their validity, conducts verification of sales (through IVA) by means of desk checks and phone and field surveys. End-User Subsidies for Energy Access: A toolkit 79 The periodic revision and adjustment of subsidy Table 13: Summary of design elements in Nigeria’s end-user levels, allocation of subsidies to companies, and other subsidy200 implementation aspects, help keep the project on track. Design element Selected option Targeting Untargeted 3.5 Analyzing EUS of Nigeria’s OBF for Eligible products Eligible products: Quality-verified OGS SHS under NEP and subsidy products of Tier 1 and above. Since 2019, the World Bank has been supporting the Nigeria level Subsidy level: fixed amount for each system Electrification Project (NEP) to increase access to electricity size/ level of service category, proportional to services for households, public educational institutions, price and micro, small, and medium enterprises (MSMEs). NEP Subsidy delivery Delivery channel: Through companies via RBF is managed by the Rural Electrification Agency (REA), the implementing agency of the Federal Government of Nigeria, Company selection: Open to all companies tasked with electrification of rural and unserved communities. meeting minimum requirements, on a rolling One of the project’s components, the Output-Based Fund basis until funds are exhausted (OBF) for SHS, promotes SHS through an RBF mechanism. It is a large and ambitious project component, with $60m Subsidy disbursement: one-off after of funding and the objective to improve energy services verification for millions of unserved and underserved households and Fund management: government agency with MSMEs. support of a grant administrator Verification Assisted by software platform drawing data RBF was implemented in two phases. In the first phase, OGS from participating companies’ CRM systems. companies were provided with a supply-side RBF ranging from 7 to 20 percent of the reference price for different Exit Transition to a more targeted subsidy structure product categories198. These grants were meant to finance investments in people, training, advertising, processes, and Key context factors for the design include: logistics. About 400,000 OGS products were sold across all 36 Nigerian states during this first phase. In the second Nigeria experienced sustained inflation since the phase, which was introduced in January 2022, RBF was onset of the COVID-19 pandemic, coupled with the increased to 40-60 percent, including a 20 percent end-user depreciation of the local currency, impacting prices of subsidy199. Over 1 million households have been connected imported OGS inputs and deteriorating people’s ability since the end-user subsidy was introduced. The RBF project to pay for them. was interrupted in December 2022 due to funds being exhausted. Very large market opportunity for OGS products, with (i) over 85 million people lacking access to electricity, This analysis refers to that second phase providing the end- and (ii) businesses and households connected to the user subsidy. electricity network facing unreliable and insufficient supply, a gap often filled with power from petrol and . diesel generator sets that are costly and highly polluting to people and the environment201. 198 The subsidy level varied based on the product category, defined by the MTF and a scale of 6 levels. Tier-1 level-1 products (the most basic) received 20% of a fixed reference price for products of that category, predetermined by a benchmarking study. Tier-5 level-6 products (higher-end products) received 7% of the reference price of products of that category. 199 20-40% of reference price given as supply-side subsidy, based on the product category, and 20% as end-user subsidy. 200 NEP’s website and NEP’s Implementation Manual. 201 World Bank, 2013. Press release : Nigeria to Expand Access to Clean Energy for 17.5 Million People 80 End-User Subsidies for Energy Access: A toolkit An OGS market classified as “emerging”, with sales 20 percent of a reference product price for each service rapidly increasing but still a large electricity access gap level, resulting in a subsidy ranging from $18 for an entry- remaining202. level Tier-1 product, to almost $1,000 for PV systems of 2-kWp and above providing Tier-5 access205,206. High penetration of low-quality OGS products into the market. GOGLA research estimates that 73 percent of The choice of a subsidy available to products of all sizes and products in the market are non-affiliate products203. proportional to price is aligned with the objective of scaling the market for the benefit of poor and wealthy populations Uneven penetration of OGS products in the country. alike. 50 percent of OGS sales since 2019 are concentrated in four out of the 36 Nigeria states (Oyo, Kano, Ogun, Subsidy delivery Ondo). These four states only represent 16 percent of the Nigerian population204. Consistent with the priority of developing a market for OGS products, and given the subsidy was untargeted, companies Targeting were chosen as a delivery mechanism, via RBF. Given the large market size, the large size of the project, and a Given the large untapped market and underdeveloped growing competitive environment, the project was open OGS distribution and service networks, NEP prioritized the to all companies meeting minimum eligibility criteria, on development of the market through: a rolling basis until funds are exhausted. Criteria were set up to ensure that companies have the capacity to deliver a supply-side RBF (of about 20-40 percent of the quality-verified products and after-sales service at scale, product’s price) to help companies expand operations and to manage the reporting and audit requirements. A and to enable them to serve more customers quickly, pre-qualification process was put in place for this purpose, managed by a Grants Administrator engaged by REA. a lower end-user subsidy (about 20 percent of price), After receiving approval from the project’s investment coupled to the supply-side RBF, to tackle widespread committee, companies get qualified through the signing of deterioration of ability to pay in the context of the a grant agreement. To date, more than 50 companies have pandemic. qualified for the OBF, including both local and international companies. For these reasons, an easy-to-implement untargeted subsidy was chosen. With regards to disbursements, qualified companies submit monthly or quarterly claims for systems sold within the Eligible products and subsidy level period. The claim includes information on each customer who purchased a system, the product’s specifications, and All quality-verified OGS products providing Tier-1 access pricing. After these claims are verified, the full subsidy is and above are eligible. A fixed unit rate in USD was defined disbursed to companies. There is no cap placed on the total for each product admitted into the program, based on amount of subsidy companies can claim. Funds are allocated its capacity and level of service (6 levels of service were to companies based on their sales, on a first-come-first-serve defined). The fixed rate of the end-user subsidy was set at basis. 202 Lighting Global/ESMAP, GOGLA, Efficiency For Access, Open Capital Advisors 2022, Off- Grid Solar Market Trends Reports 2022: State of the Sector. Washington, DC: World Bank. 203 Lighting Global/ESMAP, GOGLA, Efficiency For Access, Open Capital Advisors 2022, Off- Grid Solar Market Trends Reports 2022: State of the Sector. Washington, DC: World Bank. Affiliate products are defined as products sold by companies that are connected to any of the partner organizations involved in the semiannual GOGLA sales data collection and which share their sales data. Tbis includes GOGLA members or companies selling products that meet VeraSol quality standards. Non-affiliate products are sold by companies that are not within the matrix of companies distributing affiliate products listed above. Much less is known about the quality and level of Tier access their products provide. 204 Data collected by REA under NEP. 205 The 20% refers only to the end-user subsidy amount that was to be passed on as a price discount to customers. The total amount of the RBF, including both a supply-side and end-user subsidy, ranged from 60% of the reference price of tier-1 level-1 products (equating to $55) to 40% of tier-5 level-6 products (equating to $1,946). 206 While subsidy rates were defined as USD amounts, subsidies were paid out in Naira, at the official exchange rate on the date of disbursement. End-User Subsidies for Energy Access: A toolkit 81 A Project Management Unit (PMU) established under caused funds to be fully utilized before such progressive REA was responsible for the overall implementation of phaseout. the subsidy program. A Grant Administration firm was engaged to support subsidy implementation. This firm is Results responsible for tracking the grant allocations to companies, screening applicants against eligibility criteria at the pre- The project exceeded expectations, with almost 1.1 million qualification stage, and tracking progress toward meeting devices sold, verified, and RBF paid for, against an original the milestones and results achieved by the grantees. The target of 340,000. Participating companies indicated that the Grant Administration firm also interfaced with the IVA to subsidy made products much more affordable, made the conduct claims verification and report back to the PMU for selling of entry-level low-margin products viable, and allowed processing. them to expand distribution networks to cover more states, and into more rural and poorer areas. RBF management platform (Odyssey) Figure 24 shows analysis of sales data collected by REA207. To simplify and accelerate administrative processes, REA put The comparison of sales under phase 1 of the OBF with in place a software platform to manage the RBF activities, that of phase 2 (when RBF was increased, and the end- including receiving and treating companies’ applications user subsidy introduced) provides insights on the impact for pre-qualification, submission of claims by companies, of the subsidy in terms of the profile of products sold and supporting remote verification by the IVA, and reporting. customers reached: Verification The majority of OGS products sold in both phases were Tier-1 products of a capacity of under 50 Wp. During After receiving claims through the Odyssey platform, phase 2 though, there was a substantial shift from entry- payments to companies are made after the verification of level Tier-1 product to mid-tier Tier-1 products, probably each sale has been conducted by the IVA. The verification as a result of products becoming cheaper and thus process can be done through either (i) phone calls and more affordable. field visits to a sample of end users within the claim, or (ii) by leveraging the data in the Odyssey platform. The Most OGS products were sold to customers residing in latter verification method was introduced to accelerate the wealthiest wards of Nigeria (wealth quintile 5), but the verification process. It played an important role in the the penetration of products into the poorer quintiles exponential increase in sales during phase 2. The method is increased slightly during phase 2. This may be the based on Odyssey linking via an Application Programming result of products becoming more affordable, but also Interface (API) with the companies’ customer relationship of companies expanding their distribution networks into management (CRM) systems. Through the platform, the IVA harder-to-reach areas. verifies that for each claim made, a connection was made to an actual customer from whom payment has been received, The penetration of OGS products into rural areas and that the subsidy was used to reduce the price. increased slightly during phase 2 of the project, which is probably also a result of products being more affordable Exit strategy and companies expanding their reach. It was planned that the subsidy would be progressively phased out. The very rapid uptake and rollout of this subsidy 207 This analysis was based on sales data captured by the Odyssey platform, including date of sale, system size, and GPS coordinates of the customer for the 1.4m products sold under the project. The location of customers was overlayed with ward-level poverty mapping and rural/urban classification produced by the World Bank. 82 End-User Subsidies for Energy Access: A toolkit Figure 25: Profile of OGS products sold, and customers (DARES) project. This project will provide RBF and end- reached, by project phase user subsidies for SHS. While the project is currently under Capacity of products Wealth quintile (ward level) Urban vs rural design, it is anticipated that the subsidy structure will sold (%) of customers reached (%) customers reached (%) include (i) geographic and economic targeting to increase 2 3 7 3 3 5 penetration among the poor and hard-to-reach populations, 7 11 (ii) a more robust monitoring, reporting and verification 14 20 21 45 52 framework. 9 3 22 22 Key take-aways for designers 54 68 Untargeted subsidies are very useful to support the 60 55 48 48 acceleration of market development, allowing participating companies to scale up operations and expand their 20 networks progressively. However, the impact on reaching the poor and hard-to-reach populationswas limited. Phase 1 Phase 2 Phase 1 Phase 2 Phase 1 Phase 2 6-11 Wp Quintile 1(poorest) Accepting companies into the project on a rolling basis Urban 12-20 Wp Quintile 2 and allocating funds on a first-come-first-serve basis helps Rural 21-49 Wp Quintile 3 achieve results quickly. Notably, without proper monitoring 50-99 Wp Quintile 4 and management of funds, there is a substantial risk of >100 Wp Quintile 5(wealthiest) overcommitting funds. The project’s most important criticism was that available Verification of claims assisted by software significantly funds ran out abruptly, without proper communication reduced processing time and helped sales increase to companies. Without RBF and end-user subsidies, exponentially. More research is needed to determine the participating companies had to increase prices, switch accuracy of this verification method versus more traditional back to selling higher-end products targeted at wealthier phone and field surveys. households (instead of the entry-level low-margin products), and some had to downsize and collapse the distribution Continuous monitoring, adaptation of the design, and networks they had expended during the project. communication with companies and stakeholders are critical for the sustainability of the project. Way forward In December 2023, the World Bank approved the Nigeria Distributed Access through Renewable Energy Scale-up End-User Subsidies for Energy Access: A toolkit 83 04. Appendix This appendix provides information on additional resources for the design of subsidies for off-grid solar and clean cooking products. End-user subsidy lab website The EUSL website contains a repository for useful resources in the design and implementation of subsidies, including: Country case studies Relevant reports and resources Presentations and recordings of webinars Resources from existing subsidy projects Some of the subsidy programs cited in this document provide publicly available documentation on the design of their subsidies, such as: Rwanda REF Window 5: Operations manual Uganda EASP: Operations manual Other resources Linkages between end-user subsidies and other public finance mechanisms: Designing Public Funding Mechanisms in the Off-Grid Solar Sector Linkages between end-user subsidies and policy development: Off-Grid Solar Policy Toolkit 84 End-User Subsidies for Energy Access: A toolkit To connect with the Lighting Global team: Email: info@lightingglobal.org www.lightingglobal.org