After successive crises, government support for businesses and households remains elevated in many OECD countries. Subsidies spending increased sharply during the COVID-19 pandemic and remained high during the subsequent energy price and cost-of-living crises. During the same period, the use of indirect support measures such as tax expenditures has expanded. This chapter reviews savings initiatives in this area. The measures include reducing direct support to private enterprises, reducing tax expenditure for specific businesses expenses, and reducing support for state-owned enterprises. Reported measures also involve reducing support to lower households’ energy spending. The chapter also covers recent developments in official development assistance, where several countries are significantly reducing spending. In this area, governments are also altering sectoral or geographical allocations and leveraging private sector finance.
Restoring Public Finances
Enabling Effective Government
Abstract
Other areas of public spending include support for businesses and non-welfare support for households, including many of the energy-related support measures adopted in recent years, as well as support to foster more efficient use of energy. Other areas of public spending also include international development assistance.
After multiple crises, government spending on subsidies remains elevated in several OECD countries, and tax relief is widely used. Still, the RPF Survey shows that around two thirds of respondents have pursued measures to streamline, cap, or phase-out various types of support for businesses and households, or to improve the targeting of the support. Measures tend to be more prevalent in areas where support was deemed to be less effective, involved heavy administration, or was considered less necessary as markets have matured. The sharp increase in energy prices in 2026 has renewed pressures for energy-related support, leading a number of OECD countries to reintroduce such measures (OECD, 2026[1]).
In addition, some countries are reducing spending on international development assistance through various means.
Reform initiatives and savings measures
Reducing and streamlining direct grants and industry support; often as a result of cross-sectoral spending reviews or third-party assessments. Some are redirecting savings to higher-value uses to promote firm innovation and competitiveness.
Tightening or phasing out outdated or inefficient tax relief and increasingly undertaking systematic assessments to identify such support.
Targeted reductions in policies to provide incentives for more efficient use of energy for both businesses and households, where costs have risen or markets are maturing. In some cases, countries are scaling back incentives that are no longer considered necessary for adoption.
After a period marked by largely untargeted household support to ease the adverse effects of high energy and food prices, some governments are focusing support on those most in need.
Some respondents are reducing spending on official development assistance (ODA), including through headline reductions, targeted cuts through sectoral and geographic reprioritising, or by streamlining governance and delivery.
13.1.1. Recent trends in government spending on economic support to businesses
In 2024, direct government expenditure on support for private enterprises, measured as subsidies on production and imports across sectors, amounted to the equivalent of 1.0% of GDP in the OECD, up from 0.9% of GDP in 2019 (see Figure 13.1).
Note: General government expenditure, subsidies on production and imports across all policy areas. Data for Chile and Indonesia shows 2023 rather than 2024, and Chile is thus not included in the 2024 OECD average. No data available for Türkiye.
Source: Public finance by economic transaction - government at a glance indicators, yearly updates – based on OECD National Accounts Statistics (database) and Eurostat Government Finance Statistics (database). OECD Data Explorer.
There is considerable variation across countries. European countries tend to spend more on subsidies, with the equivalent average in 2024 being 1.5% of GDP among OECD countries in the EU. The figures include only direct government expenditure, and exclude tax incentives, below-market loans, guarantees, equity injections or other indirect spending. They capture therefore only a part of governments’ total economic support for enterprises. Due to large variety in scope and instruments used, comparable data on the total size of support remains scarce (OECD, 2025[2]).
Subsidies are commonly used in crises as a rapid response for governments to stabilise markets, protect vulnerable groups and ensure the continuity of essential services. As seen in Figure 13.2, the share of government expenditure spent on subsidies increased sharply during the COVID-19 pandemic, as most governments took swift and massive actions to support households and businesses (Koelle, Kwak and Brunnengräber, 2026[3]). One challenge is that once introduced, support schemes can be difficult to remove – even when conditions normalise and support is no longer needed. Although average OECD spending on subsidies had nearly returned to pre-pandemic levels by 2024, country-level data, as displayed in Figure 13.1, shows that subsidy spending remains elevated in many countries compared to 2019. On average in the OECD, subsidies constituted 2.2% of general government expenditure in 2024, up from 2.1% in 2019.
Subsidies for enterprises, 2007-2024, yearly averages for OECD Member countries
Note: Data for Chile and Türkiye are not included in the OECD average.
Source: Public finance by economic transaction - government at a glance indicators, yearly updates – based on OECD National Accounts Statistics (database) and Eurostat Government Finance Statistics (database). OECD Data Explorer.
Beyond direct expenditures, governments also use indirect instruments to support businesses. Tax expenditures are provisions in the tax system that reduce the amount of tax an individual or business would otherwise owe and are commonly used. Moreover, other indirect support instruments, such as below-market loans or guarantees and equity injections, have become a more common budgeting feature, especially in crises (OECD, 2023[4]). Such instruments were widely deployed during the pandemic (Moretti, Braendle and Leroy, 2021[5]), and are increasingly considered as tools for supporting policy priorities, such as the energy transition. Although indirect support instruments may be warranted when carefully designed, their effectiveness relative to direct spending outside of crisis response depends on design and context, and is not firmly established (Espinoza et al., 2022[6]).
From a budgeting perspective, the growing use of loans, guarantees and equity injections also introduces additional complexity. Although these instruments create financial assets that in theory offset their associated liabilities, they impose real costs because governments do not recover the full value of every loan, guarantee or equity stake. Some borrowers will miss payments, some guarantees will be called, and some equity holdings may lose value. As a result, the government incurs losses across the portfolio. These risks and uncertain outcomes make the true cost of such support less transparent than direct spending and increase the analytical burden on budget offices (OECD, 2025[7]). Recent OECD work highlights the need for stronger practices for valuing indirect support instruments, monitoring fiscal risks and clearly presenting them in budget documents to maintain effective scrutiny. Greater transparency on the size, scope and beneficiaries of support measures can also enable greater government accountability and public scrutiny – for example by allowing impartial third parties to assess whether support measures represent a good use of taxpayers’ funds.
Overall, maximising the value of market-intervening support involves targeted, transparent, and time bound support measures. While interventions can be warranted to address market failures and help align private incentives with broader policy goals, gaps in design and governance can limit the effectiveness of support and divert scarce public resources from more productive uses (OECD, 2025[8]). Untargeted and open-ended support are particularly prone to creating inefficiencies, as they can shield uncompetitive firms, discourage innovation, or generate unintended effects for other policy objectives. Given their discretionary nature, scaling back or better targeting unproductive support measures often represents an area of scrutiny for countries pursuing fiscal consolidation and seeking to increase fiscal space.
Ensuring fiscal sustainability while meeting new industrial policy objectives is challenging, particularly as pressures to increase subsidies continue to rise. The observed growth in subsidy spending in several countries can be linked to what has been coined as ‘the return of industrial policy’. In recent years, governments have shown increased interest in supporting industrial firms to pursue strategic goals such as national competitiveness, resilience and security. Data from the new OECD MAGIC database, leveraging data on subsidies to large manufacturing firms, indicate that in 2023, industrial subsidies reached their highest level since 2009 (OECD, 2025[2]). Furthermore, the distribution of subsidies is highly skewed. For example, government support tends to be concentrated in certain industries (Solar PV, semiconductors, aluminium) and firms with partial state ownership receive significantly more support than their private counterparts. Subsidies tend to increase firms’ international market shares but are found to have no positive effect – or even negative effects – on investment and productivity (OECD, 2025[9]). These findings may have implications from a budgetary perspective, as they suggest that the subsidy-induced increase in market share does not arise from efficiency gains but the ability of subsidised firms to lower their prices or deter competitors.
13.1.2. Reform initiatives and savings measures in economic support to businesses
Measured approved or submitted to parliament for fiscal years of 2025 and 2026
Note: Results based on 39 RPF Survey responses. The measures reported in the “Other” category include one case of unspecified savings and one reduction in support measures for private roads, which will affect mainly businesses and other entities. Data is not available for France for Question 9.
Source: 2026 OECD Survey on Restoring Public Finances, Question 9: Support to Businesses and Households, Question 10: Tax Expenditures.
Figure 13.3 provides an overview of the frequency of recent savings measures, drawing on the responses to the RPF Survey. The results show that around three fifths of respondents are taking steps to tighten economic support for businesses. Ongoing reforms focus on streamlining business grants and reducing programmes with high growth in recent years, low value for money or heavy administration. After a period of expanding support for lower CO2 emissions in industry (OECD, 2025[13]), the RPF Survey includes examples of respondents currently scaling back such programmes. Details on the scope and impact of reported measures are provided below.
Reduce direct support to private enterprises
A common strategy reported in the RPF Survey is to reduce or streamline direct business and industry support. In several cases, cross-sectoral spending reviews or assessments have been used to identify savings. Such tools can help disclose overlapping initiatives or policies that are no longer justified, especially in countries where support programmes are dispersed across multiple ministries and policy areas. The RPF Survey shows that some governments are shielding – or even channeling savings toward – higher‑priority policies such as incentives for business R&D or firm competitiveness. This indicates that several OECD countries are seeking to maximise the marginal value of public funds by redirecting spending towards growth enhancing activities.
Both Austria (see Box 13.2) and Canada are pursuing broad efforts to reduce subsidies, including support for industrial decarbonisation. Following its Comprehensive Expenditure Review, Canada is reducing or discontinuing several industrial support programmes that were found to have accomplished their goals or be overlapping with other initiatives. One notable example is the Net Zero Accelerator, which was a time-limited acceleration mechanism to support large‑scale industrial decarbonisation projects. According to federal audit findings (Office of the Auditor General of Canada, 2024[14]), the Net Zero Accelerator had administrative and performance challenges, as well as lengthy and complex application processes. After a period of declining demand, the government has decided to close the initiative. At the same time, the government established a Strategic Response Fund in September 2025, which provides flexible terms to help firms in all sectors and regions impacted by tariffs to adapt, diversify, and grow (Government of Canada, 2025[15]).
Following broad political agreement, “the Agreement on a Stronger Business Community” (Denmark's Ministry of Taxation, 2024[16]), Denmark is discontinuing and restructuring multiple business support and promotion schemes. Through a mix of spending cuts and reduced tax expenditures, the reform is projected to free DKK 900 million annually (approximately 0.03% of GDP). Finland is also reducing business subsidies and support for development projects that have been found to not promote R&D activity. The measures are part of the government’s strategy to direct business support towards R&D, where the long‑term competitiveness impacts are considered to be highest. Ireland is reviewing and refocusing support for small businesses in the Department for Enterprise, Tourism and Employment, ensuring that spending is focused on capital priorities while reducing red tape.
Some respondents are reducing support for specific sectors or types of companies. This includes, among others, media, and cultural sectors.
Belgium will restructure subsidies for local media in the French-speaking community and not support any new cultural centres.
Iceland is reducing its appropriations for film reimbursements by tightening the eligibility criteria in its large film‑production reimbursement scheme. The measure is expected to lower expenditure by ISK 1.1 billion (approximately 0.02% of GDP) and comes after several years of growth in the number of applicants to the programme (Olsberg SPI, 2024[17]).
Norway is capping CO₂ compensation support to energy‑intensive industrial enterprises and requiring that at least 40% of the funds be used for energy‑saving and other related measures. By aligning CO₂ pricing with the actual electricity cost impacts from the EU carbon market, these changes are expected to trigger new investments in emission‑reducing technologies while reducing public expenditure. The reform was estimated to deliver savings of around NOK 3.8 billion over the period 2025-2031 (equivalent to approximately 0.1% of GDP, depending on the carbon allowance price). At the same time, Norway is reducing the funding envelop of a specific Fund financing emission-reducing measures, after a period of rapid growth in these appropriations.
Other measures in this category include Korea reducing support for high‑credit companies with adequate access to private financial markets and the Netherlands reporting a general decrease in subsidy budgets across ministries.
As a result of rising debt and sustained deficits, Austria has adopted a fiscal consolidation package to restore public finances and comply with fiscal requirements from the European Union. In the “double budget” for 2025 and 2026, the government plans savings of EUR 1.3 billion (approximately 0.3% of 2025 GDP) yearly on expenditure-side subsidies, reducing support for both businesses, municipalities and households. Around half of the savings over the period are related to environmental subsidies, including support for environmentally-friendly heating systems and renovations of private homes; subsidies for the decarbonisation of production processes of large enterprises; and subsidies for electromobility. There are also savings on other transport subsidies. In the short term, cuts were targeted at subsidies where reductions could be implemented quickly. In the medium term, the goal is to both reduce budgetary costs and increase desired policy effects by making subsidies more efficient, accurate and socially just. The government has established a task force to review and restructure subsidies across policy areas, and to identify additional subsidy savings worth EUR 800 million by 2029. With this, the consolidation volume for expenditure-side subsidies is estimated to increase to EUR 2.1 billion by 2029.
Source: Federal Ministry of Finance, Austria
Reduce tax expenditures for specific businesses expenses
As part of their efforts to rationalise business support, several RPF respondents are reducing expenditures on preferential tax treatment for specific business expenses or sectors, including tax expenditures in VAT, excise duties, or corporate income tax. This follows a broader trend where governments are scrutinising their use of tax expenditures in a systematic way to reduce outdated, inefficient, or non-justified support (Box 13.3).
Japan is reviewing and rationalising tax expenditures and subsidies with low policy effectiveness. In the 2026 tax reform (Japan’s annual package of tax policy changes), 30 special tax measures for corporations were subject to review. Of these, three measures were abolished, and 18 were revised with reductions. Going forward, the cabinet plans to consistently pursue reviews starting from the initial request and proposal stages.
The Netherlands has adopted a strategy to reduce tax expenditures, anchored in regular evaluations conducted by the Ministry of Finance to assess effectiveness, efficiency, administrative feasibility, and the rationale for government intervention. Around 200 tax expenditures have been identified, of which 125 are actively monitored, representing approximately EUR 167 billion, more than 40% of total tax revenues. Many schemes have been found to be ineffective, outdated, overly complex, or poorly targeted, and the government has developed options to adjust, redesign, or phase them out.
Portugal has established a tax unit in the Ministry of Finance, responsible for evaluating and monitoring the relevance and impact of tax benefits. In 2026, a budget programme dedicated to tax expenditure management and monitoring was created, establishing concrete objectives and ambitious performance targets for quantifying tax benefits and assessing their effectiveness. The goal is to simplify the existing framework, reduce tax expenditures and strengthen the cost-effectiveness of remaining benefits.
Efforts to further streamline and strengthen the performance evaluation for tax expenditures are also underway in other OECD Member and accession candidate countries, including Costa Rica, Germany, Korea and Peru.
Source: Ministry of Finance, Japan; Ministry of Finance, The Netherlands; Ministry of Finance, Portugal
Several OECD countries, such as Denmark, Finland, Japan, and Lithuania, are reducing tax expenditures for small to medium-sized enterprises (SMEs) or certain business investments. Japan is taking several steps to progressively phase out tax expenditures deemed to be inefficient. Strategies include tightening eligibility requirements, revising tax credit rates, setting predictable expiry dates, and reviewing measures before potential renewal.
Some countries are also ending small corporate tax expenditures for specific expenses, such as Belgium abolishing tax relief for social liabilities and capital gains on company vehicles and Finland abolishing the deduction for employer educational expenses.
Other OECD countries are reducing their support for specific sectors. In some cases, this can be due to particularly beneficial schemes in the past. For example, Hungary is reducing windfall profit tax relief for banks, which was aimed to incentivise banks into investing in government securities. Latvia is abolishing excise‑duty exemption for petroleum products used in electricity generation and cogeneration, increasing the tax burden on energy‑utilities and industrial cogeneration operators. The Netherlands is reducing motor‑vehicle tax provisions for fair and circus‑operators.
OECD countries are also reviewing tax expenditures adopted to incentivise a more effective energy mix, including the reduction or abolition of tax relief for low-emission vehicles.
In 2022, Costa Rica introduced a comprehensive framework to regulate the granting, monitoring, and revocation of tax exemptions, aiming to address longstanding weaknesses in oversight. The law strengthened administrative procedures, established a clear revocation mechanism, and created a new sanctioning regime including fines equivalent to the misused exempted amount. It also mandated periodic reviews to assess whether exemptions should be maintained, modified, reduced, or eliminated, supporting fiscal policymaking. In addition, the reform provided clearer definitions of exemption types and reinforced the Treasury’s supervisory authority, improving legal certainty and control. A major outcome was the repeal of approximately 138 obsolete exemptions, significantly simplifying and reducing the scope of Costa Rica’s tax expenditure system.
Source: Ministry of Finance, Costa Rica
Reduce support to state-owned enterprises
Recent OECD analysis finds that state-owned enterprises tend to receive higher public support than their private counterparts (OECD, 2025[12]). In the RPF Survey, some respondents reported measures to rationalise such support. For example, Korea is reducing personnel and operating costs in certain state-owned enterprises, and rationalising welfare and benefit programmes. The central government is also reducing its contributions and investments in non-essential areas. In the Slovak Republic, transfers to state-owned enterprises under the Ministry of Agriculture and Rural Development are being reduced.
Chinese Taipei reports the end of extraordinary government injections into the electric power provider Taipower. Such support was used in 2023-2025 to help stabilise electricity prices during the period of high global fuel costs. As electricity tariffs were gradually adjusted, the government sought to return Taipower to a self‑sustaining funding model.
The need for effective reviews of support to inform further budgetary adjustments
Beyond the specific savings measures reported in the RPF Survey, the scale and allocative inefficiency of support for industry and other businesses can be reduced by making support subject to a clear set of rules, objectives and principles before granted. If combined with competitive and neutral allocation, this can help reduce government actions that distort competition by favouring certain beneficiaries over others (OECD, 2022[18]). One approach to competitive allocation consists of grouping all existing spending to support a given policy objective and let potential recipients compete against each other. This ensures that support goes to the most deserving and efficient recipients and has been routinely used, among others, in energy sectors. If well designed and governed, competitive structures also give governments the flexibility to adjust funding envelopes in line with changing policy priorities and fiscal constraints.
In addition, establishing clear frameworks for making support time-bound – for example with sunset clauses (automatic expiration dates) of 2, 4 or 5 years – and contingent upon systematic review before renewal, can help prevent fiscal resources from being locked into outdated or inefficient programmes. Regular renewal assessments create an opportunity to scrutinise whether each measure continues to deliver value for money or if support should be adjusted or discontinued. Adjustments could include tightening eligibility or reducing support levels. Findings from the RPF Survey show that several countries are pursuing a more systematic approach for reviewing and reassessing tax expenditures. When it comes to direct support, spending reviews remain the primary tool to reassess and adjust baseline expenditure. Recently, several OECD countries have also sought to institutionalise systemic approaches to policy evaluation, entailing structures that embed evaluation capacity across government (see Box 13.5).
Australia: The Australian Centre for Evaluation (ACE) was created in 2023 within the Department of the Treasury to strengthen evaluation practice across the federal government. It functions as a central expert hub rather than a distributed analytical service, working with departments to design and implement evaluations and to integrate evaluation planning into policy proposals. It provides guidance, training, and advisory support to local departmental evaluation teams. Its position within Treasury links evaluation to the budget and policy development process, encouraging departments to include evaluation strategies when seeking funding. While the institution is relatively new, the aim is to improve evaluation frameworks and strengthen their role in budget preparation and programme design.
Ireland: The Irish Government Economic and Evaluation Service (IGEES) was established in 2012 to strengthen analytical capacity and policymaking across the Irish civil service. It operates as a cross-government professional service of economists and analysts embedded within line departments but co-ordinated centrally by the Department of Public Expenditure. The model places strong emphasis on capability building, recruiting and training analysts who work directly with policy teams while applying common appraisal and evaluation standards. IGEES analysts undertake analysis and publish research across a wide range of policy areas. Examples of this output include economic and expenditure analysis, value-for-money assessments, programme evaluations, ex ante appraisals, social and distributional impact analysis, and wider research papers Rather than acting as a central authority, the service strengthens the analytical capability inside ministries, which has helped integrate evaluation into routine policy development and identify opportunities to reprioritise or redesign public spending.
United Kingdom: The Evaluation Task Force (ETF) was established in 2021 as a joint unit of HM Treasury and the Cabinet Office designed to strengthen the role of evaluation in central government decision making. The United Kingdom’s approach relies on a small central team positioned at the core of government, enabling it to influence major policy and spending decisions. The Task Force works with departments to design and commission evaluations, particularly for large or high-priority spending programmes, and promoting common evaluation standards across government. A distinctive feature of the model is its strong integration with the Treasury’s spending review and budget processes, where evaluation results and plans increasingly inform decisions about programme continuation, redesign, or expansion. Rather than building a large evaluation workforce, the ETF has focused on improving the strategic use of evaluation in fiscal decision making.
Although the above country examples differ in how evaluation expertise is organised, how centralised the system is, and how evaluation links to budgeting and decision making, they all illustrate ways to integrate policy evaluation into policy development and budget processes in a more systematic way.
Source: Department of the Treasury, Australia, Department of Public Expenditure, Infrastructure, Public Service Reform and Digitalisation, Ireland, and HM Treasury, United Kingdom.
13.2.1. Recent spending trends in non-welfare support for households
Government support to households expanded during the COVID-19 pandemic. While the increase was particularly prominent in social spending, other budgetary transfers and tax relief were also used to support the liquidity of households (OECD, 2021[19]). After phasing out emergency measures from the pandemic, many governments introduced new support policies to mitigate adverse impacts from the sharp increase in energy prices from 2022. Countries adopted both price measures (e.g. budgetary compensation to suppliers for price caps or reduced taxation on energy products) and measures to support incomes (e.g. transfers and tax credits to consumers). This represented a significant strain on public budgets, and on general government expenditure.
The OECD Energy Support Measures Tracker, released in June 2023, estimated the fiscal cost of energy support in 2022-2023 to be substantial. Although the level of exposure to inflationary pressures and energy price spikes differed among OECD countries, the median gross fiscal cost of support measures was estimated at 0.7% of GDP in 2022 and 0.8% in 2023 (Hemmerlé et al., 2023[20]). In a few OECD countries, support reached more than 2% of GDP in each of the two years. As governments rushed to protect firms and households, untargeted measures accounted for 77% of the total estimated cost of energy support in 2022-2023 (Hemmerlé et al., 2023[20]). Untargeted measures often entail large fiscal costs and can disincentivise energy savings.
Estimated gross fiscal costs of support measures
Note: Support measures are taken in gross terms, i.e. not accounting for the effect of possible accompanying energy-related revenue-increasing measures, such as windfall profit taxes on energy companies. Gross fiscal costs are estimates for both 2022 and 2023, reflecting a combination of official estimates and assumptions on energy prices and consumption. Measures corresponding to categories “Credit and equity support” and “Other” have been excluded. When a given measure spans more than one year, its total fiscal costs are assumed to be uniformly spread across months. For measures with no officially announced end-date, an expiry date is assumed and the fraction of the gross fiscal costs that pertains to 2022-2023 has been retained.
Source: OECD (2023[20]), “Aiming better: Government support for households and firms during the energy crisis”.
Some countries extended temporary measures introduced in 2022-2023 to alleviate rising cost-of-living pressures, for example through reduced VAT rates on essential goods such as energy and food (OECD, 2025[21]). In 2024, the fiscal costs of energy support fell sharply, as energy prices started to stabilise and countries began phasing out crisis-era compensation payments (OECD, 2025[22]).
After the past years’ successive shocks, there is a risk for crisis-related support to have become part of the normal, potentially raising public expectations for further government interventions in future crises (OECD, 2025[23]). Following disruptions in global energy markets in the spring of 2026, many OECD countries have moved quickly to provide new support measures. According to recent estimates from the 2026 edition of the OECD Energy Support Tracker, 26 OECD countries had adopted at least one support measure as of April 9 (OECD, 2026[1]). While most of these are fuel-tax cuts and other price-support measures, some countries have also used income support in the form of payments to vulnerable households or particularly exposed sectors. As the current energy shock is still evolving and the duration remains uncertain, around two-thirds of announced measures by OECD countries are temporary, meaning they carry an explicit end date or are designed as one-off interventions.
As governments rushed to protect households’ purchasing power and firms’ viability in 2022-2023, untargeted measures accounted for 77% of the total estimated cost of energy support (Hemmerlé et al., 2023[20]). Untargeted measures often entail large fiscal costs and can disincentivise energy savings. On a global scale, the fiscal cost of support for fossil fuels almost doubled in 2022, and around 90% was related to the fuel consumption of households or businesses (OECD, 2025[22]).
Gross fiscal costs of energy support measures in 2022-2023
Note: Support measures are considered targeted if their main beneficiaries are not “all households” or “all firms” or “all energy users”. The figure includes both income and price measures.
Source: OECD (2023[20]), “Aiming better: Government support for households and firms during the energy crisis”.
The 2022-2023 energy crisis showed how costly broad-based energy support can become. Although the current shock to energy prices is different in nature from the 2022-2023 crisis, past experiences can provide useful policy lessons for the design of future support measures:
1. Targeting the most vulnerable households and otherwise viable, heavily exposed firms can limit fiscal costs and maximise impact of support.
2. Embedding clear sunset clauses or automatic phase-out rules in support measures can prevent support from becoming entrenched and excessively costly.
3. Designing support in ways that maintain price signals preserves incentives to save energy, which is particularly important during an energy supply crunch.
Source: Adapted from OECD (2026[1]), “Energy prices are spiking again: New relief measures, old lessons”
In addition to energy price and cost-of-living support, many OECD countries provide financial incentives for households’ investments in energy transition and efficiency. Measures include both grants and tax expenditures for renewable energy investments, home renovations, or purchases of energy-saving appliances and electric vehicles, including through the VAT system (see Chapter 14). Such measures can be costly and disproportionately benefit higher‑income households, who are more able to invest in items such as home energy upgrades. As a result, this is an area where many respondents are reporting ongoing savings measures. Some countries are citing maturing markets and technologies having gone from niche products to mainstream options as one reason for allowing the withdrawal of support.
13.2.2. Reform initiatives and savings measures
RPF Survey respondents adopted a range of savings measures in their 2025-2026 budgets to reduce other support for households, including reducing budgetary support related to energy and other non-welfare support.
Measures approved or submitted to parliament for the fiscal years of 2025 and 2026
Note: Results based on 39 Survey responses. “Reduce other non-welfare support to households (incl. tax expenditures)” combines expenditure savings and reductions in tax expenditures for households. Data is not available for France.
Source: 2026 OECD Survey on Restoring Public Finances, Question 9: Support to Businesses and Households and Question 14: Tax Expenditures.
Reducing support lowering households’ energy spending
Following a period of rapid expansion in support policies designed to lower costs of energy and promote energy-efficient and low-emission investments, respondents have sought to rationalise such support. In addition to fiscal pressures and the growing maturity of certain technologies, respondents have tried to reduce subsidisation and focus support to vulnerable households. For example:
Austria has reduced subsidies for household heating-system replacement and private-home renovations, lowering the maximum subsidy rate from 75% to 30% following a comprehensive efficiency review, which found that lower subsidy rates can deliver similar emissions reductions at lower fiscal cost. In addition, the government ended its universal “Klimabonus” in 2025, a measure originally introduced to compensate households for CO₂ pricing (see Box 13.7).
Belgium is discontinuing multiple household level support schemes in the Flemish region, including subsidies for energy-efficient household appliances, and tightening eligibility for renovation subsidies to the most vulnerable households. These changes are expected to generate EUR 141.8 million in 2026 and follows a broader subsidy reform agenda aimed at focusing support where it is most effective.
Canada has ended its Greener Home Grants programme in response to a large number of applicants and budgetary constraints. This programme provided homeowners with grants for energy‑efficiency upgrades, such as heat pumps, insulation, and better windows.
As of 2025, Austria has removed the Klimabonus introduced in 2022 to compensate households for costs incurred from CO₂ taxation. This refund was introduced at the same time as Austria’s new CO₂ pricing system, and redistributed the funds collected through direct payment to all inhabitants in the years 2022-2024. The refund was paid out automatically once a year. Individuals who effectively reduced their CO₂ emissions would keep a larger share of the Klimabonus, and so would individuals facing challenges in transitioning to lower CO₂ alternatives, for example people living in regions with less access to public transportation. In 2025, Austria decided to keep the CO₂ tax but eliminate the rebate payments. The removal of the bonus means households bear the full cost of the tax, increasing costs associated with emitting CO₂, while also producing fiscal savings of around EUR 2 billion per year.
Source: Federal Ministry of Finance, Austria
Replacing broad energy price support with more targeted measures can help reduce fiscal costs and limit unintended interference with other policy goals, while preserving energy price signals and focusing support on those most in need. The RPF Survey includes examples of respondents that scaled back their energy price compensations after the 2022-2023 crisis, which can offer insights also in the current context of renewed support:
Czechia gradually moved from broad energy price support to means-tested social benefits. In 2022, the government established a universal subsidy scheme to compensate households and companies for high energy costs. The contribution was paid automatically over the electricity bills, regardless of consumption. In 2023, the government tightened its support policy, setting out rules for capping energy prices, and from 2024, there were no more direct interventions from the state budget. The country’s Housing Allowance, which continued to be broadly used in 2023-2024 due to high inflation, was discontinued in 2025 and integrated into the new “Super allowance” (see Chapter 3, Box 3.4), where stricter means-testing will be applied.
Slovenia ended its compensation to reduce energy costs for households in the first half of 2025, and shifted towards investment-based support, passing several measures aimed at improving the energy efficiency of households and increase resilience to potential energy crises in the future. Measures include co-financing of heat pumps, insulation, and zero-emission buildings. A specific programme targets recipients of social assistance. The aim of these measures was to reduce the risk of future state aid interventions.
Türkiye sought to contain fiscal costs by reducing the number of households eligible for subsidised electricity tariffs, excluding high-consumption households. This was done by lowering the consumption-based “Last Resort Supply Tariff” threshold for households from 5 000 kWh to 4 000 kWh as of 1 January 2026.
Drawing on lessons learned from OECD countries’ response to the 2022-2023 energy crisis, where data shows that energy support for households was largely untargeted, recent OECD analysis suggests that wider use of means-testing and income profiling could help reduce fiscal costs going forward.
One alternative to universal energy price support would be to limit payments to people already in social assistance programmes. Within countries, the burden of high energy prices falls disproportionately on poorer households because energy accounts for a larger share of their consumption baskets. Exploiting existing social transfers can be a quick and administratively simple way to target the most vulnerable households. If delinked from consumption, such approaches also have the advantage of preserving price signals. A major issue during the 2022-2023 energy crisis was that price support disincentivised energy savings. Nonetheless, a comprehensive OECD study, analysing countries’ support measures adopted in response to the 2022-2023 energy crisis, finds that OECD countries made limited use of such top-up payments. Notable exceptions included a monthly transfer to vulnerable households in Costa Rica and a one-off top-up to the means-tested energy voucher in France. The study also highlights that after the initial shock has passed, indexation of social benefits or minimum wages to inflation can be another strategy to target support and facilitate the withdrawal of universal support measures.
A more comprehensive approach to targeting involves differentiating compensation according to households’ degree of vulnerability. As opposed to a uniform top-up payment or indexation, this approach seeks to fully account for the specific vulnerabilities to the shock in question. In energy crises, this could include the inability to renovate an energy-inefficient home, limited access to cheaper forms of energy or higher-than-average energy needs due to age, illness or geographical location. A few countries have already explored categorising households to distinct categories by assessing their vulnerability to different parameters or using econometric tools to estimate energy needs (for example, United Kingdom’s “Low Income Low Energy Efficiency” indicator). Ireland provides another example of targeted support based on clearly defined vulnerability criteria. The Fuel Allowance is a payment to help households cover heating costs during the winter months and is granted based on age, income, and household composition. Eligibility requires that recipients be either aged 66 or over or receive a qualifying social welfare payment, live alone or only with other qualified people, and pass a means-test.
Several respondents are also reducing tax relief used to lower the upfront costs of energy efficient assets and households’ spending on fuel and heating, as shown by the examples below.
As solar technology is becoming increasingly affordable, some countries are reducing support to promote Photo Voltaic (PV) electricity generation. France has introduced stricter eligibility criteria for reduced taxation on PV equipment and its installation. The Netherlands is ending its tax relief for PV net-metering, a major saving that will affect both residential and commercial PV system owners as they lose the ability to offset electricity consumption with feed-in tariffs. The reform reflects concerns about higher costs for non-solar households, high fiscal costs and increasing pressure on the electricity grid. The savings are projected to be EUR 574 million from 2027, increasing to 665 million from 2032 and onwards.
Some saving measures relate to building upgrades aimed at making dwellings more energy efficient. For example, Italy is reducing its tax expenditures for building renovations, progressively lowering the deduction rate for second homes from 50% in 2024 to 30% by 2026.
A number of respondents, including Austria, Costa Rica, Finland and Norway, have reported measures to reduce tax relief for electric or low-emission vehicles. This increases purchase or ownership costs, and reduces the implicit subsidy previously granted to households. The ongoing adjustments reflect both the growing maturity of markets for electric vehicles and a need for countries to protect their vehicle tax base. As electric and hybrid vehicles transition from niche products to mainstream options, governments are phasing out incentives that are no longer considered necessary for adoption. For example, Norway is gradually phasing out electric vehicle subsidies (to be fully removed by 2027), reflecting the government’s assessment that the initial policy goal has effectively been achieved.
Other measures are more indirect but affect households through higher fuel costs. For example, Portugal is revoking its tax exemption on advanced biofuels and renewable gases, which will lead to higher end-user prices for fuel consumers. Lithuania has abolished its VAT exemption for heating fuels supplied to household energy consumers.
Reducing other non-welfare support to households
Savings on other non-welfare support to households include various reductions or abolitions of support for specific household expenses. For example, as part of the RPF Survey submissions:
Austria is reducing subsidies for electromobility.
Belgium is reducing tax expenditures for legal aid and adoption,
Finland is scaling back tax credits for domestic services and home office deductions,
Korea is reducing tax expenditure for comprehensive savings accounts,
Sweden is reducing its tax expenditure for commuting costs.
13.3.1. Recent trends in development assistance spending
Official Development Assistance (ODA) is government aid that promotes and specifically targets the economic development and welfare of developing countries. ODA expenditures are measured and monitored under the OECD Development Assistance Committee (DAC) (OECD, 2025[25]). While comparatively small, ODA is subject to scrutiny, given that funding aims to create impact in other countries using taxpayers’ resources, including institutional mechanisms and tools to ensure relevance, effectiveness and efficiency.
ODA from DAC countries1 increased in most cases between 2019 and 2023, both as a percentage of GNI and GDP2 (Figure 13.7). This was partly a result of increased spending on support measures during the COVID-19 pandemic and aid in response to Russia’s war of aggression against Ukraine (OECD, 2025[26]).
Note: Data for Panel A is ODA in grant equivalent, current prices, percentage of gross national income. Data for Panel B is ODA in grant equivalent, current prices, national currency, and GDP in nominal value, national currency, market prices.
Source: Flows by donor (ODA+OOF+Private) [DAC1] as of 9 April 2026, OECD Data Explorer and Economic Outlook 118 as of 24 February 2026, OECD Data Explorer.
Recent consolidation efforts in several DAC member countries follow a period of sustained expansion in ODA spending (Figure 13.8). From a peak of USD 229 billion in 2023 (constant 2024 prices) across all DAC members, this consolidation produced decline of around 28% in ODA on a grant equivalent basis between 2023 and 2025. 2025 marks the second consecutive year of decline (OECD, 2026[27]). According to OECD DAC projections, net ODA is expected to continue to decline in 2026 (see Box 13.9).
Over a four-year period (2019 to 2023), the composition of ODA changed. A growing share of ODA, in net terms, was directed towards in-donor refugee costs3 and support related to Ukraine (Figure 13.8). The declines over the past two years suggest that reductions have affected multiple ODA components, including a notable drop in bilateral development projects, programmes and technical co-operation.
Note: Values are constant 2024 prices. The category, “Bilateral development projects, programmes, and technical co-operation” is the remainder of total net ODA after subtracting all other components. The net debt relief grant are not displayed on the chart as the values (max USD 0.7 billion), are to small to be represented.
Source: Flows by donor (ODA+OOF+Private) [DAC1] as of 9 April 2026, OECD Data Explorer; Aid (ODA) disbursements to countries and regions [DAC2A] as of 9 April 2026, OECD Data Explorer.
ODA is projected to decline further from pre-2020 levels, based on announcements by several major donor countries, which together account for nearly two-thirds of DAC ODA. Overall, net ODA expenditure by DAC countries is projected to decline by around 5.8% in 2026, not yet accounting for additional strain from the current crisis in the Middle East (OECD, 2026[27]).
Note: Data is constant 2024. Amounts for 2025 are preliminary. Amounts for 2026 are OECD projections.
Source: Flows by donor (ODA+OOF+Private) [DAC1] as of 9 April 2026, OECD Data Explorer. See original graph at: https://www.oecd.org/en/data/insights/data-explainers/2026/04/a-historic-decline-in-foreign-aid-preliminary-2025-oda-data.html.
13.3.2. Reform initiatives and savings measures in development assistance
In line with this broader environment, several RPF Survey respondents indicated that they plan to streamline or reduce ODA expenditures during 2025-2026. Measures of this sort lower the resource envelope available for development co-operation by implementing budget cuts.
Measures approved or submitted to parliament for the fiscal years of 2025 and 2026
Note: Results based on 39 RPF Survey responses.
Source: 2026 OECD Survey on Restoring Public Finances, Question 13.1: Other Expenditure - Development Assistance.
For example, France’s ODA budget is set to decline in 2026 by approximately EUR 800 million compared to 2025 (a 18% decrease) (French Ministry of Public Action and Accounts, 2026[28]). The Slovak Republic also indicated savings through reductions in ODA as part of fiscal consolidation efforts (Ministry of Foreign Affairs and European Affairs, 2026[29]) and the United Kingdom announced in 2025 that it would reduce ODA from 0.5% of GNI to the equivalent of 0.3% of GNI from 2027 to fund an increase in defence spending.
In a context of expenditure pressures, some governments have intensified efforts to review ODA objectives, sharpen strategy and alter the sectoral or geographical allocations within ODA budgets. Respondents are reducing or phasing out ODA expenditures on selected sectors, instruments or partner countries, while protecting or expanding funding for priority areas. Shifts in spending also reflect growing humanitarian needs, increases in refugee populations and changing policy priorities. The following examples are illustrated in the RPF Survey:
Canada is making reductions concentrated on global health programming, following increased pandemic-era support (Government of Canada, 2025[30]).
Sweden reprioritised its 2025 ODA budget to reallocate around SEK 1.7 billion to Ukraine and humanitarian initiatives (around EUR 153 million), relative to sectoral assistance in other regions (Government of Sweden, 2025[31]).
Other RPF Survey respondents are seeking to use ODA funds to mobilise and leverage private-sector finance to support development objectives. ODA can be used to improve the enabling conditions for the private sector to invest, for instance through providing of guarantees, blended finance, and partnerships with financial institutions. This approach aims to increase the impact of limited ODA funds by “crowding-in” additional private sector funds, for example in areas such as infrastructure and small business finance. Two examples from the RPF Survey include:
Japan revised its Japan International Co-operation Agency (JICA) law to better address global development challenges by encouraging private sector investment. The updated law empowers the JICA to support bond issuance by companies in developing countries and to assume part of the project risks in order to promote small business lending. It also introduced outcome‑based overseas investment and loan schemes (Ministry of Foreign Affairs of Japan, 2025[32]).
Thailand’s Neighbouring Countries Economic Development Co-operation Agency (NEDA), which mobilises loans from domestic financial institutions to on-lend neighbouring countries, requests government budget support only when necessary to subsidise interest rate differentials and minimise fiscal burden. Looking ahead, NEDA plans to further expand its development assistance through public-private partnerships (PPPs), as an additional step toward reducing reliance on direct government funding4. This approach is reflected in a recent bridge construction project in Lao PDR, where NEDA combines Thai budget support with domestic borrowing and applies an Output and Performance-Based Contract (OPBC) for the maintenance phase with quarterly payments subject to agreed level of service standards.
Other RPF Survey respondents mention measures to strengthen evaluation of the effectiveness of expenditures and help prioritisation. For example, as part of its RPF Survey submission, and while not confirming reductions in ODA, Spain noted that its Independent Fiscal Institution (AIReF) had carried out a spending review on Official Development Assistance (AIReF, 2026[33]). In particular, an Office for the Evaluation of Spanish Development Co-operation has been established to define and implement the national evaluation plan, manage strategic evaluations, and co-ordinate actors across the evaluation system, while proposals to introduce annual performance reports by region and sector to strengthen results-based management have been put forward.
References
[33] AIReF (2026), Spending Review - Official Development Assistance, https://www.airef.es/en/studies-document-center/official-development-assistance/.
[16] Denmark’s Ministry of Taxation (2024), Aftale om Et stærkere erhvervsliv, https://skm.dk/media/z2vmd0ml/aftaletekst-aftale-om-et-staerkere-erhvervsliv.pdf.
[6] Espinoza, R. et al. (2022), “The State as Financier of Last Resort”, Staff Discussion Notes, Vol. 2022/003, p. 1, https://doi.org/10.5089/9798400220692.006.
[28] French Ministry of Public Action and Accounts (2026), Budget de l’État - Dépenses par mission - Aide publique au développement, https://www.budget.gouv.fr/budget-etat/mission?annee=247&loi_finances=47&type_budget=all&mission=87703&type_donnee_budget=cp&op=Valider.
[15] Government of Canada (2025), “Federal Budget 2025”, Comprehensive expenditure review, https://budget.canada.ca/2025/report-rapport/anx3-en.html#a16.
[30] Government of Canada (2025), Global Affairs Canada’s 2025-26 Departmental plan, https://international.canada.ca/en/global-affairs/corporate/reports/departmental-plan/global-affairs-2025-2026-departmental-plan.
[31] Government of Sweden (2025), Government frees up nearly SEK 1.7 billion for increased support to Ukraine and humanitarian initiatives, https://www.government.se/press-releases/2025/07/government-frees-up-nearly-sek-1.7-billion-for-increased-support-to-ukraine-and-humanitarian-initiatives.
[20] Hemmerlé, Y. et al. (2023), “Aiming better: Government support for households and firms during the energy crisis”, OECD Economic Policy Papers, No. 32, OECD Publishing, Paris, https://doi.org/10.1787/839e3ae1-en.
[24] Irish Department of Social Protection (2026), Fuel Allowance, https://www.gov.ie/en/department-of-social-protection/services/fuel-allowance/.
[11] Juhász, R., N. Lane and D. Rodrik (2023), “The New Economics of Industrial Policy”, SSRN Electronic Journal, https://doi.org/10.2139/ssrn.4542861.
[3] Koelle, M., K. Kwak and J. Brunnengräber (2026), “Macroeconomic policies and OECD policy advice during the COVID-19 pandemic: A retrospective assessment”, OECD Economics Department Working Papers, No. 1857, OECD Publishing, Paris, https://doi.org/10.1787/98c6a613-en.
[10] Millot, V. and Ł. Rawdanowicz (2024), “The return of industrial policies: Policy considerations in the current context”, OECD Economic Policy Papers, No. 34, OECD Publishing, Paris, https://doi.org/10.1787/051ce36d-en.
[29] Ministry of Foreign Affairs and European Affairs (2026), Government approves Slovakia’s development cooperation focus for 2026, https://www.mzv.sk/en/pressreleasedetail?p_p_id=sk_mzv_portal_pressrelease_detail_portlet_PressReleaseDetailPortlet&p_p_lifecycle=0&groupId=10182&articleId=57406756.
[32] Ministry of Foreign Affairs of Japan (2025), Bill to Amend Part of the JICA Act (Independent Administrative Institution Japan International Cooperation Agency Act), https://www.mofa.go.jp/mofaj/ic/dapc/pagew_000001_00001.html.
[5] Moretti, D., T. Braendle and A. Leroy (2021), “Balance sheet-based policies in COVID‑19 fiscal packages: How to improve transparency and risk analysis?”, OECD Journal on Budgeting, Vol. 2021/2, https://doi.org/10.1787/6b946136-en.
[27] OECD (2026), A historic decline in foreign aid: Preliminary 2025 ODA data, https://www.oecd.org/en/data/insights/data-explainers/2026/04/a-historic-decline-in-foreign-aid-preliminary-2025-oda-data.html.
[1] OECD (2026), “Energy prices are spiking again: New relief measures, old lessons”, OECD Publishing, Paris, https://doi.org/10.1787/a68e5c37-en.
[8] OECD (2025), Fast-tracking Net Zero by Building Climate and Economic Resilience: A Summary for Policymakers, OECD Publishing, Paris, https://doi.org/10.1787/f2c22c96-en.
[12] OECD (2025), “How governments back the largest manufacturing firms: Insights from the OECD MAGIC Database”, OECD Trade Policy Papers, No. 289, OECD Publishing, Paris, https://doi.org/10.1787/d93ed7db-en.
[22] OECD (2025), OECD Inventory of Support Measures for Fossil Fuels 2025: Policy Trends, OECD Publishing, Paris, https://doi.org/10.1787/6fff0cb7-en.
[25] OECD (2025), Official development assistance (ODA), https://www.oecd.org/en/topics/official-development-assistance-oda.html.
[26] OECD (2025), Preliminary official development assistance levels in 2024: Detailed Summary Note, OECD, Paris, https://one.oecd.org/document/DCD(2025)6/en/pdf.
[7] OECD (2025), Quality Budget Institutions: Developments in OECD countries, OECD Publishing, Paris, https://doi.org/10.1787/8e811202-en.
[13] OECD (2025), “Quantifying industrial strategies 2019-2022: Trends and priorities across 11 OECD countries”, OECD Science, Technology and Industry Policy Papers, No. 179, OECD Publishing, Paris, https://doi.org/10.1787/91e20a26-en.
[23] OECD (2025), “Sustainability of public finances in OECD countries: From fiscal consolidation to state redesign?”, OECD Papers on Budgeting, No. 2025/12, OECD Publishing, Paris, https://doi.org/10.1787/df89a38a-en.
[21] OECD (2025), Tax Policy Reforms 2025: OECD and Selected Partner Economies, OECD Publishing, Paris, https://doi.org/10.1787/de648d27-en.
[9] OECD (2025), “The Market Implications of Industrial Subsidies”, OECD Trade Policy Papers, No. 296, OECD Publishing, Paris, https://doi.org/10.1787/e40b793f-en.
[2] OECD (2025), The state of play of industrial subsidies as of 2023, OECD Publishing, Paris, https://doi.org/10.1787/753cd39f-en.
[4] OECD (2023), “Rise of non-standard policy Instruments: can budgeting catch up”, Journal of Budgeting, Delphine Moretti, https://doi.org/10.1787/ec1d2b4e-en.
[18] OECD (2022), “Subsidies, Competition and Trade”, OECD Roundtables on Competition Policy Papers, No. 289, OECD Publishing, Paris, https://doi.org/10.1787/57a34bc6-en.
[19] OECD (2021), Tax Policy Reforms 2021: Special Edition on Tax Policy during the COVID-19 Pandemic, OECD Publishing, Paris, https://doi.org/10.1787/427d2616-en.
[14] Office of the Auditor General of Canada (2024), “Strategic Innovation Fund’s Net Zero Accelerator Initiative - Innovation, Science and Economic Development Canada”, Reports of the Commissioner of the Environment and Sustainable Development to the Parliament of Canada, https://publications.gc.ca/collections/collection_2024/bvg-oag/FA1-26-2024-1-4-eng.pdf.
[17] Olsberg SPI (2024), “Economic Impact of the Iceland Film Production Incentive”, Final Report to the Ministry of Culture and Business Affairs, https://www.government.is/library/01-Ministries/Economic%20Impact%20Study%20of%20Iceland's%20Production%20Incentive%20-%20Final%20Report%202024-04-09%20v3%20CLEAN.pdf.
← 1. Not all OECD Member countries are members of the DAC.
← 2. ODA is typically reported as a share of GNI, which is the standard measure used by the OECD Development Assistance Committee (see OECD’s development finance statistics here). GNI captures income accruing to residents and is therefore the most appropriate base for assessing countries’ effort to allocate resources abroad. However, ODA is also presented as a share of GDP in this report to ensure consistency with other expenditure indicators analysed in other chapters, which are expressed relative to GDP.
← 3. Refers to the portion of ODA that donor countries report as aid spending for refugees within their own borders.
← 4. While not a member of the DAC, Thailand voluntarily reports statistics on its ODA to the OECD.