OECD Inventory of Support Measures for Fossil Fuels 2024

2. Key trends in the fiscal cost of support measures for fossil fuels

The fiscal cost of government support for fossil fuels decreased significantly in 2023 as energy supply costs fell

New OECD and IEA data show that the fiscal cost of global support for fossil fuels dropped by almost one third to USD 1 102.1 billion in 2023, down from USD 1 612.6 billion in 2022. Energy supply costs, used as a benchmark reference price, fell from record highs in 2022 in many parts of the world, thereby decreasing the estimated value of fossil fuels sold below this reference price (IEA, 2024[2]). That said, with many measures supporting production and consumption of fossil fuels still in place, the fiscal cost of support measures for fossil fuels remains elevated compared relative to the historical average (Figure 2.1).

The OECD Inventory estimates that direct transfers and tax expenditures associated with support measures for fossil fuels amounted to USD 514.1 billion in 2023. In addition, the IEA calculates that fossil fuels sold below market prices amounted to USD 616.4 billion. Decreases were significant across electricity, coal, petroleum, and natural gas (Figure 2.1).

The 2023 IEA estimates of the under-pricing of fossil fuels (USD 616.4 billion) are about half of the comparable figure for 2022 (USD 1 178 billion). Benchmark prices (based on energy supply costs) eased, particularly for natural gas, thereby decreasing the difference between the subsidised end-user prices and the benchmark prices (IEA, 2024[2]). That said, the Inventory reports a slight increase in the fiscal cost of support measures for fossil fuels compared with 2022 level (USD 514.1 billion vs USD 503.7 billion), highlighting that governments continued to support fossil fuel production and consumption through specific support measures.

The fiscal cost of direct budgetary transfers amounted to USD 269.8 billion in 2023, making them for the first time more costly than tax expenditure measures (which fell from USD 260.4 billion in 2022 to USD 244.3 billion, Figure 2.2). Most of this fiscal cost of direct budgetary transfers is due to the implementation of several lump sum transfers, support providing compensation for increased energy bills and regulated or capped energy prices to offset exceptionally high energy prices since 2022.

Figure 2.1. The fiscal cost of support for fossil fuels dropped by almost one-third in 2023

OECD-IEA combined estimate of the fiscal cost of support measures for fossil fuels (82 economies), by fuels (in USD billion, 2023)

1) The OECD-IEA combined estimate covers 82 countries, resulting from merging IEA price-gap estimates and OECD Inventory estimates of the fiscal cost of support measures for fossil fuels. Countries included in the calculations are: Algeria, Angola, Argentina, Armenia, Australia, Austria, Azerbaijan, Bahrain, Bangladesh, Belgium, Bolivia, Brazil, Brunei Darussalam, Canada, Chile, People’s Republic of China (hereafter ‘China’), Colombia, Costa Rica, Croatia, Czech Republic, Denmark, Ecuador, Egypt, El Salvador, Estonia, Finland, France, Gabon, Georgia, Germany, Ghana, Greece, Hungary, Iceland, India, Indonesia, Iraq, Ireland, Islamic Republic of Iran, Israel, Italy, Japan, Kazakhstan, Korea, Kuwait, Latvia, Libya, Lithuania, Luxembourg, Malaysia, Mexico, Moldova, Netherlands, New Zealand, Nigeria, Norway, Oman, Pakistan, Poland, Portugal, Qatar, Russian Federation, Saudi Arabia, Slovakia, Slovenia, South Africa, Spain, Sri Lanka, Sweden, Switzerland, Chinese Taipei, Thailand, Trinidad and Tobago, Türkiye, Turkmenistan, Ukraine, United Arab Emirates, United Kingdom, United States, Uzbekistan, Venezuela, Viet Nam.

2) Fiscal cost of support for the fuel “Electricity” is derived from measures providing support for electricity generation or consumption (e.g. through regulated electricity tariffs for end-users). Estimates only include the fossil fuel share in electricity generation (i.e. excluding renewables and other non-fossil fuel sources). Measures supporting electricity generation technology based on a specific fossil fuel (used as an input) are categorised under this fuel (i.e. petroleum, coal, or natural gas).

3) Fiscal cost of support measures for fossil fuels included in the Inventory are based on information reported by countries through official documentation (e.g. budget reports). Support measures for which such information is not available are excluded from the aggregate amount reported in this report. In addition, support measures in certain countries may not have been exhaustively identified.

4) The Inventory reports tax expenditures as estimates of revenue that is foregone due to a particular feature of the tax system that reduces or postpones tax payments (relative to a jurisdiction’s benchmark tax system) to the benefit of fossil fuels’ producers or users. Hence, (i) tax expenditures estimates can increase over time either because of greater concessions (relative to the benchmark tax system) or because of an increase in the benchmark itself; (ii) cross-country comparisons of tax expenditures can be misleading due to country-specific benchmark tax systems.

5) Individual support measures for fossil fuels are included in the Inventory without reference to their economic or environmental effects. No judgment is therefore made as to whether such measures are inefficient or ought to be reformed.

6) Data are expressed in constant 2023 US dollars. Data for 2023 are on a preliminary basis.

Sources: OECD Inventory of support measures for fossil fuels and IEA Fossil Fuel Subsidies Database.

Figure 2.2. The fiscal cost of direct budget transfers was larger than tax expenditures in 2023

Fiscal cost of support measures for fossil fuels in OECD & partner countries (48 countries), by support mechanism (in USD billion)

1) Countries included in the calculations are: Argentina, Armenia, Australia, Austria, Azerbaijan, Belgium, Brazil, Canada, Chile, China, Colombia, Costa Rica, Czech Republic, Denmark, Estonia, Finland, France, Georgia, Germany, Greece, Hungary, Iceland, India, Indonesia, Ireland, Israel, Italy, Japan, Korea, Latvia, Lithuania, Luxembourg, Mexico, Moldova, Netherlands, New Zealand, Norway, Poland, Portugal, Slovakia, Slovenia, South Africa, Spain, Sweden, Switzerland, Türkiye, United Kingdom, United States.

2) Fiscal cost of support measures for fossil fuels included in the Inventory are based on information reported by countries through official documentation (e.g. budget reports). Support measures for which such information is not available are excluded from the aggregate amount reported in this report. In addition, support measures in certain countries may not have been exhaustively identified.

3) The Inventory reports tax expenditures as estimates of revenue that is foregone due to a particular feature of the tax system that reduces or postpones tax payments (relative to a jurisdiction’s benchmark tax system) to the benefit of fossil fuels’ producers or users. Hence, (i) tax expenditures estimates can increase over time either because of greater concessions (relative to the benchmark tax system) or because of an increase in the benchmark itself; (ii) cross-country comparisons of tax expenditures can be misleading due to country-specific benchmark tax systems.

4) Individual support measures for fossil fuels are included in the Inventory without reference to their economic or environmental effects. No judgment is therefore made as to whether such measures are inefficient or ought to be reformed.

5) Data are expressed in nominal US dollars. Data for 2023 are on a preliminary basis.

Source: OECD Inventory of support measures for fossil fuels.

The fiscal cost of support for coal fell from record highs, while support for natural gas has becomes relatively more important

Over the last ten years, the fiscal cost of support measures for coal – the most emission-intensive fuel – has shown significant fluctuations (Figure 2.1). Between 2014 and 2018, this type of support increased from USD 25.7 billion to USD 35.9 billion. This reflected the efforts of various governments to sustain coal use amid concerns about energy security and debates about the need to maintain conventional energy sources. While volatile over 2018 to 2021, the fiscal cost of coal support measures saw a significant increase from 2020, peaking at USD 43.5 billion in 2022, the highest level recorded in the decade. This surge resulted from transitional measures adopted by many governments to ease the financial burden on vulnerable households, to counteract the impact of high oil and gas prices.

However, 2023 saw a significant reduction in the fiscal cost of coal support measures to USD 27.7 billion, a 36% decrease from the previous year. This decline was largely due to the end of the transitional measures implemented from 2022. Nevertheless, this shift could also reflect the start of a realignment of policies with international climate commitments, with increased efforts to reduce support for the most emission-intensive fuel.

While the fiscal cost of support to coal has seen a dramatic decrease, natural gas greatly increased its share in the total fiscal cost of support for fossil fuels. This growth trend for natural gas began as far back as 2018 yet grew rapidly in the past three years to from 20% of total fiscal support in 2020, to 31% in 2023. This represents a more than doubling in the last five years, from USD 145 billion in 2018 to USD 343 billion in 2023.

Considered to be the cleanest type of fossil fuel in terms of CO₂ emissions, natural gas has often been promoted as an effective transitional fuel while renewable alternatives become available at scale. While this may explain the broader trend of increasing fiscal cost of support for natural gas, it does not fully account for the recent jump in the fiscal cost of support, especially during and after the energy crisis. Rather, this recent fluctuation may be better explained by the interruption of piped natural gas to Europe from the Russian Federation resulting in a peak in global natural gas prices in 2022 (Figure 2.3; (IEA, 2024[3]; World Bank, 2024[4]).

Figure 2.3. Fossil fuel prices went down in 2023

Note: Natural gas prices are based on the TTF benchmark price. Oil prices are based on the Brent Crude benchmark. Coal Prices are based on the South African benchmark.

Source: World Bank Commodity Price Data.

Support for consumption by households and firms accounts for most of the fiscal cost of support for fossil fuels, but remains largely untargeted

While energy supply costs eased since their peak in 2022 (IEA, 2024[2]), the fiscal cost of support to fossil fuel consumption has remained high and even grew slightly from USD 412 billion in 2022 to USD 464 billion in 2023. The Consumer Support Estimate (CSE) reached its highest ever share of the total fiscal cost of support to fossil fuels, accounting for just over 90% of the fiscal cost of support, up from 87% in 2022 (Figure 2.4). At the same time, support to producers estimate (PSE) and general support estimate (GSSE) both decreased by 21% (from USD 46.1 billion to USD 36.5 billion) and 16% (from USD 15.3 billion to USD 12.9 billion) in 2023, respectively.

Figure 2.4. Support to consumption represents the vast majority of the fiscal cost of support to fossil fuels

Fiscal cost of support measures for fossil fuels in OECD and partner countries (48 countries), by type of support (in USD billion)

1) Producer Support Estimate (PSE) includes measures that support the activities of exploration and extraction, bulk transportation and storage, and refining and processing of fossil fuels. Consumer Support Estimate (CSE) includes measures that support the use of fossil fuels in power and heat generation, industrial processes, and activities outside of the energy sector, and all other final uses of fossil fuels, whether in the transport sector, the residential sector, or primary industries outside of the energy sector (e.g. agriculture and forestry). General Services Support Estimate (GSSE) includes measures benefitting producers or consumers collectively, as are measures that do not increase current production or consumption of fossil fuels but may do so in the future. Examples of GSSE measures would include public support for industry-specific infrastructure development such as public support for the construction of coal or natural-gas terminals.

2) Countries included in the calculations are: Argentina, Armenia, Australia, Austria, Azerbaijan, Belgium, Brazil, Canada, Chile, China, Colombia, Costa Rica, Czech Republic, Denmark, Estonia, Finland, France, Georgia, Germany, Greece, Hungary, Iceland, India, Indonesia, Ireland, Israel, Italy, Japan, Korea, Latvia, Lithuania, Luxembourg, Mexico, Moldova, Netherlands, New Zealand, Norway, Poland, Portugal, Slovak Republic, Slovenia, South Africa, Spain, Sweden, Switzerland, Türkiye, United Kingdom, United States.

3) Fiscal cost of support measures for fossil fuels included in the Inventory are based on information reported by countries through official documentation (e.g. budget reports). Support measures for which such information is not available are excluded from the aggregate amount reported in this report. In addition, support measures in certain countries may not have been exhaustively identified.

4) The Inventory reports tax expenditures as estimates of revenue that is foregone due to a particular feature of the tax system that reduces or postpones tax payments (relative to a jurisdiction’s benchmark tax system) to the benefit of fossil fuels’ producers or users. Hence, (i) tax expenditures estimates can increase over time either because of greater concessions (relative to the benchmark tax system) or because of an increase in the benchmark itself; (ii) cross-country comparisons of tax expenditures can be misleading due to country-specific benchmark tax systems.

5) Individual support measures for fossil fuels are included in the Inventory without reference to their economic or environmental effects. No judgment is therefore made as to whether such measures are inefficient or ought to be reformed.

6) Data are expressed in nominal US dollars. Data for 2023 are on a preliminary basis.

Source: OECD Inventory of support measures for fossil fuels.

The residential sector benefitted the most from the increased support for consumption, with its fiscal cost reaching a record USD 189 billion, a 29% increase from 2022 (Figure 2.5). Residential support has also become by far the largest share of total fiscal cost of sector support at 37%. The fiscal cost of support to manufacturers and industry also experienced an important 14% increase in 2023, reaching a record high of USD 103 billion. The growth of the fiscal cost of support in these two sectors was matched by a relative decline in the fiscal cost of support for transportation and electricity generation. The transportation sector's share of the fiscal cost of support fell to a record low of 18.8%, declining from USD 115.5 billion in 2022 to USD 96.8 billion in 2023. Similarly, the share of the fiscal cost of support for electricity generation fell to a low of 5.1%, declining from USD 27.4 billion in 2022 to USD 26.2 billion in 2023.

Figure 2.5. Support for residential use and manufacturing now represents the largest shares of the overall fiscal cost of support

Fiscal cost of support measures for fossil fuel in OECD and partner countries (48 countries), by sector (in USD billion)

1) Production sector includes measures that support the production, exploration, trade, storage, and transportation of fossil fuels. The transportation sector includes measures that support the final uses of fossil fuels in the transport sector, including domestic aviation, domestic navigation, road, and rail. Residential sector includes measures that support the final use of fossil fuels (including electricity) in the residential sector (i.e. consumption by households). The electricity generation sector includes measures that support fossil fuel-based electricity and heat generation (both main and auto producers). Manufacturing and others include measures that support the use of fossil fuels in the energy transformation sector other than electricity and heat generation, industrial and manufacturing sector, commercial and public services, agriculture, forestry, and fisheries.

2) Countries included in the calculations are: Argentina, Armenia, Australia, Austria, Azerbaijan, Belgium, Brazil, Canada, Chile, China, Colombia, Costa Rica, Czech Republic, Denmark, Estonia, Finland, France, Georgia, Germany, Greece, Hungary, Iceland, India, Indonesia, Ireland, Israel, Italy, Japan, Korea, Latvia, Lithuania, Luxembourg, Mexico, Moldova, Netherlands, New Zealand, Norway, Poland, Portugal, Slovak Republic, Slovenia, South Africa, Spain, Sweden, Switzerland, Türkiye, United Kingdom, United States.

3) Fiscal cost of support measures for fossil fuels included in the Inventory are based on information reported by countries through official documentation (e.g. budget reports). Support measures for which such information is not available are excluded from the aggregate amount reported in this report. In addition, support measures in certain countries may not have been exhaustively identified.

4) The Inventory reports tax expenditures as estimates of revenue that is foregone due to a particular feature of the tax system that reduces or postpones tax payments (relative to a jurisdiction’s benchmark tax system) to the benefit of fossil fuels’ producers or users. Hence, (i) tax expenditures estimates can increase over time either because of greater concessions (relative to the benchmark tax system) or because of an increase in the benchmark itself; (ii) cross-country comparisons of tax expenditures can be misleading due to country-specific benchmark tax systems.

5) Individual support measures for fossil fuels are included in the Inventory without reference to their economic or environmental effects. No judgment is therefore made as to whether such measures are inefficient or ought to be reformed.

6) Data are expressed in nominal US dollars. Data for 2023 are on a preliminary basis.

Source: OECD Inventory of support measures for fossil fuels.

Households have been significantly impacted by the energy crisis. Energy poverty has been rising, with millions of individuals now reported to be facing heightened difficulties in meeting essential energy needs such as heating their homes during winter, notably across Europe (European Commission[5]). Indeed, the average pre-tax electricity price for households in the European Union increased by 6.55% in 2023 over 2022 and now stands at 61.73% higher than 2021 levels (Eurostat[6]). The most vulnerable populations, generally low-income households and the elderly, are disproportionately affected due to their limited capacity to absorb these rising costs (OECD, 2022[7]). Moreover, the sharp rise in energy bills has also impacted their capacity to allocate resources to other essential needs, such as food and education. Again, low- and middle-income households are particularly vulnerable, as they devote a larger portion of their income to energy costs (OECD, 2022[7]).

Support to consumption, if well-targeted, can thus play a key role in alleviating the burden on households in times of pressure from rising energy costs. However, measures to support fossil fuel consumption were still largely poorly targeted in 2023 (85% of fossil fuels sold below market prices were untargeted, Figure 2.6). While defining and identifying the right target to support may be challenging (OECD, 2023[8]; Hemmerlé et al., 2023[9]), poorly targeted measures raise concerns related to efficiency, equity, and the environment. First, the fiscal cost of poorly targeted support measures is higher than necessary as a broader range of recipients beyond just lower income households benefit from them, raising efficiency concerns. Second, these measures often disproportionately favour wealthier households, which typically consume more energy – e.g. through larger vehicles, more spacious homes potentially equipped with air conditioning and more electrical appliances (Wooders, 2024[10]), raising equity concerns. Finally, untargeted measures distort price signals, thereby contributing to the higher continued consumption of fossil fuels.

Figure 2.6. The vast majority of fossil fuel consumption support measures by value are still untargeted

Fossil fuels sold below market prices in 49 countries (in USD billion, 2023)

1. Targeted support measures are defined as benefitting a specific category of households, firms, or energy users. Targeted support includes agriculture and household (LPG and kerosene fuels only) sectoral support.

2. Countries included in the IEA Fossil Fuel Subsidies estimates are: Algeria, Angola, Argentina, Austria, Azerbaijan, Bahrain, Bangladesh, Bolivia, Brunei Darussalam, China, Colombia, Croatia, Ecuador, Egypt, El Salvador, France, Gabon, Ghana, Hungary, India, Indonesia, Iraq, Islamic Republic of Iran, Kazakhstan, Korea, Kuwait, Libya, Malaysia, Mexico, Nigeria, Oman, Pakistan, Poland, Qatar, Russian Federation, Saudi Arabia, Slovakia, South Africa, Sri Lanka, Chinese Taipei, Thailand, Trinidad and Tobago, Turkmenistan, Ukraine, United Arab Emirates, United Kingdom, Uzbekistan, Venezuela, Viet Nam.

Source: IEA (2024[2]).

Likewise, poorly targeted support measures provided to businesses raise concerns related to competition, market efficiency, and the environment. Support to firms can embed price distortions by decreasing energy input prices which firms could pass onto their output prices. This can therefore have indirect effects on wider markets. Government support to firms can also imply more significant climate implications as it may support emission-intensive production processes, especially when supporting energy-intensive sectors (Garsous, Smith and Bourny, 2023[11]). While challenging to implement, governments should aim to target support to companies facing liquidity and solvency problems due to energy price shocks but which would otherwise be viable (OECD, 2022[12]). This strategy would mitigate against the risk of supporting inefficient firms, which can impede competition, hinder domestic productivity growth, and distort international markets (OECD, 2020[13]). In general, government support to firms should be time limited, transparent, proportionate, and non-discriminatory (i.e. applying objective and transparent criteria for determining firms’ eligibility) (OECD, 2020[13]; 2021[14]).