The Energy Subsidy Reform Assessment Framework (ESRAF) GOOD PRACTICE NOTES Toward Evidence-Based Energy Subsidy Reforms i OVERVIEW NOTE Guidance for Comprehensive Energy Subsidy Reforms Thomas Flochel Sudarshan Gooptu ii  CONTENTS Acknowledgments v About the Authors vi Acronyms and Abbreviations vi Introduction 1 Scope of the Framework 2 Structure of the Framework 3 Executive Summary 5 Energy Subsidies 5 Fiscal Cost 7 Impact of Higher Energy Prices on Households 9 Social Perceptions of Price Subsidies and Their Removal 10 Social Assistance for Coping with Higher Energy Prices 11 Impacts of Higher Energy Prices on Firms and Industrial Competitiveness 13 Impacts of Higher Energy Prices on Macroeconomy and Greenhouse Gas Emissions 14 Environmental and Health Implications of Higher Energy Prices 15 Political Context 16 Communications and Citizen Engagement 18 How to Use the Framework 19 Assessing the Subsidies 19 Modeling Impacts of Energy Price Subsidies and of Their Reform 20 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 Impact on Households .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 Impact on Firms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 Impact on the Economy and the Environment Assessing the Political Context and Preparing Communications 26 A Rapid Assessment of Energy Price Subsidies 28 Appendix: Checklist for Rapid Assessment: Analytical Issues Related to Energy Subsidy Reform 30 Endnotes 34 References 34 i OVERVIEW NOTE FIGURES Figure O.1: Energy Subsidy Reform Assessment Framework: Energy Subsidies, Impacts, and the Context of Reform 3 BOXES Box O.1: Combining Qualitative and Quantitative Tools 22 ii ACKNOWLEDGMENTS ACKNOWLEDGMENTS This is an Overview of the Energy Sector Reform Assessment Framework (ESRAF), an initiative of the Energy Sector Management Assistance Program (ESMAP) of the World Bank. ESRAF consists of 10 good practice notes, which provide a guide to analyzing energy subsidies, assessing the impacts of subsidy reform and understanding the political context for subsidy reform in developing countries. ESRAF was produced by a team led by Thomas Flochel and Sudarshan Gooptu. The principal authors of the Good Practice Notes are Sandra Bernick, Arlan Brucal, Andrew Burns, Ezgi Canpolat, Calvin Djiofack Zebaze, Santiago Enriquez, Thomas Flochel, Verena Fritz, Sophia Georgieva, Sudarshan Gooptu, Leonardo Iacovone, Gabriela Inchauste, Masami Kojima, Caterina Ruggeri Laderchi, Bjorn Larsen, Amr Moubarak, Anne Olivier, Sara Bryan Pasquier, Dinar Prihardini, Ernesto Sanchez Triana, Eva Schiffer, David Victor, Heather Worley, and Ruslan Yemtsov. The authors are indebted to Masami Kojima whose careful review of all notes and incisive comments have greatly improved the coherence of the framework. The authors thank Marianne Fay and John Panzer for initiating this project. Overall guidance for this product was provided by Marianne Fay, Paloma Anós Casero, and Rohit Khanna. The framework also benefited from the very helpful comments of Margaret Grosh, and each note was reviewed by experts from the respective Global Practices at the World Bank. David Coady from the International Monetary Fund, and Ronald Steenblik of the Organization for Economic Co-operation and Development provided helpful comments. The authors are also grateful to Jan von der Goltz and Praveen Kumar who helped prepare the concept note for the project, to Rebecca Kary for her careful editing, and to Lauren Kaley Johnson, Nansia Constantinou, and Heather Austin for creative design work. ESRAF was financed by the Energy Sector Management Assistance Program (ESMAP). ESMAP—a global knowledge and technical assistance program administered by the World Bank—assists low- and middle-income countries to increase their know-how and institutional capacity to achieve environmentally sustainable energy solutions for poverty reduction and economic growth. ESMAP is funded by Australia, Austria, Denmark, the European Commission, Finland, France, Germany, Iceland, Italy, Japan, Lithuania, Luxembourg, the Netherlands, Norway, the Rockefeller Foundation, Sweden, Switzerland, the United Kingdom, and the World Bank. iii OVERVIEW NOTE ABOUT THE AUTHORS Thomas Flochel is Energy Economist in the Energy and Extractives Global Practice at the World Bank, where he works for the Energy Sector Management Assistance Program (ESMAP), in the Sustainable Development (SD) Practice Group Vice Presidency of the World Bank. Sudarshan Gooptu is the Global Lead for Fiscal Policy, Macroeconomics and Fiscal Management Global Practice, in the Equitable Growth, Finance and Institutions (EFI) Practice Group Vice Presidency of the World Bank. ACRONYMS AND ABBREVIATIONS CGE computable general equilibrium (models) ESMAP Energy Sector Management Assistance Program ESRAF Energy Subsidy Reform Assessment Framework LPG liquefied petroleum gas iv INTRODUCTION 1 INTRODUCTION Countries around the world are finding number of countries, including several that that energy subsidies can threaten their include fossil fuel subsidy reforms in their economic, fiscal, and environmental health. intended nationally determined contributions, The benefits of energy subsidies are often have energy subsidy policies that have been captured largely by higher-income groups entrenched for decades, and they find it and specific industries. Subsidies can create difficult to navigate through the politically budget pressures on governments—pressure sensitive paths of reforms. that can even reach crippling levels over time. In some countries, spending on subsidies The Energy Subsidy Reform Assessment represents an unacceptably high proportion Framework (ESRAF) proposes a guide to of government revenue. Energy subsidies tend analyzing energy subsidies, the impacts of to grow and shrink with world fuel prices, subsidies and their reforms, and the political making them unpredictable and difficult to context for reform in developing countries. manage. Price distortions in the energy sector Reforming energy subsidies in a sustainable can jeopardize its financial sustainability and and socially responsible way requires a clear deter investment. When they keep prices understanding of a range of economic, financial, of fossil fuels artificially low in the form of social, and political factors. Subsidies have consumer price subsidies, subsidies increase implications for the health of the energy sector, their consumption, thereby contributing to the affordability as well as the quality of energy pollution and greenhouse gas emissions. service delivery, and the government’s fiscal According to one estimate, global fossil fuel position and debt sustainability. They affect subsidies in 2015 totaled US$322 billion, nearly people’s livelihoods, firms’ competitiveness, double the global subsidies to renewable macroeconomic stability, and the environment. energy in that year (IEA 2016). The impacts of reforming energy subsidies In recent years, many countries have taken are interconnected, and are therefore best advantage of low world oil prices to tackle examined in a comprehensive framework. consumer price subsidies in the energy sector. Where energy subsidies keep prices artificially The landmark voluntary commitments by low, subsidy reforms will likely increase energy governments in Paris to address climate prices. Other forms of subsidies, however, may change have also highlighted concerns have no obvious effects on prices, at least in the about greenhouse gas emissions from the near term, while subsidies to protect energy combustion of fossil fuels, to which consumer producers may even raise, rather than lower, price subsidies contribute. At least 11 countries— prices charged to consumers. The impact of Arab Republic of Egypt, Ethiopia, Ghana, India, reforming energy subsidies on households Islamic Republic of Iran, Kuwait, Morocco, and firms therefore varies considerably Rwanda, Togo, the United Arab Emirates, depending on the type of subsidy and how and Vietnam—have formally committed to it is administered. Similarly, subsidy reforms fossil fuel subsidy reforms in their intended vary in their ability to generate the savings nationally determined contributions.1 A governments can use to compensate affected 2 OVERVIEW NOTE households or firms, or redirect to other priority also enables users to focus on areas of priority programs in the government budget. The in their context and areas of specialization. political environment for reform is affected by who benefits from as well as who is harmed by Given the economic, sectoral, and political the subsidy regime, and how the distribution diversity, ESRAF is, by design, focused on of benefits and disbenefits are perceived. analysis, rather than policy recommendations. Energy subsidies in developing countries tend Policy recommendations need to be specific to benefit the rich disproportionately, but to each country and the sectors affected, with to the extent that consumer price subsidies due consideration to the country’s political, are considered essential for making energy business, and economic environment at the affordable for the poor, they may be seen time. ESRAF provides the set of tools to as vital and their removal opposed by many. design and prioritize reforms in the light of an As such, a government’s ability to deliver a assessment of their interconnected impacts. coherent communication campaign is crucial The notes are meant to be “living documents” to a reform’s success, and coordination among and will be updated from time to time; later policy makers from different government versions may capture some of the topics not yet ministries is essential for reforms to be well fully covered. Areas only tangentially touched rounded. upon in ESRAF include complementary measures to help energy consumers cope with SCOPE OF THE FRAMEWORK higher prices. Decreasing energy consumption through demand management and energy ESRAF is intended for use in developing efficiency improvement may play an essential countries and covers subsidy reforms for fossil role, and the government may also consider, fuels (oil, natural gas, and coal), electricity, on a temporary basis, targeted subsidies and district heating. Subsidies for renewable outside the energy sector, such as passenger energy are covered only as part of overall transport. Further, energy subsidy reforms subsidies for electricity, and are otherwise are often undertaken within the context of a excluded. To be useful in a wide spectrum of broader reform of the energy sector. Energy contexts, each note is general and modular. sector reforms are a vast and complex topic, Each note flags where other sectoral experts which is beyond the scope of ESRAF, but it is may provide data, where other notes may important to understand the reform context provide the analysis needed as an input, and and the role that reforming subsidies plays in where coordination with other experts may be reaching the objectives of the sector reform. necessary to complete the analysis. Guidance on how to deal with regional specificities—such as varying data quality and availability—is STRUCTURE OF THE FRAMEWORK covered to the extent possible, and illustrative Figure O.1 shows the areas covered in the 10 examples are provided. The modular structure notes of ESRAF: INTRODUCTION 3 FIGURE O.1: Energy Subsidy Reform Assessment Framework: Energy Subsidies, Impacts, and the Context of Reform Energy Subsidies: are distributed by income and other 1 | Identifying and Quantifying Energy indicators, and how consumer price Subsidies. The starting point for any subsidy reform may affect households, reform is to understand the types and depend on the consumption patterns magnitude of energy subsidies. of households. The note discusses the data and tools available to conduct the Fiscal Impacts: analysis. 2 | Assessing the Fiscal Cost of Subsidies 4 | Analyzing the Incidence of Price and Fiscal Impact of Reform. These Subsidies and the Impact of Reform subsidies often imply budgetary risks on Households: Qualitative Analysis. related to both direct fiscal costs and Asking energy consumers about how debt sustainability. The note discusses they perceive the benefits and disbenefits how to model and project fiscal scenarios of consumer price subsidies and their following reform. reforms, and how they may change their behavior and energy purchase patterns Household Impacts: in response to reform, will yield valuable 3 | Analyzing the Incidence of Price information. As with Note 3, an important Subsidies and the Impact of Reform on criterion is selecting different income Households: Quantitative Analysis. How groups and those who are considered price subsidies flowing to households particularly vulnerable. 4 OVERVIEW NOTE 5 | Assessing the Readiness of Social Safety of harmful pollutants, subsidies aggravate Nets to Mitigate the Impact of Higher air pollution with associated public health Prices. Where subsidy removal results effects in the form of premature deaths in price increases, governments need and illnesses. to be able to rely on their social safety net infrastructure to assist vulnerable Political Economy: segments of the population. 9 | Understanding the Political Economy of Reform. A good understanding of the Firm and Industrial Impacts: historical and political context of subsidy 6 | Identifying the Impacts of Higher reforms is essential to their successful Energy Prices on Firms and Industrial design. Competitiveness. Globally, more energy is consumed by firms than by any Communication: other consumer category. Ensuring the 10 | Designing Communications Campaigns competitiveness of suppliers of goods for Energy Subsidy Reform. A well- and services is among the frequently cited designed communications strategy can objectives of consumer price subsidies. reduce the political risk of reform and Higher energy prices from price subsidy enhance the conditions for success. reforms will have different short-term effects depending the characteristics of It is important to stress that energy subsidies those firms, while the efficiency of the are not synonymous with lower prices, and affected firms that outlive the subsidy subsidy reforms do not always lead to higher reforms is likely to increase in the medium energy prices. However, because consumer to long term. price subsidies—whereby prices charged to consumers are kept artificially low—are usually Economic and Environmental Impacts: the most visible form of subsidy and create 7 | Modeling Macroeconomic Impacts large economic distortions, Notes 3–8 focus and Global Externalities. Impacts on primarily or even exclusively on this form households and production sectors of subsidy. It is nevertheless important to can be combined to explore the bear in mind that other less visible forms of macroeconomic effects of higher energy subsidies can also be large and damaging, prices. Macroeconomic models, such as while escaping public scrutiny and attention general equilibrium models, offer a way from policy makers. of projecting some of the economywide The next section summarizes the main issues impacts of reform on economic growth covered in each note. It is followed by a user and inflation, as well as changes in the guide for the framework. At the end of this country’s greenhouse gas emissions. note, a checklist that can be used in a rapid 8 | Assessing the Local Environmental assessment is proposed. It aims to guide the Externalities of Consumer Price production of diagnostic reports that will help Subsidies. Where energy subsidies lead provide the analytical base to decide how to to overconsumption of fossil fuels or sequence and prioritize various aspects of a lower fuel quality and increase emissions reform. EXECUTIVE SUMMARY 5 EXECUTIVE SUMMARY ENERGY SUBSIDIES Subsidies can be provided through four main channels. Some attract much attention—with This note defines an energy subsidy as a price controls arguably being the most visible deliberate policy action by a government of all—while others, such as tax expenditures that specifically targets electricity, fossil and shifting of risk burdens, are seldom, if fuels, or district heating and that has one or ever, reported, let alone scrutinized. The more of the following effects: latter are difficult to identify, and a lack of scrutiny in turn helps perpetuate and even • Reducing the net cost of energy purchased increase these forms of subsidies. The different • Reducing the cost of energy production types of subsidies can also be categorized or delivery as consumer support, producer support, and general services support (not specifically • Increasing revenues retained by energy benefiting an identified group). Whichever way producers and suppliers one chooses to categorize energy subsidies, the eventual goal is to capture all types of Fossil fuels comprise crude oil and various subsidies in a unified framework, as follows: petroleum products, including kerosene, gasoline, diesel, and fuel oil; liquefied petroleum 1 | Direct transfer • Direct transfers to energy produ gas (LPG); natural gas; and coal. Although the of government • Cash transfers to consumers, wh focus of ESRAF is primarily on energy use funds energy consumption of fuels, where they are used as feedstocks, such subsidies are also covered. Examples include natural gas as a feedstock for fertilizer manufacture and LPG used in petrochemicals. 2 | Government- • Prices or price ceilings set by the government induced transfers • Domestic price effects of restricting import or export of energy between • Cross-subsidies among consumers or among different forms of producers and energy consumers • Mandate to purchase or supply a specific form of energy 6 OVERVIEW NOTE 3 | Forgone • Tax expenditure (such as corporate income tax reduction and sales government tax waiver) revenue • Other forgone fiscal revenues (such as royalties and production share in upstream oil and gas production) • Lower government revenue from energy suppliers: Where energy suppliers are absorbing subsidies, dividends from state-owned energy companies and corporate income tax from all energy suppliers would be lower than in the absence of subsidies 4 | Underpricing • Subsidized inputs, such as fuels and water of other goods • Lending and credit (loan guarantees, below-market provision of and services, loans) including risk • Goods and services provided by government (such as underpricing of access to government land) • Shifting of risk burdens (government assumption of price, safety, and other risks; consumer or resident assumption of risks through limits on commercial viability) • Special treatment of state-owned energy enterprises enabling undue risk-taking, amplified by soft budget constraints Who pays for the subsidies is an important are kept artificially low by the government driver of how likely they are to be retained so as to lower prices charged to consumers or reformed. While governments are often • Commercial lenders, other governments, assumed to be paying for subsidies, that is not international financial institutions, and necessarily the case, and not grasping who other financiers who have to accept debt pays, directly or ultimately, can lead to flawed cancellations in the energy sector, especially analysis and misguided policy conclusions. for state-owned companies, or provide The subsidies are paid for by the following: concessional financing • Taxpayers who pay for on and off-budget transfers, either today or in the future, Because price subsidies create distortions or make up for lower fiscal revenue due with ripple effects throughout the economy, to producer support or higher cost of they arguably represent the most serious government borrowing due to debt form of energy subsidies. In particular, price forgiveness caused by subsidies subsidies that keep end-user energy prices artificially low (also referred to as underpricing) • Energy consumers who pay higher prices have plagued policy makers in many countries, to provide producer support because their removal raises prices, an unpopular move in any country. Higher prices • Energy producers who suffer financial losses invite public resentment and political backlash, due to payment arrears or non-payment of and have varying adverse effects on different subsidy reimbursements, or whose margins segments of society with equity implications. EXECUTIVE SUMMARY 7 For this reason, ESRAF gives more attention Subsidies actually delivered to the intended to reforming consumer price subsidies than beneficiaries need to be studied with care. any other form of energy subsidies. Because of the unintended consequences, large positive price gaps (consumer price In countries with large energy price subsidies, subsidies) do not automatically translate to low how much official prices may need to increase prices for consumers. Price and availability— for subsidy removal is an important policy what consumers actually pay, and how much question. The price adjustment needed (also subsidized energy they are able to purchase— referred to as the price gap) is defined as the are two questions that need to be examined difference (gap) between a likely unit price of systematically, especially in distributional and energy (either in a competitive market setting political-economy analyses. or to achieve cost recovery) and the official or some other indicator of an artificially low FISCAL COST (or high) unit price on the local market. The total value of the subsidy for a particular The important fiscal issues are the impact form of energy is its price gap multiplied by of subsidies on the fiscal balance, how to the total units subsidized. The price gap is finance them, and how subsidy reforms positive for consumer support and negative may affect key debt and fiscal sustainability for producer support. indicators. Assessing the fiscal impacts of energy subsidy reforms is an essential Unintended consequences of energy prerequisite for designing and implementing subsidies can be considerable, and can the reforms. Energy subsidies may be provided even harm rather than benefit consumers. through various channels on the production Examples include commercial malpractice and consumption sides, and may generate (two prime examples being black marketing contingent liabilities—explicit or implicit—that and smuggling of subsidized fuels) whereby the government must monitor and manage as subsidies are captured by those engaged in part of overall macroeconomic management. criminal activities rather than the intended consumers; energy shortages; prices paid by From a macro-fiscal perspective, it is useful consumers being markedly higher than official to distinguish between explicit and implicit prices; and declining performance in the liabilities that subsidies may generate. Explicit energy sector. If energy suppliers are forced to liabilities are government commitments based operate at a loss and absorb the subsidies, the on laws and contracts, such as government financial viability of energy suppliers can be guarantees and budgetary transfers specified threatened, resulting in declining investment in budget laws. The transfers may be to in operation, maintenance, rehabilitation, and suppliers to compensate them for the price capacity expansion. Possible outcomes include gap, or to households to purchase energy. electricity and fuel shortages, an unreliable Implicit liabilities are commitments that are supply of energy, and increasing reliance on typically based on political announcements, fuel imports in countries with domestic fuel public expectations, and possible interest resources, which in turn could further increase group pressures. Examples include losses, the subsidy bill, foreign exchange shortages, nonguaranteed obligations, arrears, and and budgetary transfers. deferred maintenance of state-owned energy suppliers. 8 OVERVIEW NOTE Fluctuating international fuel prices and Governments must carefully balance exchange rates can introduce uncertainty and between the likely need to compensate increase fiscal risks. Where subsidies include losers during the reform and maintaining underpricing of fuels, rising world fuel prices or fiscal sustainability over the medium term. currency depreciation can pose an additional Where there is underpricing of energy, the threat to fiscal sustainability. Even where fuels path of energy prices toward fully market- are not underpriced, currency depreciation or based pricing mechanisms influences the higher interest rates can increase capital and fiscal and debt trajectories during the non-fuel operating expenditures, potentially transition. During this period, some fiscal increasing the magnitude of subsidies in the space is needed to ensure that the poor and absence of corresponding increases in energy vulnerable are compensated for higher energy prices. The fiscal impacts from the uncertainty prices and to address some of the concerns in key macroeconomic variables and fiscal risks of those opposed to reform. The ability of the from the government’s contingent liabilities government to redirect resources to existing should be incorporated in the fiscal analysis. or new social safety nets (Note 5) and to firms Similarly, natural disasters, such as extreme for short-term assistance (Note 6) depends weather events, can pose fiscal risks—for on the size of the fiscal space created by the example, a drought may force a large shift subsidy reform and the relative importance from hydropower, typically the least-cost of other competing priorities. source of electricity, to emergency diesel generation, among the most expensive— Policies to mitigate adverse effects of the thereby immediately creating a large price subsidy reforms on households and firms gap and corresponding contingent liabilities must therefore be designed hand in hand with for the government. the fiscal strategy. Where energy prices rise as a result of energy subsidy reforms, the prices Contingent liabilities through the balance of goods and services that use previously sheet of state-owned energy suppliers can subsidized energy. Higher prices of energy represent an important source of fiscal risk, and of other goods and services consuming especially when losses threaten their financial previously subsidized energy—such as food viability. Even when such enterprises are and transportation services—can markedly supposed to operate on a commercial basis, affect the consumption basket of the poor the government is likely to come to their rescue and other vulnerable groups, who will need if their failure is not considered acceptable to be provided with appropriately targeted and especially if their failure could result in social protection. These reforms will also affect non-delivery of essential energy services. future fiscal deficits and associated public Knowing that the government considers them sector financing requirements which, in turn, too big or too important to fail, these energy will affect the country’s gross public sector suppliers may and often do take undue risks. borrowing requirements and the affordability New infrastructure investments using public- of the government’s overall public spending private partnerships may also create claims needs in the medium term. on future public resources, calling for a careful assessment of associated fiscal risks and contingency planning by the fiscal authorities in the government. EXECUTIVE SUMMARY 9 IMPACT OF HIGHER ENERGY and type of income-generating activities, PRICES ON HOUSEHOLDS particularly those in which the poor tend to engage. These effects typically differ Consumer price subsidies for energy are significantly between rural and urban areas often introduced on social grounds. One in low-income countries. frequently cited rationale is making clean, modern energy affordable to low-income Direct impacts will depend on the share of groups to meet their basic needs. If prices household expenditures spent on subsidized are made sufficiently low, underpricing can energy source, the potential for substitution, enable many households to shift from the and the degree of nonessential consumption. traditional use of biomass, which may be the Direct effects caused by higher prices of energy cheapest source of energy albeit polluting and purchased affect all households paying for the damaging to health, to electricity or gaseous previously subsidized energy sources. The fuels. Another cited justification is to reduce urban poor are more dependent on purchased disruptions to the economy and shield it from energy than their rural counterparts—many of energy price volatility, thereby dampening whom use wood and other forms of biomass spikes in inflation. that are collected rather than purchased for cooking and heating—and hence are generally While consumer price subsidies are generally more vulnerable. However, one exception inefficient as a means of redistributing income is where biomass cannot be used as a to the poor, their removal is likely to affect substitute: energy for lighting. Especially in low-income households negatively. In most low- and lower-middle-income countries, rural cases, higher energy prices would appear to households in general and the rural poor in have a relatively limited effect on the incidence particular are less likely to use electricity, and of poverty (the number of people living below more likely to rely on kerosene for lighting. the poverty line). However, for the low-income Therefore, a higher share of rural households households now facing higher energy prices, is affected by higher kerosene prices than in their having to cut further into their budgets urban areas. to meet basic energy needs can have serious adverse consequences. All households that use subsidized energy sources for income-generating activities will For households, the two primary channels of be affected. The relevant energy sources for impacts of higher energy prices are through income-generating activities are gasoline, the changes in consumption patterns and diesel, and electricity. For those engaged effects on income streams. Both consumption in commercial activities, the extent of the and income can be affected directly by higher impact depends critically on how much of the energy prices, and indirectly through induced additional input costs they can pass through price changes, such as higher transport costs. to final consumers (Note 6). Groups that have These indirect effects, though harder to been found to be particularly vulnerable include quantify than direct effects, can be significant. fisherfolk (who use gasoline or diesel fuel in Actual impacts on people’s welfare depend fishing boats), farmers (who use diesel or on their consumption patterns, the extent to electric pumps for irrigation), all businesses that which consumers can adjust their consumption use gasoline or diesel in delivery of goods and when prices change, and the distribution services, and all businesses that use electricity 10 OVERVIEW NOTE or backup generators fueled by gasoline or particularly strong. For example, agricultural diesel. In addition, some microbusinesses rely households are likely to be more affected on subsidized LPG, such as street vendors by rising fertilizer costs and higher costs of selling food cooked using LPG. transporting their products to markets. Firm- level impacts are explored in more detail in Indirect effects will depend on the Note 6. consumption baskets of poor households and the price elasticity of demand for different Finally, higher energy prices can result in goods. Indirect effects touch all households broader impacts on the livelihoods of poor through (a) general inflation arising from people and their communities. Where clean, increases in costs of goods and services that modern energy is subsidized for household depend on subsidized energy sources, as well use, raising prices could force some households as increases in the costs of other energy sources to rely much more on the traditional use of from higher demand through substitution; and solid fuels, if not shift back altogether, with (b) lower employment in energy-intensive their attendant air pollution and adverse health activities. Higher prices of energy sources effects. Such a move affects not only the that are intermediate goods are responsible households now making greater use of solid for the first indirect effects. They are likely fuels, but exposes their communities to greater to be significant where transport fuels were air pollution. Because women and children previously underpriced, particularly diesel are often tasked with biomass collection and fuel (used in long-distance transportation). household chores, they are more vulnerable For the poor, higher food prices—from any to these adverse effects than others. combination of higher costs of transporting food to consumption centers (affected by SOCIAL PERCEPTIONS OF PRICE higher prices of gasoline or diesel), higher SUBSIDIES AND THEIR REMOVAL costs of irrigation (diesel, electricity), higher fertilizer prices (natural gas), and higher D i f fe re n t h o u s e h o l d s h ave s p ecif ic costs of operating farm equipment (diesel, vulnerability factors and will adopt divergent electricity)—would particularly be a concern. coping mechanisms and behaviors in response The urban poor tend to depend more on goods to higher energy prices. Qualitative analyses transported from elsewhere for their basic can illustrate households’ experience with needs and on public passenger transport, energy subsidies, and how they are likely to making them more vulnerable to these effects. respond to their reforms. Note 4 on qualitative Because LPG and kerosene for household analysis focuses primarily on coping with use are much less likely to be intermediate the removal of consumer price subsidies by goods—in fact, if these price subsidies are analyzing contextual factors and what it means perfectly targeted, by definition they are not for households to access sufficient energy for intermediate goods—their indirect effects are their basic needs. Households cope with the usually minor. same increases in energy prices in different ways, some by cutting back on different types All households involved in productive of spending and others by substituting different activities are likely to be affected by increases energy sources. It is important, however, not to in input costs stemming from higher energy assume low official prices and instead probe prices. In some sectors, indirect effects can be whether the government policy of keeping EXECUTIVE SUMMARY 11 the prices of (for example) liquid fuels low These qualitative findings should encompass have resulted in large-scale diversion, fuel the types and scope of subsidy- and shortages, and high prices on black markets— reform-related issues that concern different energy subsidy reforms may eliminate fuel categories of respondents. Qualitative shortages, and in extreme cases lower rather findings are not expected to be nationally than increase prices. or geographically representative, and they should not be extrapolated to the rest of the Knowing how people respond to price country to arrive at generalized conclusions. increases is critical to informing social policy Nonetheless, they can help anticipate responses. Information on how people have impacts or monitor ongoing impacts that are experienced prior price increases, what parts experienced and attributable to the specific of the household expenditures will be most reform measures, such as rising energy prices affected, and how they perceive the overall or greater availability of energy. It is advisable impact on their well-being is critical to guiding to undertake such qualitative assessments policy makers toward supporting positive periodically in order to track any changing coping mechanisms and avoiding behaviors perceptions that may emerge as the reforms with possible negative social or environmental progress in a country. consequences. Understanding their coping mechanisms with subsidies—fuel shortages SOCIAL ASSISTANCE FOR COPING and black market prices, to name a few—is WITH HIGHER ENERGY PRICES equally important; focusing only on higher prices, assuming perfect delivery of energy Where subsidies are large and take the form subsidies before the reform, could deliver of low prices charged to consumers, subsidy misguided analysis and conclusions. reforms can lead to a price shock. Price shocks for something as essential as energy can Citizen perceptions of existing social have economywide effects, and harm to the safety nets and associated institutions also poor and the vulnerable can be considerable affect how people will respond to different without appropriate mitigation. options for mitigating the impact of higher prices. These considerations should shape While the need to protect the poor from price a government’s expectations and resulting shocks is well known, often subsidies are choice of social safety net programs (Note removed without any accompanying social 5). The perceptions or narratives that citizens safety net measures.2 A recent report states attach to reforms in the energy sector, such that in only 9 out of the 28 cases analyzed were as expectations about service quality or any social safety nets used in energy subsidy accountability of providers, play a role in their reform efforts, and in an additional eight cases, overall acceptance of, or resistance to, reforms. vouchers for energy or retargeting of subsidies The level of citizens’ awareness about reforms to the poor were implemented (Clements is likely to vary considerably. Understanding and others 2013). A lack of mitigating effort this and what drives their attitudes toward is closely correlated with a lack of capacity in the reforms of energy subsidies and social government to implement social safety nets to protection is important for understanding the provide an adequate and timely response. The political-economy environment and designing lack of capacity was also evident in a global communication strategies (Notes 9 and 10). response to the food and fuel crisis of 2008–10, 12 OVERVIEW NOTE which revealed the weaknesses of social safety subsidy reforms, but as long as economies nets, especially in low-income countries. A remain exposed to shocks and fluctuations. survey conducted at the time of 146 countries For this reason, most countries already have showed that only 39 of them had expanded some type of social safety net program in targeted safety nets, while many more had place, and are typically working on reforming reduced food or fuel taxes, increased food or them to improve coverage, adequacy, and fuel subsidies, or undertaken a combination efficiency. of these measures, despite such across-the- board tax reduction and subsidy measures Governments should aim to use existing generally being regressive, distortive, costly, programs to respond to shocks specific to and difficult to reform (IMF 2012). energy subsidy reforms without undermining the longer-term objectives of building a Governments can reduce the fiscal burden coherent and sustainable social safety of energy price subsidies without harming net system. Countries around the world low-income households by reforming the operate a variety of social protection and subsidies and deploying different social labor programs and policies to help buffer safety net programs. Where the benefits of individuals from shocks, and equip them to energy subsidies are captured largely by the improve their livelihoods. Social protection and better-off and financed mostly by taxpayers, labor programs have evolved in complexity subsidy reforms can achieve significant over time, and the mix of instruments greatly savings while avoiding adverse effects on depends on the country context and the the poor and near poor. starting point of the programs. Some countries operate only a few programs, such as a single Where subsidy reforms increase government cash transfer, a contributory pension for revenue or decrease government spending, formal-sector workers, and a handful of other governments can often use the additional services, often with limited coverage. Other fiscal space to provide direct support to countries offer a number of social protection households. Which measures are desirable and labor benefits and services, managed as depends on the incidence of the subsidies coherent packages covering most in need of (Note 3), and their likely uptake by citizens support within the system. depends on popular perceptions of these programs (Note 4). An assessment of the Important considerations for compensation welfare consequences of higher prices caused measures include (a) the identification of the by energy subsidy reforms (Note 3) is required most appropriate scalable or at-scale social to determine the resource allocation to social safety net interventions to mitigate adverse safety nets needed to compensate the poor. effects on the poor, and (b) assessment The government’s fiscal space during the of the delivery chain, institutions, and transition will constrain the scope of feasible administration for providing an adequate compensatory mechanisms (Note 2). response. A variety of programs can often be used to transfer resources to the poor. To Protecting the poor against energy price guide in this decision, governments will need shocks is no different from protecting the to assess the readiness of different parts of poor against any economic shock. Safety nets the social safety net system to transfer these are needed not only in times of energy price EXECUTIVE SUMMARY 13 resources to those in need, and how to include IMPACTS OF HIGHER ENERGY those not currently served by the system. PRICES ON FIRMS AND INDUSTRIAL COMPETITIVENESS Institutional inertia and the need for robust delivery mechanisms for any social safety Although industrial competitiveness is a net program imply that social safety nets frequently stated reason for introducing cannot be adjusted and expanded overnight, subsidies, the effects of the removal of even when there is fiscal space to do so. energy price subsidies on firms have rarely Providing timely compensation that avoids been studied. In 2015, industry consumed drastic changes to the living conditions of the more energy than the transport or residential poor requires accurate sequencing of reforms, sectors, the other two major consumer assessment of market transmission channels, categories (IEA 2016). This makes firms the assessment of agility of social safety nets, largest consumers of energy relative to others and the ability to monitor the situation and in the economy. Energy subsidies are often make rapid adjustments. Hence, the design of introduced in developing countries to support mitigation measures should always start with their industrialization and the competitiveness the assessment of available social safety nets. of key industries’ exports through underpricing energy inputs. Many major producers of In the face of inadequate cash-based social crude oil, natural gas, and coal shielded their safety nets, other forms of compensation may industries from rising world fuel prices in the be used. Among them are energy subsidies decade following the first oil price hike in 2004, targeted to the poor, such as lifeline rates for introducing or expanding existing subsidies. electricity, district heating, and natural gas, and low prices set aside only for poor households Price subsidies distort incentives for firms to (for example to purchase a fixed quantity of allocate resources and to operate efficiently. LPG). An important policy question is how to Where prices are subsidized, firms do not enable the poor to meet their basic energy receive an accurate price signal. This distorts needs. Targeting is more challenging for the input choices of firms, and reduces their liquid fuels than for energy delivered through incentives to tackle energy intensity and networks—networks limit scope for illegal wastage, since investing in energy efficiency diversion. Lifeline pricing is most effective improvement when energy prices are low may if the poor are connected to the network, deliver small or even negative financial returns. each household is individually and accurately metered, and the subsidized block of service Price subsidy reforms could have disruptive is consonant with energy consumption by the impacts in the short term, while in the poor. However, they represent second-best medium term, the effects are likely to be options: compared to unconditional cash more positive. In the short term, higher energy transfers, they suffer from inefficiencies; poor prices could have large and disruptive impacts targeting in implementation, if not in design; on firms’ operating margins, thereby affecting and economic distortions. their profitability and potentially even driving some firms out of business. In the medium and long term, the removal of subsidies can not only increase economic and energy efficiency, 14 OVERVIEW NOTE but also accelerate innovation and adoption competitiveness and employment performance of more advanced technologies. are most likely to be affected include petrochemicals (natural gas, naphtha, and When faced with rising energy prices, firms LPG used as feedstocks), transport (gasoline can either pass on these costs, absorb them, and diesel), and agriculture (electricity and or adapt to them. Some firms may be able diesel), as well as small firms generally. Where to pass on price increases to the purchasers energy price increases are large and sudden, of their goods and services. Other firms may employment is likely to be affected, since have to offset higher energy costs by (a) firms seek to lower operating costs. While increasing energy efficiency and reducing adjusting wages is one avenue in the medium energy consumption; (b) switching to alternate to long term, short-term employment losses production processes that consume less of are likely. Job losses will likely be larger where the now more expensive energy; (c) switching industries are more reliant on energy as an to other forms of energy that are cheaper input and labor is immobile within or between (such as from LPG to wood); and (d) cutting industries and regions. other costs, such as labor (by reducing either the wage or working hours, or by laying off Unintended consequences of subsidies may workers). offset the adverse effects on firms of price subsidy reforms. In assessing the impact The effects of price shocks induced by of higher energy prices, it is important not subsidy reforms may therefore differ greatly to overlook the unintended consequences across industries and across firms within the of subsidies that adversely affect firms. As same industry. The relative energy intensity of an example, one out of every three firms production processes and technologies, and interviewed globally cites electricity as a major hence dependence on energy, is substantially constraint on business.3 Where subsidies different based on what industries produce. contribute to the poor financial health of power Within each industry, there are important utilities and to power outages, subsidy reforms differences due to such factors such as may improve the reliability of electricity. If firms firm size, ownership structure (or other can switch from diesel power generation to management factors), level of competition in grid electricity, they can capture significant the market, available infrastructure (electricity, savings even if grid electricity prices are raised rail, ports, and roads), proximity to markets markedly. and essential infrastructure, and exposure to foreign technology, all influencing the manner IMPACTS OF HIGHER ENERGY in which each firm responds to the price shock. PRICES ON MACROECONOMY AND GREENHOUSE GAS EMISSIONS Particular regions or sectors may be disproportionately at risk . The regional Energy price subsidies and their reforms have concentration of firms in affected sectors (as indirect and dynamic effects on consumers is often the case of countries of the former through the interaction between households Soviet Union) could imply that specific regions (both as consumers and workers) and firms. of a country are disproportionately affected by In addition to the direct impact that a price subsidy reforms, especially if labor is relatively subsidy reform can have on consumers immobile. Key sectors and firms whose (people and firms), there are indirect effects EXECUTIVE SUMMARY 15 through increases in the price of other goods adverse effects on specific population groups and services that require energy as an input. or industries. The fiscal windfall can also be These indirect effects can ripple through used to design macroeconomic policies to the economy, with potentially wide-ranging offset the short-term cyclical effects of reform, consequences for the economy—including for such as inflation, employment dislocation, and investment, industrial structure, and household increases in the cost of living of the poor. By welfare—and economic growth, which should using emission factors for various energy uses, inform the timing and pace of reform. they can also estimate changes in greenhouse gas emissions (although with a low level of Economywide models can go beyond the reliability). direct effects of price subsidy reforms and help assess their overall costs and benefits. However, CGE models are unable to reliably They provide useful information about how track the short-term and cyclical impacts the reforms will affect macroeconomic stability of policy reform, which are better captured and the structure of the economy in the long by macrostructural models. These general term, and how to prudently spend any fiscal equilibrium models should therefore not savings generated by the reform. By capturing be relied on to study direct and short-term the direct impacts of price subsidy removal impacts on households and firms, in part on the government’s budget (Note 2), the because these long-term models assume impacts on households (Note 3) and on firms the ability to adjust energy consumption (Note 6), and integrating the interrelationships pattern perfectly, an unrealistic assumption in between economic sectors in a country, these the short term. By contrast, macrostructural models can help quantify the overall economic models can be used to quickly quantify the impacts on the real sectors. They can also likely short-term macroeconomic impacts of a estimate the second-round repercussions reform measure. These models also have the of price subsidy removal on households and advantage of relatively low data requirements, firms. Different economywide modeling tools— and are typically easier to work with than in particular, computable general equilibrium CGE models. (CGE) models—can be used to quantify these macroeconomic effects. ENVIRONMENTAL AND HEALTH IMPLICATIONS OF HIGHER By examining all sectors of the economy, ENERGY PRICES general equilibrium models can be used to design integrated reform scenarios, including Energy subsidies linked to fossil fuels can reallocation of resources following the have serious local and global environmental reform, and approximations of changes in consequences. Where subsidies keep prices greenhouse gas emissions. After identifying paid by consumers artificially low without the incidence of price subsidies on households causing serious energy shortages, subsidies and firms, and the fiscal implications of a can lead to inefficient, as well as nonessential, reform package through partial equilibrium consumption of energy. Combustion of more models, these tools can be used to design and fossil fuels increases local air pollution and sequence mitigation strategies, for example, by greenhouse gas emissions. There may be other using any resulting fiscal windfall to pay down adverse environmental externalities associated debt, invest in public infrastructure, and offset with increased energy production, such as land 16 OVERVIEW NOTE and water contamination. Artificially low prices estimated 6.5 million people die from outdoor can also increase energy-dependent activities ambient and household air pollution each year and production of fuel-intensive goods with according to the Global Burden of Disease 2015 attendant environmental effects, such as (Forouzanfar and others 2016). Combustion unsustainable extraction of groundwater and of fossil fuels and solid fuels accounts for a harm from excessive application of chemical large share of these deaths. fertilizers. Quantification of contribution of energy Energy subsidies can also have positive subsidies to local air pollution should be environmental externalities. Low prices of interpreted with care. Quantifying how much gas, electricity, and district heating can reduce more fossil fuels are burned as a result of household air pollution by enabling households subsidies is challenging, because doing so to shift away from traditional use of solid fuels, requires not only own-price and cross-price such as wood, straw, crop residues, dung, subsidies, but also information about availability and coal (see Note 4 on how to capture this of the subsidized energy and the actual, rather information). Subsidies to natural gas can than the official, prices paid. Moreover, unlike reduce combustion of coal and oil products greenhouse gas emissions, quantities of fuel in the power and industrial sectors, with net consumed alone do not determine the level reductions in hazardous local air emissions, of air pollution. Combustion technologies, the while subsidies to gaseous automotive fuels state of equipment used, how well vehicles reduce particulate emissions in the transport and equipment are maintained, and how they sector. However, energy subsidies are rarely the are used (such as vehicle driving patterns) all most cost-effective and economically efficient affect emission levels, while no analyses take solution to realize such environmental gains. account of these factors in detail. Within these An understanding of the energy subsidy, air constraints, Note 8 offers the types of analyses pollution, and health linkages, along with their that may be used to estimate the health costs magnitudes, can prove beneficial in designing and benefits of energy price subsidies. energy subsidy reforms, so that measures can be identified and considered for mitigating POLITICAL CONTEXT adverse effects of subsidy removal. Policy makers understand well the need Premature deaths and increased morbidity to reform energy subsidies, but they are caused by fine particulate air pollution hampered by political-economy challenges. represent the largest economic effects of There has been significant variation in the lower energy prices from a local environmental outcomes and sustainability of energy point of view. For outdoor air pollution, low subsidy reforms. Successes have come where prices aggravate air pollution, whereas for important pro-reform actors have strategically indoor air pollution caused by household addressed the political-economy barriers use of solid fuels, subsidized prices of clean and opportunities for reform. Implementing energy help reduce air pollution, saving reforms that endure requires tools that make lives and reducing illnesses. As such, energy it easier to understand and navigate those price subsidies carry high social costs, as barriers and opportunities. well as large social benefits, depending on how the subsidized energy is being used. An EXECUTIVE SUMMARY 17 While there has been extensive research on important is understanding the developments optimal design of subsidy policies, systematic that had led to the introduction of subsidies in research on the political economy of subsidy the first place and the major policy changes reform is much thinner and yet essential to made since then, as well as a sound knowledge effective policy design and implementation. of the decision-making structure, stakeholder A political-economy perspective pays close interests, and the reform track record. attention to such factors as the ability of interest groups to organize for or against policy Reform efforts begin and succeed where reforms and how institutions, financial benefits, they tackle the political-economy challenges or disbenefits of subsidies for consumers effectively in a given context. Analysis of and energy suppliers, and energy markets the political economy of reform involves shape political behavior. In this regard, it is understanding the relative power of important not to assume automatically that stakeholders based on analysis and mapping consumers benefit from subsidies. At one end of the goals and capabilities of major interest of the spectrum, they may be the ones paying groups. It also requires understanding how the for producer subsidies, and similarly, energy groups are organized internally to advance suppliers may welcome subsidy removal if they their goals and their internal dynamics, as are partially paying for consumer subsidies. well as how different groups connect to each A political-economy framework also puts other, how strong their connections are, which an important spotlight on incentives that coalitions they form, and how their interests affect the behavior of political leaders and overlap. administrators. It helps understand the need The scope for reform is influenced by the for pragmatic approaches that adapt what interplay between the government incentives might otherwise be ideal objectives to the to adopt policies with potentially large compromises required for success. Often, the political costs and the divergent incentives question is not “How can we achieve a result of distinct interest groups. Those interactions that mirrors global best practice?” but rather will depend on the organization and political “How can we make and maintain progress on power of the groups, and will also hinge on the energy subsidy reform?” ability of the reformers within the government Most governments create energy subsidies to choose their political allies and blunt the with stated goals, which partly determine political influence of groups that could block how different interest groups react to the adoption or reverse the implementation of reform. Subsidies often begin with the goal of energy subsidy reforms, both inside and advancing a legitimate social goal. Those goals outside the government. When reformers inform how interest groups are organized, as in the government are confident of their well as the goals that need to be retained in mandate and feel secure in power, progress a reform program. A first step should be to can be made even in the face of opposition publicize the magnitude of subsidies and their from powerful groups, which can include costs (Note 10). Understanding who gets the powerful politicians. Other aspects that can subsidy and who pays for them is crucial to shift the political equilibrium toward reform understanding whether and how the subsidy include (a) the ability of the reformers to actually reaches a socially desirable target and create a strong, simple, and credible narrative who stands to lose upon its removal. Equally on the need for reform, (b) the ability of 18 OVERVIEW NOTE the government to offer credible alternative countries . Public reactions to subsidy policies delivering benefits to some or all of reform programs are highly contextual and the citizenry, (c) a sharp increase in the cost dynamic. Reforms can be successful only if of providing subsidies, and (d) increased there is an informed and supportive public external pressure to reform. that understands the rationale for reform. International experience has shown that well- COMMUNICATIONS AND CITIZEN planned and consistent communications is ENGAGEMENT critical to the success of energy subsidy reforms. International experience shows that communicating before, during, and after Active engagement with citizens helps build subsidy reform is essential to ensuring the consensus, support, and incentives for the smooth rollout of a well-planned and behavioral changes needed for subsidy executed energy subsidy reform program. reforms. A communications strategy, plan, Energy subsidy reform is not a goal in itself, but and campaign constitute a strategic effort rather a means of achieving lasting economic to listen to stakeholders’ concerns and and social progress. Communicating with perceptions, coordinate messages within the the public and other important stakeholders government, and communicate with one voice about the benefits of reform, the drawbacks critical messages tailored to the audience’s of existing subsidies, and the government’s circumstances and perceptions. Phasing out plans to protect the vulnerable from negative energy subsidies can be politically difficult, but effects of subsidy reform helps build support a number of countries have managed to build and acceptance. It can also increase trust and support and acceptance among a variety of understanding of the political decisions that stakeholders and reformed subsidies without underpin the reform. major disruptions. They have succeeded in part through well-planned and executed Some governments undertaking energy communications campaigns. subsidy reform programs either ignore communication with stakeholders or Communications is an investment, not a cost, take a top-down approach that fails to and should be planned and implemented by recognize stakeholder views and concerns. professionals before and during the stages This happens for many reasons, including a of reform. By assessing risks early, informing lack of understanding about the powerful the public in accessible ways, and explaining role communication plays in a successful the mitigation measures to protect poor and energy subsidy reform program and the vulnerable households, public understanding absence of capacity within the government and eventually goodwill can be built for a to undertake communication activities. Yet reform process. Communications campaigns communicating about energy subsidy reform— must be flexible to accommodate shifting and doing so through a holistic and effective political, social, and cultural aspects of reform, communications campaign—is crucial. and a well-planned strategic communications strategy based on empirical research will Failure to consult stakeholders and to gain greatly enhance the acceptance and the their support has stalled and sometimes effectiveness of any subsidy reform program. reversed much-needed reforms in many HOW TO USE THE FRAMEWORK 19 HOW TO USE THE FRAMEWORK ESRAF is intended to guide the production Although these less visible forms of subsidies of comprehensive evidence to underpin a all too frequently escape scrutiny, they can reform of energy subsidies. Applying the full be large in magnitude. Examples include tax analysis suggested in all notes is not always expenditures and shifting of risk burdens feasible, nor is it always necessary. One can from energy suppliers to the government or choose areas to focus on, depending on the consumers. The note discusses how to identify objectives and context. Factors affecting different forms of energy subsidies, how to the choice of coverage include the purpose quantify them where possible, and the policy of the analysis, timeline to complete it, issues to consider when reforming subsidies audience, and data availability, as well as the in developing countries. characteristics of the subsidized products and types of subsidies to be reformed. This section For a complete analysis, practitioners should provides guidance on how to pull together complement the price gap approach with the selected facets of analysis. In addition, a an inventory of energy subsidies, in order to minimum recommended analysis is outlined, capture subsidies that do not directly affect which includes the following steps: prices. For subsidies that are not amenable to quantification without considerable data 1 | Assessing the subsidies collection or subjective judgment, noting their presence and recommending a review alone 2 | Modeling the impacts of energy price would be useful. A comparison with peers— subsidies and of their reform neighboring countries, similar economies, 3 | Assessing the political context and or similar energy markets—can help gain a preparing communications better understanding of whether a country’s subsidy policies are reasonable, how they can Where time and resources are limited, the be reformed, and what is achievable in the analysis may be confined to a rapid assessment near and long term. of energy price subsidies. For fuel price subsidies, the subsidy delivery ASSESSING THE SUBSIDIES channels should be reviewed to understand who the winners and losers are and the The starting point of any analysis is political economy of the reform (Note identification and measurement of energy 9). The delivery channels determine the subsidies. effectiveness of the subsidies at reaching their intended beneficiaries and influence NOTE 1 provides a definition of energy the prices actually paid by consumers as well subsidies and extensive categorization of the as quantities available for purchase, both of different forms of subsidies for different energy which should be considered when estimating products. This will ensure that the analysis does the impact on households (Note 3) and firms not fail to capture subsidies that do not give (Note 6). If energy suppliers are forced to the appearance of influencing energy prices operate at a loss and absorb the subsidies, or do not appear as line items in the budget. 20 OVERVIEW NOTE the financial viability of energy suppliers may challenge to winning public acceptance of be threatened. Deteriorating financial health higher prices (Note 9). Finally, a hidden cost can prompt budgetary transfers, subsidized analysis enables decomposition of the price loans, and other forms of assistance to failing gap into underpricing and different types of state-owned energy suppliers, affecting the operational inefficiency, providing an indication government’s fiscal position (Note 2). Capture of the scope for reducing costs, and therefore by unintended beneficiaries, which may not be subsidies (Note 1). apparent without analysis, compromises the objectives of subsidies, but will likely lessen For nonprice subsidies, the fiscal analyst the adverse effects of subsidy reforms (Notes should refer to section 4 of Note 1 on subsidies 3 and 6), while introducing new dimensions arising from tax expenditure items (such as to the reform’s political economy (Note 9). tax-rate reductions, accelerated depreciations, allowance for additional deductions, tax For power sector subsidies, Note 1 also details credits, and tax deferrals), and other support how to estimate the level of cost recovery measures (such as underpricing of permits against which to benchmark existing tariffs and cash transfers earmarked for energy to be able to measure the price gap. Where purchase). Special attention should be paid to to set the benchmark performance varies by these forms of subsidies in the fiscal analysis country and the maturity of the power market. (Note 2), and the fiscal and energy experts will need to work closely together to enter the Electricity, natural gas, and district heating subsidies appropriately in the fiscal model, share characteristics that distinguish them as different forms of subsidies have different from liquid fuels. Each officially connected fiscal implications. consumer is metered and billed, enabling targeted assistance to those with low MODELING IMPACTS OF ENERGY consumption, which serves as a proxy PRICE SUBSIDIES AND OF THEIR for poverty, however imperfect (Note 5). REFORM Inaccurate metering, infrequent reading of meters, and multiple connections to a single Modeling of subsidies and their reforms is meter—which can be assessed through confined largely to subsidies that lower prices qualitative methods (Note 4)—all compromise paid by consumers. Several approaches exist the ability to send appropriate price signals to model changes in energy prices, although and deliver targeted assistance. Poor quality estimation of the actual changes in energy of service, measured by the frequency prices faced by consumers and quantities and duration of service interruptions, is a purchased under the subsidy regime may prevalent problem in a number of developing carry large uncertainties. By contrast, subsidies countries. Qualitative tools (Note 4) can help that are not directly linked to price levels assess perceptions of service quality, and are seldom, if ever, modeled, especially in information on such perceptions can help developing countries. shape a communications strategy (Note 10). It generally takes much longer to improve the Once energy price subsidies are quantified, quality of service for network energy than for their impacts on the government’s fiscal liquid fuels, and the time lag between price balance, households, firms, the economy, increases and quality improvement poses a and the environment, as well the impacts HOW TO USE THE FRAMEWORK 21 of reform on each, can be estimated. tests for a wide range of circumstances to Impacts are interdependent. For example, see how government debt is likely to evolve understanding the consumption patterns in a world with uncertainty, as well as project of households and productive sectors, as contingent liabilities in different reform well as the share of subsidies financed by scenarios. the government and how that financing fits within the government’s fiscal infrastructure, Weak financial sustainability of state- are inputs to a macroeconomic model. This owned energy suppliers creates contingent model, in turn, will calculate the indirect effects liabilities and poses fiscal risks that can be of higher energy prices on households. The measured using the government’s balance design of qualitative and quantitative tools to sheet. Access to financial data from state- estimate impacts on people can also inform owned enterprises is needed to go beyond each other. The fiscal impacts of a reform will estimation of aggregate fiscal risks and identify depend on whether mitigation measures will the specific sources of such risks and how be taken. How to target mitigation measures to handle them. Such analysis requires deep depends on who is harmed by the reform, sector or utility knowledge and calls for and how many resources are available for close collaboration with the energy expert reallocation to mitigation. It is therefore crucial conducting the assessment of subsidies. that all analysts start with a good grasp of In general, government liabilities can be what energy subsidies exist, and carry out characterized in two-by-two matrices with analyses in an integrated and collaborative direct or contingent, and simultaneously, manner. explicit or implicit, as the key categories. NOTE 2 addresses a series of questions that The approach uses a fiscal risks matrix on the should guide the analysis of fiscal sustainability, liabilities side, and a fiscal hedge matrix on the and points to models and tools to address assets side, with the latter capturing different each question. The essential task of the fiscal sources of potential revenues that can cover analysis is to measure the fiscal burden of the government’s obligations. The exercise the energy subsidies, and model how the involves mainly matching the different existing fiscal balance and debt sustainability will subsidies to different types of liabilities. evolve with the reform. This requires modeling As the analysis moves from evaluating the energy subsidies within an integrated status quo to modeling reform scenarios, medium-term fiscal framework. Building the fiscal analysis should integrate the fiscal such a unified analytical framework requires impact of any planned expansion in social having a good grasp on the budget, financing, safety nets from the social protection expert intergovernmental fiscal system, and asset- (Note 5), because such expansion will increase liability management in a country. Assessing budgetary expenditures and affect the fiscal sustainability requires forming a view about position of the government. In turn, the fiscal how outstanding stocks of liabilities are likely analysis can determine how many resources to evolve over time. In the best case where can sustainably be allocated to mitigation the data are available, such assessment can programs in the short and medium term. be carried out within a spreadsheet-based stochastic fiscal sustainability model.4 Such a model can generate scenarios and run stress 22 OVERVIEW NOTE IMPACT ON HOUSEHOLDS since household responses (such as shifting consumption away from goods with higher Notes 3 and 4 explore quantitative and prices to lower-price substitutes, increasing qualitative tools, respectively, for assessing energy efficiency to reduce consumption, the impact of consumer price subsidies and and eliminating nonessential consumption) of their removal on households, while Note 5 will reduce the initial adverse effects. describes tools for assessing the readiness of social safety nets to mitigate the impact of The different issues that a quantitative higher energy prices on households. assessment of the distributional impacts of price increases would seek to address are NOTE 3 provides guidance on calculating how outlined in Note 3, and will require the energy energy subsidies affect people at different analyst to provide information on the types income levels and how energy price increases and delivery channels of the subsidies. The may affect them. Broadly speaking, one can note outlines important practical difficulties distinguish three types of analyses: (a) general related to performing this type of analysis, in equilibrium analyses, incorporating both the particular with respect to estimating quantities direct and the indirect welfare effects of higher consumed which can be challenging in many energy prices; (b) limited general equilibrium, circumstances on account of the nature of data incorporating only a subset of the indirect collected in household expenditure surveys. effects; and (c) partial equilibrium approaches, focusing only on the direct effects of higher NOTE 4 provides guidance on the use energy prices on household expenditures. of qualitative research tools. These can The latter two are commonly considered complement quantitative analysis (Note 3) the short-term effects of reforms prior to in understanding the distributional impacts longer-term adjustments by households and of consumer price reforms on households economic agents affected by higher energy (see box O.1). Focus group discussions and prices. They can also be interpreted as an in-depth interviews are the most commonly upper bound on longer-term adverse effects, used tools. They rely on open-ended questions BOX O.1: COMBINING QUALITATIVE AND QUANTITATIVE TOOLS Combining qualitative and quantitative tools can provide a comprehensive understanding of the poverty and welfare impacts of energy subsidy reforms. They can be integrated in a parallel, sequential, or iterative way. In parallel approaches, the quantitative and qualitative research teams work separately, but compare and combine findings during the analysis. Where there are resources to conduct a household survey, qualitative findings can be used to inform the design of the survey and help define survey questions. Sequential and iterative approaches, on the other hand, require varying degrees of dialogue between the qualitative and quantitative research at all phases of the research (Rao and Woolcock 2003). In the context of energy subsidy reforms, parallel approaches are frequently used. For example, qualitative assessments can show the value of collecting seasonal data in energy expenditures, thus informing the design of new survey modules, and conversely, available survey data can inform qualitative research, such as by highlighting specific locations or categories of respondents to target in focus group discussions, or the types of issues to be discussed. HOW TO USE THE FRAMEWORK 23 and give voice to citizens to describe their own understand the coverage and adequacy of experience in dealing with welfare impacts. existing social safety nets, as well as their capacity to be expanded, if needed. Four Focus group discussions and in-depth options are available to governments: (a) interviews also allow exploration of consumer keep and not make changes to the existing perceptions and attitudes to reforms and program, (b) increase benefits, (c) expand mitigation measures, and the social and coverage, and (d) launch new programs. political factors driving these perceptions An important aspect is the delivery system and attitudes. As such, they can provide inputs within a given country’s existing social safety to the design of a communications campaign net system. The framework for the delivery (Note 10), and should be designed to serve systems is organized around five important both purposes. These tools are geared toward interrelated components: (a) the delivery chain; a relatively rapid qualitative assessment, (b) institutions and governance; (c) information although other tools involving ethnography systems platform; (d) citizen interface; and (e) and participant observation, among others, performance monitoring, evaluation, learning, can also be used. The latter require a longer and adaptation. For each component, the note timeframe and deeper immersion of the outlines the detailed questions that structure researcher in the field. the analysis, data sources, and primary steps to perform the analysis. Specific areas of investigation include household use of energy and the direct In general, the assessment of existing programs impacts that subsidies have on their livelihoods, should begin in parallel with the assessment of including whether they have had to develop impact on households, and particularly analysis coping mechanisms to deal with unintended of which segments of the population will be consequences of subsidies (such as energy affected and how (Note 2). Combined, these shortages and high prices of liquid fuels on two areas of investigation form the basis for black markets) in addition to coping with evaluating options for delivering more benefits price changes following the subsidy reform. through social safety nets. The note describes where the data on existing social programs can Note 4 provides detailed guidance on how to be found and some tools that can be used to design focus group discussions and in-depth model mitigation scenarios. It is also useful to interviews, how to adapt these to the context, capture questions on people’s perceptions of and how to select a sample. Sample guides social protection programs in the qualitative for each method are provided in the technical tools (Note 4), given enrollment in social safety annexes. Section 4 of the note provides a nets carries a stigma in some countries. The step-by-step checklist of how to conduct the assessment of the fiscal cost of compensation research, while section 5 gives some tips and under different scenarios can then be provided advice on analyzing and validating research to the fiscal analysis (Note 2). findings. NOTE 5 provides guidance on the analysis IMPACT ON FIRMS of social protection options available to NOTE 6 looks at the impacts on firms, an area governments to mitigate the adverse little studied and for which a broad body of effects of higher prices. The main goal is to research does not exist. The note proposes a 24 OVERVIEW NOTE framework to think through the channels of IMPACT ON THE ECONOMY AND firm-level impacts and some approaches to THE ENVIRONMENT their assessment. NOTE 7 proposes a series of models to Firms form the largest consumers of energy estimate impacts of reform on macroeconomic and, as such, the impact of price subsidies and indicators, such as economic growth, gross their reforms on firms merit close attention. domestic product by expenditure or by Prices that are kept artificially low distort firms’ industry, the balance of payments, government production choices, particularly in energy- fiscal aggregates, labor markets, and financial intensive sectors. While efficiency gains from markets (inflation, interest rates, and exchange reforming price subsidies can be expected at rates). the sector level in the long term, firms may be heavily impacted in the short term, with social A first approach is using input-output models and political consequences that governments and social accounting matrix multiplier models need to foresee and be prepared to manage. to take into account interactions between different sectors of the economy, providing The analytical approach and tools proposed insights into the indirect effects of price provide diagnostics on three important issues. subsidy reform on household expenditures First, they allow estimation of the distortionary and cost structures of firms. effects of existing energy price subsidies and the potential efficiency gains from their Alternatively, macrostructural models based removal. Second, the approach helps identify on econometrically estimated relationships those sectors and firms potentially at risk from can be used to track the short-term and higher prices induced by the price subsidy cyclical impacts of the initial price shock reform in the short term, and quantify the more realistically. However, they do not allow potential effects, both direct and indirect, the same level of sectoral or product detail as through industry linkages following different CGE, or partial equilibrium models. transmission channels. Third, the proposed Data permitting, CGE models have generally analysis helps identify some of the potential been the tool of choice for analysis of the longer-term benefits of subsidy reforms. long-term effects of energy price subsidy Alternative approaches for assessing the reforms because they capture the many impact of price subsidy reforms on firms complex direct and indirect effects of these or economic sectors are proposed for reforms on the structure of the economy while countries with different types of data. A full allowing for stimulation of price changes for methodology is proposed for application individual energy products. However, CGE where high-quality, firm-level panel data are models require a lot of data manipulation and available, and alternative approaches for careful calibration of the input-output table countries with less detailed, firm-level data or and social accounting matrix upon which they only industry-level data. The note has several are based in order to break down the impact annexes, including a checklist for each step of price subsidies to different energy products. and some mathematical formulations. Note 7 provides detailed guidance on how to disaggregate input-output tables and social HOW TO USE THE FRAMEWORK 25 accounting matrix into the energy products fuels (solid biomass or coal), subsidy removal of interest and how to simulate different types would worsen air pollution and health damage. of energy subsidies using a CGE model. This requires close collaboration with the energy As a general rule of thumb, the pollutant expert and analysis from Note 1. Integrating most damaging to health is fine particulate distributional analysis of households in the matter. Natural gas and LPG contribute little CGE model calls for a close collaboration with to fine particulate pollution, making them the poverty expert and analysis from Note 3. suitable for use by households for cooking and CGE models can also be used to project long- heating. For outdoor air pollution, the focus of term fiscal impacts of reforms, integrating the Note 8 is therefore largely on subsidies that choice of how the government can spend any increase the consumption of coal, gasoline, additional fiscal space created by the reform. diesel, and fuel oil. For indoor air pollution, This should be informed by the fiscal analysis the focus is relative affordability of solid fuels of Note 2. Finally, CGE models can be adapted (coal and solid biomass) on the one hand and to estimate the impact of higher energy prices household energy that is clean at the point of on global greenhouse gas emissions, with the use (electricity, natural gas, LPG, and district help of an environmental specialist. heating) on the other. Main contributors to fine particulate air pollution are (a) mobile sources NOTE 8 focuses on economic valuation of using gasoline and diesel; (b) household use of health effects of air pollution associated with solid fuels for cooking and heating; (c) small energy price subsidies. This note involves and numerous stationary sources, such as multiple fields and disciplines and attempts diesel backup power generators; and (d) large to provide clarity by breaking the assessment stationary sources, such as power generation down into several distinct steps, each with plants and factories consuming coal, diesel, its own methodologies. The assessment and fuel oil. involves analyzing (a) the effect of artificially low prices on levels and patterns of energy Recent meta-analyses of the price elasticities consumption (section 4), (b) air emissions of energy demand by type of fuel and energy from the combustion of fossil fuels (section 5), provide a basis for assessing the effect of (c) human exposure to air emissions (section subsidies on energy consumption. Cross-price 6), (d) health effects of exposure to particulate elasticities may be applied in sectors and to air pollution from the combustion of both fuels where significant fuel substitution may fossil fuels and solid biomass (section 7), be expected. Where automotive gasoline or and (e) monetary valuation of health effects diesel is subsidized, using country-specific (section 8). urban transport-environment models is recommended, if available, because of the Price subsidies aggravate air pollution complexity of air emissions from motor when they increase the combustion of fossil vehicles. However, as with estimation of the fuels, with one important exception. Where impact of higher energy prices on household households are currently using subsidized expenditures and energy use patterns (Note LPG, natural gas, or electricity for cooking and 3), a lack of information on the prices actually heating, or subsidized district heating for space paid and the extent of energy shortages heating, and the alternative to consuming the present challenges. More seriously, with the subsidized energy is the traditional use of solid exception of sulfate-based secondary fine 26 OVERVIEW NOTE particles from small sources, fine particulate ASSESSING THE POLITICAL levels in the atmosphere depend only weakly CONTEXT AND PREPARING on fuel consumption and much more on the COMMUNICATIONS technical states, as well as the operating characteristics of stationary and mobile Political-economy factors are central to the sources burning the fuel. Yet taking account of success and failure of energy subsidy reforms, these factors is beyond the scope of any study, but they are difficult to address for at least two forcing analysts to make vastly simplifying reasons. One is the sheer complexity of these assumptions. factors and their dependence on the country’s historical path. The other is that these factors The note advocates the use of intake fractions themselves are politically sensitive to study to estimate population exposure to fine and utilize. particulate matter, an approach taken in some recent studies of the global burden of Any political-economy analysis requires disease. The intake fractions are combined analyzing and mapping the goals and with the relative risk functions for major health capabilities of major interest groups, both outcomes of air pollution from the Global organized and unorganized. Note 9 provides Burden of Disease Project (Forouzanfar and a framework for understanding the relative others 2016) to estimate the health effects power of those groups affected by the reform. from energy subsidies. The note proposes a The analysis aims to achieve three goals: geographic-demographic scale consisting of • A political explanation for the country’s (a) urban areas with populations of more than current energy subsidies 100,000, (b) urban areas with populations of less than 100,000, and (c) rural areas. It • An understanding of the history of reform, then discusses the availability of monitoring including how successes or failures in past measurement data and alternative options reforms can inform what may be possible for estimating ambient concentrations of for future reforms fine particulate matter at the proposed • An assessment of the prospects and geographic-demographic scale, as well as identification of elements of most suitable approaches that can be used to deal with data strategies for reform in the future scarcity. The proposed method for economic valuation of mortality caused by air pollution NOTE 9 identifies the information required uses a cross-country transfer method of the for the analysis, and then describes how to value of statistical life. Finally, methods for obtain that information—initially with desk incorporating the valuation of increased research and then with structured interviews. illnesses are outlined, although morbidity is generally found to constitute a relatively The political-economy analysis draws on the minor share of health costs of air pollution. information gleaned from other analyses in ESRAF. The stated goals of the subsidy schemes, the initial period and context when the subsidies were introduced, the evolution of the subsidy policy since introduction, type and size of subsidies, the mechanisms used HOW TO USE THE FRAMEWORK 27 to deliver subsidies, and who ultimately pays upstream work on communication defined for them can be obtained from Note 1; their below in parallel with the analysis of impacts. cost to the government from the fiscal analysis in Note 2; and who receives them from the NOTE 10 on communications discusses how energy, poverty, and firm analysts (Notes to raise awareness about the subsidies and the 1, 3, and 4 for households, and 6 for firms government’s intentions, and build acceptance and economic sectors). Finally, the social for the reform, thereby lowering its political protection specialist can outline options cost. It covers several important aspects available to mitigate adverse impacts on the of communications campaigns, including poor and the vulnerable (Note 5). the tools for (a) identifying and consulting stakeholders, (b) conducting research on Three types of information inform a forward- opinions, and (c) delivering tailored messages. looking political-economy analysis of the Communications campaigns implemented subsidies and their reform. The first is the in support of energy subsidy reform can decision-making structure, including the take many forms. Note 10 outlines common institutional setup, the de facto control over elements and proven practices that transcend subsidies, and the influence of interest groups country borders and political contexts. over these decisions. In addition, the note recommends taking a historical perspective The note provides step-by-step guidance on how these structures have evolved. Second for an approach that starts with establishing is the stakeholder interests and dynamics. The the objectives, timeline, and governance for note sketches matrices to map out the ability the communications campaign. The second of groups to organize for or against the reform step involves mapping the main internal and and their relative benefits and disbenefits external stakeholders according to their level due to the subsidies. A simple framework of importance and influence over the reform then allows inference of the conditions that agenda. This should be carried out in parallel affect the likelihood of a successful reform with the political-economy analysis (Note 9). under each case. Third, the note proposes Next, the note provides some instructions retracing the track record with reform and on how to conduct opinion research to mapping it out to understand historical factors understand stakeholder views. More detailed influencing success. information on these qualitative tools can be found in Note 4. A first step toward reforms often requires that the subsidies and their costs become Note 10 provides additional guidance on better known, as well as social safety nets coordinating, through a range of institutional that will be used to mitigate adverse effects, bodies, a common approach to messages something that can be achieved through about energy subsidy reforms among communications campaigns. The importance internal stakeholders. The fourth step of communications is often overlooked, provides guidance on creating and pretesting receiving the attention of reformers in the compelling messages to address external government at a late stage. While it comes last stakeholders’ views and build awareness. in ESRAF, it is recommended that governments A simple set of characteristics of what begin, as early in the process as possible, the makes a good message is outlined. The fifth step concerns the messenger. In particular, 28 OVERVIEW NOTE the note explains how to assign credible household data to estimate households’ messengers and spokespeople depending on consumption of subsidized energy products, the audience, and recommends media training. perhaps focusing only on the direct effects The sixth step identifies the best channels of reforms on prices and people’s welfare— for communicating messages to different although even such a limited assessment stakeholders. The seventh step outlines how would be difficult where there is rampant illegal to set a set of measurable goals to track diversion and black marketing of subsidized progress and ensure that the communications fuels or where metering of network energy campaign is achieving its objectives. consumption is poor. An assessment of existing social safety nets, in particular their A RAPID ASSESSMENT OF ability to mitigate the adverse effects of higher ENERGY PRICE SUBSIDIES prices, is among governments’ most pressing concerns. A list of existing programs should be Where resources and time are scarce and drawn up, including their basic characteristics underpricing of energy is the subsidy of (targeting, coverage, generosity, leakage, interest, a rapid assessment of the magnitude and administrative features, such as whether of subsidies may be carried out through the comprehensive social registries exist), and price gap approach. The amount by which compared to the findings of the distributional prices will need to rise to eliminate the analysis. The qualitative analysis can be left for price subsidies is a necessary input to all a more complete assessment, especially if it subsequent analyses of the impacts of reform requires some on-the-ground data collection. on households (Note 3), firms (Note 6), and While the firm-level data needed to estimate the economy (Note 7). The total value of the impacts on firms may not be available, a subsidy for a given form of energy, measured simple analysis of energy intensity by sector by the price gap multiplied by the total units using an input-output table can identify the subsidized, is an input to the fiscal analysis sectors that are most likely to be affected. The (Note 2), although price gaps do not usually macroeconomic context will help define the capture tax expenditures and other fiscal right timing and pace of the reform. losses. For the political context, a full stakeholder A rapid diagnostic examination of price gaps analysis may not be feasible. A rapid may miss several dimensions of effects of assessment should map out the different subsidies and their reforms. However, where actors and interest groups, who is paying for price subsidies have material effects on the the subsidies, how subsidies are delivered, government budget, it is useful to assess who is benefiting indirectly or even illegally the government’s fiscal balance and debt from the price subsidies, and the self-interests situation to estimate the fiscal implications of and capabilities of the interest groups. In continuing with or reforming the energy price addition, basic information on the history subsidies. Resources and time permitting, a of the subsidies is vital, as well as the stated more detailed model may be developed to goals at the time of their creation, which partly project the fiscal implications of different determine how different interest groups react reform scenarios, including the costs of to the reform. A review of the press coverage different mitigation policies. In addition, a rapid and government communication about the assessment will rely on analysis of existing subsidies would be a good starting point. HOW TO USE THE FRAMEWORK 29 The checklist in the appendix aims to guide a team preparing diagnostic reports to help provide the evidence base to decide how to sequence and prioritize various aspects of a reform. While these questions may not all be addressed by a rapid assessment, they will help ensure that important main aspects are considered at the onset. OVERVIEW NOTE APPENDIX: CHECKLIST FOR RAPID ASSESSMENT: ANALYTICAL ISSUES RELATED TO ENERGY SUBSIDY REFORM A. THE NATURE AND MAGNITUDE OF SUBSIDIES 1 | What are the various subsidies on liquid fuels (LPG, gasoline, kerosene, diesel, and fuel oil), natural gas, coal, electricity, and district heating, and what are their values, individually and combined? a. Subsidies to consumers (as well as other benefit transfers conditional upon energy consumption) b. Producer subsidies c. Financial cost of subsidies under current policies at projected benchmark prices d. Size of subsidies allocated to renewable energy as a point of comparison 2 | How are various energy products priced? a. Description of the pricing principles (including taxation) and mechanisms, including legal and administrative frameworks b. Price controls and smoothing mechanisms, if any c. Whether pricing and taxation principles are being followed, and if not, degree of departure d. Benchmarking against relevant international comparators 3 | Who pays for the subsidies? a. Tax payers (for on- and off-budget transfers for subsidies and liabilities due to guarantees, soft budget constraints, tax expenditures, and other losses of government revenue) b. Consumers of the subsidized energy (for producer support) c. Producers of the subsidized energy (for consumer support) d. Lenders and other financing institutions (debt forgiveness, concessional financing) 4 | For what, where, to whom, and how are the subsidies delivered? For what a. Forms of energy subsidized b. If a liquid fuel, all of one type (for example, subsidy to all gasoline), or a particular fuel grade (subsidy to gasoline with a research octane number of 90 but not higher octane numbers) 30 APPENDIX: CHECKLIST FOR RAPID ASSESSMENT: ANALYTICAL ISSUES RELATED TO ENERGY SUBSIDY REFORM Where At the port of entry, upstream production, ex-refinery, at power or heat generation plant gate, wholesale, retail To whom a. All consumers, targeted consumers, or for a specified use (such as natural gas used in vehicles, but for no other purposes) b. All energy suppliers or a select few (state-owned suppliers only or both state-owned and private) How c. Subsidized prices provided to purchasers of energy i. Compensated by monetary reimbursements to energy suppliers ii. Compensated by provision of subsidized inputs, loans, tax expenditures, and other concessions made to energy suppliers iii. Covered by financial losses suffered by energy suppliers d. Cash transfers, in-kind payments, and other benefits provided directly to consumers that are ear-marked for consumption of specific energy types e. Benefits provided to energy suppliers without lowering consumer prices, enabled by the following: i. Higher prices paid by consumers or taxpayers ii. Fiscal concessions, subsidized loans, shifting of risk burden from suppliers to government or consumers, and other benefits provided to energy suppliers 5 | How does actual delivery compare to the official delivery mechanism? a. Timeliness of subsidy reimbursement to energy suppliers, the size of reimbursement compared to the official guidelines, and the adequacy of the reimbursement b. Effectiveness of targeting c. Actual prices paid by consumers versus official prices set d. Amount of subsidized energy that should be available versus the amount actually available to eligible beneficiaries e. Intended versus actual beneficiaries: Percentage of intended beneficiaries reached and degree to which ineligible beneficiaries capture the subsidies 6 | How does the country compare with similar countries on the nature and volume of energy subsidies? 31 OVERVIEW NOTE B. ECONOMIC AND SOCIAL CONSEQUENCES OF SUBSIDIES 7 | What is the macro-fiscal impact of subsidies? a. Evidence of pressure on fiscal balance and contribution to debt, and pressure on external balance b. Evidence of volatility in fiscal management due to volatility of total subsidies from year to year c. Potential fiscal revenue from reducing producer subsidies, such as lower tax expenditures, or from reducing consumer subsidies absorbed by energy suppliers, such as higher dividends from state-owned energy suppliers and higher corporate tax revenue d. Macroeconomic impact of removing price subsidies 8 | Do subsidies help the poor and the vulnerable? a. Distribution of subsidies by wealth status and their importance in the budgets of the poor b. Spatial (including urban and rural) distribution of subsidies, and other relevant disaggregation c. Amount captured by households versus amount captured by nonresidential consumers, and the percentage of total subsidized energy captured by poor households or other consumer categories of interest 9 | Do subsidies help improve the competitiveness of industries producing tradeable goods and services? a. What is the energy intensity of the main sectors of the economy? Are some of the most energy-intensive sectors also open to foreign competition? b. What is the level of energy efficiency of firms in these sectors? 10 | What are the opportunity costs of energy subsidies? a. Evidence of low public and private investment in energy infrastructure, such as energy shortages; international benchmark comparisons; and impact on economic growth b. Evidence of rationing of public expenditure in priority areas that could lead to higher economic growth and faster human development c. Loss in long-term growth and employment from economic distortions caused by subsidies d. Efficiency losses due to overconsumption of energy and misallocation of resources 11 | What might other consequences of subsidies be? a. Evidence of exacerbation of social harms resulting from excess production and consumption of energy, such as air pollution and greenhouse gas emissions 32 APPENDIX: CHECKLIST FOR RAPID ASSESSMENT: ANALYTICAL ISSUES RELATED TO ENERGY SUBSIDY REFORM b. Estimation of the extent of diversion, black market activities, smuggling, and other forms of corruption c. Health and other benefits of making modern clean household energy affordable C. CONSIDERATIONS FOR REFORM OPTIONS AND PRIORITIES 12 | What factors support or hinder reform? a. Characteristics of the energy sector: importer or exporter status, consumption share of different forms of energy, quality of energy service delivery, market concentration of energy suppliers, sector performance, sector governance, performance and other characteristics of state-owned energy suppliers, energy efficiency in the economy b. Dependence of the economy or specific sectors on subsidized energy c. Public perception of, and level of awareness about, energy subsidies d. Lessons from previous experience of reforms e. Opportunities for and hindrances to subsidy reform due to exogenous developments, such as world fuel price movements, extreme weather events, global financial crisis, and sharp currency depreciation f. Political support for reforms g. Key constituencies likely to benefit from or suffer losses as a result of reforms h. Performance of social safety nets and support systems to protect the vulnerable population 13 | What adverse effects of consumer price subsidy reforms call for mitigation measures? a. The poor and near-poor who will be adversely affected through direct and indirect effects of subsidy reforms b. Appropriate monetary and fiscal policy responses to short-term macroeconomic impact on output, inflation, and other macroeconomic indicators c. Competitiveness of industries and commercial firms reliant on subsidized energy and mitigation mechanisms to help them adapt to new input prices 33 OVERVIEW NOTE ENDNOTES 1 The registry of intended nationally determined contributions, which become nationally determined contributions upon entry into force, is available at http://www4.unfccc.int/ ndcregistry/Pages/All.aspx. Additional information at regional, country, and sectoral levels can be found at indc.worldbank.org. 2 Social safety nets are noncontributory transfers in cash or in kind targeted to the poor and vulnerable. Social safety nets span various forms of support: conditional and unconditional cash transfers, public works, fee waivers, food rations and feeding programs, and social services. Social safety nets are part of broader social protection and labor policies, which also include contributory transfers (such as pensions) and employment promotion (such as retraining or job search assistance). If the cash transfer or any other benefits are conditional on purchasing a specific type of energy, this framework categorizes such measures as energy subsidies. 3 Responses from more than 130,000 firms around the world are tabulated and regularly updated at http://www.enterprisesurveys.org/data/exploretopics/infrastructure. 4 Stochastic models, however, are unlikely to be suitable in countries with material dependence on oil revenue combined with subsidies on petroleum products, because oil revenue calculation is complex and not amenable to incorporation in a stochastic model. REFERENCES Clements, Benedict, David Coady, Stefania Fabrizio, Sanjeev Gupta, Trevor Alleyne, and Carlo Sdralevich, eds. 2013. Energy Subsidy Reform: Lessons and Implications. Washington, DC: International Monetary Fund. Forouzanfar, M. H., A. Afshin, L. T. Alexander, H. R. Anderson, Z. A. Bhutta, S. Biryukov, M. Brauer, R. Burnett, K. Cercy, F. J. Charlson, A. J. Cohen, and others. 2016. “Global, Regional, and National Comparative Risk Assessment of 79 Behavioural, Environmental and Occupational, and Metabolic Risks or Clusters of Risks, 1990–2015: A Systematic Analysis for the Global Burden of Disease Study 2015.” Lancet, 388(10053):1659–724. IEA (International Energy Agency). 2016. World Energy Outlook 2016. Paris: OECD/IEA. IMF (International Monetary Fund). 2012. “Fiscal Monitor Update January 2012.” Fiscal Affairs Department, IMF, Washington, DC. Rao, V., and M. Woolcock. 2003. “Integrating Qualitative and Quantitative Approaches in Program Evaluation.” In François Bourguignon and Luiz Pereira da Silva (eds.), The Impact of Economic Policies on Poverty and Income Distribution: Evaluation Techniques and Tools. Washington, DC: World Bank and Oxford University Press. 34 ENDNOTES Energy Subsidy Reform Assessment Framework LIST OF GOOD PRACTICE NOTES NOTE 1 Identifying and Quantifying Energy Subsidies NOTE 2 Assessing the Fiscal Cost of Subsidies and Fiscal Impact of Reform NOTE 3 Analyzing the Incidence of Consumer Price Subsidies and the Impact of Reform on Households — Quantitative Analysis NOTE 4 Incidence of Price Subsidies on Households, and Distributional Impact of Reform — Qualitative Methods NOTE 5 Assessing the readiness of Social Safety Nets to Mitigate the Impact of Reform NOTE 6 Identifying the Impacts of Higher Energy Prices on Firms and Industrial Competitiveness NOTE 7 Modeling Macroeconomic Impacts and Global externalities NOTE 8 Local Environmental Externalities due to Energy Price Subsidies: A Focus on Air Pollution and Health NOTE 9 Assessing the Political Economy of Energy Subsidies to Support Policy Reform Operations NOTE 10 Designing Communications Campaigns for Energy Subsidy Reform 35 36