## wpiea2021171-print-pdf

## Source details

**Canonical URL:** [wpiea2021171-print-pdf](https://www.imf.org/-/media/files/publications/wp/2021/english/wpiea2021171-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2021/english/wpiea2021171-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2021/english/wpiea2021171-print-pdf.pdf.json)

---

### I. Introduction — purpose and proposed fiscal package
- Finland has cut greenhouse gas (GHG) emissions by about a fifth since 1990 and aims to become emissions neutral by 2035.
- Proposed fiscal package:
  - Scale up and harmonize carbon pricing across fuels/sectors to a target effective carbon price of €125 ($150) per tonne by 2030.
  - Combine carbon pricing with revenue-neutral feebates (tax-subsidy schemes).
  - Target transport, industry, buildings, forestry, and agriculture; assume power largely decarbonized by 2030.
- Distributional and labor-tax interaction:
  - A €125 per tonne carbon price imposes a burden on the average household of 1.8 percent of consumption (relative to a baseline with current carbon pricing).
  - Carbon pricing revenues could be used to reduce labor tax rates and boost hours worked to largely offset household burden.

### II. Recommended economy-wide and sectoral fiscal instruments (summary)
- Economy-wide
  - Impose sector-specific carbon surcharges so effective carbon prices (surcharge + fuel/carbon taxes + EU ETS, net of domestic-environment taxes) reach target levels.
  - Set target prices to reach €125 per tonne by 2030 and progressively harmonize across fuels/sectors.
  - Recycle revenues via broad reductions in labor income taxes to balance distributional impacts and support employment.
- Road transport
  - Introduce vehicle-level feebates: sliding fees/rebates on vehicles with above/below average CO2/km.
  - Raise diesel tax to the gasoline tax.
  - Feebates can promote EVs aggressively without fiscal cost or new tax burden on the average motorist.
- Industry
  - Introduce firm-level feebates based on emission rates relative to industry average to incentivize cleaner production and limit leakage concerns.
- Buildings
  - Combine energy efficiency regulations with: (i) tax-subsidy schemes to shift from natural gas to electric/clean fuel heating; (ii) feebates for efficient appliances and lighting.
- Land use / Forestry
  - Nationwide landowner feebate: CO2 price × (baseline carbon storage − current carbon storage).
  - Use satellite/aerial imagery and sampling for approximate carbon inventories; design revenue-neutral and no burden on average landowner.
- Agriculture
  - Farm-level feebates: CO2 price × (farm emissions per acre − industry average per acre) × acreage.
  - Use input-based estimates and default emissions factors; consumer-level 'sin' taxes could reinforce diet shifts.

### III. Rationale for combining scaled carbon pricing and feebates
- Carbon pricing scaling and harmonization:
  - Scale up via carbon surcharges to align effective prices (domestic taxes + EU ETS) to targets; define effective prices net of taxes warranted by domestic environmental externalities.
  - Progressive harmonization improves cost effectiveness.
- Role and advantages of feebates:
  - Revenue-neutral sliding fees on above-average emitters, rebates for below-average emitters.
  - Provide flexibility, strong mitigation incentives, avoid large tax burdens on average households/firms, avoid fiscal burden for government, and often use existing administrative capacity.
- Sectoral fit:
  - Transportation: finer incentives for lower-emission vehicles.
  - Industry: cleaner production with less leakage concern.
  - Buildings: reinforce electric heating and efficient appliances.
  - Land use: incentives for carbon storage in forests and peat land.
  - Agriculture: promote lower-emission practices; consumer taxes may be needed to avoid leakage.

### IV. Finland’s emissions targets, projections, and sinks
- National pledges and implied levels:
  - GHG neutral by 2035.
  - 2035 target: 21 million tonnes (Mt) of CO2 equivalent (including ETS and effort-sharing) — a reduction of 70 percent below 1990 levels (71 Mt CO2 equivalent).
  - 2035 target implies a cut of 55 percent below 1990 levels in 2030 (assuming a linear pathway).
  - By 2050 pledged reduction to 80 percent below 1990 levels (14 Mt CO2 equivalent).
- Historical and projected emissions:
  - GHG emissions: 56 Mt CO2 equivalent in 2018; government projection 53 Mt CO2 equivalent in 2019.
  - WEM scenario: projected 2035 emissions 39 Mt CO2 equivalent; with planned-but-not-implemented measures 36 Mt CO2 equivalent.
- LULUCF sink:
  - LULUCF emissions: −10.3 Mt CO2 equivalent in 2018; −17.4 Mt CO2 equivalent in 2019.
  - Further absorption mainly from increasing average forest age; methodological variability raises questions about accounting for the 2035 neutrality target.
  - Once forest age is optimized, additional forestland does not absorb additional carbon; other sinks (e.g., direct air capture) will be needed for remaining annual emissions.

### V. Emissions structure and sectoral shares (2018)
- Sources of GHGs (2018):
  - Fossil fuel CO2: 75%
  - Industrial processes: 10%
  - Agriculture: 12%
  - Waste: 3%
- Sector share of fossil fuel CO2:
  - Energy (mostly power/district heating): 44%
  - Industry (and construction): 17%
  - Transport: 28%
  - Other sources (mainly commercial/residential buildings and farm machinery): 11%
- Fuel shares of fossil fuel CO2:
  - Coal/peat: 17%
  - Oil: 65%
  - Natural gas: 18%
- Power generation by fuel type:
  - Coal/peat: 18.0%
  - Natural gas: 5.1%
  - Oil: 0.9%
  - Hydro: 0.1%
  - Other renewable: 60.6%
  - Biomass: 15.4%

### VI. Coverage, EU ETS interactions, and carbon pricing context
- EU ETS coverage:
  - 43 percent of economy-wide emissions in Finland are covered by the EU ETS (just under the EU average of 45 percent), mainly power/district heat and large industrial firms.
- Market Stability Reserve (MSR) and price-floor options:
  - MSR withdraws allowances when banked allowances exceed thresholds; as of March 2021, allowance prices are around €45 per tonne.
  - Emissions reductions from overlapping domestic policies would be offset elsewhere under a fixed EU cap, though MSR operation reduces that offset to some degree.
  - A more robust mechanism is an exogenous price floor ramping up over time (via minimum auction price or MSR price trigger) with permanent allowance withdrawals as needed.
- Free allocations and BCA:
  - Energy intensive, trade exposed (EITE) industries receive free allowances; the European Commission is considering replacing this with a border carbon adjustment (BCA).
- Carbon and fuel taxes (2018 OECD-based estimates used in analyses):
  - Effective carbon tax (gasoline): €77 per tonne
  - Effective carbon tax (diesel): €75 per tonne
  - Effective carbon tax (heating fuels for buildings): €62 per tonne
  - Additional fuel taxes equivalent to carbon charges:
    - gasoline: €234 per tonne
    - diesel: €124 per tonne
    - heating fuels for buildings: €29 per tonne
    - coal/peat use by industry: €10 per tonne
    - natural gas use by industry: €31 per tonne
  - Taxes refer to 2018; taxes revised up in 2021 (see source tables).

### VII. Projected impacts, scenarios, and uncertainties
- Global context:
  - Global fossil fuel CO2 and other GHGs need to fall by about 25 percent below 2018 levels by 2030 to contain warming to 2°C, or 50 percent below for 1.5°C, and continue declining thereafter.
  - Global CO2 emissions in 2020 projected about 8 percent lower than in 2019; projections suggest global emissions in 2030 about 20 percent above 2018 levels.
- Finland BAU and policy shortfall:
  - BAU scenario: emissions flat after 2020 as energy intensity improvements offset GDP expansion.
  - A €125 carbon price combined with current measures still leaves a shortfall relative to national/sectoral objectives.
- Illustrative impacts of a uniform €125 per tonne carbon price (ramped to 2030 and frozen thereafter; coordinated across sectors):
  - 2035 emissions would remain at 29 Mt CO2 equivalent, or 38 percent above the target level — leaving an emissions gap of about 8 Mt CO2 equivalent (assumes carbon pricing covers 85 percent of future GHG emissions).
  - Extra revenue raised: 0.3 percent of GDP in 2030 (relative to revenue under current effective carbon pricing, accounting for pre-existing fuel tax base erosion).
  - Economic efficiency costs: about 0.2 percent of GDP (consumer/producer surplus losses net of government revenue gains, accounting for pre-existing taxes and emissions pricing).
  - Price effects in 2030 relative to 2030 levels with current carbon prices frozen:
    - Natural gas prices increase by 6.5 percent.
    - Electricity prices increase by 11.5 percent.
    - Gasoline prices increase by 8 percent.
- Uncertainties and policy implications:
  - Large uncertainties in carbon prices consistent with emissions neutrality and in political acceptability of high carbon prices.
  - Feebates can act as insurance if carbon pricing emissions impacts are smaller than projected and provide reinforcement to achieve objectives.
  - Carbon prices well above €125 per tonne could further close gaps but risk public backlash.

### VIII. Sectoral detail—transportation, buildings, industry, agriculture, forestry
- Transportation targets and measures:
  - EU new passenger vehicle fleetwide standard: 95 g/km in 2021 with a 37.5 percent reduction in 2030 relative to 2021.
  - Finland targets by 2030:
    - (i) 50 percent reduction in transportation emissions relative to 2005 (100 percent cut by 2045);
    - (ii) biofuels share rising from 20 percent in 2030 to 30 percent in 2030;
    - (iii) 250,000 EVs and 50,000 natural gas vehicles by 2030 (current: 29,000 EVs and 9,000 NGVs).
  - Road transport accounts for 95 percent of transportation emissions.
  - Fuel excise rates: gasoline €0.70 per liter; diesel €0.45 per liter.
  - Vehicle taxation: registration fees rise from 3 percent to 50 percent of vehicle price as CO2 rises 0 g/km → 360 g/km; 2019 average registration fees by fuel type: gasoline 15%, diesel 19%, CNG 9%, plug-in EVs 4%, full EVs 3% (average 15% across vehicles).
  - EV subsidy: €2,000; registration taxes raise 0.1 percent of GDP.
  - Proposed transport feebate formula: {CO2 price} × {vehicle’s CO2/km − sales fleet average CO2/km} × {average lifetime vehicle use (discounted km)}.
    - Illustration: feebate price €800 per tonne + lump-sum €5,500 to maintain revenue would yield subsidy €3,700 for an EV and fee €12,000 for a 200 g CO2/km vehicle (assumptions: fleet average 115 g/km; discounted lifetime 100,000 km).
    - Progressive feebate shifting new sales to 100 percent EVs by 2035 would reduce road fuel emissions about 50 percent below otherwise projected 2035 levels (assumes 8 percent fleet replacement per year; EV new sales start at 2 percent and rise linearly to 100 percent by 2035).
  - Diesel specifics:
    - Diesel emits about 16 percent more CO2 per unit than gasoline.
    - Raising diesel tax by €0.172 per liter (to gasoline level) would increase pump prices for diesel by 19 percent; reduce diesel consumption about 8 percent (0.6 Mt CO2e); raise revenues 0.1 percent of GDP; increase lifetime cost of average diesel vehicle by about €1,400.
- Buildings, heating, and appliances:
  - Space heating emissions declined from 4 to 2.5 Mt CO2 equivalent between 2015 and 2018; projected to decline to 1.7 Mt CO2 equivalent by 2030.
  - Policy intentions: phase out oil heating in government buildings by 2025; aim to phase out fossil fuel heating in all buildings by early 2030s via grants and energy efficiency.
  - Feebates and interim oil tax revenues could fund heat pump/hydrogen boiler subsidies.
  - Product-level feebate formula example: {CO2 price} × {CO2 per unit of energy} × {energy consumption rate − industry-wide average energy consumption rate}.
- Industry feebates and examples:
  - Industry feebate formula: {CO2 price} × {CO2/output − industry-wide average CO2/output} × {output}; applies to direct emissions.
  - Potential target industries and 2018 shares of nationwide GHGs:
    - Metals: about 5%
    - Forest products: about 5%
    - Chemicals: about 4%
    - Cement/lime production: about 3%
  - Feebates reduce first-order transfer payments and can be built on EU ETS monitoring procedures; EU state aid compatibility uncertain.
  - Illustrative production-cost impacts at €50/ton CO2:
    - Steel: integrated production cost increase ≈ €100/ton of steel; alternatives ≈ €20/ton. Feebate yields ±€50–€30 per ton depending on technology assumptions.
    - Cement: traditional production cost increase ≈ €50/ton cement (about 40%); feebates provide subsidies to efficient/CCUS plants in illustrative ranges.
- Forestry feebates:
  - Feebate fee: {CO2 rental price} × {baseline stored carbon − current stored carbon}; rental payments equate carbon price × interest rate × number of years; partial exemptions for timber harvested for long-lived products may be warranted.
  - Monitoring via satellite, aerial imagery, LIDAR, and ground sampling; cost reductions by sampling infrequently and leveraging approvals.
- Agriculture: feebates and consumer measures
  - Agriculture emissions stable at 8–9 Mt CO2 equivalent since 2005 and projected to remain until 2030 under current policy.
  - Peat land responsible for half agricultural GHGs while accounting for 10 percent of farmland.
  - Feebates formula: {CO2 price} × {CO2 equivalent/acre − industry average CO2 equivalent/acre} × acres.
  - Illustrative feebate at €20 per tonne CO2e — subsidies/fees per acre and percent of farm income:
    - Plant-based agriculture: −€14 per acre (18 percent of farm income)
    - Poultry farms: −€10 per acre (6 percent of farm income)
    - Pig farms: €11 per acre (7 percent of farm income)
    - Cattle farms: €14 per acre (14 percent of farm income)
    - Dairy farms: €29 per acre (18 percent of farm income)
    - Negative values = subsidy; positive = fee.
  - Consumer-level measures (taxes on meat/dairy) could reinforce diet shifts but risk leakage and political resistance.
- Leakage and competitiveness instruments:
  - Estimated simple average leakage rate from energy price increases ≈ 0.2; Finland’s leakage rate cited as 0.2; EU-wide policies leakage rate cited as 0.13.
  - Feebates may better limit leakage/competitiveness issues than free allowances; BCA could be more efficient and raise revenue but has administrative and design challenges.

### IX. Equity, household incidence, revenue use, and labor-tax reforms
- Channels of household incidence:
  - (i) direct energy price effects; (ii) indirect price effects on other consumer goods; (iii) wage changes in trade-exposed sectors; (iv) recycling of carbon revenues.
- Energy price and household exposure:
  - Retail electricity prices rise 11.5 percent under €125/ton carbon price scenario (reflects share of zero-carbon fuels and coal phase-out).
  - Natural gas prices increase by 6.5 percent; gasoline by 8 percent.
  - Average household energy and fuel consumption share: 6.3 percent of total household consumption (2015 data).
- Distributional incidence of a €125 per tonne CO2 carbon tax in 2030 (prior to revenue use):
  - Average household burden: 0.9 percent of consumption, following a U-shape across income distribution.
  - Bottom quintile: purchasing power decrease of 0.9 percent of consumption.
  - Middle-income households: largest incidence ~0.97 percent of consumption.
  - Top 20 percent: purchasing power erosion of 0.9 percent of consumption; almost half driven by lower wages.
- Revenues and compensating reforms:
  - Carbon tax collects around 0.65 percent of consumption in revenues in 2030 (relative to BAU).
  - Revenues fall short of average incidence by 0.25 percent of consumption.
  - Small reductions in marginal labor tax rates using carbon revenues can fully offset household incidence at modest additional fiscal cost.
  - Long-run labor tax reform: marginal tax rates reduced by roughly 0.8 percentage point for most workers (assumptions: long-run labor supply elasticity 0.5; flat labor demand).
  - Hours worked: carbon tax alone leads to hours worked decline ~0.6 percent; using revenue to reduce marginal labor taxes reverses decline.
  - Alternative lump-sum transfers would reduce inequality—e.g., fully compensating bottom quintile via around 30 percent of carbon revenues improves ex-post progressivity but may increase inactivity traps and lower overall labor supply relative to marginal-rate cuts.
- Incidence mechanics and modeling approach (refer Appendix II):
  - Uses Leontief production assumption, input-output tables, exporter wage pass-back, quasi-linear utility, and constant-rate-of-progressivity (CRP) tax schedule fitted parameters γ = 0.106 (s.e. 0.002) and α = −0.965 (s.e. 0.041), regression R-squared = 0.95.
  - Embodied carbon facts: EITE imports embodied carbon 14 million tons in 2015 (30 percent of domestic CO2); around 35 percent of domestic carbon emissions in 2015 were embodied in exports.

### X. Industry burden conceptual analysis (Appendix I) and illustrative sector impacts
- Conceptual decomposition:
  - Emissions reduction decomposed into reduced emissions intensity (∆E_int) and reduced output (∆E_out) summing to ∆E_tot.
  - Under carbon pricing: total burden = second-order efficiency cost (resource cost of cleaner production) + first-order transfer payment (payments for remaining emissions).
  - Free allocation offsets transfer payment but loses efficacy under deep decarbonization.
  - Feebates reduce first-order transfer payments and can lower overall burden, especially at moderate abatement levels.
  - Numerical example: if ∆E_tot/∆E_int = 1.5 then feebate burden is 21%, 50%, and 90% of carbon-pricing burden for emissions reductions of 25%, 50%, and 75% respectively.
- Illustrative impacts (examples preserved from source calculations):
  - Steel:
    - Traditional integrated steel: ≈2 tons CO2/ton steel; alternatives ≈0.3–0.4 tons CO2/ton.
    - At €50/ton CO2: integrated production cost increases ≈€100/ton steel; alternatives ≈€20/ton.
  - Cement:
    - Traditional process ≈1 ton CO2/ton cement; process emissions ≈70% of cement emissions.
    - At €50/ton CO2: traditional production cost increase ≈€50/ton cement (~40%); more efficient plants ≈€30/ton; CCUS-fitted plants ≈€8–25/ton.
  - Feebates can substantially change relative price incentives across plant types in these sectors.

### XI. Policy synthesis and conclusions
- Use a mix of instruments: strengthen and harmonize carbon pricing while deploying sectoral feebates, targeted subsidies, BCAs, and regulations to handle leakage, distributional effects, monitoring limits, and political acceptability.
- Strengthening carbon pricing to higher, more uniform levels is the most cost-effective single approach to reach emissions neutrality goals, but complementary sectoral measures are required to close remaining gaps and manage uncertainty.
- Finland as a prototype: aiming for emissions neutrality by 2035—"15 years ahead of most other countries"—offers lessons applicable elsewhere.

*Source: wpiea2021171-print-pdf (excerpt).*

### References .............................................................................................................

### wpiea2021171-print-pdf - References ................................................................................................................................35

### I. Introduction — purpose and proposed fiscal package
- Finland has cut greenhouse gas (GHG) emissions by about a fifth since 1990 and has set a goal of becoming emissions neutral—annual CO2 equivalent emissions must be offset by annual absorption from carbon sinks—by 2035.
- Finland introduced a carbon tax in 1990 and already faces significant carbon pricing, but further measures are needed to achieve emissions neutrality by 2035.
- Paper proposes a package of fiscal measures that: scales up and harmonizes carbon pricing; combines carbon pricing with revenue-neutral feebates (tax-subsidy schemes); targets transport, industry, buildings, forestry, and agriculture; assumes power will be largely decarbonized by 2030.
- Target effective carbon price: raise all effective carbon prices to €125 ($150) per tonne by 2030.
- Estimated household impact: a €125 per tonne carbon price imposes a burden on the average household of 1.8 percent of consumption (relative to a baseline with current carbon pricing).
- Proposed revenue use: carbon pricing revenues could be used to reduce labor tax rates while boosting hours worked to largely offset household burden.

### II. Recommended economy-wide and sectoral fiscal instruments (summary)
- Economy-wide
  - Impose carbon surcharges on fuels by sector to raise the effective carbon price—the sum of the surcharge, fuel/carbon taxes, and pricing from the EU ETS (net of taxation warranted for domestic environmental externalities)—to a target level, progressively harmonized across fuels/sectors.
  - Set target prices to reach €125 per tonne by 2030.
  - Embed carbon pricing as part of broader fiscal reform involving broad reductions in labor income taxes to boost the economy in a distributionally balanced manner.
- Road transport
  - Introduce feebates for passenger vehicles: sliding scale of fees/rebates applied to vehicles with above/below average CO2/km.
  - Incentives can be set aggressively to promote electric vehicles without a fiscal cost or new tax burden on the average motorist.
  - Raise the diesel tax to the gasoline tax.
- Industry
  - Introduce feebates: sliding scale of fees/rebates on firms with emission rates above/below the industry average emission rate.
  - Feebates provide incentives for cleaner production processes while lessening competitiveness and emissions leakage concerns.
- Buildings
  - Supplement energy efficiency regulations with: (i) tax-subsidy scheme promoting shift from natural gas heating systems to electric or other clean fuel systems; (ii) feebates to promote more efficient appliances and lighting.
- Land use
  - Introduce a nationwide feebate applied to landowners equal to a CO2 price times the difference between carbon storage on their land in a baseline year and carbon storage in the current year.
  - Approximate carbon inventories can be assessed using satellite and aerial imagery and on-the-ground sampling.
  - Design is fiscally neutral and promotes increasing storage with no burden on the average landowner.
- Agriculture
  - Introduce feebates: sliding scale of fees/rebates applied to the difference between a farm's emissions per acre and the industry average emissions per acre scaled by farm-level acreage.
  - Emissions can be estimated based on farm-level inputs and default emissions factors.
  - A shift from beef and dairy to poultry and crop-based production could be reinforced by 'sin' taxes at the consumer level.

### III. Rationale for combining scaled carbon pricing and feebates
- Carbon pricing scaling and harmonization:
  - Carbon pricing can be scaled up through carbon surcharges that raise effective carbon prices—from the surcharge, fuel taxes, and the EU ETS—to a target carbon price by fuel and sector.
  - Effective carbon prices should be defined net of taxes warranted by domestic environmental considerations.
  - Progressive harmonization across fuels/sectors enhances cost effectiveness.
- Role of feebates:
  - Feebates are revenue-neutral sliding scales of fees on above-average emitters and rebates on below-average emitters.
  - Advantages of feebates:
    - Provide a more flexible and cost-effective approach than regulations.
    - Provide strong mitigation incentives.
    - Avoid large tax burdens on the average household or firm.
    - Avoid fiscal burdens for the government.
    - Can often build off existing administrative capacity.
  - Sectoral fit:
    - Transportation: finer incentives for lower-emission vehicles relative to current tax system.
    - Industry: encourage cleaner production with less concern about leakage.
    - Buildings: reinforce shift to electric heating and efficient appliances.
    - Land use: comprehensive incentives for carbon storage in forests and peat land, designed to be fiscally neutral.
    - Agriculture: promote less emissions-intensive practices; indirect taxation would need reinforcement at consumer level to avoid leakage.

### IV. Finland’s emissions targets, projections, and sinks
- National pledges and targets:
  - Finland pledged to become GHG neutral by 2035.
  - By 2035 annual GHG emissions should be 21 million tonnes (Mt) of CO2 equivalent, including both the ETS and ‘effort sharing’ sectors—a reduction of 70 percent below 1990 levels (71 Mt CO2 equivalent).
  - Finland’s 2035 target implies a cut of 55 percent below 1990 levels in 2030 (assuming a linear pathway), consistent with the recently strengthened EU emissions target.
  - By 2050 Finland pledged to reduce emissions to 80 percent below 1990 levels (reaching 14 Mt CO2 equivalent).
- Historical and projected emissions:
  - GHG emissions in 2018 were 56 Mt CO2 equivalent and government projections estimate 53 Mt CO2 equivalent in 2019.
  - With existing mitigation policies (WEM scenario), projected emissions for 2035 are 39 Mt CO2 equivalent, or 36 Mt CO2 equivalent with measures that are planned but not yet implemented.
- LULUCF (land use, land use change, and forestry) sink:
  - LULUCF emissions were -10.3 Mt CO2 equivalent in 2018 and are estimated at -17.4 Mt CO2 equivalent in 2019.
  - Further increase in absorption will largely come from increasing the average age of forests given limited scope for expanding forest land.
  - Note: methodological challenges and year-to-year variability in estimating the LULUCF sink; it is an open question how variability will be accounted for (e.g., multi-period averages) when assessing achievement of the 2035 neutrality target.
  - Once forest age is optimized, additional forestland does not absorb additional carbon—other sinks (e.g., direct air capture) will be needed to offset remaining annual emissions.

### V. Emissions structure and sectoral shares (2018)
- Sources of GHGs (2018):
  - Fossil fuel CO2: 75%
  - Industrial processes: 10%
  - Agriculture: 12%
  - Waste: 3%
- Sector share of fossil fuel CO2:
  - Energy (mostly power/district heating): 44%
  - Industry (and construction): 17%
  - Transport: 28%
  - Other sources (mainly commercial and residential buildings and farm machinery): 11%
- Fuel shares of fossil fuel CO2:
  - Coal/peat: 17%
  - Oil: 65%
  - Natural gas: 18%
- Power generation by fuel type:
  - Coal/peat: 18.0%
  - Natural gas: 5.1%
  - Oil: 0.9%
  - Hydro: 0.1%
  - Other renewable: 60.6%
  - Biomass: 15.4%

### VI. Coverage and market signals
- EU ETS coverage:
  - 43 percent of economy-wide emissions in Finland are covered by the EU ETS (just under the EU average of 45 percent), primarily power/district heat generation and large industrial firms (both CO2 and process emissions).
- EU allowance prices:
  - EU allowance prices have a history of volatility though prices have risen and become more stable recently (see referenced Figure 5).

### VII. Projected impacts and uncertainties
- Global context:
  - Global fossil fuel CO2 and other GHGs need to fall rapidly—by about 25 percent below 2018 levels by 2030 to contain future warming to 2°C, or 50 percent below for the 1.5°C target and continue to decline rapidly thereafter.
  - Global CO2 emissions in 2020 are projected to be about 8 percent lower than in 2019 due to lower GDP and structural shifts; projections suggest global emissions in 2030 will be about 20 percent above 2018 levels.
- Finland BAU projection:
  - In a business as usual (BAU) scenario (current mitigation policies frozen), IMF staff project emissions are flat after 2020 as trend reductions in the energy intensity of GDP offset expanding GDP.
- Policy shortfall:
  - A €125 carbon price combined with other current measures still leaves a shortfall relative to national and sectoral mitigation objectives.
- Uncertainties:
  - Large uncertainties about the carbon prices consistent with Finland’s emissions neutrality goal and acceptability constraints on carbon pricing.
  - Feebates can serve as insurance if emissions impacts of carbon pricing are smaller than projected and provide reinforcing measures to achieve stated objectives.

*Source: wpiea2021171-print-pdf - References .............................................................................................................*

### introduction of the Market Stability Reserve (MSR) which withdraws allowances (sometimes

### Introduction of the Market Stability Reserve (MSR)

### MSR and EU ETS interactions
- The Market Stability Reserve (MSR) withdraws allowances (sometimes permanently) when the amount of banked allowances exceeds a threshold level.
- As of March 2021, allowance prices are around €45 per tonne.
- With a fixed cap on emissions at the EU level, emissions reductions from overlapping policies in Finland would be offset tonne-for-tonne by extra emissions in other EU countries (via a decline in the ETS allowance price).
- This problem is, to some degree, offset if the MSR is operating.
- A more robust and transparent mechanism would be to underpin the EU ETS with an exogenous price floor ramping up over time.
- Price floors might be implemented through allowance auctions with a minimum auction price or making the MSR subject to a price trigger (see Flachsland and others 2018)—either way, allowances should be permanently withdrawn from the system whenever needed to prevent the price falling below the floor.

### Free allocations, BCA, and EITE industries
- Energy intensive, trade exposed (EITE) industries (e.g., iron/steel, cement, chemicals, refineries, pulp/paper) are granted free allowance allocations to address competitiveness and leakage concerns.
- The European Commission is considering replacing this mechanism with a border carbon adjustment (BCA).

### Finland’s effort-sharing sector and targets
- 57 percent of current GHGs in Finland are from the effort-sharing (i.e., non-ETS) sector, primarily transport, buildings, and agriculture.
- Finland is currently required to reduce emissions in the effort-sharing sector by 39 percent to 21 Mt CO2 equivalent by 2030 relative to 2005 levels, though this target will be revised following the EU’s recently enhanced mitigation pledge.
- Even the current target is quite stringent and is unlikely to be attained without stronger policy action.

### Process emissions trajectory
- Process emissions also include fluorinated gases used in refrigerants and aerosols.
- These emissions are projected to drop from current levels of about 1.2 Mt CO2 equivalent to about 0.4 Mt CO2 equivalent in 2030 (MOE 2020, Figure 1.6) due in part to EU regulations (e.g., on emission limits for vehicle air conditioners).

### Carbon and fuel taxes in Finland
- According to OECD estimates (see Table 3), the effective carbon tax for gasoline, diesel, and heating fuels for buildings is currently:
  - gasoline: €77 per tonne
  - diesel: €75 per tonne
  - heating fuels for buildings: €62 per tonne
- Additional fuel taxes are equivalent to carbon charges of:
  - gasoline: €234 per tonne
  - diesel: €124 per tonne
  - heating fuels for buildings: €29 per tonne
  - coal/peat use by industry: €10 per tonne
  - natural gas use by industry: €31 per tonne
- Substantial taxation of road fuels is warranted by non-carbon externalities.
- Finland’s carbon tax is at the top end of many carbon pricing schemes—only Sweden has significantly higher prices (for non-EU ETS emissions).
- Taxes in Table 3 refer to 2018—for comparability, these are used for the core analyses in this paper. The taxes have been revised up in 2021. See Table A1.

*Source: wpiea2021171-print-pdf - introduction of the Market Stability Reserve (MSR) which withdraws allowances (sometimes*

### 4. In addition, carbon tax rates shown above are for Finland's effort sharing sector.

### 4. In addition, carbon tax rates shown above are for Finland's effort sharing sector.

### Transportation: targets, taxes, and vehicle policy
- EU fleetwide average CO2 emission standard for new passenger vehicles: 95 g/km in 2021 with a 37.5 percent reduction in 2030 compared to the 2021 standard.
- Finland targets by 2030:
  - (i) a 50 percent reduction in transportation emissions relative to 2005 levels (and a 100 percent cut by 2045);
  - (ii) a share of biofuels in transportation rising from 20 percent in 2030 to 30 percent in 2030;
  - (iii) 250,000 EVs (either full or plugin) and 50,000 natural gas vehicles on the roads (currently there are 29,000 and 9,000 of these vehicles respectively).
- Road transport accounts for 95 percent of transportation emissions.
- Fuel excise rates:
  - Gasoline tax equivalent: €0.70 per liter.
  - Diesel tax equivalent: €0.45 per liter.
  - Annual propelling-force tax on diesel passenger vehicles averages €420 per annum.
- Retail gasoline prices for Finland are about at efficient levels; diesel fuel taxes fall short of their (second-best) efficient levels.
- Vehicle taxation and incentives:
  - Registration fees rise from 3 percent to 50 percent of the vehicle price as CO2 emission rate rises from 0 g/km to 360 g/km.
  - Average registration fee paid on gasoline, diesel, CNG, plug-in EVs and full EVs in 2019 was 15, 19, 9, 4 and 3 percent of vehicle prices respectively—and 15 percent of the vehicle price averaging over all vehicles.
  - Example: for a vehicle with price €30,000, the average registration fee would have amounted to €4,500.
  - Registration fees currently raise modest revenues of 0.1 percent of GDP.
  - EV subsidy: €2,000.
  - Annual taxes rise from €53 to €650 as CO2 emission rate rises from 0 g/km to 360 g/km and above—the average tax across all vehicles is expected to be about €220 in 2020, equivalent to about €2,050 over the vehicle life.
- Proposed transport feebate (vehicle-level):
  - Feebate formula: {CO2 price} × {vehicle’s CO2/km ─ sales fleet average CO2/km} × {average lifetime vehicle use (in discounted km)}.
  - Illustration: a feebate with a price of €800 per tonne of CO2, combined with a lump-sum tax of €5,500 to maintain revenue, would:
    - provide a net subsidy of €3,700 for an EV;
    - apply a fee of €12,000 to a vehicle with 200 g CO2/km.
  - Assumptions noted: fleet average emission rate of 115 g/km and discounted lifetime driving of 100,000 km.
  - A feebate that progressively shifted new sales to 100 percent EVs by 2035 would reduce road fuel emissions about 50 percent below otherwise projected levels for 2035 (assumes 8 percent of the fleet is replaced each year, initial 2 percent of new vehicle sales are EVs rising linearly to 100 percent by 2035).
- Diesel policy:
  - Per unit of fuel use diesel vehicles emit about 16 percent more CO2 emissions than gasoline.
  - Raising the road diesel tax by €0.172 per liter (to increase it to the level of gasoline tax) would:
    - increase current pump prices for diesel by 19 percent;
    - reduce diesel fuel consumption about 8 percent, or 0.6 Mt CO2e;
    - raise revenues of 0.1 percent of GDP;
    - increase the lifetime cost of using an average diesel vehicle by about €1,400.
  - Note: a lower tax might be warranted for off-road diesel uses via rebates.

### Buildings, heating, and appliances
- Emissions from space heating:
  - Declined from 4 to 2.5 Mt CO2 equivalent between 2015 and 2018.
  - Projected to decline to 1.7 Mt CO2 equivalent by 2030.
- Policy intentions:
  - Phase out oil heating in government buildings by 2025.
  - Aim to phase out use of fossil fuel heating in all buildings by the early 2030s (through grants for heating conversions and energy efficiency in residential buildings).
- Agriculture:
  - Agricultural emissions have remained relatively stable at 8-9 Mt CO2 equivalent since 2005 and on current policies are projected to remain at about this level till 2030.
  - Envisioned measures: promote carbon storage in soils and peatlands; reduce food waste by an aspirational target of 50 percent by 2030; nutrition programs to promote plant-based diets.
- Feebates and other instruments for buildings and appliances:
  - Feebate structures to encourage phase out of oil-based space heating and greater energy efficiency.
  - Revenues from interim tax on oil heating technologies could fund subsidies for electric heat pumps or hydrogen boilers.
  - Product-level feebate formula: {CO2 price} × {CO2 per unit of energy} × {energy consumption rate ─ industry-wide average energy consumption rate} (example metric for refrigerators: kWh/cubic foot cooled).

### National carbon pricing: role, targets, and impacts
- Carbon pricing rationale (listed functions):
  - promotes across-the-board behavioral responses to reduce energy and shift to cleaner fuels;
  - automatically minimizes the costs of these responses;
  - levels the playing field for clean technology investments;
  - mobilizes government revenue;
  - generates domestic environmental benefits (reductions in local air pollution mortality, traffic congestion);
  - is easily scaled up from a technical perspective (institutional capacity for the EU ETS and carbon tax well established).
- Current situation: ETS and carbon tax together comprehensively cover fossil fuel CO2 emissions in Finland, but effective prices vary substantially across sectors and fuels and do not ramp up predictably.
- Recommended approach:
  - Scale up carbon pricing as much as practically feasible and move towards more even carbon pricing across fuels and sectors by:
    - Setting a target trajectory of future effective carbon prices for each fuel in each sector;
    - Imposing annual surcharges on fuels equal to the difference between the yearly target price and the prevailing effective carbon price (combined effect of domestic carbon tax, fuel tax, and ETS carbon price, less any tax warranted by domestic environmental externalities);
    - Progressively harmonizing target carbon prices across fuels.
- Precedents and illustrative target:
  - Carbon Price Floor in the UK (national variable tax set three years in advance equal to difference between exogenous target price and projected EU ETS price).
  - Netherlands scheme: planned target price rising from €30 per tonne in 2021 to €125 per tonne in 2030.
  - Consideration for Finland: a carbon price, uniform across fuels and sectors, that ramps up from current effective levels to €125 per tonne by 2030.
- Impacts of increasing nationwide carbon pricing in Finland to €125 per tonne in 2030 (and freezing thereafter), applied in a coordinated way across sectors:
  - 2035 emissions would remain at 29 Mt CO2 equivalent, or 38 percent above the target level, leaving an emissions gap of about 8 Mt CO2 equivalent (calculation assumes carbon pricing covers 85 percent of future GHG emissions).
  - Extra revenue raised: 0.3 percent of GDP in 2030 (relative to revenue in 2030 under current levels of effective carbon pricing and accounting for erosion of bases for pre-existing fuel taxes).
  - Economic efficiency costs: about 0.2 percent of GDP (losses in consumer and producer surplus in fossil fuel markets, net of government revenue gains, accounting for pre-existing fuel taxes and emissions pricing).
  - Price effects in 2030 relative to 2030 levels with current carbon prices frozen:
    - Natural gas prices increase by 6.5 percent.
    - Electricity prices increase by 11.5 percent.
    - Gasoline prices increase by 8 percent.
- Complementary measures and risks:
  - Carbon pricing needs reinforcement by sectoral measures (feebates, regulations) because uncertainties over emissions impacts of carbon pricing increase at higher price levels.
  - Carbon prices in excess of €125 per tonne could further close emissions gaps but risk public backlash (example: France’s planned increase suspended at €45 per tonne due to protests).

### Sectoral feebates and industry application
- Feebates overview:
  - Revenue-neutral tax-subsidy schemes: fees for above-average emission rates and rebates for below-average emission rates.
  - Can be applied by finance ministries and build on existing tax/subsidy/EU ETS administration.
  - More flexible and potentially more cost-effective than regulations without extensive credit trading.
- Transportation feebate (vehicle-level) advantages:
  - Absolute tax incentives depend only on emission rate differences, not vehicle prices.
  - Automatically maintain revenue neutrality via updating fleet average emission rate.
  - Require no new data/administration beyond recalibration.
  - CO2 price in feebate can be adjusted to meet EV penetration targets (requires legislation).
- Industry feebate formula and scope:
  - Fee = {CO2 price} × {CO2/output ─ industry-wide average CO2/output} × {output}.
  - Applies to direct emissions (fuel combustion and processes); indirect emissions in electricity are priced under the EU ETS.
  - Potential target industries: metals, forest products, chemicals, cement/lime production.
  - These industries accounted for about 5, 5, 4, and 3 percent of 2018 nationwide GHGs in Finland respectively.
  - For highly concentrated industries, a modified feebate taxing the difference between a firm’s average emission rate and a target emission rate (progressively reduced over time) could be used.
- Compatibility and considerations:
  - Feebates avoid a first-order tax burden on the average producer, helping alleviate competitiveness and leakage concerns.
  - Scheme could build off existing monitoring procedures under the EU ETS.
  - Whether feebates are consistent with EU state aid rules is not entirely clear; operationally equivalent to uniform emission rate regulation with credit trading.

### Equity and incidence of carbon pricing
- Channels determining household incidence:
  - (i) direct price effect of higher energy prices;
  - (ii) indirect price effect of higher prices for other consumer goods;
  - (iii) changes in wages in trade-exposed sectors that cannot pass forward higher energy input costs;
  - (iv) how households benefit from recycling of carbon pricing revenues.
- Interaction with Finland’s progressive income tax schedule further affects overall incidence.
- Carbon pricing has a relatively modest impact on energy prices in Finland (illustrated earlier: natural gas 6.5 percent, electricity 11.5 percent, gasoline 8 percent for a €125 per tonne price in 2030).

*Source: Excerpt from the provided IMF chapter/section content.*

### 7.6 and 7.9 percent,

### wpiea2021171-print-pdf - 7.6 and 7.9 percent,

### Energy price impacts
- Retail electricity prices would rise by 11.5 percent, reflecting the share of zero-carbon fuels in generation and the phase out of coal.
- Coal price impacts are large, but coal accounts for a small share of energy consumed by households (excluding district heating).
- The standard deviation of the quarterly gasoline retail price index since 2010 has been around 6.5 percent.
- Carbon pricing increases exclude biomass.
- Figure reference: "Energy Price Impact of Carbon Price, 2030 (Percent Increase Relative to BAU)."

### Household energy consumption and budget shares
- On average, energy and fuel consumption accounts for 6.3 percent of total household consumption in Finland (2015 data).
- Example calculation method: "Pre-tax direct impact = {budget share} × {percentage increase in energy price}." If the budget share for a certain energy product is 5 percent and the price rises 10 percent, the pre-tax real income decrease equals 0.5 percent.
- Eurostat tables on the structure of consumption expenditure for 48 aggregated categories by income quintile for 2015 are used and assumed to apply for 2030.
- For 2015 in Finland, the ratio of total household consumption expenditure as estimated from the household budget survey to the national accounts was 0.90.
- Household energy consumption follows an inverted U-shaped pattern across the income distribution; middle income households spend relatively more on road fuels and solid fuels, including firewood. Electricity and district heat consumption are broadly equal as a share of total household consumption.

### Indirect price effects, input-output treatment, and assumptions
- Indirect price effects are calculated assuming increases in energy production costs are fully and immediately passed forward onto domestic prices of goods and services.
- A supply-use table for Finland for 2016 is used to calculate sectoral price responses and these price impacts are assumed the same in 2030.
- The calculations assume the price of non-energy intermediate and final imports stays constant when domestic energy prices increase, i.e., no Border Carbon Adjustment (BCA).
- Under a full BCA with export rebates, the price of imports increases in proportion to their carbon content; estimating that incidence accurately would require inter-country input-output tables with country-sector specific carbon emission intensity of production.
- Assuming constant household budget shares and energy intensities of production will lead to some overestimate of carbon tax incidence in 2030 as energy efficiency will continue to improve.

### Wage effects and tax interaction
- Households are affected by changes in wages; higher energy prices lead to different wage changes across sectors given different energy intensity of production, labor income shares, and exposure to trade.
- Sectoral wage changes multiplied by the share of output that is exported are matched to household survey data to calculate impacts by household income group.
- Two channels by which income tax interacts with price and wage effects:
  - Effect on after-tax consumption: consumer price increases only apply to the post-tax share of income (example: a household with effective average tax rate of 30 percent faces higher consumer prices only on 70 percent of its income).
  - Effect through the effective marginal tax rate: declines in earnings reduce tax liabilities by the marginal tax rate.
- The progressivity of Finland's income tax schedule means higher-income households are relatively more shielded from the incidence through higher average and marginal tax rates.

### Distributional incidence of a €125 per ton CO2 carbon tax in 2030 (prior to revenue use)
- A carbon tax of €125 per ton of CO2 in 2030 imposes an average burden on households (relative to BAU) of 0.9 percent of consumption and follows a U-shape pattern.
- Bottom quintile: purchasing power would decrease by 0.9 percent of total consumption. Main channels: higher prices for electricity and district heating, lower wages, and higher prices for other goods and services.
- Top 20 percent: purchasing power erosion of 0.9 percent of consumption, with almost half of the incidence driven by lower wages.
  - Explanations:
    - Workers in energy intensive and trade exposed sectors (manufacturing of chemicals, paper and machinery) tend to be in households with higher incomes.
    - The share of labor income increases along the income ladder.
- Middle income households: largest incidence impact of around 0.97 percent of consumption, mainly from lower wages and higher prices for electricity and heating.
- Income taxes reduce the total burden more for higher income households than for lower income households.

### Revenues and compensating reforms
- In the scenario under consideration, the carbon tax collects around 0.65 percent of consumption in revenues in 2030 (relative to BAU).
- The average incidence is 0.9 percent of consumption, so revenues fall short of the average incidence by 0.25 percent of consumption.
- Policy implication: the incidence arising from the carbon tax can be fully offset at fairly modest additional fiscal cost by transferring proceeds and a modest top up from general funds back to households (for example, via labor tax cuts).
- The compensating reform would fully protect all household income groups and workers from higher energy prices (analysis assumes wages fully respond to energy price increases over the next 10 years so that it doesn’t lead to higher unemployment).

### Labor tax reform design and macro-labor effects
- Small reductions in marginal labor tax rates can compensate households for higher prices and lower wages due to the carbon tax.
- In the long run (about five years), with wages determined by constant returns to labor and capital, the labor income tax reform would reduce marginal tax rates by roughly 0.8 percentage point for most workers compared to the current schedule.
  - Assumptions: long-run labor supply elasticity of 0.5; labor demand is assumed to be flat; marginal tax rate schedule estimated from Survey on Income and Living Conditions 2016.
- Using carbon tax revenues to reduce marginal labor income tax rates would have a positive impact on hours worked and counteract adverse employment effects of higher energy prices.
- In the scenario considered, hours worked decline by around 0.6 percent as the carbon tax is passed forward to higher domestic consumer prices and passed back through lower wages in exporting firms.
- If tax reform uses revenue to compensate households for initial welfare loss, the decline in hours worked is reversed and hours worked do not change on net.

### Alternative transfer schemes
- Lump-sum transfers of carbon tax proceeds to households would further reduce inequality; consumption in the bottom quintile would increase on net by around [text truncated in source].

*Source: wpiea2021171-print-pdf - 7.6 and 7.9 percent, IMF.*

### 0.9 percent on average and decrease by 0.6 percent of consumption for households in the top

### wpiea2021171-print-pdf - 0.9 percent on average and decrease by 0.6 percent of consumption for households in the top

### Distributional impacts and revenue recycling
- A carbon pricing reform would raise costs unevenly: "0.9 percent on average and decrease by 0.6 percent of consumption for households in the top quintile."
- Alternative compensation design:
  - Fully compensate the bottom quintile through transfers using "around 30 percent of the carbon pricing revenues."
  - Compensating the bottom quintile this way "would improve the ex-post progressivity of the reform, but would increase inactivity traps."
  - Increasing inactivity traps "could lower overall labor supply compared with full recycling of carbon tax revenues through marginal tax rate reductions."
- Labor tax design trade-offs:
  - The reform scenario analyzed "assumes labor tax cuts that exceed the carbon tax revenues themselves to compensate everyone for the increase in carbon pricing."
  - Under some compensation approaches there "would be no employment benefits since marginal tax rates would not be reduced."
  - An alternative that lowers average tax rates (rather than marginal rates) "encourages labor force participation, but not extra hours worked on the job." Empirical evidence cited indicates "most of the labor supply elasticity is due to the former response."

### Carbon pricing, leakage, and competitiveness
- Border Carbon Adjustment (BCA) context:
  - The European Commission is considering an EU-level BCA.
  - A BCA can be implemented as an import tax/export subsidy or a requirement for importers to purchase allowances (possibly from a separate pool).
  - Free allowance allocation under the EU ETS currently addresses competitiveness for EITE industries but "will lose viability with deeper decarbonization of industry."
- Empirical leakage estimates:
  - The simple average leakage rate from increases in energy prices across the sample is "around 0.2."
  - "Finland is 0.2 (a consequence of high trade openness in embedded carbon) while the leakage rate for the EU-wide policies is 0.13."
- Instruments comparison:
  - Feebate: "may be a more efficient instrument for limiting leakage and competitiveness concerns than free allowance allocation" because it imposes a smaller burden and does not divert revenues.
  - BCA: "could be more efficient still" because it raises costs for imported goods, promotes the full range of mitigation responses, and "in fact it raises additional revenue."
- Key BCA design issues summarized (administration, mitigation incentives, revenue use), including considerations such as:
  - Extending coverage beyond EITIs increases administrative burdens and provides little additional mitigation for some product classes.
  - Measuring embodied carbon by exporting country "complicates administration... but provides incentives for governments of overseas exporting countries to strengthen mitigation policy."
  - Rebating domestic exporters reduces competitiveness concerns but "modestly reduces revenue from domestic carbon pricing."
  - Using some revenue for international climate finance/rebates "could help with CBDR-CR and lower risk of retaliation/WTO ruling."

### Forestry: promoting carbon storage and feebates
- Main nationwide channels to increase carbon storage: "(i) reducing deforestation; (ii) afforestation; and (iii) enhancing forest management (e.g., planting larger trees, fertilizing, tree thinning, increasing rotation lengths)."
- Feebate design for forestry:
  - Fee formula: {CO2 rental price} × {carbon storage on their land in a baseline period ─ stored carbon in the current period}
  - Feebates can reward all three channels and be designed "to be revenue-neutral in expected terms."
  - Payments should be rental payments (not large upfront payments) because "changes in carbon storage may not be permanent."
  - Rental payments should equal "the product of the carbon price times, the interest rate, and the number of years in a period."
  - Partial exemptions may be warranted "for timber harvested for wood products (e.g., furniture, houses) because the carbon emissions (released at the end of the product life) will be delayed."
- Practical monitoring:
  - Forest carbon inventories use satellite monitoring, aerial photography, and on-the-ground tree sampling.
  - LIDAR and low-level aerial photography can estimate wood volume more cheaply than full ground sampling.
  - Administrative costs might be reduced by "limiting sampling to once every several years" and building off regulatory approval procedures.

### Agriculture: emissions pricing, feebates, and consumer measures
- Emission sources and mitigation levers:
  - Reducing livestock herds reduces methane and nitrous oxide emissions; shifting from livestock to crop production lowers emissions because "CO2 equivalent emissions per acre... are much lower for crops than livestock."
  - Use of peat land is important: peat land is "currently responsible for half the GHG emissions from agriculture in Finland but accounting for only 10 percent of farmland."
- Pricing and leakage concerns:
  - A proxy price on agricultural emissions could be set using livestock, feed, fertilizer, acreage, default emissions factors, and conversion conventions, but "due to import competition, this may do little to promote consumer responses, could cause significant emissions leakage, and may face political resistance."
  - Fees may be largely passed back to farmers rather than to consumers, implying limited diet shifts and potential leakage via imports.
- Feebates for agriculture:
  - Feebate formula: {CO2 price} × {CO2 equivalent/acre ─ industrywide average CO2 equivalent/acre} × {acres under production}
  - Feebates "would cost-effectively promote all behavioral responses for reducing emissions per acre but with no first-order tax burden on the average farmer."
  - Integration into existing tax collection and Ministry of Agriculture and Forestry data is feasible.
  - Consumer-level measures: "taxes on meat and dairy products" could reinforce demand responses but may need gradual introduction, subsidies for plant-based alternatives, and communications to enhance acceptability.
- Illustrative feebate impact at a price of "€20 per tonne of CO2 equivalent":
  - Subsidies/fees per acre and percent of farm income:
    - Plant-based agriculture: -€14 per acre (18 percent of farm income)
    - Poultry farms: -€10 per acre (6 percent of farm income)
    - Pig farms: €11 per acre (7 percent of farm income)
    - Cattle farms: €14 per acre (14 percent of farm income)
    - Dairy farms: €29 per acre (18 percent of farm income)
  - Note: "Negative values denote a subsidy, positive values a fee."

### Policy synthesis and conclusions
- Use multiple, complementary mitigation instruments given uncertainties over individual tools and feasibility.
- "Strengthening carbon pricing, through higher and more harmonized pricing across sectors, would be the most effective and cost-effective way to make headway on Finland’s emissions neutrality goal."
- Sectoral supporting measures (feebates, targeted subsidies, consumer measures, BCAs) are needed to address leakage, distributional effects, monitoring constraints, and acceptability.
- Finland's mitigation strategy could serve as a prototype: Finland aims to be one of the first countries to become emissions neutral—"15 years ahead of most other countries that have pledged emissions neutrality targets." Lessons from Finland may inform mitigation strategies elsewhere.

*Source: wpiea2021171-print-pdf (excerpt).*

### REFERENCES

### REFERENCES

### Key cited works (selection)
- ACEA, 2018. ACEA Tax Guide 2018. European Automobile Manufacturers Association. Available at: www.acea.be/uploads/news_documents/ACEA_Tax_Guide_2018.pdf.
- Arregui, Nicolas, Christian Ebeke, Jan-Martin Frie, Daniel Garcia-Macia, Dora Iakova, Andy Jobst, Louise Rabier, James Roaf, Chen Ruo, Anna Shabunina, and Sebastian Weber. Forthcoming. “EU Climate Change Mitigation: Sectoral Policies”. EUR Departmental Paper, IMF.
- Batini, N, 2019. “Macroeconomic Gains from Reforming the Agri-Food Sector: The Case of France”. International Monetary Fund WP No. 19/41.
- IMF, 2019. Fiscal Policies for Paris Climate Strategies—From Principle to Practice. Washington, DC.
- IPCC, 2018. Global Warming of 1.5 O C. Inter-Governmental Panel on Climate Change. Available at: https://www.ipcc.ch/sr15.
- OECD, 2019. Taxing Energy Use 2019: Using Taxes for Climate Action, OECD Publishing, Paris.
- World Bank, 2020. “State and Trends of Carbon Pricing 2020” (May), World Bank, Washington, DC.

### Additional referenced topics and data sources
- Energy and emissions: IEA, 2020. Global Energy Review 2020: The Impacts of the Covid-19 Crisis on Global Energy Demand and CO2 Emissions.
- Carbon markets and policy analysis: Böhringer et al. (EMF 29), Flachsland et al. (2018), Branger and Quirion (2014), Hirst (2018), ICAP (2020).
- Distributional and tax design: Diamond and Mirrlees (1971); Parry (2004); Parry, Veung and Heine (2015); Heathcote, Storesletten, and Violante (2017).
- Sector-specific studies: van Reijven and others (2016) data used for industry boxes; VTT, 2018 (Scenarios for Finland until 2050); EAT–Lancet Commission (Walter Willett et al., 2019).

*Appendix I and referenced bibliography as provided in the source document.*

### APPENDIX I. BURDEN OF CARBON MITIGATION POLICIES ON INDUSTRIES — Conceptual analysis (summary)
- Framework:
  - The industry burden from carbon mitigation policies is decomposed into effects from reduced emissions intensity (∆E_int) and reduced output (∆E_out), summing to total emissions reduction ∆E_tot.
  - Three marginal-cost curves depicted: marginal cost of reducing emissions through reducing domestic industry output (upper curve), reducing emissions intensity of output (middle curve), and the envelope of these two curves (lower curve).
- Two components of burden under carbon pricing:
  - Second-order efficiency cost (red triangle): resource cost of adopting cleaner but costlier production methods.
  - First-order transfer payment (blue rectangle): payments to government for emission allowances covering remaining emissions.
  - Free allowance allocation offsets the transfer payment component, but becomes less effective at deeper decarbonization as the efficiency cost rises relative to the transfer payment.
- Comparison with feebates/regulation:
  - A feebate reduces emissions intensity with (approximately) no impact on output because it does not charge for remaining emissions.
  - The feebate’s efficiency cost equals the carbon-pricing efficiency cost times ∆E_tot/∆E_int.
  - Overall burden is generally lower under feebates, especially at relatively low abatement levels, because there is no transfer payment.
  - Numerical example: if ∆E_tot/∆E_int = 1.5 (i.e., two-thirds of emissions reduction under carbon price comes from reduced emissions intensity) then the burden of the feebate is 21, 50, and 90 percent of that under carbon pricing for emissions reductions of 25, 50, and 75 percent respectively.

### Illustrative impacts on production costs — Steel
- Baseline processes and emissions:
  - Traditional integrated steel production: about two tons of CO2 per ton of steel.
  - Alternatives (electrified scrap-based, CCUS, direct reduction): produce about 0.3–0.4 tons CO2 per ton of steel.
- Price and cost impacts at a carbon price of €50/ton of CO2:
  - Increase in cost of integrated production: about €100/ton of steel through the first-order transfer payment (about one sixth of recent steel prices).
  - Increase in cost under alternative technologies: about €20/ton of steel.
- Feebate example (assumptions from text):
  - Assuming industry average emission rate of 1 ton of CO2 per ton of steel:
    - Integrated production cost would increase €50 per ton of output.
    - Alternative technologies would receive a subsidy of about €$30 per ton of output.
  - Note: technology switching incentives vary across plants and depend on local fuel and electricity prices; retrofitting existing plants is more likely than scrapping and rebuilding.

### Illustrative impacts on production costs — Cement
- Baseline processes and emissions:
  - About 90 percent of cement is produced using traditional kilns; process produces about 1 ton of CO2 per one ton of cement.
  - Process emissions contribute about 70 percent of total cement emissions.
- Alternatives and abatement potential:
  - State-of-the-art plants: currently about 10 percent of production; largely eliminate non-process emissions.
  - CCUS options:
    - Post-combustion: reduce emissions about 55 percent; increase capital costs by about 25 percent.
    - Oxy-combustion: reduce emissions about 85 percent; increase capital costs by about 100 percent.
- Price and cost impacts at a carbon price of €50/ton of CO2:
  - Increase in cost of traditional production: about €50 per ton of cement, or about 40 percent.
  - Increase in price of more efficient plants: €30 per ton of output.
  - Increase in price of CCUS-fitted plants: €8–25 per ton of output.
- Feebate implications (as presented):
  - A feebate with price $50/ton of CO2:
    - Increases cost of traditional production by €5 per ton of cement.
    - Provides a subsidy to more efficient plants of €10 per ton of output.
    - Provides a subsidy to CCUS-fitted plants of €18–35 per ton of output.

*Appendix I figures, calculations, and referenced data points are reproduced exactly as presented in the source.*

### APPENDIX II. HOUSEHOLD INCIDENCE OF A CARBON TAX

### APPENDIX II. HOUSEHOLD INCIDENCE OF A CARBON TAX

### Model structure and key assumptions
- Consumer price changes follow the approach proposed by Coady (2006).
- Production and pass-through:
  - Representative firm profits Πj = pj f(Aj, Ej, Lj) − qAj − rEj − wjLj (equation (1) notation preserved).
  - Leontief production is assumed (Yj = f(Aj, Ej, Lj)) so input shares are fixed.
  - Producer and user prices related by q = p + t, with t general sales and excise taxes, tariffs, and subsidies.
  - A carbon tax τ directly affects the user price of energy r and indirectly prices of other domestic inputs q (see equation (2) notation).
  - For domestic suppliers:
    - Higher input costs are fully passed on through higher producer prices p^d_j.
    - Profits are constant through a no-arbitrage condition.
    - User prices of imported goods are not affected.
  - These assumptions imply ∂p^d_j/∂τ = ∂q^d/∂τ α a_j + ∂r/∂τ e_j (equation (3) notation).
  - Price-change system in percentage terms leads to ∂q̇^d/∂τ = ∂ṙ/∂τ Σ K K, with K K = (I − α Λ)^{-1} (Leontief inverse) (equation (4) notation).
- Exporters:
  - Perfect substitutability with foreign goods implies higher input costs are passed back to wages.
  - No labor reallocation between exporters and domestic suppliers.
  - Wage response for exporters: ∂ẇ^*_j/∂τ = −∂ṙ/∂τ Σ (I + Θ α Λ) λ^{-1} (notation preserved; λ diagonal matrix of labor intensity).
- Policy scope assumptions used in empirical implementation:
  - Only the price of energy and domestically produced non-energy goods are affected by the carbon tax.
  - Only wages paid by exporters are reduced.
  - User prices of imported goods are held fixed (no pass-through).

### Household preferences, labor supply, and welfare incidence
- Utility and consumption aggregation:
  - Quasi-linear utility: Ui(Ci, Li) = Ci − 1/(1+1/e) Li^{1+1/e} (equation (5) notation; e is elasticity of labor supply).
  - Consumption Ci is a CES bundle of j goods with elasticity of substitution σ and skills-specific expenditure shares ω^j_i (equation (6) notation).
  - Ideal price index Pi for skills group i given as in text (notation preserved).
- Budget constraint and labor supply:
  - Budget: Pi Ci = wi Li − T(wi Li) (equation (7) notation).
  - Labor supply first-order condition: L^*_i = [ (wi/Pi) (1 − T′(wi Li^*)) ]^{e} (equation (8) notation).
  - Log-change identity for labor supply: d log L^*_i = e d log(wi) − e d log(Pi) + e d log(1 − T′(wi Li^*)) (equation (9) notation).
- Welfare incidence formula:
  - Consumer incidence of a carbon tax change for skill i given by (equation (10) notation):
    - (1 − mtr) d ẇ_i − (1 − atr) Σ_j ω^j_i d ṗ_j
  - Interpretation preserved:
    - First term: percentage change in wages multiplied by one minus the marginal tax rate (1 − mtr).
    - Second term: percentage changes in consumer prices, weighted by final consumption budget shares ω^j_i, multiplied by one minus the average tax rate (1 − atr).
  - With quasi-linear utility, changes in consumer surplus dUi^* are equivalent to changes in income; welfare changes expressed as a share of pre-tax real earnings wi/(Pi Li^*_i).

### Empirical implementation and data
- Energy price changes:
  - Use energy price changes resulting from a carbon tax from the model developed by Parry and others (2014).
- Non-energy price changes:
  - Calculated from an input-output table for Finland using equation (2) and matched to consumption tables by income quintiles from Eurostat.
- Wage changes:
  - Wage changes for exporters by sector calculated from equation (3) and matched to household survey data to calculate changes in wages throughout the income distribution.
- Tax schedule and parameters:
  - Marginal and average tax rates for Eq. (10) estimated using a constant rate of progressivity (CRP) tax schedule fitted to 2017 Income Distribution Survey data accessed via the LIS platform.
  - CRP functional form: T(wi Li) = wi Li − (1 − α)/(1 − γ) (wi Li)^{1 − γ}.
  - Estimated parameters: γ = 0.106 (standard error 0.002) and α = −0.965 (standard error 0.041).
  - Regression fit: R-squared = 0.95.
- Labor supply and compensating reform:
  - Changes in labor supply reported in main text obtained using equation (9).
  - Compensating tax reform calculated using Proposition 2 in Tsyvinski and Werquin (2019) by replacing the exogenous wage change with (d ẇ_i − 1/(1 − γ) Σ_j ω^j_i d ṗ_j).

### Relevant quantitative facts and sectoral statistics (as presented)
- Embodied carbon and trade:
  - Embodied carbon in EITE imports was 14 million tons in 2015, or 30 percent of the size of domestic CO2 emissions.
  - In 2015, around 35 percent of domestic carbon emissions in Finland where embodied in exports.
- Estimated CRP parameters and fit:
  - γ = 0.106 (standard error 0.002)
  - α = −0.965 (standard error 0.041)
  - Regression R-squared = 0.95
- Table A1. Carbon Taxes by Major Fuel Type in Finland, €/tonne (sector/fuel type 2018 2021 — values preserved as in source):
  - Electricity
  - Coal and other solid fuels00
  - Industry
  - Coal and other solid fuels2536
  - Natural gas7487
  - Road transport
  - Gasoline311337
  - Diesel199224
  - Buildings
  - Diesel92104
  - Note: "Cabon tax is the sum of domestic carbon tax and fuel tax (on energy content). See Table 3." Source entries: MOF (2021), MOF (updated), OECD (2019) (updated).

### Key implications for policy design and incidence analysis
- Incidence channels to evaluate:
  - Direct energy price increases (affecting consumer price index and consumption bundles).
  - Indirect pass-through to non-energy domestic goods via input-output linkages (Leontief framework and Leontief inverse K K).
  - Wage effects for exporters (wage reductions for exporter-paid labor under perfect substitutability assumption).
- Distributional adjustments from tax system:
  - Welfare losses from wage declines are mitigated by lower tax liabilities by one minus the marginal tax rate.
  - Welfare losses from price increases are mitigated because consumption expenditures are post-tax, adjusted by one minus the average tax rate.
- Practical implementation notes used in the empirical exercise:
  - Use sectoral input-output linkages and consumption shares by income quintile to map price changes to households.
  - Use household microdata to estimate tax schedule parameters and to match wage change exposures across the income distribution.
  - Compensating tax reforms can be derived analytically (Tsyvinski and Werquin (2019) Proposition 2) by accounting for price-change-induced effective wage changes.

*Source: APPENDIX II. HOUSEHOLD INCIDENCE OF A CARBON TAX (from the supplied PDF content).*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2021/english/wpiea2021171-print-pdf.pdf_