## Response to the 2022 Energy Crisis (sipea2023017)

## Source details

**Canonical URL:** [Response to the 2022 Energy Crisis (sipea2023017)](https://www.imf.org/-/media/files/publications/selected-issues-papers/2023/english/sipea2023017.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/selected-issues-papers/2023/english/sipea2023017.pdf.md)
- [Structured JSON version](/-/media/files/publications/selected-issues-papers/2023/english/sipea2023017.pdf.json)

---

### Background and context
- Global mitigation benchmarks: cut CO2 by 50 and other GHGs by 25 percent by 2030 compared to 2019 levels; rapid decline to net zero near mid-century (IPCC 2021).
- Over 85 countries, representing around 80 percent of current emissions, have communicated ‘net-zero’ targets by mid-century.
- Under current policies, global emissions expected to increase significantly to 2030 despite higher energy prices and the Covid-19 slowdown.
- Belgium climate risks and emissions:
  - Annual economic costs of heat-related productivity/health losses and flooding projected up to one percent of GDP by 2050.
  - Belgium’s emissions in 2019: 19 percent below 2005 levels.
  - 2019 sectoral shares: industry 34 percent, transportation 21 percent, buildings 20 percent, power generation 13 percent, agriculture 8 percent.
  - Sectoral changes since 2005: power generation emissions fell by 38 percent; industry by 21 percent; buildings by 21 percent. Transportation and agriculture emissions remained relatively constant.
  - Electricity generation in 2019: nuclear 45 percent, natural gas 27 percent, wind 10 percent, remainder solar/bioenergy/coal.
  - Belgium committed to phase out nuclear power by 2035 (one reactor already decommissioned).
- Under current policies (no tightening) Belgium’s total emissions are estimated to increase to four percent above 2019 levels by 2030, driven by higher power-sector emissions-intensity as nuclear is phased out and natural gas generation rises (partly offset by increased wind generation).
- Energy-intensity of GDP expected to fall due to efficiency improvements from higher natural gas and oil prices and newer technology.

### Projected gaps vs. targets (selected exact entries)
- EU Effort Sharing Regulation (ESR) target: reduce non-ETS emissions to 47 percent below 2005 levels by 2030.
- Projected decline by 2030 under current policies: 11 percent below 2005 levels.
- NECP additional policies: overall emissions reduced by 28 percent (non-ETS: -35 percent; ETS: -22 percent) but gap to ESR remains.
- Selected table entries reproduced exactly:
  - ETS sectors, emissions: 2019 = -27% (relative to 2005). CPAT: -20% (WEM). EEA: -14% (WEM), -19% (WAM).
  - Power, emissions: 2019 = 38% (relative to 2005). CPAT: -10% (WEM). EEA: 4% (WEM), -4% (WAM)1/.
  - Industry, emissions: 2019 = -21%. NECP: -33% in 2030, -79% in 2050 for non-ETS industry. CPAT: -24% (WEM). EEA: -25% (WEM), -29% (WAM).
  - Non-ETS sectors, emissions: 2019 = 11%. ESR: -47% in 2030. NECP: -35% in 2030, -85 to 87% in 2050; Flanders (-40% in 2030, -85% in 2050); Wallonia (-35% in 2030); Brussels (-40% in 2030). CPAT: -10% (WEM). EEA: -10% (WEM), -35% (WAM).
  - Transport, emissions: 2019 = 3%. EU: -55% reduction by 2030 relative to 20212/. NECP: -27% in 2030, -100% in 2050; Wallonia (-24% by 2030). CPAT: -6% (WEM). EEA: 4% (WEM), -32% (WAM).
  - Buildings, emissions: 2019 = -21%. NECP: no explicit federal target; Wallonia (-29% by 2030). CPAT: -18% (WEM). EEA: -24% (WEM), -44% (WAM).
  - Agriculture, emissions: 2019 = -4%. NECP: -23% in 2030, -47% in 2050. CPAT: 1% (WEM). EEA: -8% (WEM), -18% (WAM).
  - Economy-wide, emissions: 2019 = -19%. NECP: no explicit federal target, Wallonia (-30% by 2020 and -95% by 2050, compared to 1990), Brussels (-40% in 2030, ‘approach net zero’ by 2050), Flanders (no explicit target). CPAT: -16% (WEM). EEA: -13% (WEM), -28% (WAM).
  - Power, renewable share: 2019 = 9.9%. EU: 25% of gross final energy consumption by 2030. NECP: 17.5% of energy consumption by 2030. CPAT: 16.7% (WEM).
- Notes:
  - 1/ Power in the EEA analysis includes petroleum refining and manufacture of solid fuels, while these activities are included in industry for the IMF analysis.
  - 2/ Reflects proposals in the Fit for 55 package.
  - 3/ WAM = ‘with additional measures’; WEM = ‘with existing measures’.

### EU and Belgian policy framework (key points)
- EU ETS:
  - Covers large emissions sources from energy, industry, and within-EU aviation; currently covers about 41 percent of total EU GHGs.
  - Cap declines by 2.2 percent each year presently; expected to increase to 4.3–4.4 percent annually to comply with ‘Fit-for-55’.
  - EU allowance prices rose to around €80 per tonne in September 2021 and have since hovered around this level.
  - Market Stability Reserve (MSR) withdraws allowances when banked allowances exceed a threshold.
  - EITE industries historically granted free allowances; expected to be phased out as Carbon Border Adjustment Mechanism (CBAM) is introduced.
- New EU ETS for road transport, buildings, and other sectors (provisional agreement December 2022):
  - Introduction in 2027 (or 2028 if energy prices are ‘exceptionally high’).
  - Price of allowances not exceeding €45 per tonne.
  - Allowance cap set to an annual, linear decline of 5 percent per year from 2024 emission levels.
  - Exemption for member states with a national-level carbon tax equal to or above the ETS allowance price.
  - Portion of ETS revenues to support vulnerable households and small businesses through the Social Climate Fund.
- Other EU directives: ETD, Renewable Energy Directive, Energy Efficiency Directive, Energy Performance of Buildings Directive, CO2 emissions performance standards for road transportation.
- Division of competencies in Belgium:
  - Regions: electricity and gas distribution, district heating, onshore renewables, waste, public transport, urban/rural planning, agriculture, building/vehicle efficiency (except federal buildings), vehicle registration.
  - Federal: tax policy, offshore renewables, nuclear energy, energy R&D, energy pricing policy, national rail.
  - Coordination forums (e.g., National Climate Commission) exist but lack binding coordination.

### Analysis focus and modelling methodology
- Evaluation of fiscal policies: economy-wide carbon pricing; sectoral policies (feebates); policies to reduce relative price of electricity; energy-related social protection; policy response to the energy crisis.
- Modelling uses IMF’s Carbon Pricing Assessment Tool (CPAT), developed jointly by IMF and World Bank staff. Results broadly align with EEA (2022) and Leuven (2022).
- CPAT projections cited as CPAT: -X% (WEM) or similar throughout the assessment.

### Key federal price-based policy recommendations (summary)
- Implement an economy-wide carbon tax for non-ETS sectors (transportation, buildings, agriculture, and excluded industrial emissions):
  - Tax should progressively increase to around €100 per tonne by 2030.
  - Phase in as international energy prices fall and ahead of the EU-wide ETS for buildings and transportation.
  - Revenues used to compensate vulnerable households and trade-exposed firms, and increase productivity; direct support for firms phased out as CBAM is phased-in.
  - Extend tax to non-carbon GHGs (e.g., methane) if administrative and political costs manageable.
- Introduce a domestic price floor for ETS sectors equal to the non-ETS sector carbon tax to increase investor certainty and equalize emissions prices economy-wide.
- Implement feebates for power and transportation to promote decarbonization as nuclear is phased out and accelerate vehicle electrification; couple transport feebates with removal of the company car regime.
- Reduce fees and levies on electricity consumption that are not directly related to delivering electricity to end-users; finance such measures through the general budget.
- Transition from the current social tariff to fiscally-sustainable, income-based social protection with benefits phasing out as income or wealth rises.
- Promote EU-level dialogue to harmonize ETS prices and include sectors under a single trading scheme.

---

### Excises, subsidies, and effective carbon prices

### Current excise, VAT, and tax structure (key figures)
- Excises on unleaded petrol and diesel: €193 and €160 per tonne of CO2 (€0.45 per liter of fuel), respectively, when used for transportation.
- Diesel used for non-transportation excise: significantly reduced to around €0.02 per liter.
- Refund on excise for professional transportation diesel: the refund was €0.08 per liter most recently.
- Natural gas and coal excises have maximums of €0.31 and €0.37 per GJ (equivalent to around €4.3 per tonne of CO2) and are exempt in many cases.
- Electricity excise: €0.50 per MWh (commercial use) and €2.93 per MWh (residential use).
- Biofuel content of blended fuels taxed at fossil fuel rate.
- VAT: standard rate 21 percent; reduced to 6 percent for natural gas, district heating, and electricity (reduced rates currently temporary but may be made permanent).

### Fiscal cost of fossil-fuel subsidies and distributional shares
- Reduced excise duties (natural gas and heating oil) bulk of subsidies at €10.5 billion in 2019 (around 2.2 percent of GDP).
- Preferential tax treatment for company cars and road fuel subsidies: around €2.5 billion and growing.
- Social tariff coverage: around 20 percent of households.
- Approximate subsidy allocation shares:
  - Heating and electricity: around 40 percent.
  - Transportation: around 30 percent.
  - Industry: around 25 percent.
- Although total subsidies decreased from 2015 to 2019, recent policy decisions and international price developments likely reverse the trend.

### Effective carbon tax rates and sectoral gaps
- Road transportation and power sector carbon taxes: broadly in line with EU average and slightly above OECD average, but below neighboring countries and below levels needed for emissions-reductions targets.
- Industry, buildings, and agricultural/fish/forestry sectors: particularly low effective carbon tax rates relative to EU, OECD, and neighbors.
- Drivers of low prices:
  - Industry: low natural gas and coal excise rates; about 20 percent of industrial sector emissions not covered by the ETS.
  - Buildings and agriculture: reduced excise rates for diesel (when used for heating) and natural gas.

### Response to the 2022 energy crisis (measures and implications)
- Energy shares of total household consumption for the lowest 20 percent:
  - Electricity: 3.9 percent
  - Natural gas: 2.1 percent
  - Oil products: 3.5 percent
- Federal and regional measures cost about 1.7 percent of GDP and include:
  - Reduced VAT for natural gas, electricity, and district heating.
  - Reduced excises on petrol and diesel.
  - Additional cash transfers.
  - Expanded eligibility for the social tariff.
- Targeting and distortion:
  - Roughly 1/3 of measures are targeted and distort price signals.
  - Roughly 2/3 are untargeted and not price distortionary.
- Government plans a flexible excise rate that declines for base consumption as natural gas and electricity prices rise; medium-term objective to improve targeting via matching tax, household size, and third-party data.

### Rationale and design principles for economy-wide carbon pricing
- Advantages:
  - Across-the-board incentives to reduce energy consumption and shift to cleaner fuels.
  - Equalizes marginal abatement costs across the economy (minimizes mitigation costs).
  - Mobilizes revenue for vulnerable households/firms or productivity-enhancing uses.
  - Domestic co-benefits: reduced local air pollution mortality.
  - Administratively straightforward, building on existing excise system.
- Design principles:
  - Cover emissions comprehensively.
  - Establish predictable prices and align stringency with mitigation goals; announce price path well in advance.
  - Couple non-ETS carbon price with a price floor for ETS sectors.
  - Use revenues to support vulnerable households/firms and improve productivity.

### International comparators (selected effective tax prices, € per tonne)
- Austria: 35 (2023) to 55 (2025) — Heat, Transport (price fixed until 2026, then to ETS).
- Denmark: 47 (2022) to 101 (2030) — All (lower rate for ETS sectors).
- Finland: 77 (2022, Transport), 53 (2022, other) — Heat, Transport, Industry.
- France: 45 (2022) — Heat, Transport, Industry (ETS sectors exempt).
- Germany: 30 (2022) to 55 (2025) — Heat, Transport (fixed until 2026 then subject to ETS).
- Ireland: 34-41 (2022) to 100 (2030) — Heat, Transport, Industry.
- Portugal: Previous years’ avg. ETS price — Heat, Transport, Industry.
- Sweden: 110 — Heat, Transport.

---

### Modelling results and carbon-price scenarios (CPAT outputs)

### General findings (selected)
- A carbon price progressively reaching €75 per tonne:
  - Reduces overall emissions to 25 percent below 2005 levels in 2030 (or 12 percent below a Business-as-Usual (BAU) scenario).
  - Raises revenue of 0.6 percent of GDP.
  - Economic efficiency costs of 0.2 percent of GDP; welfare benefits net of consumer welfare losses shown as small.

### Modeled sectoral emissions outcomes in 2030 (relative to 2005 levels) at various carbon prices (selected figures)
- All (aggregate):
  - Carbon price 0: -19
  - 25: -15
  - 50: -20
  - 75: -22
  - 100: -25
  - 125: -28
- Non-ETS:
  - 0: -47
  - 25: -11
  - 50: -10
  - 75: -14
  - 100: -17
  - 125: -19
- Transport:
  - 0: -27
  - 25: -3
  - 50: -6
  - 75: -8
  - 100: -10
  - 125: -12
- Buildings:
  - 0: -21
  - 25: -18
  - 50: -21
  - 75: -25
  - 100: -28
  - 125: -31
- Agriculture:
  - 0: -23
  - 25: -4
  - 50: 1
  - 75: -8
  - 100: -9
  - 125: -10
- ETS (aggregate):
  - 0: -21
  - 25: -18
  - 50: -23
  - 75: -27
  - 100: -30
  - 125: -33
- Power:
  - 0: -39
  - 25: -10
  - 50: -11
  - 75: -11
  - 100: -11
  - 125: -11
- Industry:
  - 0: -21
  - 25: -24
  - 50: -30
  - 75: -34
  - 100: -39
  - 125: -44

### Other modeled outcomes (selected values)
- Renewable share:
  - 0 price: 17.5
  - 25: 9.6
  - 50: 17
  - 75: 18
  - 100: 18
  - 125: 19
- Revenue raised (% GDP):
  - 25: 0.2
  - 50: 0.3
  - 75: 0.6
  - 100: 0.8
  - 125: 1.1
- Efficiency costs (% of GDP):
  - 25: -0.1
  - 50: -0.2
  - 75: -0.3
  - 100: -0.4
  - 125: -0.6
- Welfare benefits (% GDP) (net of consumer welfare losses):
  - 75: -0.1
  - 100: -0.2
- Deaths averted (annual, selected):
  - 25: 50
  - 50: 72
  - 75: 93
  - 100: 114
  - 125: 132

### Limitations and sectoral notes
- Power sector reductions limited by Capacity Remuneration Mechanism commitments and tax applying only above EU ETS prices.
- Agricultural emissions mainly methane and nitrous oxide; price responsiveness at low carbon prices modest and deep abatement costly.
- Even €125 per tonne alone does not achieve mitigation targets—sectoral reinforcing policies required.

---

### Distributional impacts, revenue recycling, and compensation

### Distributional impacts (headline figures)
- Indexation effect: current wage and social benefit indexation automatically protects the average household that uses electricity and/or natural gas for heating since higher energy taxes increase inflation.
- Ignoring indexation:
  - Loss in purchasing power for poorer households: 1.8 percent of total consumption.
  - Loss in purchasing power for wealthier households: 1.5 percent of total consumption.
  - Losses higher for rural vs. urban households across all income deciles.
- After revenue recycling and targeted compensation:
  - Around 25 percent of carbon pricing revenue needed to fully compensate households in the lower four income deciles.
  - Leaves around 0.6 percent of GDP in revenue for other objectives (e.g., reducing income taxes).
- Energy-intensive firms:
  - Would see a 1–2   percent increase in input costs due to higher electricity prices.
  - Firms receive an income tax deduction for increased input costs, resulting in a 25 percent (Belgium CIT rate) lower change in post-tax costs.

### Revenue recycling and compensation recommendations
- Use portion of carbon tax revenues to compensate vulnerable households via income-based support phased out with rising income or wealth.
- Design principles:
  - Income-based cash transfers (not conditional on energy expenditure) to preserve price signals.
  - Phase out support gradually with rising income to avoid unemployment traps.
  - Do not condition support on energy expenditure to avoid subsidizing inefficiency.
- Firm support:
  - Consider reductions in corporate income tax, investments in productive public infrastructure and R&D, and/or rebates based on output (not energy inputs).
  - Rebates based on output to be phased out as CBAM is phased in to ensure WTO compatibility.

### Recommended carbon pricing sequencing and design
- Phase in carbon tax for non-ETS sectors through excise regime, reaching €100 per tonne by 2030.
- Introduce domestic carbon pricing prior to planned EU-wide ETS for buildings and transport (proposed 2027).
- Price path design: announce in advance with small annual increases (e.g., €10 per tonne each year); adjust for inflation.
- Administrative approach: phase in via excise carbon tax element to avoid volatility, build on existing excise regime, be compatible with future EU ETS, and avoid leakage.
- Consider extending tax to non-carbon GHGs (e.g., methane) if manageable; proxy taxes as medium-term solution.

### Price floor for ETS sectors
- Consider price floor aligned with non-ETS carbon tax to provide investor certainty and equalize abatement costs.
- Implementation could mirror U.K. Carbon Price Floor: national variable tax set for three years in advance equal to the difference between an exogenous target price and projected EU ETS price.
- Netherlands example: price floor levels likely below EU ETS and thus not binding.

---

### Sectoral policies: feebates, electricity pricing, and targeted measures

### Feebates and general sectoral guidance
- Feebates: revenue neutral tax-subsidy schemes financing subsidies to clean activities while taxing dirty ones; more flexible and potentially more cost-effective than emissions-rate regulations.
- Feebates do not raise average prices for consumers or costs for firms and can complement regulations by rewarding beyond-standard performance.
- Design guidance: be flexible, technology-neutral, accept differing implicit carbon prices across sectors during transition.

### Power generation: context, feebate design, and impacts
- Nuclear: 45 percent of generation; legislated phase-out by 2035. First reactor (17 percent of nuclear capacity) decommissioned in 2022; others planned for 2023, 2025, and 2035.
- Capacity Remuneration Mechanism (CRM):
  - First CRM auction (2021) awarded 4.4 GW for delivery in 2025: 3.6 GW from natural gas and rest primarily co-generation/demand-side response.
  - Annual CRM auctions planned to ensure capacity one and four years ahead.
- Offshore wind: federal auction to develop 3.15 to 3.5 GW in Princess Elisabeth Island; increasing interconnections to import clean electricity from Norway and Denmark.
- Feebate formula (generator-level):
  - {CO2 price} × {CO2/kWh – industry-wide average CO2/kWh} × {electricity generation}
- Practical implementation:
  - Build off EU ETS monitoring; exempt nuclear; smooth industry-wide average with five-year average; apply through refundable credits/taxes or direct excise/subsidy.
- Illustrative 2022 impacts with €100 per tonne:
  - Fee equivalent to 3 cents/kWh for natural gas.
  - Subsidy of 1.5 cents/kWh for renewables.
- Considerations:
  - Feebates reduce electricity emissions-intensity relative to BAU but national reductions may reduce ETS price and cause leakage if MSR not operating.
  - CRM maintains capacity while feebate promotes low-carbon generation.

### Existing clean energy support critique
- Green Certificate Scheme:
  - Cost of certificates: €2.4 billion in 2020.
  - Contributed to growing debt for Elia, the transmission operator.
  - Regressive: wealthier households have larger rooftop solar shares.
  - Financed partly through surcharges on electricity bills, raising electricity prices relative to fossil fuels.
- Recommendation: reconsider or replace the Green Certificate Scheme; feebates could improve incentive efficiency without increasing prices.

### Reducing fees and levies on electricity to promote electrification
- Electricity price relative to heating oil, natural gas, diesel, gasoline higher in Belgium than neighbors.
- Energy portion of electricity costs on par with France, Germany, Netherlands; differences driven by network recovery charges, levies, taxes.
- Recommendation: gradually remove charges not directly related to delivering electricity and finance them from general budget to better align electricity prices with generation costs; pair with enhanced renewable and efficiency policies.

### Industry
- Industrial emissions: 35 percent of total.
- Main sub-sectors: chemicals, refining of oil and coke, manufacture of non-metallic minerals and metals.
- About 20 percent of industrial emissions outside ETS due to installations below 25,000 tonnes CO2/year threshold and biomass exclusions.
- Feebate example for industry:
  - {CO2 price} × {CO2/output – industry-wide average CO2/output} × {firm output}
- Design trade-offs:
  - Applied to fuel combustion and process emissions; avoids first-order burden on average producer; less efficient since output not reduced.
  - Can build off EU ETS monitoring; pivot-point exogenous or sub-sector specific; permit trading across sub-sectors to avoid leakage.

### Buildings
- Buildings: 21 percent of total emissions; residential buildings 75 percent of building emissions.
- Energy source shares (Belgium vs EU):
  - Oil: 32 percent (Belgium) vs. 21 percent (EU)
  - Renewables: 8 percent (Belgium) vs. 20 percent (EU)
  - Electricity: 20 percent (Belgium) vs. 25 percent (EU)
- Heating: 83 percent of building energy use for space and water heating.
- Biomass: 7 percent of household heating in 2020.
- Building stock: 80 percent built prior to energy standards across Belgium; 94 percent in Brussels.
- 45 percent of households not connected to natural gas grid.
- Heating oil shares (houses with oil boilers): 16 percent (Brussels), 24 percent (Flanders), 50 percent (Wallonia).
- Distributional note: lowest income quartile spent roughly eight percent of total consumption on natural gas, heating oil, electricity vs. 3.6 percent for highest quartile in 2018.
- Heating economics (illustration assumptions for Figure 12):
  - Air-source heat pumps represent 95 percent of existing heat pumps.
  - Discount rate 7 percent; interest rate 4.5 percent; grant covers 70 percent of equipment and installation costs; systems last 20 years.
  - Fuel prices pre-crisis: EUR 0.054 and 0.034 per kWh (natural gas; without and with social tariff), EUR 0.060 and 0.045 per kWh (heating oil), EUR 0.290 and 0.189 per kWh (electricity).
- Impacts:
  - Carbon tax raises all-in natural gas heating costs by about 28 percent, 38 percent for oil, and 4 percent for a heat pump.
  - Removing fees/levies on electricity consumption (outside network costs) reduces heat pump costs by 20 percent.
  - Efficient pricing mix: carbon tax, no social tariff, and network fees/levies aligned with neighbors would align fossil-fuel and electricity heating costs.
  - Grants and no-interest loans can reduce upfront costs; social tariff currently makes natural gas heating significantly cheaper than heat pumps even with full-capital subsidy.
- Renovation rate:
  - Current renovation rate: 1 percent annually; needed: 2.5 to 3 percent to meet emissions targets.
- Additional proposals:
  - Feebates for household heating and appliances; property-tax linkage to energy use per square meter or EPC scores (administration/regressivity caveats).
  - Federal feebate options limited by information and competency constraints.

### Transportation
- Road transport emissions: passenger vehicles 75 percent; road freight majority of remainder.
- Car ownership: around one car for every two people.
- EV and plug-in hybrid share of new car sales: 5 percent in 2021.
- EV charging infrastructure: 8.8 charging points per 100 km of road in Belgium; Germany 26; Netherlands 64. Less than 10 percent of Belgian stations are ‘fast’.
- Regional registration and annual circulation taxes: Flanders and Walloon registration taxes vary by vehicle CO2 emissions; Brussels considering similar policy.
- Federal income tax credit: 15 percent of purchase price up to €5,150 for EVs purchased by individuals available until 2024.
- Heavy commercial vehicle scheme: charges per kilometer.
- Vehicle tax design issues:
  - Registration/annual taxes expressed on a lifetime basis apply modest taxes on high-emission vehicles.
  - Relative price of low-emission non-electric vehicles not decreased relative to average-emission vehicles in Brussels or Wallonia.
  - Fleetwide EU emission standards can offset demand shifts to low-emission vehicles.
- Proposed vehicle feebate formula:
  - {CO2 price} × {the vehicle’s CO2/km – the industry average CO2/km} × {the average lifetime km driven per vehicle}
- Feebate illustration: at €400 per tonne CO2:
  - Same EV subsidy as present.
  - Tax of €3,000 to a vehicle with 200 g CO2/km (an increase of around €1,000 for Brussels and Wallonia).
- Advantages:
  - Incentivizes low-emission non-electric vehicles.
  - Could replace company car regime; can be introduced unilaterally via PIT credits, ideally harmonized with regions.
  - Subsidies decline as average fleet emissions fall; revenue neutrality maintained with fleet updates.
- Company car regime:
  - Creates incentive to accept a car over cash; recent reforms: restrict company car eligibility to EVs by 2026 and introduce mobility budget.
  - Mobility budget take-up low.
  - Regime is regressive, contributes to congestion (~12 percent of total vehicle fleet), and fiscally costly (~1 percent of GDP).
  - Recommendation: eliminate company car regime and replace with feebate policy.

### Broader transport pricing and externalities
- Externalities: congestion, road damage, accidents.
- Distance-, location-, and time-based charges best-suited to address externalities, especially as EV penetration rises.
- Existing heavy duty truck distance-based charge examples road damage pricing.
- Administrative feasibility of distance/time/location charges improving with GPS tools.
- Congestion charges within regional competencies; should be coupled with reductions in road fuel taxes and potentially vehicle fees.
- PAYD vehicle insurance promoted via fiscal incentives to better price accident externalities.
- Transition to distance/time-based pricing shifts revenue from federal to regional governments but maintains robust revenue base.

---

### Implementation sequencing, EU dialogue, and expected impacts

### Recommended reform pathway (practical sequencing)
- Phase in carbon price on non-ETS activities via excise regime; announce predictable price path (e.g., €10 per tonne increases annually).
- Introduce domestic carbon price floor for ETS-covered sectors aligned with non-ETS carbon tax.
- Implement reforms as international energy prices fall to avoid excessive post-tax prices and restore pre-crisis relative prices.
- Shift energy taxation toward fossil fuels and away from electricity; recycle revenues to support vulnerable households/firms and reduce distortionary taxes or fund productivity measures.
- Complement carbon pricing with sectoral policies (feebates, targeted subsidies/regulations) where pricing alone is insufficient.

### EU-level dialogues and reforms suggested
- Promote EU dialogue to:
  - Extend EU ETS so power, industry, transport, and buildings subject to one aggregate cap with common emissions price.
  - Extend member-state ability to re-allocate emissions reductions across sectors to lower mitigation costs.
  - Extend ability to trade ESR quotas between states to promote cost-efficiencies (political agreement reached in late November; details not yet clear).
  - Establish exogenous and escalating price floor for both EU ETSs to provide investment signals and allow national measures to lower EU emissions.

### Expected outcomes of full recommendation set
- Reduce emissions to at least 20 percent below business-as-usual levels by 2030 (33 percent below 2005 levels).
- Raise substantial revenue (at least 0.6 percent of GDP) to support vulnerable households and firms and increase productivity.

### Complementary fiscal measures from annexed studies (selected)
- Mobility budget: minimum € 3,000 per year and maximum of the lesser of one fifth of gross remuneration and €16,000 per year.
- Leuven (2022) findings:
  - Carbon-tax-based energy taxation would reduce transportation and building emissions by 11.3 percent and 14.1 percent, respectively, by 2030 in a €100 per tonne scenario.
  - Revenue: €2.7 billion by 2030.
  - Recommended package: carbon price, align excise taxes with revised Energy Tax Directive, recycle revenues via direct income transfers to vulnerable households.
- NBB (Cornille et al 2021) findings:
  - Effective carbon tax rates low for industry and buildings; link tax partly to carbon content.
  - Green Certificate Scheme not cost-effective; promote renewables via carbon pricing and capital subsidies.
  - Reducing electricity price relative to natural gas supports heat pump investment.
  - Environmental tax revenues should not be earmarked but spent across government objectives.
- National Carbon Pricing Debate (Climact 2018) principles:
  - Revenue neutrality; credible long-term investment signal; complementary policies required.
  - Sector-specific options: buildings compensation, transport smart pricing, industry phased approaches, agriculture targeted coverage.

### Modelling approach and key parameters (appendix notes)
- CPAT dependencies: energy price projections (IMF WEO + IEA WEO Stated Policies), GDP from IMF WEO, price/income elasticities, autonomous energy-efficiency improvements.
- Selected elasticity parameters (Belgium, key modelling parameters):
  - Natural gas elasticity w.r.t income: Transport 0.5, Residential 0.3, Industry 1.0; w.r.t price: Transport -0.4, Residential -0.3, Industry -0.4.
  - Gasoline income elasticity: Transport 0.7, Industry 0.5; price elasticity: Transport -0.3, Residential -0.4, Industry -0.3.
  - Diesel income elasticity: Transport 0.6, Residential 0.5, Industry 0.5; price elasticity: Transport -0.2, Residential -0.2, Industry -0.6.
  - Electricity income elasticity: Transport 1.2, Residential 0.8, Industry 0.8; price elasticity: Transport -0.1, Residential -0.2, Industry -0.2.
  - Coal (industry) income elasticity 0.5; price elasticity -0.2.

*Italic source attribution: Prepared by Nate Vernon (IMF Fiscal Affairs Department), with comments from Mark Horton, Ian Parry, the National Bank of Belgium, and Belgium’s Federal Planning Bureau; February 10, 2023 (selected excerpts from sipea2023017).*

### 1. Response to the 2022 Energy Crisis _____________________________________________ 10

### 1. Response to the 2022 Energy Crisis

### Background
- Limiting global warming to 1.5 to 2°C requires cutting CO2 by 50 and other GHGs by 25 percent by 2030 compared to 2019 levels, followed by rapid decline to net zero near mid-century (IPCC 2021).
- Over 85 countries, representing around 80 percent of current emissions, have communicated ‘net-zero’ targets by mid-century.
- Under current policies, global emissions are expected to increase significantly to 2030 despite higher energy prices and the Covid-19-induced economic slowdown.
- Belgium’s primary climate-related economic risks: heat-related productivity and health losses and flooding, with annual economic costs projected to be up to one percent of GDP by 2050.
- Belgium’s emissions in 2019 were 19 percent below 2005 levels.
  - Sectoral shares of 2019 emissions: industry 34 percent, transportation 21 percent, buildings 20 percent, power generation 13 percent, agriculture 8 percent.
  - Sectoral changes since 2005: power generation emissions fell by 38 percent; industry by 21 percent; buildings by 21 percent. Transportation and agriculture emissions remained relatively constant.
- Belgium’s nuclear generation: 45 percent of electricity in 2019; natural gas 27 percent; wind 10 percent; remainder split among solar, bioenergy, and coal.
- Belgium committed to phase out nuclear power generation by 2035 (one reactor already decommissioned).
- Under current policies (no tightening), Belgium’s total emissions are estimated to increase to four percent above 2019 levels by 2030, driven by higher power-sector emissions-intensity as nuclear is phased out and natural gas generation rises (partly offset by increased wind generation).
- Energy-intensity of GDP is expected to fall due to energy efficiency improvements from higher natural gas and oil prices and newer, more efficient technology.

### Projected Gaps vs. Targets
- Under the EU Effort Sharing Regulation (ESR), Belgium must reduce non-ETS emissions to 47 percent below 2005 levels by 2030.
- Projected decline by 2030 under current policies: 11 percent below 2005 levels.
- Additional policies in Belgium’s National Energy and Climate Plan (NECP) reduce overall emissions by 28 percent (non-ETS: -35 percent; ETS: -22 percent) but do not close the gap to ESR requirements.
- Table 1 summary (selected entries reproduced exactly as in source):
  - ETS sectors, emissions: 2019 = -27% (relative to 2005). CPAT: -20% (WEM). EEA: -14% (WEM), -19% (WAM).
  - Power, emissions: 2019 = 38% (relative to 2005). CPAT: -10% (WEM). EEA: 4% (WEM), -4% (WAM)1/.
  - Industry, emissions: 2019 = -21%. NECP: -33% in 2030, -79% in 2050 for non-ETS industry. CPAT: -24% (WEM). EEA: -25% (WEM), -29% (WAM).
  - Non-ETS sectors, emissions: 2019 = 11%. ESR: -47% in 2030. NECP: -35% in 2030, -85 to 87% in 2050; Flanders (-40% in 2030, -85% in 2050); Wallonia (-35% in 2030); Brussels (-40% in 2030). CPAT: -10% (WEM). EEA: -10% (WEM), -35% (WAM).
  - Transport, emissions: 2019 = 3%. EU: -55% reduction by 2030 relative to 20212/. NECP: -27% in 2030, -100% in 2050; Wallonia (-24% by 2030). CPAT: -6% (WEM). EEA: 4% (WEM), -32% (WAM).
  - Buildings, emissions: 2019 = -21%. NECP: no explicit federal target; Wallonia (-29% by 2030). CPAT: -18% (WEM). EEA: -24% (WEM), -44% (WAM).
  - Agriculture, emissions: 2019 = -4%. NECP: -23% in 2030, -47% in 2050. CPAT: 1% (WEM). EEA: -8% (WEM), -18% (WAM).
  - Economy-wide, emissions: 2019 = -19%. NECP: no explicit federal target, Wallonia (-30% by 2020 and -95% by 2050, compared to 1990), Brussels (-40% in 2030, ‘approach net zero’ by 2050), Flanders (no explicit target). CPAT: -16% (WEM). EEA: -13% (WEM), -28% (WAM).
  - Power, renewable share: 2019 = 9.9%. EU: 25% of gross final energy consumption by 2030. NECP: 17.5% of energy consumption by 2030. CPAT: 16.7% (WEM).
- Notes:
  - 1/ Power in the EEA analysis includes petroleum refining and manufacture of solid fuels, while these activities are included in industry for the IMF analysis.
  - 2/ Reflects proposals in the Fit for 55 package.
  - 3/ WAM = ‘with additional measures’; WEM = ‘with existing measures’.

### EU and Belgian Policies
- EU Emissions Trading Scheme (ETS):
  - Covers large emissions sources from energy, industry, and within-EU aviation; currently covers about 41 percent of total EU GHGs.
  - Cap declines by 2.2 percent each year presently; expected to increase to 4.3–4.4 percent annually to comply with ‘Fit-for-55’.
  - EU allowance prices rose to around €80 per tonne in September 2021 and have since hovered around this level.
  - Drawbacks: volatile emissions prices; overlapping national policies in ETS sectors can lead to full leakage (extra emissions in other EU countries via decline in ETS allowance price).
  - Market Stability Reserve (MSR) withdraws allowances when banked allowances exceed a threshold.
  - Energy intensive, trade exposed (EITE) industries historically granted free allowances; expected to be phased out as Carbon Border Adjustment Mechanism (CBAM) is introduced.
- New EU ETS for road transportation, buildings, and other sectors (provisional agreement December 2022):
  - Introduction in 2027 (or 2028 if energy prices are ‘exceptionally high’).
  - Price of allowances not exceeding €45 per tonne.
  - Allowance cap set to an annual, linear decline of 5 percent per year from 2024 emission levels.
  - Exemption for member states with a national-level carbon tax equal to or above the ETS allowance price.
  - Portion of ETS revenues to support vulnerable households and small businesses through the Social Climate Fund.
- Other EU directives and regulations relevant to Belgium:
  - Energy Taxation Directive (ETD): provides minimum excise duties on energy products and electricity.
  - Renewable Energy Directive: sets renewable energy targets for electricity, heating, cooling and transportation; legally binding as of 2021.
  - Energy Efficiency Directive: provides binding energy consumption reductions relative to a baseline.
  - Energy Performance of Buildings Directive: requires long-term renovation strategies and phases in minimum energy performance standards.
  - CO2 emissions performance standards for road transportation: progressively stringent; require all new passenger vehicles and vans to be zero-carbon by 2035.
- Division of competencies in Belgium:
  - Regions control: electricity and natural gas distribution (including tariffs), district heating networks, onshore renewable energy, waste management, public transportation, urban and rural planning, agriculture, building and vehicle efficiency requirements (except federal buildings), and vehicle registration.
  - Federal government controls: tax policy, offshore renewable energy, nuclear energy, energy R&D, energy pricing policy, and the national rail system.
  - Coordination forums exist (e.g., National Climate Commission) but do not ensure binding coordination.

### Analysis Focus and Methodology
- The paper evaluates fiscal policies to promote cost-effective emissions reductions: economy-wide carbon pricing; sectoral policies such as feebates; policies to reduce relative price of electricity to support electrification; and energy-related aspects of social protection, including the policy response to the energy crisis.
- Analysis builds on IMF, European Commission, OECD, IEA, Belgian institutions, and others.
- Modelling uses the IMF’s Carbon Pricing Assessment Tool (CPAT), developed jointly by IMF and World Bank staff. Results generally align with EEA (2022) and Leuven (2022).
- CPAT projections cited as CPAT: -X% (WEM) or similar throughout the assessment.

### Key Policy Recommendations (federal, price-based reforms)
- Implement an economy-wide carbon tax for non-ETS sectors (transportation, buildings, agriculture, and excluded industrial emissions):
  - Tax should progressively increase to around €100 per tonne by 2030.
  - Phase in as international energy prices fall and ahead of the EU-wide ETS for buildings and transportation.
  - Revenues used to compensate vulnerable households and trade-exposed firms, and increase productivity; direct support for firms phased out as CBAM is phased-in.
  - Extend tax to non-carbon greenhouse gases (e.g., methane) if administrative and political costs are manageable.
- Introduce a domestic price floor for sectors covered by the existing EU ETS (industry and power) equal to the non-ETS sector carbon tax to increase investor certainty and equalize emissions prices economy-wide.
- Implement feebates for power and transportation sectors to promote decarbonization as nuclear power is phased out and to accelerate electrification of vehicle fleet.
  - Transportation feebates should be coupled with removal of the company car regime.
  - Consider feebates for industry, building heating, and appliances to promote cost-efficient, deeper decarbonization and housing stock renovation.
- Reduce fees and levies on electricity consumption for charges not directly related to delivering electricity to end-users to allow for more cost-reflective electricity pricing and promote switching from fossil fuels to electricity-powered activities.
- Transition from the current social tariff to a fiscally-sustainable, income-based social protection system with benefit levels that phase out as total income or wealth increases to improve employment incentives and decarbonization.
- Promote EU-level dialogue to harmonize ETS prices and include all sectors under a single trading scheme.

_Italic source attribution: Prepared by Nate Vernon (IMF Fiscal Affairs Department), with comments from Mark Horton, Ian Parry, the National Bank of Belgium, and Belgium’s Federal Planning Bureau; February 10, 2023 (selected excerpts from sipea2023017)._

### 10.      Excises are Belgium’s main instruments to (indirectly) price carbon emissions, but

### Excises are Belgium’s main instruments to (indirectly) price carbon emissions, but current rates result in substantially different carbon prices across fuels and activities

### Current excise, VAT, and tax structure: key facts
- Excises on unleaded petrol and diesel are €193 and €160 per tonne of CO2 (€0.45 per liter of fuel), respectively, when used for transportation.
- Excise rates for diesel used for non-transportation (e.g., heating, commercial uses) are significantly reduced to around €0.02 per liter.
- A portion of the excise charged on transportation diesel for professional use is refunded (the refund was €0.08 per liter most recently).
- Natural gas and coal excises have maximums of €0.31 and €0.37 per GJ (equivalent to around €4.3 per tonne of CO2) and are exempt in many cases.
- Electricity excise is €0.50 per MWh (commercial use) and €2.93 per MWh (residential use).
- Biofuel content of blended fuels is taxed at the rate of the fossil fuel (e.g., diesel with biofuel is fully taxed at the diesel rate).
- VAT: standard rate is 21 percent but has been reduced to 6 percent for natural gas, district heating, and electricity (reduced VAT rates currently temporary but may be made permanent).

### Fiscal cost of fossil-fuel subsidies and distributional facts
- Reduced excise duties, especially for natural gas and heating oil, made up the bulk of subsidies at €10.5 billion in 2019 (around 2.2 percent of GDP).
- Preferential tax treatment for company cars and road fuel subsidies are significant at around €2.5 billion and growing.
- Social tariff coverage has expanded (now around 20 percent of households) and higher per-unit natural gas subsidy has increased its cost.
- Overall subsidy allocation (approximate shares):
  - Heating and electricity: around 40 percent of subsidies.
  - Transportation: around 30 percent of subsidies.
  - Industry: around 25 percent of subsidies.
- Although total subsidies decreased from 2015 to 2019, recent policy decisions (expanded social tariff coverage and reduced VAT on natural gas, district heating, and electricity) and international price developments are likely to reverse the recent trend.

### Effective carbon tax rates and sectoral gaps
- Combined effect of excise, VAT, and social protection systems yields:
  - Road transportation and power sector carbon taxes broadly in line with the EU average and slightly above the OECD average, but below neighboring countries and below levels needed for emissions-reductions targets.
  - Industrial, buildings, and agricultural/fish/forestry sectors have particularly low effective carbon tax rates relative to EU, OECD, and neighbors.
- Reasons for low carbon prices in particular sectors:
  - Industry: low natural gas and coal excise rates and about 20 percent of industrial sector emissions not covered by the ETS.
  - Buildings and agriculture: reduced excise rates for diesel (when used for heating) and natural gas.

### Response to the 2022 energy crisis (Box 1): measures and implications
- Energy shares of total household consumption for the lowest 20 percent of the income distribution:
  - Electricity: 3.9 percent
  - Natural gas: 2.1 percent
  - Oil products: 3.5 percent
- Federal and regional measures introduced cost about 1.7 percent of GDP and include:
  - Reduced VAT for natural gas, electricity, and district heating.
  - Reduced excises on petrol and diesel.
  - Additional cash transfers.
  - Expanded eligibility for the social tariff.
- Targeting and distortion:
  - Roughly 1/3 of measures are targeted and distort price signals.
  - Roughly 2/3 are untargeted and not price distortionary.
- Policy intent:
  - Government plans to introduce a flexible excise rate that declines for base levels of consumption as natural gas and electricity prices rise.
  - Improving targeting mechanisms through matching tax, household size, and third-party data is a medium-term objective.

### Rationale for economy-wide carbon pricing (sections 14–16)
- Advantages of carbon pricing over alternative mitigation instruments:
  - Provides across-the-board incentives to reduce energy consumption and shift to cleaner fuels, discriminating only by carbon content.
  - Equalizes marginal abatement costs across the economy (minimizes mitigation costs).
  - Mobilizes revenue that can support vulnerable households and firms or be used to improve productivity.
  - Generates domestic environmental co-benefits (e.g., reduced local air pollution mortality).
  - Administratively straightforward and can build on existing excise tax system.
- Design principles:
  - Cover emissions comprehensively.
  - Establish predictable prices and align stringency with mitigation goals (announce price path well in advance).
  - Couple non-ETS carbon price with a price floor for ETS sectors to equalize abatement costs across the economy.
  - Use revenues to support vulnerable households/firms and improve productivity.

### International practice and comparators (excerpt from Table 2)
- Prices for non-ETS sectors in various countries (effective taxes, € per tonne):
  - Austria: 35 (2023) to 55 (2025) — Heat, Transport (price fixed until 2026, then to ETS).
  - Denmark: 47 (2022) to 101 (2030) — All (lower rate for ETS sectors).
  - Finland: 77 (2022, Transport), 53 (2022, other) — Heat, Transport, Industry.
  - France: 45 (2022) — Heat, Transport, Industry (ETS sectors exempt).
  - Germany: 30 (2022) to 55 (2025) — Heat, Transport (fixed until 2026 then subject to ETS).
  - Ireland: 34-41 (2022) to 100 (2030) — Heat, Transport, Industry.
  - Portugal: Previous years’ avg. ETS price — Heat, Transport, Industry.
  - Sweden: 110 — Heat, Transport.
- Many European peers use tax-based carbon pricing for non-ETS sectors; some are introducing ETS price floors for power/industry.

### Modeling results and scenarios (Table 3 and discussion)
- General finding: A carbon price progressively reaching €75 per tonne
  - Reduces overall emissions to 25 percent below 2005 levels in 2030 (or 12 percent below a Business-as-Usual (BAU) scenario).
  - Raises revenue of 0.6 percent of GDP.
  - Economic efficiency costs of 0.2 percent of GDP are offset by improved health, reduced congestion, and other welfare benefits (welfare benefits net of losses to consumer welfare are shown as small).
- Modeled sectoral emissions outcomes in 2030 (relative to 2005 levels) at various carbon prices (selected figures from Table 3):
  - All (aggregate):
    - Carbon price 0: -19
    - 25: -15
    - 50: -20
    - 75: -22
    - 100: -25
    - 125: -28
  - Non-ETS:
    - 0: -47
    - 25: -11
    - 50: -10
    - 75: -14
    - 100: -17
    - 125: -19
  - Transport:
    - 0: -27
    - 25: -3
    - 50: -6
    - 75: -8
    - 100: -10
    - 125: -12
  - Buildings:
    - 0: -21
    - 25: -18
    - 50: -21
    - 75: -25
    - 100: -28
    - 125: -31
  - Agriculture:
    - 0: -23
    - 25: -4
    - 50: 1
    - 75: -8
    - 100: -9
    - 125: -10
  - ETS (aggregate):
    - 0: -21
    - 25: -18
    - 50: -23
    - 75: -27
    - 100: -30
    - 125: -33
  - Power:
    - 0: -39
    - 25: -10
    - 50: -11
    - 75: -11
    - 100: -11
    - 125: -11
  - Industry:
    - 0: -21
    - 25: -24
    - 50: -30
    - 75: -34
    - 100: -39
    - 125: -44
- Other modeled outcomes (selected values from Table 3):
  - Renewable share:
    - 0 price: 17.5
    - 25: 9.6
    - 50: 17
    - 75: 18
    - 100: 18
    - 125: 19
  - Revenue raised (% GDP):
    - 25: 0.2
    - 50: 0.3
    - 75: 0.6
    - 100: 0.8
    - 125: 1.1
  - Efficiency costs (% of GDP):
    - 25: -0.1
    - 50: -0.2
    - 75: -0.3
    - 100: -0.4
    - 125: -0.6
  - Welfare benefits (% GDP) (net of consumer welfare losses):
    - 75: -0.1
    - 100: -0.2
  - Deaths averted (annual, selected):
    - 25: 50
    - 50: 72
    - 75: 93
    - 100: 114
    - 125: 132
- Limitations and sectoral notes:
  - Power sector emissions reductions are limited because capacity commitments to natural gas through the Capacity Remuneration Mechanism and the tax only applying above EU ETS prices constrain reductions.
  - Agricultural emissions are primarily methane and nitrous oxide; price responsiveness at low carbon price levels is modest and deep abatement is more costly.
  - Even relatively high carbon prices (e.g., €125 per tonne) do not, by themselves, achieve mitigation targets—reinforcing sectoral policies are required.

### Recommended reform pathway and policy design (section 19 and earlier)
- Practical reform recommended:
  - Phase in a carbon price on non-ETS activities.
  - Introduce a domestic carbon price floor for ETS-covered sectors (the tax applies only if price levels exceed the EU ETS price).
  - Implement reforms as international energy prices fall to avoid excessively high post-tax prices while restoring pre-crisis relative prices (e.g., natural gas vs. coal).
- Policy design considerations:
  - Shift energy taxation from electricity toward fossil fuels (e.g., progressively rising excise rate on natural gas and other fossil fuels, potentially through carbon pricing).
  - Announce price paths well in advance to allow households and firms to adjust (support investments in clean technologies with longer payback periods).
  - Use carbon pricing revenues to support vulnerable households and firms, and to reduce distortionary taxes or fund productivity-enhancing measures.
  - Complement carbon pricing with sectoral policies where carbon pricing alone is insufficient to meet targets.

*Source: IMF staff analysis in sipea2023017 (chapter section on excises, subsidies, and carbon-pricing reform in Belgium).*

### 20.      The distributional impacts of a €100 per tonne carbon price are small considering

### 20.      The distributional impacts of a €100 per tonne carbon price are small considering extensive indexation in Belgium

### Distributional impacts and compensation needs
- Current wage and social benefit indexation automatically protects the average household that uses electricity and/or natural gas for heating since higher energy taxes increase inflation.
- Ignoring the impact of indexation:
  - Loss in purchasing power for poorer households: 1.8 percent of total consumption.
  - Loss in purchasing power for wealthier households: 1.5 percent of total consumption.
  - Losses are higher for rural vs. urban households across all income deciles.
- After revenue recycling and targeted compensation:
  - Around 25 percent of carbon pricing revenue is needed to fully compensate households in the lower four income deciles.
  - This leaves around 0.6 percent of GDP in revenue to allocate for other objectives, such as reducing income taxes.
- Energy-intensive firms:
  - Would see a 1–2   percent increase in input costs due to higher electricity prices.
  - Firms would receive an income tax deduction for their increase in input costs, which would result in a 25 percent (i.e., the Belgium CIT rate) lower change in post-tax costs.

### Policy recommendations on revenue recycling and household support
- Use a portion of carbon tax revenues to compensate vulnerable households through income-based support that is phased out as total income (labor and capital income) or wealth increases.
- Design principles for social support:
  - Provide income-based cash transfers (not conditional on household or firm energy expenditure) to preserve price signals and allow households to allocate budgets optimally.
  - Phase out support gradually with rising income to avoid reinforcing unemployment traps.
  - Do not condition support on energy expenditure, as this subsidizes energy inefficiency and raises costs of investing in emissions reductions.
- Firm support:
  - For energy-intensive firms, consider reductions in the corporate income tax, investments in productive public infrastructure and R&D, and/or rebates based on output (rather than energy inputs) to maintain incentives for emissions-intensity reductions.
  - Rebates based on output would need to be phased out as CBAM is phased in to ensure WTO compatibility.

### Recommended carbon pricing design and sequencing
- Phase in a carbon tax for non-ETS sectors through the excise regime, reaching €100 per tonne by 2030.
- Introduce domestic carbon pricing for non-ETS sectors prior to the planned EU-wide ETS for buildings and transportation (proposed to be introduced in 2027) because:
  - Non-ETS emissions reductions are needed immediately to meet Belgium’s ESR.
  - Final design decisions for the EU-wide ETS for buildings and transport are pending, increasing investment uncertainty.
  - Competitiveness and leakage concerns are less pronounced for transportation and buildings.
  - The expected decline in international energy prices provides an opportunity to phase-in carbon pricing without increasing post-tax prices.
- Price path design:
  - Announce price path in advance with small annual increases (e.g., €10 per tonne each year).
  - Adjust for inflation to maintain real value.
- Phasing and administrative considerations:
  - Phase in through an excise carbon tax element to avoid volatility, build on existing excise duty regime, be compatible with future EU ETS, and avoid leakage from regional mitigation policies.
  - The tax could be extended to non-carbon GHGs (e.g., methane) if administrative, compliance and political costs can be managed; proxy taxes could be used in the medium-term to address barriers.

### Price floor for ETS sectors
- Consider a price floor for ETS sectors (power and industry) aligned with the carbon tax for non-ETS sectors to:
  - Provide greater certainty for investors.
  - Minimize abatement costs across the economy and equalize abatement costs.
  - Recycle revenues generated from industry to protect against competitiveness concerns.
- Implementation example:
  - Apply as a surcharge resembling the U.K. Carbon Price Floor (national variable tax set for three years in advance equal to the difference between an exogenous target price and projected EU ETS price).
  - Note: Netherlands is implementing a similar scheme at price floor levels likely below EU ETS and thus not binding.

### EU-level dialogues and reforms suggested
- Promote dialogue at the EU level to strengthen decarbonization policy and synergies:
  - Extend the EU ETS so aggregate emissions from power, industry, transport, and buildings are subject to one aggregate cap with a common emissions price across sectors.
  - Extend the ability of member states to re-allocate a greater share of emissions reductions from transport/buildings to power/industry to lower mitigation costs.
  - Extend the ability to trade ESR quotas between states to promote cost-efficiencies (political agreement reached in late November; details not yet clear).
  - Establish an exogenous and escalating price floor for both EU ETSs to provide investment signals and allow overlapping national measures to lower EU emissions.

### Reinforcing carbon pricing with feebates and sectoral policies
- Feebates (revenue neutral tax-subsidy schemes) are recommended as a relatively efficient reinforcing policy:
  - Finance subsidies to relatively clean activities while taxing relatively dirty ones.
  - More flexible and potentially more cost-effective than emissions-rate regulations.
  - Do not raise average prices for consumers or costs for firms, making them less politically challenging.
  - Can complement regulations by rewarding going beyond standards.
- Sector guidance:
  - Design sectoral instruments flexibly and technology-neutral where possible.
  - Accept differing implicit carbon prices across sectors as needed during decarbonization.

### Power generation: context, feebate design, and impacts
- Context and capacity:
  - Nuclear power currently makes up 45 percent of generation.
  - Nuclear phase-out legislated in 2003 and expected to be complete by 2035.
  - First reactor (17 percent of nuclear capacity) was decommissioned in 2022; others planned for 2023, 2025, and 2035.
  - Capacity Remuneration Mechanism (CRM):
    - First CRM auction (in 2021) awarded 4.4 GW for delivery in 2025, with 3.6 GW from natural gas and the rest primarily from co-generation and demand-side response.
    - Annual CRM auctions planned to ensure capacity one and four years ahead.
  - Federal auction to develop 3.15 to 3.5 GW in new offshore wind generation in Princess Elisabeth Island; increasing interconnections to import clean electricity from Norway and Denmark.
- Feebate formula and implementation:
  - Under a feebate scheme, generators could be subject to a fee given by:
    - {CO2 price} × {CO2/kWh – industry-wide average CO2/kWh} × {electricity generation}
  - Practical implementation:
    - Build off existing EU ETS monitoring procedures.
    - Exempt nuclear given planned phase-out.
    - Use a smoothing mechanism for the industry-wide average emissions-intensity (e.g., a five-year average).
    - Apply through refundable credits and taxes in corporate or personal income (PIT) tax regime, or as a direct excise/subsidy scheme.
    - Interacts well with CRM: CRM maintains capacity incentives while feebate promotes low-carbon electricity.
- Illustrative impacts (2022, with a €100 per tonne price):
  - Fee equivalent to 3 cents/kWh for natural gas.
  - Subsidy of 1.5 cents/kWh for renewables.
  - Fees on fossil-fuel generation would decrease as nuclear is phased-out because the average emissions-intensity of electricity would increase; pivot point could be set exogenously to decline to meet renewable targets.
- Emissions and system considerations:
  - Feebates can significantly reduce the emissions-intensity of electricity compared to business-as-usual.
  - Switching towards renewables must be combined with grid management, demand-side response, and storage to address intermittency.
  - National emissions reductions via feebates will not fully reduce EU-wide emissions if the MSR is not operating, because falling Belgian emissions would reduce the ETS price and cause leakage to other countries and to Belgium’s industrial sector.

### Existing clean energy support and recommendation
- The Green Certificate Scheme:
  - Cost of certificates: €2.4 billion in 2020.
  - Has contributed to growing debt levels for Elia, Belgium’s electricity transmission system operator.
  - Regressive because wealthier households have larger shares of rooftop solar and therefore residential renewable electricity production.
  - Includes arbitrary design parameters (such as minimum level of support) and is partly financed through surcharges on electricity bills, contributing to higher electricity prices relative to fossil fuels.
- Recommendation:
  - Reconsider or replace the Green Certificate Scheme; feebates could replace it to improve efficiency of incentives while not increasing prices.

*Source: IMF staff estimates.*

### 33.      Reducing fees and levies on electricity would promote electrification by lowering the

### 33.      Reducing fees and levies on electricity would promote electrification by lowering the

### Electrification and electricity pricing
- The price of electricity relative to heating oil, natural gas, diesel and gasoline is generally higher in Belgium than in neighboring countries.
- The energy portion of electricity costs (i.e., the cost excluding all fees, levies, and taxes) is on par with France, Germany, and the Netherlands (CREG 2021, Deloitte 2022); differences arise from charges to recover network costs, fees, levies, taxes, and (potentially) market power.
- Levies include charges to fund a heating premium, green certificates, and decommissioning of the Mol-Dessel nuclear site (Deloitte 2022).
- Recommendation: Gradually remove charges that do not directly relate to delivering electricity to end-users and instead finance those measures through the general budget to promote climate objectives and economic efficiency by better aligning electricity prices with generation costs.
- Note: Such removal should be coupled with enhanced policies to promote renewable energy and energy efficiency to avoid promoting polluting electricity use.

### Industry
- Industrial emissions contribute 35 percent to Belgium’s total emissions.
- Main sub-sectors: chemicals, refining of oil and coke, manufacture of non-metallic minerals and metals.
- Emissions are roughly split between direct fuel combustion and industrial processes.
- About 20 percent of industrial emissions are outside of the ETS due to:
  - installations below the ETS threshold of 25,000 tonnes of CO2 per year, and
  - emissions from biomass being excluded from the ETS.
- Most excluded emissions occur in the chemical and food sectors and are currently not covered by domestic mitigation policies, except voluntary regional agreements.
- Policy proposal: Feebates to reinforce incentives for cleaner production processes in carbon-intensive industries. Example formula:
  - {CO2 price} × {CO2/output – industry-wide average CO2/output} × {firm output}
- Design features and trade-offs:
  - Applied to emissions from fuel combustion and direct and process emissions.
  - Avoids a first-order burden on the average producer (they pay no charge on their remaining emissions), alleviating competitiveness and leakage concerns versus full pricing of remaining emissions.
  - Less efficient in reducing emissions because it does not result in a decrease in output.
  - Could build off existing EU ETS monitoring procedures and operate in tandem with voluntary agreements.
  - Pivot-point (industry-wide average emissions-intensity) could be set exogenously or be unique to each sub-sector, with permit trading across sub-sectors to promote cost-effectiveness and avoid leakage.

### Buildings
- Buildings make up 21 percent of total emissions.
- Residential buildings account for 75 percent of building emissions.
- Energy source shares (Belgium vs EU averages):
  - Oil: 32 percent (Belgium) vs. 21 percent (EU)
  - Renewables: 8 percent (Belgium) vs. 20 percent (EU)
  - Electricity: 20 percent (Belgium) vs. 25 percent (EU)
- Most energy use in buildings is for space and water heating: 83 percent.
- Biomass use for household heating: 7 percent of total household heating in 2020.
- Building stock characteristics:
  - 80 percent of constructions occurred prior to energy standards across Belgium and 94 percent in Brussels.
  - 45 percent of households are not connected to the natural gas grid.
  - Heating oil shares of houses heated using an oil boiler: 16 percent (Brussels), 24 percent (Flanders), 50 percent (Wallonia).
- Distributional note: Households in the lowest income quartile spent roughly eight percent of total consumption on natural gas, heating oil, and electricity, compared to 3.6 percent for the highest quartile in 2018.
- Heating cost dynamics and policy interactions:
  - Illustration assumptions for Figure 12: air-source heat pumps (95 percent of existing heat pumps in Belgium are air-source), a discount rate of 7 percent, interest rate of 4.5 percent, grant covers 70 percent of equipment and installation costs, all systems last for 20 years. Fuel prices assumed at pre-crisis levels: EUR 0.054 and 0.034 per kWh (natural gas; without and with social tariff, respectively), EUR 0.060 and 0.045 per kWh (heating oil), and EUR 0.290 and 0.189 per kWh (electricity).
  - A carbon tax raises all-in natural gas heating costs by about 28 percent, 38 percent for oil, and 4 percent for a heat pump.
  - Removing fees and levies charged on electricity consumption (outside of those covering network costs) reduces heat pump costs by 20 percent.
  - Efficient pricing mix: carbon tax, no social tariff, and network fees and levies in line with neighboring countries would align costs of fossil-fuel and electricity-based heating.
  - Revenue gains could be recycled to support impacted households or subsidize low-carbon heating.
  - Grants and no-interest loans could help reduce upfront costs.
  - Note: The social tariff currently makes natural gas heating significantly cheaper than heat pumps, even if the entire capital cost of a heat pump is paid for by government.
- Renovation rates and barriers:
  - Current renovation rate: 1 percent annually; needs to increase to 2.5 to 3 percent to meet emissions targets.
  - Barriers: liquidity constraints (especially lower income households), cost-benefit mismatches between owners and renters, unawareness/uncertainty of energy savings, lack of consistency in EPC criteria across regions, lack of knowledge of long-term EPC requirements, and potential supply constraints in the construction sector.
- Additional policy proposals:
  - Feebates for household heating and appliances. Example feebate formulas:
    - Heating: carbon tax on heating fuels combined with subsidies for electric or clean-fuel heating systems and energy-efficiency renovations.
    - Appliances: {CO2 price} × {CO2 per unit of energy} × {energy consumption per unit – industry-wide energy consumption per unit} × {number of units}
  - Property taxes could be linked to energy use per square meter or EPC scores to mimic a feebate, but would increase administration and could be regressive since poorer households generally live in worse-insulated dwellings.
  - Federal policy options to implement feebates are limited by lack of information and competency.

### Transportation
- Road transportation emissions: passenger vehicles 75 percent, road freight makes up most of the remainder.
- Car ownership: around one car for every two people (similar to EU average and neighboring countries).
- EV and plug-in hybrid share in new car sales: 5 percent in 2021.
- EV charging infrastructure: 8.8 charging points per 100 km of road in Belgium (ACEA 2022b); Germany 26 per 100 km; Netherlands 64 per 100 km.
- Less than 10 percent of charging stations in Belgium are ‘fast’.
- Registration and annual circulation taxes are applied at the regional level; Flanders and Walloon registration taxes vary based on vehicle CO2 emissions-rate; Brussels is considering such a policy.
- Federal income tax credit: 15 percent of the purchase price up to €5,150 for electric vehicles purchased by individuals available until 2024.
- Nationwide scheme: charges heavy commercial vehicles per kilometer travelled.
- Existing issues with vehicle tax design:
  - Registration and annual circulation taxes expressed on a lifetime basis apply only modest taxes on high emission vehicles in Flanders, Brussels, and Wallonia.
  - Relative price of low-emitting, non-electric vehicles is not decreased relative to average-emission vehicles in Brussels or Wallonia.
  - EU emission rate standards applied to fleetwide averages can offset demand shifts to low-emission vehicles.
- Proposed feebate for vehicles:
  - Formula: {CO2 price} × {the vehicle’s CO2/km – the industry average CO2/km} × {the average lifetime km driven per vehicle}
  - Illustration: A feebate with price of €400 per tonne of CO2 would:
    - Provide the same EV subsidy as at present,
    - Apply a tax of €3,000 to a vehicle with 200 g CO2/km (an increase of around €1,000 for Brussels and Wallonia).
  - Advantages:
    - Incentivizes purchase of low-emission non-electric vehicles (currently not subsidized).
    - Could replace the company car regime; can be introduced unilaterally by the federal government via personal income tax credits, but harmonization with regions is ideal.
    - Subsidies for EVs decline over time as the average fleet emission rate declines.
    - Manufacturers are penalized for any increase in fleet emissions in response to higher EV sales shares.
    - Feebates automatically maintain revenue neutrality as average fleet emission rate updates.
    - Feebates do not require new data or administrative capacity relative to the existing emission rate program.
- Company car regime:
  - Creates incentive to be compensated with a car rather than cash; recent reforms include restricting company car eligibility to EVs by 2026 and introducing a mobility budget.
  - Take-up rate of the mobility budget is reported to be low.
  - Company car regime is regressive, contributes to congestion (around 12 percent of the total vehicle fleet), and is fiscally costly (estimated at around 1 percent of GDP).
  - Recommendation: Ideally eliminate the company car regime and replace it with a feebate policy.

*Source: IMF staff analysis as presented in the provided chapter excerpt.*

### 49.      Broader reforms using other fiscal instruments could address other transportation

### 49.      Broader reforms using other fiscal instruments could address other transportation

### Transportation externalities and pricing instruments
- Transportation externalities include congestion, road damage, and accidents.
- Distance-, location- and time-based (i.e., congestion) charges are presented as best-suited instruments to address transportation externalities rather than fuel taxes.
- The existing heavy duty truck distance-based charge is cited as an example of pricing road damage through a distance-based mechanism.
- Importance of moving toward distance/time-based tax systems increases with EV penetration because:
  - EVs contribute to external costs but driving is not discouraged through road fuel taxes (nor are fiscal revenues gained).
- Administrative feasibility of distance/location/time-based charges is improving with technological improvements, such as improved and lower cost global positioning system (GPS) tools.
- For congestion charges, these policies would be within regional competencies but should be coupled with reductions in road fuel taxes (and potentially vehicle registration and annual circulation fees).
- Moving towards more efficient distance and time-based pricing would result in a revenue shift from federal to regional governments but still provide the general government with a robust source of revenue from transportation as the base of traditional fuel tax systems is eroded.
- To more accurately price accident-related externalities, pay-as-you-drive (PAYD) vehicle insurance could be promoted through fiscal incentives at the household or insurance company level; PAYD charges a premium per kilometer travelled rather than a lump-sum amount and therefore incentivizes reductions in driving levels.

### Summary of policy conclusions on carbon pricing and sectoral policies (E. Conclusion and Summary of Policy Recommendations)
- Carbon pricing coupled with reinforcing sectoral policies would promote emissions reductions at limited economic and administrative costs.
- Carbon pricing should be the centerpiece of Belgium’s decarbonization policy because it is most efficient and can be implemented through the existing excise regime.
- Price-based, sectoral policies, such as feebates, provide a complement when exceedingly high carbon prices are needed to meet emissions reduction targets (e.g., in the buildings and transportation sector) or significant non-climate considerations (e.g., nuclear phase-out and industrial competitiveness concerns).

### Recommended policy priorities and projected impacts
- Priority: carbon pricing in non-ETS sectors and a price floor for the existing EU ETS are ideally prioritized.
- Expected outcomes of the full set of recommendations:
  - Reduce emissions to at least 20 percent below business-as-usual levels by 2030 (33 percent below 2005 levels).
  - Raise substantial revenue (at least 0.6 percent of GDP) that could be used to support vulnerable households and firms and increase economic productivity.

### Complementary fiscal measures and sectoral recommendations (from annexed studies)
- Mobility budget note:
  - The mobility budget allows employers to provide workers with in-kind, tax preferred benefits for low-emissions company cars and other mobility options (e.g., bicycles) and cash for public transport and accommodation near the place of employment.
  - The mobility budget is a minimum of € 3,000 per year and maximum of the lesser of one fifth of gross remuneration and €16,000 per year.
- Leuven (Leuven 2022) study findings and recommendations:
  - Revising energy taxation to be based on carbon content would reduce transportation and building emissions by 11.3 percent and 14.1 percent, respectively, by 2030 in the context of a reference scenario for a €100 per tonne carbon tax).
  - The tax would raise significant revenues (€2.7 billion by 2030); revenues could be recycled to improve equity, maintain reduced VAT on electricity, and/or promote economic efficiency (for example, by reducing labor taxes).
  - Recommended policy package: a carbon price, aligning excise taxes on fuels with the soon-to-be-revised Energy Tax Directive, and recycling revenues through direct income transfers to vulnerable households.
  - Sectoral-level recommendations: removing preferential VAT treatment for fertilizer, pesticides, meat, and aviation; a tax on pesticides and fertilizers; phasing out or reducing the reimbursement for commercial diesel; phasing out the preferential tax treatment of company cars and fuel cards; increasing the aviation departure tax; introducing excises on LPG, compressed natural gas (CNG), and kerosene; assessing taxation of nuclear energy rents and potentially extending to other energy sources with high upfront and low operating costs; taxation of inland waterway transport.
- National Bank of Belgium (NBB) (Cornille et al 2021) study findings:
  - Effective carbon tax rates are currently relatively low for industry and buildings in Belgium; tax rates should be partly linked to the carbon content of fuel consumption.
  - The Green Certificate Scheme is likely not cost-effective; promoting renewables should be via carbon pricing and subsidies on the capital cost of low-carbon, residential technology.
  - Reducing the price of electricity relative to natural gas, which is partly high due to extensive surcharges for electricity consumers and low natural gas taxes, would support investment in heat pumps.
  - Current support for heat pumps is relatively low, with subsidies for condensing boilers that use natural gas or oil.
  - Incidence analysis conclusions: low-income households would benefit from general income support as carbon prices are introduced, as well as subsidies for clean technology, including concessionary access to financing.
  - Environmental tax revenues should not be earmarked but rather spent considering the full set of government objectives.
- National Carbon Pricing Debate (Climact 2018) principles and sectoral options:
  - Three principles for introducing carbon pricing: (i) revenue neutrality (potentially via reducing labor taxes or electricity charges and levies, or lump-sum transfers), (ii) send a credible, long-term signal to direct investments and input choices toward greener options, (iii) complementary policies are needed to maximize benefits and address barriers.
  - Buildings: carbon pricing is efficient for greening buildings but regressive without compensating measures; suggested compensations include lump-sum transfers or energy vouchers to vulnerable citizens, renovation programs, support to SMEs.
  - Transport: carbon price should cover all fossil fuels (biomass component taxed using emission factor equivalent); special care for diesel used for commercial purposes due to leakage risk. Two options: (i) include a carbon price on top of existing excise and limit final price of diesel for commercial purposes to that of neighboring countries, or (ii) introduce carbon pricing using a ‘smart’ road pricing system (requires time to implement). Company car policies should be aligned with emissions reduction goals.
  - Industry: non-ETS emissions are 17 percent of total industrial sector emissions—35 and 65 percent are from processes and energy, respectively. Two options: (i) price all fossil fuel emissions except those at risk of carbon leakage (limit carbon prices to current price difference with neighboring countries) and price process emissions conservatively; (ii) reform voluntary agreements of Walloon and Flemish regions to incorporate an explicit carbon price and impose a carbon price on non-participants.
  - Agriculture: carbon pricing should cover energy-related emissions from non-stationary sources (offroad vehicles and machinery) and use a similar approach for stationary sources (mostly agricultural greenhouses) as for non-ETS industry; pricing non-CO2 emissions is complicated by measurement difficulties.

### Modeling approach and key parameters (Appendix D)
- CPAT model primary dependencies:
  - Energy price projections (and relative prices across energy sources), driven by international prices and domestic policies.
  - Responsiveness of energy use to prices and income (elasticities).
  - Energy efficiency improvements from autonomous technological change.
- Forecasted international energy prices use the average of the IMF’s World Economic Outlook and the International Energy Agency’s World Energy Outlook (Stated Policies Scenario).
- Forecasted GDP is sourced from the IMF’s WEO.
- Selected elasticity parameters from Table 1 (Belgium: Key Modelling Parameters):
  - Natural gas: Elasticity with respect to income — Transport 0.5, Residential 0.3, Industry 1.0; Elasticity with respect to price — Transport -0.4, Residential -0.3, Industry -0.4.
  - Gasoline: Elasticity with respect to income — Transport 0.7, Industry 0.5; Elasticity with respect to price — Transport -0.3, Residential -0.4, Industry -0.3.
  - Diesel: Elasticity with respect to income — Transport 0.6, Residential 0.5, Industry 0.5; Elasticity with respect to price — Transport -0.2, Residential -0.2, Industry -0.6.
  - Electricity: Elasticity with respect to income — Transport 1.2, Residential 0.8, Industry 0.8; Elasticity with respect to price — Transport -0.1, Residential -0.2, Industry -0.2.
  - Coal: Industry elasticity with respect to income 0.5 and with respect to price -0.2; Transport and Residential marked NA.

*International Monetary Fund: Selected Issues Paper — Belgium (sipea2023017), excerpted section as provided.*

---


_Source: https://www.imf.org/-/media/files/publications/selected-issues-papers/2023/english/sipea2023017.pdf_