OECD Economic Outlook, Volume 2026 Issue 1

2. From energy shocks to stronger resilience

Introduction

The disruption in energy supply and rise in energy prices linked to the evolving conflict in the Middle East have brought energy affordability and supply security back to the centre of the policy debate. Higher energy prices are once again fuelling inflation, raising pressure on household budgets and firms’ costs - all at a time when fiscal space is limited. Governments are therefore facing a familiar policy question: whether to cushion the short-term economic and social effects of the energy shock and how to do so without adding to already strained public budgets or weakening incentives to save energy.

The government policy responses to the crisis show how difficult this balance can be. Many governments have acted rapidly, often rolling out measures such as fuel tax cuts, subsidies and other price interventions. Such broad-based measures can provide timely relief during emergencies as they can be introduced quickly and do not require data and administrative systems to identify vulnerable households and firms in real time. These features also enhance their political acceptability during crises, as they avoid politically difficult eligibility decisions and help reduce the risk of coverage gaps. However, they often tend to be costly, hard to unwind, and weaken incentives to save energy, as the 2022-23 energy crisis showed.

The lessons from the 2022-23 crisis remain highly relevant for today’s energy crisis. Support measures are more effective and less costly when they are targeted at vulnerable households and viable, exposed firms, are temporary by design, and preserve incentives to save energy and reduce reliance on fossil fuels. At the same time, repeated crises have underscored that strengthening resilience to energy shocks requires more than short-term support measures.

This chapter draws on the OECD’s Energy Support Measures Tracker to document government policy responses to the 2026 energy crisis and identifies policy priorities. Looking ahead, policy makers have at their disposal a range of demand-, supply-side, and governance and regulatory tools to improve crisis responses, strengthen future crisis preparedness, and enhance longer‑term energy system resilience (Table 2.1):

• Crisis response focuses on managing the immediate shock. This includes coordination of the use of strategic energy stocks, activating emergency-demand restraint measures, and providing support to vulnerable households and exposed but viable firms while preserving energy-saving incentives. Clear and transparent rules for reviewing support measures with explicit sunset clauses or phaseout rules can help contain fiscal costs and support the timely withdrawal of emergency support measures.

• Crisis preparedness focuses on improving the capacity of the energy system and the public administration to anticipate, coordinate and execute responses to shocks. This involves ensuring that energy stockholding systems are adequately designed, governed and ready to be deployed when disruptions occur. It also hinges on improving data availability and continuity to identify vulnerabilities, inform decision making, as well as digitalised benefit-delivery systems to provide timely and targeted support, and advanced analytics to enhance operational triage, anomaly and fraud detection in addition to short-term forecasting of caseloads and budget needs. Finally, it calls for expanding flexible electricity demand and ensuring storage capacity to ease system stress.

• Structural resilience requires policy reforms to durably lower vulnerability to fossil-fuel supply disruptions and other energy system shocks. This hinges on pursuing two complementary and mutually reinforcing objectives: higher energy efficiency and broader diversification across energy sources, technologies and locations. Electrification of end uses can underpin both objectives by enabling the integration of a wider range of primary energy sources into final demand, provided that grid upgrades and regulatory reforms support improvements in electricity systems’ operation, monitoring, control, and flexibility.

Table ‎2.1. Policy options to improve responses and strengthen resilience to energy shocks

Policy function	Demand-side tools	Supply-side tools	Governance and regulatory tools
Crisis response	Targeted support; emergency demand restraint; demand response activation; public communication on support measures and energy saving	Strategic stock releases (internationally coordinated); mobile generation; rapid repair mobilisation; activation of cross-border emergency assistance arrangements	Crisis units; timely emergency communications; rapid data sharing (e.g. on inventories); review and phase-out rules for support measures; international coordination for strategic stock releases; high-frequency budget monitoring and reporting (spending, coverage, take‑up); transparent fiscal accounting of contingent liabilities (e.g. guarantees to utilities)
Crisis preparedness	Pre-defined demand-restraint protocols.	Building or improving energy reserve systems and critical equipment stockpiles; stress tests of critical energy infrastructure	Stress tests of support delivery systems; linking information and databases to better identify vulnerable households and firms; data continuity rules; pre-defined protocols for activating, recalibrating and withdrawing support without delay; ex ante fiscal governance frameworks, including appropriations, contingency reserves, and rapid reallocation mechanisms; audit and control protocols to limit leakage and fraud; mechanisms for rapid ex post evaluations and reconciliation of crisis spending
Structural resilience	Energy efficiency improvements; electrification of end‑uses; incentives and demand response mechanisms to increase demand flexibility (e.g. dynamic electricity tariffs)	Diversified energy supply mix; cross-border interconnections; decentralisation of energy assets and resources; grid reinforcements; backup capacity; electricity storage	Resilience embedded in planning; permitting and grid-connection reform; real-time power system monitoring, data exchange and control; pre-arranged and operational cross-border emergency assistance arrangements; integration of “insurance value” of resilience (i.e. avoided losses from outages and supply disruptions) and risk metrics into energy projects’ appraisal and budget prioritization; medium‑term budget frameworks for resilience investment (multi‑year envelopes for grids and flexibility services); systematic tracking of public spending and tax expenditures relating to energy system resilience

Energy prices have surged but exposure differs across countries, firms and households

The conflict in the Middle East has roiled energy markets, as discussed in Chapter 1 of this Outlook. Since late February, the near halt in shipments through the Strait of Hormuz and damage to key energy infrastructure in the region have drastically disrupted the flow of crude oil, oil products and liquefied natural gas. With few alternative transport routes for products originating in the Gulf of Persia, and limited scope to ramp up alternative supply sources in the short term, the disruption has quickly fed into prices (Figure 2.1). The effects are spreading beyond energy markets, with business surveys beginning to point to rising input price inflation (see Chapter 1).

Strategic oil reserves can provide an important buffer against supply disruptions, but current inventory levels provide limited protection against a prolonged supply disruption. As detailed in Chapter 1, the closure of the Strait of Hormuz since late February 2026 has removed an estimated 12.8 million barrels of oil per day (mb/d) from global oil supply, which is equivalent to around 12% of global daily consumption (IEA, 2026d). Gulf production alone has fallen by 45% from pre-war levels. While exports have been partially rerouted through alternative ports and pipelines, and Atlantic Basin producers have stepped up supply, these measures fall short of compensating for the scale of Gulf losses. IEA member countries have released 164 million barrels of oil from emergency stocks out of a total commitment of 400 million barrels. The IEA projects a cumulative deficit of 900 million barrels through September 2026 – assuming conflicts cease by early June – and that rebuilding depleted stocks might take up to three years on top of underlying demand growth (IEA, 2026d).

Figure ‎2.1. Energy prices have spiked

Note: Shows the evolution of U.S. Gulf Coast Ultra-Low Sulfur No 2 Diesel Spot Price for Diesel; TTF Natural Gas Price for Europe and LNG for Asia. Latest data refers to 26 May 2026.

Source: LSEG; and OECD calculations.

IEA member countries are required to ensure oil stock levels equivalent to no less than 90 days of net imports. Some countries, such as Japan and Korea, hold well in excess of those levels. Net exporters, also tend to hold strategic reserves, even absent an IEA commitment. The US Strategic Petroleum Reserve (SPR) adds around 20 days of additional coverage on top of commercial stocks. However, large headline stock levels do not necessarily guarantee immediate market relief, as refinery capacity, tanker availability and product bottlenecks can constrain how quickly strategic releases translate into physical supply. The picture is considerably less reassuring for many emerging market economies (Figure 2.2). In particular, many Southeast Asian economies hold stocks covering approximately one month of consumption. These economies combine high import dependence with tighter fiscal positions and greater currency vulnerability, amplifying the speed and severity with which supply shocks transmit to households and public finances.

Overall, energy price increases are putting pressure on household energy bills and business costs, with adverse consequences for growth and welfare for most countries. Exposure to the current shock differs markedly across countries and sectors. At the country level, vulnerability depends not only on direct import dependence on the Middle East, but also on indirect exposure through domestic and international supply chains, and the availability of inventory buffers (see Box 1.2 in Chapter 1).

Firms in energy-intensive sectors, such as transport, petrochemicals and metals, are likely to be most affected. However, even sectors with limited direct energy use — including machinery, construction and services — remain exposed through production networks that rely on energy from the Middle East. More broadly, vulnerability also depends on firms’ ability to substitute away from affected inputs, pass higher costs on to customers, and absorb temporary liquidity pressures. Limited diversification, tight liquidity and constrained financing amplify firms’ vulnerability, especially for smaller businesses.

Household vulnerability varies widely, both across and within countries. Energy costs account for a larger share of consumer spending in some countries than in others, ranging from around 17% in Latvia to around 5% in the United Kingdom (Figure 2.3 Panel A). Cross-country differences also depend on how strongly international energy-price shocks pass through to retail prices, for example because of differences in energy taxation and subsidies, countries’ energy mixes and the frequency with which retail contracts are adjusted. Within advanced countries, the burden of high energy prices falls disproportionately on poorer households, because energy accounts for a larger share of their consumption baskets (Figure 2.3 Panel B). Income is therefore a key determinant of household vulnerability, but it is not the only one. Housing energy efficiency, the energy source used, households’ energy needs and geographic location, including whether they live in rural areas, also shape the extent to which they are affected by higher energy prices (Hemmerlé et al., 2023).

Figure ‎2.2. Many large oil importers have significant short‑term buffers, but the level of coverage varies

Days of oil stock coverage based on net imports (IEA) and consumption (non-IEA), latest data available

Note: Oil stocks include all petroleum (primary products and refined products), excluding naphtha and international marine bunkers. Refined products are converted to crude‑oil equivalent. Days of cover are calculated relative to net imports adjusted for stock changes. Further details on methodology are available in the IEA oil stock data documentation. The dashed line denotes the IEA 90‑day emergency stockholding requirement, which applies only to IEA member countries. IEA data refers to December 2025. For non‑IEA economies (green bars), fuel‑reserve coverage refers to domestic consumption based on official government statements and briefings published in 2026 by national authorities. The figures for non‑IEA countries are indicative estimates and may vary over time and by fuel product.

Source: IEA; Bangladesh MPEMR/BSS; Government of India Press Information Bureau; Indonesia Cabinet Secretariat; Myanmar MOI; Pakistan parliamentary statements; Philippines DOE/PNA; Thailand PRD; Vietnam MOIT; and OECD calculations.

Statlink

Figure ‎2.3. The share of energy in household consumption differs across countries

Note: Latest available information. Data in Panel A are based on weights used for national consumer price indices (harmonised index of consumer prices for countries in the euro area) and generally include fuels for transport equipment as well as electricity, natural gas and other fuels.

Source: OECD Consumer Price database; CEIC; Eurostat; Instituto Brasileiro de Geografia e Estatística; India, Ministry of Statistics and Program Implementation; United Kingdom, ONS; Reserve Bank of Australia; Statistics Canada; Statistics of Japan; Statistics Korea; Turkish Statistical Institute; US Bureau of Labor Statistics; and OECD calculations.

Statlink

Many countries have acted quickly to provide relief

Many governments across the OECD have moved quickly to shield households and firms from rising fuel costs (see Table 2.2). As documented by the OECD Energy Support Measures Tracker, as of 18 May 2026, 31 OECD countries and 15 key non-OECD countries (out of the 55 countries plus the European Union covered by the Tracker) have announced measures. Energy price relief accounts for the largest share of recorded measures, primarily delivered through tax reductions and regulated price caps (Figure 2.4). Regulatory measures and income support via subsidies tied directly to energy bills or broader fiscal transfers have also been used, though to a lesser extent. Among the measures announced by OECD countries with a direct fiscal cost, around 78% are explicitly temporary, i.e. they have an explicit end date or have been designed as one-off interventions. Around 44% are targeted, with most measures focusing on industrial sectors most exposed to fuel price increases such as agriculture and road freight transport. Around 37% of measures are both temporary and targeted. This is somewhat higher than the share recorded during the comparable period of the 2022–23 energy crisis, when around 30% of measures announced within the first three months of the crisis (March–May 2022) were both temporary and targeted.

Figure ‎2.4. Classification of energy support measures

Count of measures

Note: The chart classifies support measures along two dimensions. First, measures are grouped by type of support: energy price relief, income support, credit and equity support, or other measures. Second, measures are classified by implementation mechanism, such as tax measures, regulated or capped prices, budgetary transfers, regulatory actions, or financial support instruments. The non-OECD economies included are Argentina, Brazil, Bulgaria, China, Croatia, India, Indonesia, Malaysia, Morocco, the Philippines, Romania, South Africa, Thailand, Ukraine, and Viet Nam. Data as of 18 May 2026.

Source: OECD Energy Support Measures Tracker 2026.

Statlink

Fuel tax reductions – the most widely used instrument so far – have been used in 21 OECD countries and 10 of the non-OECD countries. These measures lower the tax component of retail prices directly, thereby reducing the price faced by consumers and can be implemented quickly through existing systems. The scale of relief, however, varies considerably across countries. For instance, Latvia reduced diesel excise duties by around 15% (EUR 0.07 per litre), while Ireland temporarily lowered them by EUR 0.22 per litre, broadly comparable to the excise duty cuts implemented in Italy. Some countries achieved larger reductions through combined measures. Spain cut excise burdens sharply while simultaneously reducing VAT on fuels from 21% to 10%, resulting in pump price reductions of around EUR 0.30 to 0.40 per litre. Poland and Slovenia reduced fuel excise duties to the minimum levels permitted under EU law, with Poland complementing these cuts with a temporary VAT reduction and Slovenia with the suspension of environmental levies.

Direct price interventions have been used more selectively than fuel tax measures, with 13 OECD and 9 non-OECD countries having rolled them out. Of these, 8 OECD and 5 non-OECD countries have implemented both direct price intervention and tax cuts. These include retail price caps, ceilings on wholesale prices passed through to retail, and per-litre subsidies designed to limit the extent to which price increases reach final users. Some countries have opted for direct and transparent reductions in retail fuel prices. Greece has introduced a per‑litre subsidy for diesel applied directly at the pump, equivalent to about EUR 0.20 per litre, and Hungary imposed fixed maximum pump prices for petrol and diesel specifically for vehicles with domestic license plates. In contrast, Japan and Korea have intervened upstream in the pricing chain. Japan has subsidised fuel wholesalers to keep the nationwide average retail price of gasoline near JPY 170 per litre, with similar support extended to diesel, heavy oil and kerosene, while Korea has imposed a ceiling on refiners' wholesale prices for gasoline and diesel.

Direct income support has also been deployed, though its scope varies considerably across countries. So far, direct income support has been implemented in 14 OECD and 6 non-OECD countries. Such measures can take the form of energy‑related income support, which links transfers to energy use or bills thus lowering the average cost of energy consumption. Sweden has introduced nationwide compensation for electricity and gas bills as part of a SEK 4.7 billion (USD 502 million) relief package. Payments will be based on actual household energy consumption in January and February 2026 and delivered automatically through the social insurance system. In other cases, income support has been provided without being explicitly tied to energy consumption, offering households greater disposable income to absorb higher living costs. Greece has introduced a targeted fertiliser‑cost subsidy for farmers that refunds 15% of fertiliser invoice values for a limited period, alongside a digital fuel wallet for vulnerable households. Some measures have been designed to reflect both fuel type and consumer vulnerability. The United Kingdom targeted aid for off-grid and heating-oil-dependent households, while Portugal introduced a EUR 15 reimbursement per bottled-gas cylinder, distributed through local parish councils.

Beyond price relief and income support, governments have also introduced regulatory measures. Australia and New Zealand temporarily relaxed domestic fuel quality specifications to broaden import options, with Australia allowing higher sulphur levels in petrol and New Zealand aligning its standards with Australian specifications. The United Kingdom strengthened regulators' ability to challenge price increases through stronger monitoring and mandatory data disclosure. Portugal and Spain introduced consumer protections, prohibiting supplier disconnections during billing disputes and requiring flexible contract conditions respectively.

Some countries introduced fuel rationing or purchase limits. These include India, which directed all domestic LPG supply to household use, Indonesia, which capped subsidised fuel purchases at 50 litres per vehicle per day, and the Slovak Republic, which limited the value of subsidised fuel purchases to EUR 400 per transaction. Several governments also sought to reduce consumption directly. Korea launched a nationwide energy-saving campaign and mandated odd-even driving restrictions for public-sector employees, and Indonesia required civil servants to work from home and reduce international travel. Chile froze regulated public transport fares through end-2026, Lithuania halved domestic rail fares, and several Australian states offered free public transport for up to three months.

Export restrictions have been less common among advanced countries but feature extensively in some emerging markets. China and Korea used administrative controls, introducing export-licensing requirements on refined fuels and caps on exports of refined petroleum products and naphtha respectively. India and Argentina applied fiscal disincentives, with India imposing export duties on diesel and aviation turbine fuel and Argentina raising crude oil export duties to 8% when Brent exceeds USD 80 per barrel. Viet Nam and Romania required producers to redirect uncommitted crude to domestic refineries or seek prior authorisation before exporting fuels. Hungary and the Slovak Republic temporarily imposed outright export bans on crude oil or key refined products to preserve domestic supply.

The announced fiscal costs of measures vary across countries. Korea announced a supplementary budget of approximately USD 17 billion covering direct energy cost relief, consumer vouchers for lower-income households, fuel subsidies for farmers and fishermen, and grants to local governments. Germany's two-month reduction of the energy tax on gasoline and diesel will provide around USD 1.8 billion in relief. Spain announced a USD 5.7 billion package spanning tax cuts, income transfers and sector-specific subsidies. Canada suspended its federal fuel excise tax at an estimated cost of USD 1.7 billion. France announced at least USD 74 million in targeted support for road transport, agriculture and fishing, though several measures remain uncosted.

Comparing the fiscal measures countries implemented so far in 2026 with those announced during the first three months following Russia’s full-scale invasion of Ukraine (March - May 2022) reveals three broad patterns (Figure 2.5). First, in a small number of countries, including France, the Netherlands, New Zealand and the United Kingdom, universal price relief measures that featured prominently in 2022 are absent from the 2026 policy mix to date. Instead, all or most of the 2026 measures take the form of targeted interventions, such as transfers and subsidies directed at vulnerable households and the most exposed industrial sectors. Second, countries like Austria, Estonia, Poland and the Slovak Republic have moved in the opposite direction. Targeted transfers that characterised most of their 2022 response have given way to universal price caps and fuel tax cuts in 2026. For a third group, including Chile, Greece, Ireland, Italy, Lithuania, and Spain, the 2026 response broadly mirrors the 2022 approach in structure with measures combining universal price controls with targeted industrial sector and household support.

Table ‎2.2. Selected energy support measures announced in response to the 2026 energy crisis

Examples drawn from the 2026 OECD Energy Support Measures Tracker, cutoff 18 May 2026

Type of support	Mechanism of support	Examples from the Tracker
Energy price relief	Tax measures	France: Waived the excise tax on agricultural non‑road diesel for the month of April and May, targeting farmers registered as agricultural operators. Ireland: In March, increased the Diesel Rebate Scheme ceiling for transport operators from EUR 0.075 to EUR 0.12 per litre. The increase was backdated to January 2026 and applies through June. Norway: Removed the road‑usage tax on petrol and diesel to lower pump prices, and on mineral oils used in professional fishing and hunting. This is applied from April to September 2026.
	Reduced, regulated or capped marginal energy prices	Japan: Subsidised fuel wholesalers to keep the nationwide average retail price of gasoline near JPY 170 per litre, with subsidy rates reviewed weekly. Similar subsidies apply to diesel, heavy oil and kerosene, and at a reduced rate for aviation fuel. Korea: Introduced a ceiling on refiners’ wholesale gasoline and diesel prices, limiting maximum supply prices to gas stations.
Income-support (energy related)	Budgetary transfers	Ireland: Extended the fuel‑allowance season by one month in spring 2026 to support low‑income households with higher heating expenses. Greece: Offered a monthly digital fuel wallet to vulnerable households. This provides a fixed allowance designated for fuel purchases.
	Reduced, regulated or capped average energy prices	Portugal: Granted low‑income households a EUR 15 discount per bottled‑gas cylinder, up to two cylinders per month.
Income support (non-energy related)	Tax measures	New Zealand: Increased the In‑Work Tax Credit for working families by NZD 50 per week from April 2026 to March 2027 to provide temporary income support, with the measure operating as a price‑triggered mechanism that automatically ends if 91‑octane petrol prices fall below NZD 3.00 per litre for four consecutive weeks.
	Budgetary transfers	Greece: Provided temporary cash reimbursements to farmers for fertiliser purchases during a two‑month 2026 support window. The reimbursements amount to 15% of the cost of invoices during the support window. Spain: One-off, automatic financial aid to farmers to offset fertiliser‑cost increases. United Kingdom: Provided one‑off direct payments to means-tested vulnerable households reliant on heating oil (and LPG in Scotland) in March 2026.
Credit and equity support	Deferred tax payments	France: Deferred social‑security contributions and tax deadlines for agriculture, fishing, and transport firms.
Non‑budgetary measures	Export bans	Hungary: Imposed a temporary export ban on crude oil, gasoline, and diesel with no specified expiry date. Slovak Republic: Restricted diesel exports by banning shipments in containers and tanker trucks.
	Regulatory reforms and safeguards	Germany: Adopted legislative changes expanding antitrust powers in fuel markets to strengthen oversight of pricing behaviour. Slovak Republic: Introduced a EUR 400 spending cap per diesel transaction during a 30‑day emergency period to limit excessive purchases. Portugal: Prohibited electricity and gas disconnections in billing‑dispute cases beginning in March 2026 to protect consumers. New Zealand: Temporarily aligned fuel‑quality specifications with Australian standards to widen import options for domestic suppliers.

Note: Support measures are classified along two dimensions: (i) support type, which captures the economic channel (energy price relief, income support (energy-related or not energy-related), credit and equity support, export restrictions, or other), and (ii) mechanism, which identifies how the measure is implemented (e.g. tax reductions, regulated or capped prices, budgetary transfers, or financial support instruments). Energy price relief is classified based on whether it affects the marginal price (the cost of an additional unit of energy) or the average price (total expenditure per unit consumed). Income support is defined as energy-related when it is conditional on energy use or bills (e.g. energy vouchers), thus reducing the average price of energy, and as non-energy-related when it is not tied to energy consumption (e.g. general cash transfers). Data as of 18 May 2026.

Source: 2026 OECD Energy Support Measures Tracker. https://www.oecd.org/en/data/tools/oecd-energy-support-measures-tracker.html.

Figure ‎2.5. Countries have not converged towards a targeted approach following the 2022 energy crisis

Share of total fiscal energy support measures, OECD countries

Note: The chart covers fiscal energy support measures only. It compares the share of targeted energy support measures implemented during the March–May 2022 energy crisis with measures announced as of 18 May 2026. For countries below the 45-degree line, the share of targeted measures is higher in 2026 than in 2022; the opposite holds for countries above the line. Targeted measures are defined as fiscal instruments directed at a specific subgroup of households or firms. Countries that reported no measures in at least one of the two periods are excluded from the chart.

Source: OECD Energy Support Measures Tracker 2022/23; and OECD Energy Support Measures Tracker 2026.

Statlink

Lessons from the 2022-23 energy crisis

The 2022-23 energy crisis showed how costly broad-based support can become. Across 41 OECD and non-OECD countries, announced support measures totalled about USD 400 billion in 2022 and USD 405 billion in 2023 in gross terms. In the median OECD economy, this corresponded to gross fiscal costs of 0.7% of GDP in 2022 and 0.8% in 2023, with support exceeding 2.5% of GDP in some countries. Untargeted support measures accounted for nearly 80% of the estimated total fiscal costs (Figure 2.6), while energy price relief measures alone accounted for over half of total spending. Although the current shock differs in its initial source of disruption from the 2022-23 crisis, similar policy design issues arise as fossil-fuel price spikes are putting pressure on household budgets, firms’ costs and inflation.

Figure ‎2.6. The bulk of support measures in 2022-23 were untargeted

Gross fiscal costs of support measures, 2022-23

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 price and income measures. The gross fiscal cost does not account for the impact of accompanying energy-related and revenue-increasing measures, such as the taxation of windfall profits.

Source: OECD Energy Support Measures Tracker 2022-23.

Statlink

The current energy crisis comes at a time when public finances are strained. In many countries, budget deficits are still high, debt is elevated, and debt servicing costs are higher than in 2022-23. In 2025, across OECD countries, the average general government gross debt as a share of GDP stood at 111% of GDP – up from 110% in 2022 and 73% in 2007. Energy support measures add to other fiscal pressures, including ageing, higher defence spending in some countries, extreme weather events, and the energy transition.

Broad-based measures, while easier and faster to implement, will further weigh on budgets. However, in countries where social benefits or public sector salaries are closely linked to inflation, interventions that lower energy prices may generate some offsetting fiscal effects. The fiscal implications also differ across instruments. Cuts in VAT or other ad valorem taxes partly forgo revenue that would otherwise rise automatically with higher energy prices, whereas income support and reductions in per-unit taxes imply more direct budgetary costs or revenue losses. Broad and untargeted price measures postpone rather than remove inflation pressures and have higher fiscal costs than targeted measures – in particular if they are long-lasting. They are also politically difficult to withdraw, even once the crisis abates. In this respect, three lessons from the 2022-23 energy crisis stand out: support should be targeted, temporary and preserve incentives to save energy (Hemmerlé et al., 2023). These lessons also matter for structural resilience to future energy supply shocks, since broad, open-ended or price-suppressing emergency measures can weaken incentives to save energy and delay investment in energy efficiency or more diversified energy supply (as discussed below).

Targeting support measures to help contain fiscal costs

Given limited fiscal space in most countries, support should focus on preventing severe hardship among the most vulnerable households and avoiding severe hardship for the most affected and otherwise viable firms. However, identifying in real time firms for which the energy price spike creates only temporary liquidity stress rather than solvency problems is challenging. Using pre-crisis benchmarks and exposure thresholds is one way to determine eligibility. The use of declining and capped support and requiring co-payments rather than providing full compensation might also avoid support for unviable companies. Any support for firms should not come at the expense of business dynamism or distort competition in domestic markets by protecting incumbents or providing unfair advantages to some firms. SMEs may warrant particular attention as they tend to be particularly vulnerable to energy price shocks (Hemmerlé et al., 2023). Governments may, however, also want to target large, energy-intensive firms or industries for strategic or national security reasons.

Emergency top-up payments to existing welfare recipients are often the simplest and fastest way to support households most in need. However, in the 2022-2023 energy crisis, only around 10% of the total costs of energy-price relief measures exploited existing social transfer programmes to deliver support (Hemmerlé et al., 2023). As lower-income households are generally the most exposed to high energy prices, income remains the main basis for targeting. Yet, among low-income and near-poor households, exposure can differ depending on factors other than income such as housing energy efficiency, geographical location and social benefits already received. Existing benefits systems, therefore, provide a useful starting point to identify and reach households in need of protection but may need to be complemented with additional information to calibrate support and reach vulnerable households that existing benefit systems may miss.

The 2022-23 crisis showed that targeting support is feasible (Hemmerlé et al., 2023). Costa Rica, for instance, provided income support directly into bank accounts of poor households that were registered in a digital database that recorded all past and present beneficiaries entitled to social transfers. In Ireland, household support was channelled mainly through existing welfare schemes, especially higher regular and lump-sum payments to recipients of an existing means-tested fuel allowance. Ireland also provided targeted support to SMEs and the self-employed whose energy bills had increased by 50% or more in a reference period, covering 40% of the increase in the reported energy bill, up to a monthly cap of EUR 10 000. Likewise, Italy granted energy intensive firms tax credits covering 20% of electricity and gas costs when prices rose more than 30% above 2019 levels. This shows that, even in a crisis, support can be targeted by relying on existing data and administrative systems, and setting clear eligibility thresholds.

Some measures introduced by OECD countries in 2026 move in a similar direction, for instance, through one-off payments to vulnerable households, temporary extensions of existing allowances, and more targeted support for exposed sectors such as farming and transport. Even so, broad-based measures still account for a larger share of the current policy mix (Figure 2.7).

Ensuring support is temporary

A second lesson is that support should be temporary. The fact that support is set to automatically expire or needs to be explicitly reviewed can help limit costs and keep crisis measures focused on acute energy price pressures, thereby avoiding unnecessary and long-lasting social payments or subsidies. The 2022‑23 crisis showed that some governments kept broad support measures in place even well after energy prices abated, adding to fiscal costs and blurring the case for continued intervention.

New support could therefore come with clear end dates or automatic phase-out rules linked, for instance, to retail energy prices falling back to a pre-defined benchmark. For example, Germany’s support for households in 2022-23 was tied to contract prices, so the value of support declined automatically as retail energy prices fell, helping to contain fiscal costs as market conditions improved. For firms, Bulgaria’s measures in 2022-23 covered 75% of electricity costs for SMEs, but only for four consecutive months in 2022, at the firms’ choice.

Most measures introduced by OECD countries in 2026 also include explicit support windows, one‑off interventions or temporary extensions of existing schemes. Others, however, lack clear expiry dates or phase-out rules.

Figure ‎2.7. Energy support measures announced as of 18 May 2026

Count of total fiscal energy support measures, OECD countries

Note: The figure covers fiscal energy support measures only. Targeted measures are defined as fiscal instruments directed at a specific subgroup of households or firms. Extensions or parameter adjustments of existing measures are consolidated into a single measure.

Source: OECD Energy Support Measures Tracker 2026.

Statlink

Preserving incentives to save energy

A third lesson is that support should preserve energy-saving incentives while supporting diversification away from fossil fuels. Measures lowering energy prices for households and firms weaken incentives to reduce energy demand – which is particularly problematic when facing an energy supply crunch. Furthermore, such measures can discourage improvements in energy efficiency and the diversification of energy supply, two key levers for enhancing countries’ resilience to energy supply shocks (as discussed below). By contrast, direct income transfers can cushion hardship for vulnerable households while preserving price signals and thereby incentives to save energy. For firms, support can be linked to requirements and incentives that encourage innovation and improve energy efficiency, including in industrial processes.

When governments reduce energy bills directly (e.g. through energy bill vouchers), support that applies only up to a basic level of energy consumption is preferable to support that covers the entire energy bill. This limits relief to basic energy needs without weakening incentives to reduce consumption above that basic level. Because information on basic energy needs is often unavailable, many countries during the 2022-23 crisis based their support (often price caps) on average electricity consumption thresholds instead. Setting such thresholds at a sufficiently low level and clearly communicating them to households would help preserve incentives to save energy while protecting a minimum level of energy consumption.

During the previous energy crisis, some countries designed support in this direction. In the Netherlands, for example, energy price relief applied different consumption thresholds in warmer and colder months. Hence, energy usage above the seasonal threshold continued to face market prices, preserving incentives to save energy throughout the year. Croatia provided an electricity-cost subsidy for energy-intensive firms conditional on an energy audit and on investments that either substantially reduced greenhouse gas emissions or increased the share of renewable energy in electricity consumption to at least 60%.

In the current 2026 response by OECD countries, one-off payments, capped discounts or income support tend to preserve energy-saving incentives. However, most measures introduced so far have taken the form of fuel-tax reductions or direct price interventions that lower marginal fuel prices, thereby weakening incentives to curb energy use.

Enhancing administrative capacity to deliver targeted and timely support

Energy shocks remain a recurrent macroeconomic, social and fiscal risk. The experience of the 2022‑23 episode and the current energy crisis show that effective crisis responses hinge not only on fiscal space, but also on administrative capacity to deliver support rapidly and effectively. Insufficient administrative capacity and a lack of data are key reasons why governments often defaulted to broad-based and price‑cap measures during the 2022-23 crisis, inflating costs and undermining the efficiency of the support provided (Hemmerlé et al., 2023). Where digital infrastructure and administrative capacity are weaker, governments may therefore be structurally more constrained in their ability to target support, even when targeted measures would be preferable.

Once implemented, broad-based measures also risk becoming entrenched. In 2024, when energy prices were significantly below the crisis peak, untargeted measures still accounted for over 70% of the fiscal cost of consumption support for fossil fuels across OECD countries (OECD, 2025e).

Better crisis preparedness – through fit-for-purpose institutions, granular and high-quality data systems and stress-tested delivery mechanisms – is therefore key to ensuring support is timely, temporary and targeted. Making strides in this direction requires progress on three fronts: i) better shock anticipation capacity to prepare and trigger timely, rule-based responses; ii) stronger coordination among relevant government agencies to align objectives across finance, energy and social policy; iii) greater execution capacity to deliver targeted support at speed, monitor take-up in real time and unwind the support when shocks dissipate.

Shock anticipation

Crisis preparedness starts with a strong capacity to anticipate shocks and plan effective responses. This entails:

• ensuring that emergency energy stockholding systems are adequately designed, governed and ready to be deployed when disruptions occur;

• being able to identify people and firms more exposed to shocks and supply chain disruptions;

• and regularly stress-testing the eligibility and delivery systems.

Preparedness for energy supply shocks depends on energy reserve systems with clear legal authority over stock releases, reserve levels calibrated to countries’ vulnerabilities, and stocks that are physically accessible and operationally releasable when needed. Pre-arranged international co-ordination mechanisms can help align stock releases across countries and make collective responses more effective.

Assessing households and firms’ exposures to shocks hinges on linking information and databases to monitor and quantify the channels through which higher energy prices affect households and firms. Key factors include income, household composition, housing and heating characteristics, location, and existing benefit receipt (Hemmerlé et al., 2023). For firms, relevant information may include indicators of energy use, payroll and tax records, and use of existing support schemes. Opt-in schemes, where firms apply for support, can elicit firms to provide the required information and lower administrative complexities.

Granular data and robust digital infrastructure are another important component of anticipatory capacity. For example, smart meters and network data can show where electricity consumption is concentrated and where peak demand is rising most sharply, while also helping operators detect outages and local bottlenecks more quickly (World Bank, 2025; IEA, 2025a). Smart-meter rollout, however, remains uneven across countries: for example, in the European Union, rollout in 2024 exceeded 80% in 14 member states but remained below 30% in 10 others, while in the United States advanced meters accounted for about 72% of all electric meters in 2022 (CEER, 2025; EIA, 2023).

Regular scenario exercises and stress tests of eligibility and delivery systems can help reveal whether support programmes can be activated and scaled up quickly, whether eligibility rules are operationally robust, and where financing or administrative bottlenecks are most likely to emerge (Coudouel, Fuselli and Saidi, 2024). Similar exercises can also test whether energy emergency stocks can be released quickly during a crisis and whether bottlenecks in transport, storage and distribution channels could delay this process.

If supply disruptions risk turning into physical shortages, governments also need pre-defined rationing or curtailment protocols (Haas, Kozluk and Giuliana, 2022). These should be a last resort. To limit economic and social damage, priority should go to critical users, based on transparent criteria such as social criticality, wider supply-chain effects, the scope for fuel switching, and the risk that interruption would cause lasting damage or long restart delay (OECD/IEA, 2005). This in turn requires clear legal authority, real‑time information and regular simulation exercises.

Government coordination

An effective crisis response also requires strong coordination across public institutions. Responses to energy shocks cut across responsibilities of different ministries (such as finance, energy, social policy, housing and digital administration) and can demand joint action by regulators, independent agencies and subnational authorities. Without clear coordination and responsibilities, emergency responses risk being fragmented, delayed, and costly.

Crisis preparedness requires clear demarcation of roles and responsibilities among the agencies involved in emergency responses, common definitions of vulnerability, interoperable databases and pre‑defined protocols for activating, recalibrating and withdrawing support without delay. This can involve pre‑existing inter-ministerial crisis units, shared real-time information systems, and inter-agency protocols that clarify who leads on data sharing, eligibility assessment, financing and communication during an emergency (OECD, 2025c). Linked administrative and IT systems can improve coordination and alignment across social programmes and connect energy support with other services, such as housing and social assistance. This can reduce duplication and hasten implementation, while strengthening outreach to vulnerable households who might otherwise be missed (Hemmerlé et al., 2023).

Capacity to deliver support

Ultimately, delivery capacity determines whether timely support can be delivered to the households and firms that need it most. In a crisis, topping up existing social programmes is usually faster, cheaper to administer and less error-prone than implementing new schemes from scratch. Where legal and technical conditions allow, automatic enrolment through existing payment channels can reduce delays, administrative burdens and non-take-up by relying on information governments already hold. However, reliance on existing programmes does not by itself eliminate the risk of non-take-up. In Belgium, for example, between 37% and 51% of eligible working-age people do not take up social assistance, and those with the fewest resources are the least likely to respond to simple outreach interventions (OECD, 2024b). This reinforces the case for better outreach, linked administrative data, simpler enrolment procedures and automatic delivery where feasible (see Box 2.1).

Advanced analytics, including AI tools, can further strengthen execution through operational triage, anomaly and fraud detection and short-term forecasting of caseloads and budget needs (OECD, 2024a). However, these systems should complement rather than replace robust and well-tested administrative processes and clear accountability rules for decision making. Their use in public administration also requires strong safeguards for data quality, privacy, cybersecurity, transparency and accountability (OECD, 2019).

Box ‎2.1. Automatic activation and delivery: country examples of targeted support in practice

Experience from the 2022-23 energy crisis shows that pre-defined triggers and linked administrative systems can help governments reduce delays, limit non-take-up and avoid relying on broad-based measures during a crisis. This is particularly valuable when energy prices rise sharply and support needs to be deployed quickly.

The United Kingdom’s Cold Weather Payment illustrates the value of a standing scheme with pre‑defined triggers and automatic delivery. Eligible households receive a payment when temperatures in their area are recorded or forecast to be 0°C or below for seven consecutive days, with eligibility linked to existing benefit receipt and local weather data. This reduces the need for case‑by‑case applications during severe cold spells and provides a clear example of how automatic activation rules can support timely assistance.

Italy’s social bonus provides an example of automatic delivery through linked administrative systems. The discount is applied automatically to eligible electricity, gas and water bills for households below a means-tested threshold based on income, assets and household composition. Eligibility is then verified through cross-checks across social-security and utility-related databases. This helps to limit the non‑take-up that often arises when households need to apply separately for support.

Denmark offers a further example of how linked data can improve targeting. By combining income data with information from administrative housing registers on households’ main heating source, the authorities were able to identify households most exposed to the surge in heating costs and distribute heating cheques automatically. The crisis measure provided a one-off payment of DKK 6 000 to more than 400 000 low-income households whose heating source was particularly exposed to the surge in prices. This shows that even when a support measure is introduced as an emergency response, pre‑existing data systems can still make targeting more operational.

Source: Hemmerlé et al. (2023)

Strengthening the resilience of energy systems

Energy and the macro-economy

Energy shocks can have real macroeconomic consequences, going beyond social impacts. For example, André et al. (2023) find that following an energy price shock firms reduce capacity utilisation and experience lower productivity in the short term. The repeated nature of such shocks calls for better‑designed emergency support and crisis management, but also for reducing structural exposure to energy price volatility and supply disruptions.

This is becoming more important as modern energy demand is increasingly being shaped by structural rather than cyclical developments. The IEA World Energy Outlook 2025 projects the world’s energy demand rising by 8 to 15% over the coming decades, driven by increasing need for mobility, heating, cooling, and other household and industrial uses in addition to rising demand from data centres and artificial intelligence (AI)-related services. Rapid electrification, the evolving concerns about energy security (including the supply of critical minerals) and the increasing share of global demand from emerging economies are among the main forces shaping the global energy system.

Strengthening the resilience of energy systems to shocks lowers the frequency and fiscal costs of future interventions and is therefore becoming a more pressing policy priority. Making progress in this direction rests on two complementary strategic axes:

• Improving energy efficiency can lower overall energy demand, limiting exposure to volatile energy markets and fossil‑fuel import dependence while buttressing competitiveness through lower energy costs.

• Greater energy supply diversification, in turn, reduces vulnerability associated with single points of failure by avoiding excessive reliance on a narrow set of fuels, technologies, suppliers or transit routes.

Pursuing both objectives can be mutually reinforcing. Energy efficiency gains reduce the scale of supply that must be secured, while diversification strengthens the reliability of the remaining supply. The electrification of the energy system can support both energy efficiency and diversification, with generation being based on a wide range of primary energy sources. These resilience measures can reduce the economic losses from future supply disruptions, but they also involve upfront costs and, in some cases, higher system costs. Governments should prioritise measures that ensure the largest reduction in potential disruption costs.

Improving energy efficiency

Energy efficiency investments reduce the energy intensity of production and consumption, thereby reducing the total costs of energy use (fuel, operating expenditure, but also capital expenditure on generation, networks, and storage), while potentially enhancing productivity and international competitiveness. Reduced energy demand can diminish reliance on high-risk import chokepoints, exposure to volatile energy markets, and inflationary pass-through from energy shocks.

OECD economies have made sizeable gains in energy efficiency. In some OECD economies, efficiency improvements since 2000 have saved an amount of energy equivalent to around one-fifth or more of current energy consumption with respect to a no-efficiency gain baseline (Figure 2.8). By reducing the baseline demand for energy, higher energy efficiency makes it easier to respond to an energy crisis (as rationing becomes less stringent and costly, strategic reserves last longer and blackout risks diminish because of lower peak loads).

Figure ‎2.8. Energy efficiency gains since 2000 have resulted in energy savings in many OECD countries

Share of energy consumption avoided between 2000-2024 through energy efficiency improvements

Note: Savings are the difference between observed energy consumption and the energy consumption that would have occurred had energy intensity remained at its 2000 level, with output growing at the same rate as observed during the 2000-2024 period. A higher share indicates greater cumulative efficiency gains relative to each country's 2000 baseline. The reference year is 2023 for Austria, Belgium, Czechia, Finland, Greece, Hungary, Japan, Korea, Spain and the United Kingdom; 2022 for Australia, Denmark, France, Germany, Italy, Latvia, Poland, Portugal, Switzerland and the United States; and 2021 for Brazil, Canada and the Netherlands.

Source: International Energy Agency (IEA), Energy Efficiency Indicators, 2024.

Statlink

High and volatile energy prices provide incentives for firms and households to invest in energy efficiency, but market failures can hinder such investments from reaching economically efficient levels. The policy mix should therefore focus on providing clear price signals while addressing the market failures hindering energy-efficiency investment. These include imperfect information about future energy savings, behavioural inertia and upfront financing costs (Allcott and Greenstone, 2012; Gerarden, Newell and Stavins, 2017). Policies to address these market failures cover energy performance standards, energy efficiency mandates, fuel economy standards, finance mechanisms, and information campaigns, among others. Information and awareness campaigns aimed at addressing information gaps and overcoming behavioural biases can be particularly useful in helping households and firms identify low‑cost energy‑saving opportunities and encouraging behavioural adjustments.

In most countries, adoption of energy efficiency policies remains uneven, reflecting in part differences in countries’ industrial structure and the potential for energy savings in each sector (Figure 2.9). Even so, substantial untapped potential remains in many sectors, in part because policy stringency has often lagged technological progress. Many appliances sold today are only around half as efficient as the best available models. For example, the efficiency of best-in-class lightbulbs doubled over the past 15 years, while minimum energy performance standards increased by only around 30% (IEA, 2025a). At the same time, improving energy efficiency involves trade‑offs because of upfront outlays or investments.

Figure ‎2.9. Progress in adoption varies across energy efficiency policies

Number of countries, 2010-2024

Note: Data are based on the OECD Climate Actions and Policies Measurement Framework (CAPMF), which is built on data from the OECD, IEA, World Bank and others. The heatmap illustrates the 13 energy efficiency policies tracked by the CAPMF across 105 countries (2010-2024). The colour scale ranges from dark orange (no countries have adopted the policies) to dark blue (highest number of adopted policies). MEPS are minimum energy performance standards.

Source: OECD (2026).

Statlink

Stronger minimum energy performance standards for buildings (including existing buildings when they are renovated, sold or rented out), combined with accelerated retrofit programmes, can improve buildings' energy efficiency. These efficiency gains lower energy bills, reduce dependence on fossil fuels and limit the need for costly additions to generation and network capacity. In 2024, buildings accounted for about 30% of global final energy consumption and contributed to about 20% of the growth of global energy demand between 2019 and 2024. Furthermore, energy for space cooling has seen the fastest growth of any end-use in buildings since 2000, growing over 4% per year (IEA, 2025a). The IEA estimates that if all air conditioners bought since 2019 had been the most efficient model available, the world could have avoided electricity demand growth equivalent to that from data centres over the same period (IEA, 2025a). This illustrates how tighter standards, clearer labelling, and incentives (including financial incentives) to replace inefficient equipment can help avoid locking in higher peak electricity demand and larger future needs for generation, networks and storage, while lowering energy costs for owners and tenants (IEA, 2024).

However, building energy efficiency improvements require upfront investments and present trade-offs with housing affordability. Financing constraints remain a major barrier to retrofits, especially for low-income households. In addition, information gaps and uncertainty about expected savings can also delay take-up. Policy therefore needs to strengthen incentives to invest, ease access to finance and address information barriers through, for instance, energy audits and labelling. Targeted grants and concessional or low-cost loans can then help ease remaining financing constraints where they are most binding (OECD, 2023). Targeted grants are likely to be more important where borrowing constraints are tight, while concessional or low-cost loans can play a larger role where households’ repayment capacity is stronger. Slovenia illustrates such a mixed approach: the Eco Fund provides loans and grants for building insulation and cleaner heating systems for households more broadly, while the Energy Poverty Mitigation programme provides fully funded retrofits for vulnerable households and the Social Climate Plan supports housing‑efficiency improvements for low-income groups.

In industry, the priority is to support adjustments that durably lower energy intensity and thus reduce exposure to volatile fossil-fuel prices in addition to lowering energy costs. The contribution of industry to economy-wide energy intensity varies markedly across countries, reflecting differences in economic structure and the deployment of energy saving technologies (Figure 2.10). Globally, between 2019 and 2025, industry accounted for about two-thirds of global final energy demand growth. In the same period, progress in reducing industrial energy intensity slowed sharply to 0.5% per year, down from 2% per year over the last decade (IEA, 2025a). This is weighing on economy-wide energy efficiency progress.

Figure ‎2.10. Use of energy to create value added differs among countries and sectors

Value added to total energy consumption, by sector, 2022-2023

Note: Each bar represents total value added divided by total final energy consumption for a given country, decomposed by sector's contribution to aggregate value added. A higher value indicates that more value added is generated per unit of energy consumed, i.e. lower economy-wide energy intensity. Cross-country differences reflect both sectoral composition and differences in energy use within sectors. Data cover the 2022-2023 average.

Source: International Energy Agency (IEA), Energy End-uses and Efficiency Indicators database 2025; and OECD calculations.

Statlink

The 2022-23 energy crisis showed that emergency energy savings and fuel switching can help absorb an acute supply shock, but durable gains require lasting changes. This would require firms to invest in, for instance, process efficiency, energy management, more efficient motors and pumps, and electrification of low-temperature heat where this is technically feasible and economically justified (IEA, 2025a). This might require policy support to overcome market failures hindering such investments. Electrification can also help diversify the industrial energy mix (as discussed below). In practice, firms may underinvest in energy efficiency because energy-saving investments compete with other capital spending, expected payback can be uncertain, and financing constraints can be binding, especially for smaller firms. This in turn points to a role for policies to reduce information and implementation barriers, including energy audits, labels and standards for industrial equipment, and loan guarantees or other financing tools for smaller firms when needed. Adoption of standards and labels for electric motors, for example, remains much less widespread than for appliances, despite their relevance to industrial energy use (Figure 2.9).

Stable and predictable policy frameworks would help to direct resources towards enhancing energy efficiency, as policy uncertainty delays capital upgrading and hampers behavioural changes. Clear price signals, minimum efficiency standards and labels, and targeted grants or tax incentives, where market failures or financing constraints are binding, are policy options to help crowd in energy efficiency investment. Evidence from Europe suggests that policy packages combining carbon pricing with sector‑specific regulations and faster permitting for clean infrastructure can strengthen energy‑security gains and distribute them more broadly across countries than carbon pricing alone (Dolphin, 2024).

The focus of policy should be on the timing of energy use as well as on reducing annual consumption. Demand-response programmes, smart metering, time-of-use tariffs and dynamic pricing can shift electricity demand away from hours when electricity is most scarce and costly. This reduces peak stress, improves system security margins and can postpone some network and generation investment. The IEA reports that households can typically save 5-15% on electricity costs under dynamic tariffs, while flexible demand can provide system capacity at far lower cost than building new capacity (IEA, 2025c). Evidence from the United Kingdom and the United States also suggests that real-time feedback and pre‑event communications can reduce peak demand by about 3% (Department for Energy Security & Net Zero, 2024; Thayer et al., 2016). This makes demand flexibility a useful complement to supply‑side investment as it can improve resilience and affordability while containing investment needs.

Boosting diversification

Diversification across energy sources, technologies, and locations strengthens energy system resilience. Through diversification, the system becomes less vulnerable to any single point of failure — whether driven by geopolitical disruptions, extreme weather, or cyberattacks. However, diversification entails a trade‑off: it raises upfront costs while lowering expected losses under uncertainty.

Diversification across fuels and supply sources remains a first line of defence against external shocks. Governments cannot entirely eliminate supply risks, but they can reduce their macroeconomic consequences by limiting exposure to any single chokepoint or supplier. For oil and gas importers, this means reducing reliance on a narrow set of suppliers, routes and infrastructures often in geopolitically unstable regions, while maintaining emergency reserves and demand-restraint plans that can be activated when disruptions occur (IEA, 2025b). This can require investments in import infrastructure that can switch between supplies easily, including access to alternative terminals and to alternative pipelines or LNG routes (OECD, 2025b). Regulatory reforms that facilitate access to alternative network infrastructure and ease permitting requirements on supply‑security grounds would enable faster progress in this area. Recent European experience has shown that access to LNG imports from a wider range of suppliers, together with stronger pipeline interconnections, helped countries redirect flows and reduce dependence on Russian gas, thereby strengthening resilience to supply disruptions (European Commission, 2025).

The degree of energy diversification varies widely across countries (Figure 2.11). The Herfindahl- Hirschman Index, a measure of market concentration, suggests that some economies rely on a broad mix of energy sources, while others remain dependent on one or two dominant fuels. In several cases, low diversification of energy sources reflects heavy reliance on oil and gas; in others, it reflects the dominant role of coal, nuclear or some specific renewable sources. This matters for resilience because economies with a narrower supply mix are generally more exposed when the dominant source is disrupted or becomes more expensive. At the same time, concentration is not equally risky in all cases. For instance, reliance on abundant domestic sources, such as geothermal and hydropower in Iceland, may pose smaller energy security risks than dependence on imported fossil fuels routed through a small number of suppliers or transport corridors.

Figure ‎2.11. Degree of energy diversification varies across countries

Herfindahl-Hirschman Index (HHI), based on total energy supply, 2023-2024 average

Note: HHI is calculated as the sum of squared shares of each energy source in total energy supply (TES); it ranges from 0 (perfect diversification) to 1 (complete concentration). Negative TES values (exports, stock draws, transformation outflows) are excluded, therefore shares are computed on gross positive inflows only.

Source: International Energy Agency (IEA) data; World Energy Balances 2025; and OECD calculations.

Statlink

Diversification across technologies is equally important. Different technologies are exposed to different hazards and failures, so a more diversified technology mix lowers the risk that one shock disrupts a large share of supply (IEA, 2022). A resilient power system therefore relies on a mix of generation technologies and operating arrangements. This means combining dispatchable capacity with storage capacity, variable renewables and other flexibility mechanisms, so that weaknesses in one part of the system can be offset by strengths in another (IEA, 2022; IEA, 2026b). For example, when wind or solar output is low, dispatchable generation, storage and flexible demand can help keep supply and demand in balance. Conversely, when fuel supplies are disrupted or thermal plants fail, renewables can reduce pressure on the rest of the system. This becomes more important as electrification advances and risks of extreme weather events intensify.

Italy and the United Kingdom provide two recent examples of how diversification increasingly depends on regulatory design as well as new capacity. In Italy, the Testo Unico FER simplified renewable permitting, while the new MACSE framework uses long-term auctions to support investment in electricity storage and system flexibility. The United Kingdom’s Clean Power 2030 strategy illustrates the same broader approach by combining expansion of low-carbon generation with planning and grid reform and greater use of flexibility resources such as storage and demand response.

Geographic diversification within the energy system also matters. Concentration of domestic generation, storage or network assets in a small number of critical nodes can create single points of failure generating cascading losses. Recent evidence from Ukraine underscores the resilience gains from geographically dispersing assets, decentralising the energy system, and distributing emergency reserves and critical equipment across locations rather than concentrating them in a few large sites (IEA, 2026b). The same logic applies more broadly to electrical substations, major energy plants, storage terminals and transport corridors.

Decentralised systems are better able to isolate shocks, maintain essential services, and restore supply more quickly after disruption. These benefits are not without costs, however. More distributed configurations can involve the loss of some economies of scale and higher costs for network monitoring and coordination. For policymakers, this means giving greater weight – in network planning – to modularity, redundancy and local backup where the resilience gains are likely to be largest, especially where weather hazards, sabotage risks or cross-border infrastructure vulnerabilities are high. This also calls for sufficient backup capacity and alternative supply options, particularly where systems depend heavily on hard‑to‑replace equipment or routes (IEA, 2022; IEA, 2026b).

Electrification can help diversify energy sources and make the energy system more resilient

Electrification can underpin energy diversification and strengthen energy security by making final energy demand less dependent on the direct use of a narrow set of fuels and suppliers. Unlike end‑uses tied directly to a single fuel, electricity can be produced from a wide range of primary energy sources, ranging from fossil fuels and renewables to nuclear and geothermal. The share of electricity in final energy consumption differs across countries (Figure 2.12), implying that in some economies the resilience of the power system is already more central to overall energy security than in others. Among OECD countries it ranges from just above 10% in Lithuania and Latvia to more than 40% in Norway and Iceland, with an OECD average of 24%. Where electric alternatives are already more mature, notably in buildings and passenger road transport, electrification can also reduce direct fossil-fuel use and often improve energy efficiency (IEA, 2024).

Electrification does not automatically strengthen resilience. Resilience gains attributable to electrification are larger when the electricity mix itself is diversified. As electricity becomes more central to final energy demand, the reliability and diversification of the power system become more important for macroeconomic stability. More than 2 500 GW of renewable, storage and large‑load projects are currently stalled in grid connection queues worldwide, underlining how network and regulatory bottlenecks can hold back otherwise viable investment (IEA, 2026a). IEA analysis suggests that annual grid investment needs to rise by around 50% by 2030, after years in which grid investment lagged behind investment in generation (IEA, 2026c).

Electrification can also strengthen resilience not only by allowing for a more diversified energy mix, but also by shifting the structure of the system towards one that is more modular, decentralised and more robust to supply and physical disruptions. This will require reinforcing and modernising transmission and distribution systems so they can carry larger volumes of electricity, connect a growing number of smaller and more dispersed generators, and continue operating under stress. More distributed energy system configurations, however, also involve cost trade-offs (as highlighted above).

Stronger cross-border interconnections can support diversification by allowing countries to exploit complementarities across energy sources, smooth local shocks and share generation capacity. However, such interconnections require significant up-front investments. Increasing storage capacity can enhance resilience by providing rapid response during disturbances and supporting local continuity of supply, though its value will depend on system needs and costs. Distributed generation and storage, including microgrids, can likewise reduce single points of failure and help maintain critical services when the main grid is disrupted (IEA, 2026b; Anderson et al., 2017).

Figure ‎2.12. Share of electricity in final energy consumption

Percentage of total final energy consumption, 2023

Note: Electricity share calculated as electricity consumption divided by total final energy consumption. Total final energy consumption covers energy consumed by end users including for heating, cooling, transport, and industrial processes, excluding conversion and transformation losses.

Source: International Energy Agency (IEA) World Energy Balances 2025.

Statlink

Realising the resilience gains from electrification depends as much on regulation and system management as on physical investment. There is increasing recognition that regulatory barriers are hindering grid investment and its expansion (OECD, 2025d). Current regulatory frameworks have been designed for conventional, centralised energy systems. As a result, they are ill-suited to deliver the investment required to strengthen energy systems’ resilience through higher decentralisation, flexibility, and diversification (OECD, 2026a).

More broadly, this points to regulatory priorities covering grid investment, system flexibility and the reliable operation of more decentralised power systems. Reducing permitting and regulatory barriers to grid reinforcement and more efficient grid use, cross-border interconnections, storage, and flexible-demand solutions can help investment keep pace with rising electricity demand and with the need for greater resilience. Key reform areas include streamlining siting and permitting procedures, allowing remuneration for new services (such as balancing, storage and flexibility services), improving grid‑connection rules – where “first-come, first-served” approaches encourage speculative applications and long queues – and adapting regulatory frameworks that still favour traditional capital expenditure over more flexible and grid‑enhancing solutions (OECD, 2025a). The blackout in the Iberian Peninsula in April 2025 illustrates the risks that can emerge when technical, operational and regulatory arrangements do not keep pace with the evolving nature of the power system (ENTSO-E, 2026).

More specifically, the resilience benefits of decentralised electricity systems depend on regulatory reforms allowing small and dispersed electricity-generating assets to participate in the system. In many countries, the rights and obligations of households or firms that both consume and generate electricity (i.e. prosumers) remain unclear. These ambiguities raise legal and financial risks. Clarifying definitions, rights and obligations and ensuring proportionate cost-allocation rules for these new business models are prerequisites for turning the potential benefits of decentralisation into actual resilience gains (OECD, 2025a). For instance, market and regulatory frameworks should recognise the value of the services microgrids and dispersed generating assets can provide when connected to the main grid, including flexibility and peak load reduction. Moreover, network tariffs should be designed to ensure that grid users share infrastructure costs and that the services they provide (such as balancing, storage and flexibility services provided by distributed assets) are valued correctly.

Removing the regulatory and technical obstacles that still limit digital monitoring and control is also a reform priority (OECD, 2025a). That is because electricity systems with more distributed generation, storage and flexible demand increase real-time management, monitoring and control needs along with robust cybersecurity standards. Better deployment of these tools can help system operators manage power networks more efficiently, detect local bottlenecks sooner and integrate more distributed resources without relying only on costly physical expansion. Clearer rules on access to metering and system data can enlarge the scope to use data for flexibility and operational decision-making. Developing common standards for data exchange and more coordinated cybersecurity frameworks will help safely integrate digital solutions and realise their full benefits (EDSO, 2024; Monaco et al., 2024). The policy challenge is therefore not simply to electrify faster but also to align the pace of electrification with the expansion, regulation and digital management of the power system on which it increasingly depends.

References

Allcott, H. and M. Greenstone (2012), “Is There an Energy Efficiency Gap?”, Journal of Economic Perspectives, Vol. 26/1, pp. 3-28, https://doi.org/10.1257/jep.26.1.3.

Anderson, K. et al. (2017), Increasing Resiliency Through Renewable Energy Microgrids, Journal of Energy Management, https://docs.nlr.gov/docs/fy17osti/69034.pdf.

André, C. et al. (2023), “Rising energy prices and productivity: short-run pain, long-term gain?”, OECD Economics Department Working Papers, No. 1755, OECD Publishing, Paris, https://doi.org/10.1787/2ce493f0-en.

CEER (2025), Annual report 2024, Council of European Energy Regulators, https://www.ceer.eu/wp-content/uploads/2025/03/CEER.AR2024_Final.pdf.

Coudouel, A., S. Fuselli and M. Saidi (2024), Stress Testing Adaptive Social Protection Systems in the Sahel, Washington, DC: World Bank, https://doi.org/10.1596/40814.

Department for Energy Security & Net Zero (2024), Monitoring smart meter energy savings using the National Energy Efficiency Data-Framework, https://assets.publishing.service.gov.uk/media/672df39f5437e298ae64ce94/monitoring-smart-meter-energy-savings-using-NEED.pdf.

Dolphin, G. (2024), “The Energy Security Gains from Strengthening Europe’s Climate Action”, Departmental Papers, Vol. 2024/005, p. 1, https://doi.org/10.5089/9798400265655.087.

Durmus Senyapar, H. and R. Bayindir (2025), “The Energy Hunger Paradox of Artificial Intelligence: End of Clean Energy or Magic Wand for Sustainability?”, Sustainability, Vol. 17/7, p. 2887, https://doi.org/10.3390/su17072887.

EDSO (2024), Supporting interoperability in electricity sector: how to steer the European regulatory framework?, https://www.edsoforsmartgrids.eu/content/uploads/2024/10/edso-innovation-brief-regulatory-framework-for-interoperability.pdf.

EIA (2023), How many smart meters are installed in the United States, and who has them?, https://www.eia.gov/tools/faqs/faq.php?id=108&t=3.

ENTSO-E (2026), Grid Incident in Spain and Portugal on 28 April 2025, https://eepublicdownloads.blob.core.windows.net/public-cdn-container/clean-documents/Publications/2025/iberian-blackout/Final%20Report%20on%20the%20Grid%20Incident%20in%20Spain%20and%20Portugal%20on%2028%20April%202025.pdf.

European Commission (2025), Security of Gas Supply,https://energy.ec.europa.eu/topics/energy-security/security-gas-supply_en

Gerarden, T., R. Newell and R. Stavins (2017), “Assessing the Energy-Efficiency Gap”, Journal of Economic Literature, Vol. 55/4, pp. 1486-1525, https://doi.org/10.1257/jel.20161360.

Haas, J., T. Kozluk and S. Giuliana (2022), Emergency plans and solidarity: Protecting Europe against a natural gas shortage, https://oecdecoscope.blog/2022/10/21/emergency-plans-and-solidarity-protecting-europe-against-a-natural-gas-shortage.

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.

IEA (2026a), Electricity 2026, IEA, Paris, https://www.iea.org/reports/electricity-2026.

IEA (2026b), Energy system Resilience, IEA, Paris, https://www.iea.org/reports/energy-system-resilience.

IEA (2026c), Global Energy Review 2026, IEA, Paris, https://www.iea.org/reports/global-energy-review-2026.

IEA (2026d), Oil Market Report - May 2026, IEA, https://www.iea.org/reports/oil-market-report-may-2026.

IEA (2025a), Energy Efficiency 2025, IEA, Paris, https://iea.blob.core.windows.net/assets/23a80bb2-6985-4507-ab99-c1d700f6548b/EnergyEfficiency2025.pdf.

IEA (2025b), Gas Market Lessons from the 2022-2023 Energy Crisis, IEA, Paris, https://www.iea.org/reports/gas-market-lessons-from-the-2022-2023-energy-crisis.

IEA (2025c), The Value of Demand Flexibility, IEA, Paris, https://www.iea.org/reports/the-value-of-demand-flexibility.

IEA (2024), More efficient and flexible buildings are key to clean energy transitions, IEA, Paris, https://www.iea.org/commentaries/more-efficient-and-flexible-buildings-are-key-to-clean-energy-transition.

IEA (2022), Climate Resilience for Energy Security, IEA, Paris, https://www.iea.org/reports/climate-resilience-for-energy-security.

Lavappa, P. and J. Kneifel (2015), Energy Price Indices and Discount Factors for Life-Cycle Cost Analysis – 2015 Annual Supplement to NIST Handbook 135, National Institute of Standards and Technology, https://doi.org/10.6028/nist.ir.85-3273-30.

Monaco, R. et al. (2024), “Digitalization of power distribution grids: Barrier analysis, ranking and policy recommendations”, Energy Policy, Vol. 188, p. 114083, https://doi.org/10.1016/j.enpol.2024.114083;

OECD (2026a), Foundations for Growth and Competitiveness 2026, OECD Publishing, Paris, https://doi.org/10.1787/40a7532f-en.

OECD (2026b), OECD Economic Outlook, Interim Report March 2026: Testing Resilience, OECD Publishing, Paris, https://doi.org/10.1787/d4623013-en.

OECD (2026c), The Energy Crisis and Energy Efficiency: What countries are doing to promote energy efficiency and where gaps remain, https://www.oecd.org/en/data/insights/data-explainers/2026/05/the-energy-crisis-and-energy-efficiency-what-countries-are-doing-to-promote-energy-efficiency-and-where-gaps-remain.html.

OECD (2025a), Diagnostic Toolkit for Reducing Regulatory Barriers to Solar, Wind and Pumped Hydro Storage in the European Union: Empowering Policymakers at National, Regional, and Local Levels, OECD Publishing, Paris, https://doi.org/10.1787/15f4aed4-en.

OECD (2025b), Economic Security in a Changing World, New Approaches to Economic Challenges, OECD Publishing, Paris, https://doi.org/10.1787/4eac89c7-en.

OECD (2025c), Managing Emerging Critical Risks: Case Studies and Cross-Country Synthesis Report, OECD Publishing, Paris, https://doi.org/10.1787/1f9858ea-en.

OECD (2025d), OECD Economic Outlook, Volume 2025 Issue 2: Resilient Growth but with Increasing Fragilities, OECD Publishing, Paris, https://doi.org/10.1787/9f653ca1-en.

OECD (2025e), OECD Inventory of Support Measures for Fossil Fuels 2025: Policy Trends, OECD Publishing, Paris, https://doi.org/10.1787/6fff0cb7-en.

OECD (2024a), “Governing with Artificial Intelligence: Are governments ready?”, OECD Artificial Intelligence Papers, No. 20, OECD Publishing, Paris, https://doi.org/10.1787/26324bc2-en.

OECD (2024b), Modernising Access to Social Protection: Strategies, Technologies and Data Advances in OECD Countries, OECD Publishing, Paris, https://doi.org/10.1787/af31746d-en.

OECD (2023), Brick by Brick (Volume 2): Better Housing Policies in the Post-COVID-19 Era, OECD Publishing, Paris, https://doi.org/10.1787/e91cb19d-en.

OECD (2019), “Using digital technologies to improve the design and enforcement of public policies”, OECD Digital Economy Papers, No. 274, OECD Publishing, Paris, https://doi.org/10.1787/99b9ba70-en.

OECD/IEA (2005), Saving oil in a hurry, OECD/IEA, https://www.oecd.org/content/dam/oecd/en/publications/reports/2005/04/saving-oil-in-a-hurry_g1gh5691/9789264109421-en.pdf.

Thayer, D. et al. (2016), Is Behavioral Energy Efficiency and Demand Response Really Better Together?, ACEEE Summer Study on Energy Efficiency in Buildings, https://www.aceee.org/files/proceedings/2016/data/papers/2_1222.pdf.

World Bank (2025), Electricity Distribution Improvement and Strengthening through Commercialization and Transformation, https://documents1.worldbank.org/curated/en/099042825213541541/pdf/BOSIB-c6777fb6-6213-45dd-91ca-8c8e0e03c232.pdf.