## Balancing Environmental, Fiscal, and Welfare Impacts of Transportation Decarbonization in France

## Source details

**Canonical URL:** [Balancing Environmental, Fiscal, and Welfare Impacts of Transportation Decarbonization in France](https://www.imf.org/-/media/files/publications/wp/2024/english/wpiea2024145-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2024/english/wpiea2024145-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2024/english/wpiea2024145-print-pdf.pdf.json)

---

### Executive summary — context, targets, and current situation
- France committed to reducing emissions by 50 percent below 1990 levels by 2030 and to net zero by 2050.
- Global compatibility: limiting warming to 1.5 to 2°C requires cutting global CO2 by 50 and other greenhouse gases by 25 percent by 2030 relative to 2019.
- France sets five-year carbon budgets at the sector level and coordinates climate analysis and policy through the General Secretariat for Ecological Planning (SGPE).
- Transportation is the largest source of emissions in France at around one-third of total emissions; the sector’s share rose from one-quarter to around one-third as other sectors reduced emissions faster.
- Around 90 percent of sectoral emissions come from road fuels.
- The transportation sector currently raises nearly 1.5 percent of GDP in revenue; under current policies fiscal revenues are projected to fall from 1.4 to 0.9 percent of GDP as driving shifts to lightly taxed electricity.
- Declining EV costs create an opportunity to accelerate emissions reductions with potentially limited welfare impacts, but existing policies alone are insufficient to meet sectoral targets.

### Quantitative findings on emissions, targets, and projections
- Transportation accounted for 32 percent of total emissions in 2022, equal to 130 million tons (Mt) of CO2 equivalent.
- Sectoral carbon budgets and provisional SGPE targets:
  - Transportation sector annual carbon budget: 113 Mt for 2024-2029 and 94 Mt for 2029-2033 (subject to revision; provisionally set at 100 and 83 Mt for 2024-2028 and 2029-2033).
  - Road transportation sector target: reduce road transportation sector emissions to 81 Mt by 2030 (SGPE 2023).
  - Provisional carbon budget: 75 Mt by 2030.
- Projected emissions under current policies:
  - Road transportation emissions projected to fall to 94 Mt by 2030 (a 22 percent reduction relative to 2023 levels), still exceeding the carbon budget and SGPE 2023 target.
  - Passenger and light vehicle emissions decline from 88 Mt in 2023 to 66 Mt by 2030.
  - Heavy goods vehicle emissions reduction: 17 percent by 2030 (relative) versus 22 percent for passenger vehicles by 2030.
- EV uptake:
  - EV (including plug-in hybrids) share of new passenger car sales: 25 percent in 2023 (4.1 percent of total passenger vehicle stock in 2023).
  - Share of new heavy goods vehicle registration electric: 6.2 percent in 2023.
  - Under existing policies: over half of new car sales are EVs by 2030 and 100 percent by 2035; around 40 percent of used cars are EVs in 2030.
  - SGPE 2023 target: EV share of passenger vehicle sales of 66 percent and 15 percent of the stock by 2030; 51 and 50 percent of new sales for light and heavy-duty commercial vehicles, respectively.

### Fiscal impacts and revenue projections
- Current road-transportation revenue components:
  - Excises on diesel and petrol for road transportation (~1 percent of GDP) and VAT on fuel and vehicle purchases (~0.4 percent of GDP) — total around 1.4 percent of GDP in 2023.
  - Malus and driving tolls raise relatively little revenue; EV and charging subsidy costs around 0.1 percent of GDP.
- Projected revenue dynamics:
  - As EV share increases, revenue falls by about 0.07 percentage points per year, reaching 1 percent of GDP by 2029.
  - Projected fiscal revenue decline from electrification: about 0.5 to 0.6 percent of GDP over the next decade under current policies (different statements preserve both figures from source).
- Revenue from reform options:
  - Transitioning from bonus/malus to a revenue-neutral feebate and applying a separate feebate to heavy goods vehicles: fiscal savings of 0.1 percent of GDP (estimated fiscal savings of €3.2 billion).
  - Gradually increasing domestic carbon pricing (used as price floor for EU ETS II): raises revenue by 0.1 percent of GDP by 2030.
  - Road user and congestion charges calibrated to externalities: increase road transportation revenue to 2.5 percent of GDP when applied at efficient levels.
  - Efficient taxation packages generate additional revenue of 0.8 to 1.7 percent of GDP, depending on design.
  - One-cent per kilometer road user charge raises revenue of about 0.15 percent of GDP; congestion charge raises revenue by an additional 0.2 percent of GDP.
  - Carbon price of €44.6 per ton contributes €0.10 per liter of diesel and petrol each; a rebate of around €0.15 per liter applies for diesel used for heavy goods transport.

### Welfare, externalities, and efficient pricing conclusions
- First-best principle: price environmental costs by equalizing social and private prices — i.e., carbon price for climate damages; per-ton local pollutant taxes; real-time congestion charges; distance-based insurance premiums for accidents; per-kilometer axle-weight charges for road damage.
- Representative 2030 marginal environmental damages (2023 constant prices):
  - Marginal cost of congestion: €0.10 per km.
  - Road accidents: €0.015 per km.
  - Road maintenance (heavy goods): €0.04 per km.
  - Global warming: €242 per ton of CO2.
  - Local air pollution: €19,000 per ton of NOx and €38,000 per ton of direct PM 2.5.
- Welfare impacts of reforms:
  - Combined efficient road user charges (RUC) and energy taxes yield net welfare benefits of around 0.4 percent of GDP per year (equivalent to over USD 15 billion or 7 percent of annual transportation costs).
  - Efficient RUC and energy taxes reduce driving-related externalities and pollution; environmental benefits exceed economic costs by 0.4 percent of GDP.
  - Efficient RUC reduces driving by 19 percent for passenger vehicles in the model; efficient RUC applied at scale allows France to achieve road transportation mitigation targets.
  - Feebate yields small emissions and fiscal benefits but is unlikely alone to meet carbon budgets.
  - Increased broad-based EV subsidies generate fiscal costs and welfare losses; doubling broad-based subsidies leads to revenue loss of 0.4 percent of GDP by 2030 and increased economic efficiency costs.

### Evaluated policy options and modeled impacts (seven policies)
- Policy 1 — Retain domestic carbon price:
  - EU ETS II applied on top of domestic carbon price of €44.6 per ton.
  - Fiscal revenue and emissions reductions; carbon-price coordination options noted.
- Policy 2 — Introduce linear feebate:
  - Convert malus/bonus to a revenue-neutral, linear €250 per ton of CO2 feebate with separate schedules by vehicle category; increases by €10 per ton annually after 2030.
  - Reduces emissions by a couple percentage points; saves 0.1 percent of GDP in revenue relative to baseline.
- Policy 3 — Increased EV subsidies:
  - Increase average subsidies to €5,000 per passenger and €130,000 per heavy goods vehicle in 2025, phased out through 2040; reduces emissions but with significant budgetary and efficiency costs (fiscal cost of around 0.3 percent of GDP in 2030 relative to existing policies).
- Policy 4 — Efficient RUC:
  - Distance-based charge equal to marginal external cost of non-congestion externalities plus congestion charge on peak driving (12 percent of total driving); remove bonus/malus.
  - Aligns emissions with sectoral carbon budget; large reductions in driving; generates substantial revenue.
- Policy 5 — Efficient energy taxes:
  - Adjust electricity and fossil fuel taxes to capture marginal external cost of driving (scaled down for fuel economy improvements and standards); remove bonus/malus.
- Policy 6 — Efficient RUC and fuel taxes:
  - Combine policies 4 and 5 but scale down fuel taxes relative to policy 5.
- Policy 7 — Partial efficient RUC and fuel taxes:
  - Policy 6 but at half of efficient level for political acceptability; generates around two-thirds of full welfare benefits.

### Implementation sequencing, administrative feasibility, and interim measures
- Administrative path recommended:
  - Start with administratively feasible mechanisms such as annual circulation fees varying with distance driven and based on odometer readings and regular pre-payments for additional kilometers.
  - Build administrative capacity over time for targeted, distance-based road user charges and detailed distributional analysis.
  - Use proxy measures (odometer-based annual fees, frequent tolling on congested roads, peak-hour rate increases) while preparing for GPS/time-location-driver specific real-time charges.
- Revenue recycling and equity:
  - Couple revenue-enhancing measures with robust revenue recycling: investment in public transportation, cash transfers targeted at those most impacted, means-tested EV subsidies, reduced public transit fees, and improving accessibility of public transportation.
  - Exemptions can be used to bolster political acceptability but should be designed carefully to avoid undermining objectives (e.g., EV exemptions shift vehicle mix without reducing congestion).

### Modeling approach, key assumptions, and limitations
- Technoeconomic model disaggregates light-duty vehicles (LDVs) and heavy-duty vehicles (HDVs), new and used vehicles, ICEVs and EVs; TCO (total cost of ownership) NPV per-kilometer framework drives demand via logit choice.
- Key elasticities and technical assumptions:
  - Driving elasticities to post-tax marginal cost: -0.3 (light-duty) and -0.2 (heavy-duty).
  - Fuel economy elasticities: -0.4 (light-duty ICEVs) and -0.3 (heavy-duty ICEVs).
  - Useful life: 15 years (passenger and light), 8 years (freight).
  - VKT per vehicle (2022): 13,500 (passenger and light), 53,000 (freight).
  - EV purchase share (2023): 25% (passenger), 7.6% (light), 6.2% (freight).
  - Fuel/energy prices (2023): gasoline passenger USD 2.03 per liter, diesel passenger USD 1.94 per liter, diesel freight USD 1.41 per liter, electricity passenger USD 0.22 per kWh, electricity freight USD 0.14 per kWh.
  - Shadow carbon price for malus and weight penalty assumed at €150 per ton for ICE purchases with pivot point declining 4 percent per year.
  - EV subsidies assumed €2,5 00 per passenger and €50,000 for heavy goods (text preserves original formatting), phased out by 2035 in some simulations.
- Limitations:
  - Modal shifts not explicitly modelled; vehicle ownership levels exogenous; general equilibrium impacts (employment, trade) not captured; some policy interactions (e.g., supercredits) not captured.
  - Sensitivities: EU fuel economy standard welfare costs highly sensitive to EV prices (±25 percent changes can materially change welfare costs).

### Key policy recommendations and trade-offs
- Market-based measures that apply to all driving (road user/congestion charges, fuel taxes) should be the centerpiece for efficiency and fiscal resilience because they tax driving directly and apply to EVs and ICEVs alike.
- Complementary measures: promote teleworking, modal shifts, expand public transportation, and maintain targeted EV support for equity.
- Near-term, politically feasible reforms with measurable benefits:
  - Transition to a linear feebate (revenue-neutral feebate of €250 per ton pivot at 97 g CO2/km): estimated fiscal savings of €3.2 billion (0.1 percent of GDP) and modest emissions reductions.
  - Gradual increase in domestic carbon price to €90 per ton by 2030 and use as price floor for EU ETS II: raises revenue by 0.1 percent of GDP and yields net welfare improvements (estimated net welfare improvements of €1 billion in a noted statement).
  - Refine and target EV subsidies (means-tested and focused on equity) rather than broad-based expansion to avoid large fiscal and efficiency costs.
- Full efficient RUC and energy tax implementation achieves sectoral carbon budget and delivers environmental benefits net of economic costs of around 0.4 percent of GDP; but requires administrative capacity, phased implementation, and revenue recycling to protect vulnerable households and firms.

### Electricity and infrastructure implications
- Electricity demand from road transportation under scenarios:
  - Current policies: grows from 4 TWh in 2023 to 31 TWh in 2030.
  - Efficient RUC and energy tax scenario: road transport electricity demand reaches 20 TWh in 2030.
  - Increased EV subsidy: road transport electricity demand increases to nearly 40 TWh in 2030.
- National electricity use was around 450 TWh in 2023.
- Complementary grid measures recommended: smart metering, real-time electricity pricing, and demand-side management.

*Source: IMF Working Paper No. WP/2024/145 — Balancing Environmental, Fiscal, and Welfare Impacts of Transportation Decarbonization in France.*

### Executive Summary ......................................................................................................

### Executive Summary

### Context and national targets
- France is committed to reducing emissions by 50 percent below 1990 levels by 2030, followed by a decline to net zero by 2050.
- Compatible with global goals, limiting warming to 1.5 to 2°C requires cutting global emissions of carbon dioxide and other greenhouse gases by 50 and 25 percent, respectively, by 2030 compared to 2019 levels.
- France sets five-year carbon budgets at the sector level and coordinates climate analysis and policy through the General Secretariat for Ecological Planning (SGPE).

### Current situation: transportation emissions and fiscal implications
- Transportation is the largest source of emissions in France at around one-third of total emissions.
- The transportation sector’s share of national emissions grew from one-quarter to around one-third as other sectors reduced emissions faster.
- Declining electric vehicle (EV) costs create an opportunity to accelerate sectoral emissions reductions with potentially limited welfare impacts.
- The transportation sector currently raises nearly 1.5 percent of GDP in revenue.
- Under current policies, as driving shifts to relatively lightly taxed electricity, fiscal revenues generated by the sector are projected to fall from 1.4 to 0.9 percent of GDP.
- The economic cost of driving-related externalities, like congestion, is projected to worsen by half a percent of GDP by 2030.

### Assessment of existing policies
- France has a strong existing policy package, including electric vehicle purchase subsidies, EU passenger vehicle standards, fuel taxes, and public transportation investment.
- Despite this package, emissions from road transportation are projected to exceed France’s sectoral target.
- Existing policies are welfare improving but not sufficient to achieve targeted emissions reductions or to prevent a structural decline in sectoral fiscal revenue.
- Key interactions across overlapping EU-wide and domestic policies necessitate a standardized framework to assess fiscal, social, and emissions trade-offs.

### Quantitative findings on reform options and impacts
- Transitioning from the current bonus/malus system to a revenue-neutral feebate, and applying a separate feebate to heavy goods vehicles:
  - Would save 0.1 percent of GDP in revenue.
  - Would reduce emissions by a couple percentage points.
- Gradually increasing domestic carbon pricing and using it as a price floor for the forthcoming EU ETS II:
  - Provides similar revenue and environmental benefits to feebates.
- Increased EV subsidies:
  - Reduce emissions.
  - Come with significant budgetary and economic efficiency costs; subsidies should therefore focus on equity and distributional objectives and build off existing means-tested components.
- Road user and congestion charges calibrated to the size of externalities:
  - Align emissions with the sectoral carbon budget.
  - Increase road transportation revenue to 2.5 percent of GDP.
  - Yield environmental benefits net of economic costs reaching 0.4 percent of GDP.

### Policy conclusions and recommendations
- Changes to existing policies can strengthen fiscal and decarbonization incentives but are unlikely, by themselves, to achieve sectoral targets nor fully stop revenue loss.
- An integrated policy approach is recommended:
  - Fine-tune existing incentives (fuel taxes, feebates, bonus/malus adjustments).
  - Introduce distance-based road user charges and congestion charges to price driving externalities directly.
  - Coordinate carbon pricing reforms with EU ETS II to reinforce price signals.
  - Limit EV subsidies to targeted, means-tested measures aimed at equity and distributional objectives.
  - Couple revenue-enhancing measures with robust revenue recycling, including investment in public transportation and cash transfers targeted at those most impacted by reforms.
- Administrative transition path:
  - Start with administratively feasible mechanisms such as annual circulation fees varying with distance driven and based on odometer readings.
  - Build administrative capacity over time to implement more targeted, distance-based road user charges and to perform detailed distributional impact analysis.

*Source: IMF Working Paper — Balancing Environmental, Fiscal, and Welfare Impacts of Transportation Decarbonization in France*

### conclusion.

### conclusion.

### Overview of Transportation Emissions and Policies
- One-third of national emissions in France come from the transportation sector, with around 90 percent of sectoral emissions from road fuels.
- Road transportation emissions per capita and electric vehicle sales are similar to the average of other advanced EU countries.
- Sectoral emissions are projected to decline by 20 percent from 2023 to 2030 due to existing policies and declining EV costs, but still exceed the carbon budget and provisional SGPE 2023 target by 15 percent.
- The benefits of less congestion, pollution, and other externalities are expected to more than offset the economic costs, although economic costs increase as passenger vehicle emissions-standards tighten.
- There is generally overcharging for pollution costs and underpricing of driving-related externalities across vehicle types.
- Fiscal revenue from road transport is projected to decline by 0.5 percent of GDP by 2030 as driving shifts to relatively lightly taxed electricity.

### Transportation Sector Emissions and Targets
- Transportation is the largest emitting sector at 32 percent of total emissions in 2022 and 130 million tons (Mt) of CO2 equivalent.
- Transportation’s share of national emissions has increased from 26 percent in 2000.
- Reductions in emissions are due to improvements in fuel efficiency, which have more than offset increases in driving distances.
- The carbon budget for 2024-2029 and 2029-2033 are 23 and 36 percent below 2023 levels, respectively, implying a much larger annual emissions reduction than the 0.4 percent achieved since 2000.
- Transportation sector annual carbon budget is set at 113 million tons of CO2 equivalent (Mt) for years 2024-2029 and 94 Mt for 2029-2033 but subject to a revision that is provisionally set at 100 and 83 Mt for 2024-2028 and 2029-2033, respectively.
- Road transportation composes 90 percent of transportation sector emissions:
  - Passenger vehicles emit 68 Mt (17 percent of national emissions).
  - Heavy goods vehicles emit 30 Mt (7 percent).
  - Light commercial vehicles emit 20 Mt (5 percent).
  - Buses emit 3 Mt (1 percent).
- Passenger vehicles and light commercial vehicle emissions remain below pre-COVID levels while emissions from heavy good transport have returned to pre-COVID levels.
- Emissions per capita in France are around 1.8 tons, compared to an EU average of 1.6; France’s emissions per capita declined from 2.24 in 2002 at an annual rate of around one percent.
- Heavy goods vehicle distance travelled per vehicle in France is around 70 percent above the EU average.
- Passenger and light vehicle ownership is among the highest in Europe and above all comparators except for Italy.
- Emissions-intensities of passenger and light vehicles in France are much better than the EU average and most peers.
- EVs (inclusive of plug-in hybrids) share of new passenger car sales was 25 percent in 2023, up from 11 percent in 2020; EVs comprised 4.1 percent of total passenger vehicle stock in 2023.
- Share of new heavy goods vehicle registration that are electric was 6.2 percent in 2023.
- Public charging stations have rapidly expanded since 2020.
- French passenger ICEVs show relatively high fuel economy and French consumers purchase lighter ICEVs and more gasoline powered vehicles (gasoline has a 15 percent lower carbon content than diesel).
- Light commercial vehicles average around 150 g per km and have better fuel economy in France than other European countries.
- France’s road transportation sector target: reduce road transportation sector emissions to 81 Mt by 2030, amounting to a 40 percent reduction below a scenario with current policies (SGPE 2023).
- SGPE 2023 projected emissions reductions by source:
  - Passenger vehicles: 28 Mt total reduction, with EVs and more fuel efficient ICEVs contributing 14 Mt, modal shifts 5 Mt, carpooling 3 Mt, biofuels 3 Mt, and teleworking/sobriety 3 Mt.
  - Light and heavy goods: 17 Mt total reduction, with improvements in emissions-intensity and zero-emissions vehicles contributing 11 Mt, sobriety 7 Mt, modal shifts 4 Mt, load/logistics optimization 3 Mt, and biofuels 3 Mt.
- France targets an EV share of passenger vehicle sales of 66 percent and 15 percent of the stock by 2030, and 51 and 50 percent of new sales for light and heavy-duty commercial vehicles, respectively.

### Key Considerations Related to Road Transportation Policies
- Policy objectives span fiscal, efficiency, environmental, and distributional dimensions.
- Fiscal:
  - Government currently raises substantial revenue from existing transportation policies, but revenue is projected to decline and spending on subsidies to slightly increase.
  - Maintaining current levels of revenue requires increasing taxes on electricity for EV charging and vehicle purchases and/or imposing charges per distance driven; if introduced in isolation, such adjustments could disincentivize EV purchases.
  - Subsidies for EV purchases and charging stations have a negative impact on the fiscal balance and are projected to increase unless phased out.
  - Policy reforms that promote decarbonization without new charges on vehicle purchases and driving will reduce revenue due to erosion of existing tax bases.
  - Price-based policies (ETS with auctioned allowances, carbon and fuel taxes, distance-based charges) are revenue raising.
  - Revenue from decarbonization policies can be used to support vulnerable households and firms.
- Economic efficiency and environmental costs:
  - Objective may be to reduce transportation externalities by changing behavior on fuel economy (extensive margin) and driving amounts (intensive margin).
  - Practical priority: achieve intended behavior change at minimal economic cost, generally via broad-based pricing policy (e.g., economy-wide carbon pricing).
  - Alternative objective: maximize revenue from road fuel taxes, which is achieved when households and firms do not adjust behavior, but this conflicts with environmental goals.
- First-best welfare-maximizing policy is to adequately price environmental costs by equalizing social and private prices using charges equal to the marginal cost of the externality:
  - (i) Climate change damages: a carbon price equal to the carbon price implicit in France’s mitigation target.
  - (ii) Domestic health costs of local pollution: a tax per ton of local pollutant emitted, with tax levels varying based on exposure of vulnerable populations.
  - (iii) Congestion: a charge set at the marginal cost of congestion in real-time.
  - (iv) Road accidents: insurance premiums that vary according to distance driven and driver risk.
  - (v) Road damage: a per kilometer charge based on axle weight for heavy goods vehicles.
- GPS and tolling technology improvements make these measures practically implementable with the exception of local pollutant and (potentially) driver risk charges/premiums, which could be priced through annual vehicle registration fees that vary by pollution characteristics and distance driven or via higher fuel excises.
- Distributional impacts are not comprehensively quantified but acknowledged: households and firms differ in share of consumption allocated to road transportation and abatement costs.

### Road Transportation Sector Policies
- Policies categorized as:
  - Market-based measures that impose a charge on emissions.
  - Market-based measures that do not charge emissions.
  - Subsidies.
  - Regulations.

#### Market-Based Policies that Charge for Emissions
- Advantages: environmental, fiscal, and economic; minimize economic costs for a given emissions reduction; follow “polluters pay” principle.
- Challenges: politically difficult due to higher energy costs; importance of using revenues to compensate vulnerable households/firms and invest in public goods or reduce other taxes.
- France’s relevant measures:
  - Road fuel excises on diesel and petrol include separate carbon and non-carbon components and a partial rebate for heavy goods vehicles; VAT of 20 percent applies; road fuel excises in France are above EU averages.
  - France’s electricity excise is about one-third that of petrol and diesel on an energy equivalent basis but still two- and four-times EU averages for households and firms, respectively.
  - A second EU ETS (EU ETS II) will be introduced in 2027 or 2028 and apply to transportation emissions in addition to buildings and industrial emissions not covered in the original ETS. It is unlikely to impact road transportation fuel prices this decade since the ETS price is capped at €45 per tonne until 2030 and France could either remove its domestic carbon price when EU ETS II is introduced or keep the domestic tax and receive an exemption.
  - Distance-based toll on heavy goods vehicles is paid to road concessionaire and provides decarbonization incentives since it is higher for more polluting vehicles and reduces distance driven.
  - Revised Eurovignette requires differentiated heavy goods tolls based on CO2 and local air pollutant emissions and potentially other factors, but France has stated it will not introduce such differentiated tolling before 2031 due to contractual obligations under current road concessions.

#### Market-Based Policies that do not Charge on Emissions
- Characteristics: less effective than emission-charging policies; incentivize reductions in emissions-intensity but not quantity; politically more acceptable; do not raise revenue and erode fuel tax base; can exacerbate driving-related externalities by lowering marginal cost of driving.
- France’s measures:
  - Malus écologique: one-time sliding scale levies on newly registered passenger cars purchased by individuals with emissions above 118 grams of CO2 per kilometer; weight penalty charges ICEVs (but not plug-in or battery EVs) for each kg above 1,600.
  - A relatively small number of vehicles are covered by malus regimes (estimated to be 10 percent for the weight malus and a similarly low amount for the CO2 malus prior to recent changes) since thresholds are set above the average vehicle purchase and quickly escalating tax schedules shift purchases to non-taxed vehicles.
  - Company cars are subject to an annual tax based on CO2 emissions-intensity and Euro grade, both applying only to ICEVs.
  - Heavy goods vehicles pay an annual tax according to weight and number of axles, without an environmental component; France does not have a malus system for heavy goods vehicles.
  - EU passenger fuel economy standards require manufacturers to meet increasingly stringent targets and allow pooling across manufacturers.

#### Subsidies and Other Policies
- Subsidies: politically acceptable, can support equity, but fiscally costly and less efficient in reducing emissions; can address market failures and promote equity if well-targeted.
- France’s subsidies and other supportive policies:
  - Bonus écologique: subsidy for purchase of passenger zero-emissions vehicles of up to €7,000 varying by buyer income, vehicle purchase price, and other characteristics.
  - Heavy goods vehicles can receive a scrappage bonus of up to €50,000.
  - Regional grants of up to €6,000 and €13,000 for purchase of zero-emission passenger and heavy goods vehicles, respectively.
  - Means-tested leasing program providing concessionary EV leasing rates of €100-150 per month.
  - As of 2024, all EV subsidies are conditional on emissions-intensity of production, which excludes about 35 percent of EVs sold in the French market in 2023.
  - Grants for residential EV charging stations and public charging points, including at condominiums and employers.
  - Green industrial investment subsidized through a green investment tax credit of 20 to 45 percent.
  - Other policies: low emissions zones in over ten French cities, public investment in rail and cycling infrastructure, increased biofuel blending requirements, and electricity sector planning for grid resilience and increased demand.

*IMF Working Paper — conclusion (wpiea2024145-print-pdf).*

### 0.10 per liter of diesel and petrol, respectively, comes from the carbon price of €44.6 per

### Balancing Environmental, Fiscal, and Welfare Impacts of Transportation Decarbonization in France

### Policy Instruments and Details
- Carbon pricing and rebates:
  - Carbon price of €44.6 per ton of CO2 contributes €0.10 per liter of diesel and petrol, respectively.
  - A rebate of around €0.15 per liter for diesel used for heavy goods transport.
- EU ETS II:
  - Capped at €45 per ton (in real 2020 prices) until the end of 2029 and allowances will linearly decline at 5 percent per year from 2024 emission levels.
- Value-Added Tax:
  - 20 percent applied to diesel and unleaded petrol, and all vehicle purchases.
- Distance charge:
  - €0.13 per km levied through by road concession operators, with rates varying by axle rate and pollution.
- Market-based levies on vehicles:
  - Malus écologique: one-time levy on newly registered passenger vehicles; levies increase with emissions-intensity above 118 grams of CO2 per kilometer.
  - Weight penalty: one-time levy on newly registered passenger vehicles; levy increases with weight above 1,600 kilograms; EVs are exempt.
  - Annual CO2 tax: annual tax on company cars; tax increases with emissions-intensity but does not consider distance driven.
  - Annual atmospheric pollutant tax: annual tax on company cars; tax increases with worsening Euro-grade but does not consider distance driven.
- Emissions standards:
  - Passenger and light vehicle targets for new vehicles: 95 g per km for passenger and 147 for vans in the period 2020-2024; 93.6 for passenger and 153.9 for vans in 2025-2029; 49.5 for passenger and 90.6 for vans in 2030-2034; and all zero-emission after 2034.
  - Heavy goods emissions standard: 45 percent reduction in emissions per km for new vehicles by 2030 relative to 2019-2020 levels, 65 percent for 2035, and 90 percent for 2040.
- Subsidies and complementary policies:
  - Bonus écologique: means-tested subsidy up to €7,000 for EVs with costs below €47,000 (€4,000 for those above an income threshold); conversion premium up to €6,000; additional €1,000 if living or working in a low emissions zone; regional governments can provide additional grants.
  - Retrofit bonus: subsidy of up to €50,000 for replacing an ICEVs heavy goods vehicle with a zero-emissions vehicle, along with accelerated depreciation.
  - Social leasing program: subsidized leasing EV rates of €100-150 per month for low-income households that drive more than 15 kilometers to work.
  - Charging subsidies: tax credit of €330 per individual charger; up to 50 percent subsidy for shared chargers at condominiums and 40 percent at employers; reduced VAT of 5.5 percent for installation; regional governments can provide additional grants.
  - Green investment tax credit and grants: 20-45 percent tax credit for investment in manufacturing batteries and other low carbon technologies, as well as direct state aid through grants.
  - Biofuel blending mandate: 14% biofuel blending, with at least 3.5% being advanced biofuels.
  - Low emissions zones: 11 cities currently have zones where the use of highly polluting vehicles is restricted; it is expected that 32 areas will have such zones in place by 2025.

### Modeling Approach and Assumptions
- Model scope and disaggregation:
  - Technoeconomic model forecasts road transportation emissions and revenue; separate calculations for light-duty vehicles and heavy-duty vehicles; further disaggregation between used and new vehicles and ICEVs and EVs.
- Explicitly modelled policies and key parameter assumptions:
  - Malus and weight penalty: assumed to be a shadow carbon price of €150 per ton for ICE purchases with the pivot point declining by four percent per year.
  - EV subsidies: assumed to be €2,5 00 per vehicle for passenger and light vehicles and €50,000 for heavy goods and reducing annually to be fully phased out by 2035.
  - Charging station subsidies: assumed to cover 50 percent of charging costs and be phased out by 2030.
  - EU ETS II: assumed to remain at €45 per ton from 2027-29 and then increase by €10 per ton starting in 2030 but with France’s fuel tax reduced by €45 when the ETS is introduced.
  - EU fuel economy standards: modelled as the shadow carbon price needed to achieve the standard.
  - Heavy goods tolls: €0.13 per km and not differentiated by CO2 intensity prior to 2031.
  - Energy taxes: held at 2023 levels and indexed to inflation.
- Additional exogenous reductions:
  - To account for non-modelled decarbonization levers, an additional six and three Mt reduction in passenger/light commercial and heavy goods, respectively, are assumed to occur exogenously.

### Emissions Projections and Targets
- Projected emissions under current policies:
  - Road transportation emissions projected to fall to 94 Mt by 2030 under current policies but still exceed the carbon budget.
  - Passenger and light vehicle emissions decline from 88 Mt in 2023 to 66 Mt by 2030.
  - Over half of new car sales are EVs by 2030 and 100 percent by 2035; around 40 percent of used cars are EVs in 2030.
  - Heavy goods vehicles emissions reduction: 17 percent by 2030 (relative) versus 22 percent for passenger vehicles by 2030.
- Comparison with other estimates:
  - SGPE 2023 finds 2030 road transportation emissions of 114 Mt under existing policies; 35 and 52 percent of emissions reduction needed to reach 2030 targets for commercial and passenger vehicles, respectively, are achieved under existing policies.
  - EEA estimates existing policies result in 2030 emissions of 95 Mt and 82 Mt by 2035.
  - The paper’s model results are broadly in line but slightly lower than the EEA over the medium-term and lower than SGPE 2023.

### Fiscal Revenue and Welfare Impacts
- Revenue projections:
  - Current revenue from road transportation estimated around 1.4 percent of GDP: excises on diesel and petrol for road transportation (~1 percent of GDP) and VAT on fuel and vehicle purchases (~0.4 percent of GDP).
  - Malus and driving tolls raise relatively little revenue and EV and charging subsidy costs around 0.1 percent of GDP.
  - As EV share increases, revenue falls by about 0.07 percentage points per year, reaching 1 percent of GDP by 2029.
  - Decline in revenue from 2019 to 2023 attributed to lack of indexing excises to inflation, improved vehicle fuel-intensity, and less driving relative to income.
- Welfare and efficiency:
  - Economic efficiency costs from households and firms buying more expensive but cleaner vehicles and driving less are around 0.1 to 0.5 percent of GDP out to 2030 but increase afterwards as EU fuel economy standards become more stringent.
  - Benefits from reduced driving-related externalities and pollution more than offset economic costs; most environmental benefits derive from reduced driving-related externalities in the short-term.
  - Pollution benefits from reduced fossil fuel use increase over time as the social cost of carbon increases (up from €54 in 2020 to €500 in 2040, French Strategie 2019).
  - Pollution from passenger ICEVs tend to be priced above efficient levels while driving-related externalities for all vehicles, except for freight EVs, are undercharged; this undercharging worsens as EV share grows.

### Distributional Impacts and Equity Considerations
- Distributional impacts (data needs and findings):
  - Middle income and rural households likely lose the largest share of purchasing power from existing transportation fuel taxes, with impacts ranging from around 3.5 percent of consumption for the average middle income, rural household to 2.5 percent for wealthier, urban households.
  - Low electricity tax in transportation is regressive if EV ownership is concentrated among wealthier households; increasing electricity excise would be regressive for households that spend a greater share of their budgets on non-transport electricity.
  - Energy taxes can turn progressive if revenues are used to disproportionately favor the poor and rural communities.
- Distributional uncertainty for EV subsidies and feebates:
  - The distributional impact of means-tested EV subsidies (including social leasing) and bonus/malus depends on factors including affordability after incentives, whether energy efficiency is a normal good, pass-through of subsidies and taxes, differential local pollution exposure, and how net subsidy is financed.
  - One France-specific study finds bonus/malus disproportionately benefits the middle class if the net subsidy is financed through a uniform tax but is progressive when financed with a tax in proportion to income.
  - EV ownership data by income in France is not readily available; international evidence suggests initial EV uptake is skewed to wealthier households.

### Key Sensitivities and Model Limitations
- Model caveats:
  - Modal shifts are not explicitly modelled, vehicle ownership levels are exogenous, and general equilibrium impacts like changes in employment and trade are not captured.
  - Some policy interactions and the impact of “supercredits” for zero-emissions vehicle sales are not captured.
  - Differences across models are difficult to assess due to lack of documentation.
- Sensitivity examples:
  - Economic efficiency costs of the EU fuel economy standards are highly sensitive to EV prices: if EV prices are 25 percent cheaper than the baseline assumption by 2030 then economic efficiency costs never exceed one percent of GDP; if EV prices are 25 percent more expensive than the baseline assumption then economic efficiency costs reach nearly two percent of GDP.

*Source: IMF Working Paper — Balancing Environmental, Fiscal, and Welfare Impacts of Transportation Decarbonization in France*

### Box 1. Estimating Welfare Impacts and Efficient Tax Levels

### Box 1. Estimating Welfare Impacts and Efficient Tax Levels

### Methodology for Welfare Impacts
- Welfare impacts are calculated as the reduction in environmental externalities relative to the economic efficiency costs induced by the policy.
- Environmental externalities from driving considered: lost time from traffic congestion; health and property damages from vehicle accidents (excluding costs internal to the driver); damaged roads and road maintenance; health impacts from local air pollution; and economic costs from global warming (Parry et al 2014).
- Environmental benefits = change in km driven and pollutant emitted × marginal environmental damage (marginal environmental cost per km driven and pollutant emitted follows Black et al 2023a).
- Economic efficiency costs follow Harberger (1964) and capture:
  - Cost of purchasing more expensive (and/or smaller) fuel-efficient vehicles.
  - Driving less relative to a scenario without the policy.
- For non-pricing policies (EV subsidies, fuel economy standards) a shadow price is estimated by dividing the policy amount by abated fuel use.
  - Illustrative example: if an EV subsidy is €5,000; discounted lifetime VKT per vehicle is 100,000 km; average new ICEV emits 100 grams per km → vehicle emits 10 tons of CO2 over its lifetime → marginal abatement cost of the policy is €500 per ton of CO2 or €1.20 per liter of petrol.

### Illustrative Environmental Damage and Policy Price Comparisons
- Representative 2030 environmental damage estimates (using 2023 constant prices, marginal costs):
  - Marginal cost of congestion: €0.10 per km (after accounting for behavioral responses).
  - Road accidents: €0.015 per km.
  - Road maintenance (only for heavy goods): €0.04 per km.
  - Global warming: €242 per ton of CO2.
  - Local air pollution: €19,000 per ton of NOx and €38,000 per ton of direct PM 2.5.
- Example passenger diesel marginal policy price evolution:
  - Increase in marginal cost per unit of fuel due to existing policies: €1.36 per liter in 2024 (largely due to the excise and EV subsidy).
  - €1.79 per liter in 2030 (increase due to the EU fuel economy standard).
  - €5.68 per liter in 2035 as the fuel economy standard tightens.
- 2024 comparison for passenger diesel:
  - Implied price on pollution from fuel standard, bonus/malus, and fuel tax (scaled to behavioral response from improved fuel economy): €1.12 per liter.
  - External cost of pollution (climate and local air pollution): €0.85 per liter.
  - Price on driving (fuel tax scaled to behavioral response from reduced driving): €0.25 per liter.
  - External cost of driving (congestion and accidents): €1.76 per liter.
- Summary pattern:
  - Overcharging for pollution and undercharging for driving for passenger vehicles in 2024, projected to grow as pollution-related disincentives increase (fuel economy standard), driving-related charges remain stable, and value of time, mortality, and pollution costs grow.
  - Both pollution and driving related externalities from diesel freight vehicles are underpriced.
  - Driving related externalities from freight EVs are underpriced.
  - Driving related costs from passenger EVs are significantly underpriced.

### Efficient Taxing Principle
- An efficient tax regime would:
  - Price pollution-related externalities equal to their costs.
  - Price distance-related externalities using a road user charge.

### Evaluated Reform Options (description summary of Table 2)
- Seven road transportation policy options analyzed, using the existing policy package as a baseline. EU level policies are maintained in all cases and, unless noted, existing policies (e.g., EV subsidies) are also maintained.
- Policy 1 — Retain domestic carbon price:
  - The EU ETS II is applied on top of the existing domestic carbon price of €44.6 per ton.
  - Objectives: Fiscal revenue, emissions reductions.
- Policy 2 — Introduce a linear feebate:
  - Convert the malus/bonus regime to a linear, revenue-neutral, €250 per ton of CO2 feebate with a separate schedule by vehicle category; increases by €10 per ton annually after 2030.
  - Objectives: Fiscal revenue, emissions reductions for heavy goods.
- Policy 3 — Increased EV subsidies:
  - Increase average subsidies to €5,000 per passenger and €130,000 per heavy goods vehicle in 2025 and phased out through 2040.
  - Objectives: Ease political constraints, equity.
- Policy 4 — Efficient road user and congestion charge (RUC):
  - Introduce a distance-based charge equal to the marginal external cost of non-congestion externalities and a congestion charge limited to driving during peak hours (12% of total driving); remove the existing bonus/malus.
  - Objectives: Address externalities including emissions, fiscal revenue.
- Policy 5 — Efficient energy taxes:
  - Adjust taxes on electricity and fossil fuels to capture the full marginal external cost of driving but scaled down for fuel economy improvements and fuel standards; remove the existing bonus/malus.
- Policy 6 — Efficient RUC and fuel taxes:
  - Combine policy 4 and policy 5 but scale down fuel taxes relative to policy 5.
- Policy 7 — Partial efficient RUC and fuel taxes:
  - Policy 6 but at half of the efficient level for political acceptability purposes.
- Note on determining efficient fuel tax levels:
  - Marginal external cost per liter or kWh of fuel is scaled downward to account for the driving portion of the fuel price elasticity (excluding the portion of behavioral response that results in more fuel-efficient vehicles) and further downward adjustments given the incentive to improve emissions-intensity from the EU fuel economy standard and reduce driving through road user charges on heavy goods and, potentially, passengers.

### Impact of Potential Reforms and Policy Trade-offs
- Transitioning to road user and congestion charges at economically efficient levels would:
  - Allow France to achieve road transportation mitigation targets.
  - Provide a resilient source of fiscal revenue.
  - Significantly reduce transportation externalities, such as congestion.
- Implementation considerations:
  - Building administrative capacity to implement distance and congestion charges is required.
  - Proxy distance charges and higher energy taxes in the interim could ease transition and provide near-term fiscal and externality benefits.
- Smaller adjustments to existing fiscal system (e.g., converting bonus/malus to a revenue-neutral feebate) provide valuable but marginal fiscal and environmental benefits and are unlikely to close the gap to emissions targets.
- Complementary measures needed to address distributional impacts and build public support:
  - Income-based support.
  - Reduced public transit fees.
  - Improving accessibility of public transportation.
  - Detailed distributional analysis is recommended.

### Quantitative Impacts on Costs and Total Cost of Ownership (TCO)
- Effects on purchase prices:
  - Feebates and subsidies directly impact purchase price of vehicles and slightly lower cost of driving as vehicles become more energy efficient.
  - Feebate adjusts relative price of low vs. high emissions vehicles without changing price of the average vehicle (PHEVs would be lightly subsidized under a revenue-neutral feebate).
  - EV subsidy reduces price of zero-emissions vehicles and leaves price of ICEVs unchanged.
  - EU fuel economy standard increases prices for ICEVs above the standard and decreases prices for vehicles below the standard; impact grows over time as the standard tightens.
- Effects on driving and fuel costs:
  - Efficient energy and distance taxes and retaining the domestic carbon price directly increase driving costs (with exceptions on uncongested roads).
  - The domestic carbon price increases driving costs for ICEVs and does so by around 4 percent after accounting for behavioral responses.
  - Efficient taxes increase costs for all vehicles since they imply a higher charge for distance driven regardless of vehicle type.
  - Distance and energy taxes raise fuel costs for passenger EVs more than for ICEVs given light existing taxation of electricity.
- Example fuel price and cost changes under efficient policies:
  - Under the efficient fuel tax, gasoline prices are estimated to remain relatively stable (existing taxes near efficient levels).
  - Passenger diesel prices increase by 60 percent.
  - Freight diesel prices increase by 5 percent.
  - Passenger electricity prices increase by over 100 percent.
- Road user charge example (in 2025 when introduced in isolation):
  - Efficient road user charge results in €0.14 and €0.18 per vehicle-km for passenger and heavy goods vehicles on congested roads, respectively.
  - €0.015 and €0.05 per vehicle-km on uncongested roads for passenger and heavy goods vehicles, respectively.
  - Efficient energy tax examples: €0.72, €1.10, and €0.85 per liter of petrol, diesel for passenger vehicles, and diesel for heavy goods, respectively; and €0.28 per kWh (note: truncated in source).
- TCO impacts:
  - Fuel costs are estimated to be 10 percent of TCO for the average passenger EV and 20 percent of TCO for the average passenger ICEV.
  - Fuel costs approach 50 percent of TCO for freight ICEVs and 30 percent for freight EVs because freight vehicles are driven more.

*Source: Box 1. Estimating Welfare Impacts and Efficient Tax Levels (from provided PDF content).*

### 0.01 per kWh for passenger and heavy goods EVs.

### wpiea2024145-print-pdf - 0.01 per kWh for passenger and heavy goods EVs.

### Total cost of ownership and energy prices (Figure 10; 2030)
- Driving cost components shown for Passenger and Heavy Goods vehicles: Vehicle before policy; Vehicle tax/subsidy (includes fuel economy standards and the bonus/malus); Fuel; Other costs (maintenance, insurance, and range anxiety); Total.
- Panel values (selected plotted numbers shown): 1.6, -2.6, 1.5, -2.6, 2.8, -2.5, 1.5, -4.6, 1.5, -1.3, 1.4, -1.3, 1.7, -1.4, 1.7, -1.4, 6.7, 4.1, 6.9, 4.1, 6.5, 4.2, 6.6, 4.0, 15.2, 13.6, 8.1, 8.4, 14.7, 13.7, 10.8, 10.2, 35.0, 33.1, 35.1, 33.0, 36.1, 33.3, 34.9, 30.9, 41.8, 42.8, 35.7, 38.5, 41.5, 43.0, 38.1, 39.9.
- Total vehicle cost scales in charts up to USD thousands with axes labeled 0, 100, 200, 300, 400, 500 (USD thousands).
- Energy prices panel (USD per kilometer) includes values and ranges with gasoline, diesel, electric series and reform scenarios: 0.00, 0.10, 0.20, 0.30, 0.40, 0.50 (USD per kilometer) and second series from -0.20 up to 1.00 (USD per kilometer).

### Emissions reductions (summary of text and Figure 11)
- All policies except those with the efficient road user charge lead to emissions exceeding the medium-term carbon budget.
- Short-run: policies yield varying reductions over next five years; emissions levels somewhat converge by 2035 due to the EU’s fuel economy standards.
- Largest near-term decline in emissions: policies with efficient RUC (road user charge) due to large reductions in driving.
- Linear feebate and carbon price: emissions generally follow existing policies but with reductions of one to three percent.
- Increased EV subsidy, efficient energy tax, and partial road user charges: emissions reductions of 5-  6 percent.
- Policies that only impact new vehicle purchases (EV subsidy and feebate) have slower impacts because of slow vehicle fleet turnover.
- Under most policies, emissions reductions are predominantly driven by purchases of less emissions-intensive vehicles rather than reduced driving.
- Existing policies lead to a 25 percent improvement in fuel economy of heavy goods vehicles and 32 percent improvement for passenger vehicles in 2030 relative to 2023 levels, but driving distances increase.
- Feebate, efficient energy taxes, and EV subsidies cause larger fuel economy improvements for heavy goods vehicles (up to 34 percent in the case of larger subsidies).
- Efficient RUC reduces driving by 19 percent for passenger vehicles but does not encourage switching from ICEVs to EVs nor reductions in emissions-intensity of ICEVs (the charge does not vary by carbon content).
- Energy taxes, including carbon prices, encourage both reductions in emissions-intensity and driving distances.
- Driving distance drivers: population growth ("two percent in 2030 relative to 2023"), vehicle ownership per capita ("eight and nine percent for passenger and nine heavy goods, respectively"), and distance driven per vehicle ("eight and four percent, respectively") contribute to changes.
- EVs assumed to be driven slightly more than ICEVs based on evidence cited.

### EV sales and stock (Figure 11 panels)
- EV sales by 2030 under increased EV subsidies: 60 percent of new passenger vehicle sales; 49 percent of new heavy goods vehicle sales.
- EV sales by 2030 under existing policies: 54 percent of new passenger vehicle sales; 37 percent of new heavy goods vehicle sales.
- Targets referenced: 66 percent (passenger) and 50 percent (heavy goods).
- Linear feebate: significant growth in heavy goods EV sales.
- Maintaining domestic carbon price: small increases in EV uptake.
- Efficient RUC and energy tax regimes: passenger vehicle EV sales are lower because they increase taxes on EVs and the EV subsidy is assumed removed; maintaining the EV subsidy would result in five percent higher passenger EV sales.
- Stock lags sales by about 20 percent for passenger vehicles and about 10 percent for heavy goods vehicles on average, given slow turnover; heavy goods vehicles turnover about twice as quickly.

### Fiscal impact (Figure 12 and text)
- Efficient taxation generates additional revenue of 0.8 to 1.7 percent of GDP, depending on design.
- Road user charges and energy taxes provide resilient revenue streams since they apply to all driving and tax rates increase with the value of time and statistical life.
- Road user charge generates relatively more revenue and only slowly declines over time due to income elasticity of driving below one.
- Combination of efficient RUC and energy taxes: revenue primarily from the distance portion rather than pollution.
- Carbon price raises around 0.1 percent of GDP more than existing policies in 2030.
- Revenue-neutral feebate replacing existing bonus/malus results in higher revenue since bonus/malus projected to cost around 0.1 percent of GDP in 2030.
- Increased EV subsidy leads to a fiscal cost of around 0.3 percent of GDP in 2030 relative to existing policies.
- EV subsidy parameters: increased to €5  ,000 per passenger and €130,000 per heavy goods vehicle in 2025 and phased out through 2040 (text preserves exact spacing and punctuation from source).
- One-cent per kilometer road user charge raises revenue of about 0.15 percent of GDP; congestion charge raises revenue by an additional 0.2 percent of GDP.
- Emissions targets met once the road user charge reaches 5 and 7 cents per kilometer with and without the congestion charge, respectively.
- Figure 12 panel values (percent of GDP, total revenue 2023 baseline and components): 0.95 percent, 1.04 percent, 1.03 percent, 0.65 percent, 2.60 percent, 1.62 percent, 2.46 percent, 1.77 percent; chart components include Fuel taxes, Malus/feebate, EV subsidy, Road user charge, Congestion charge, VAT, Total, Total revenue (2023).

### Welfare impact (Figure 13 and text)
- Welfare metric evaluated relative to the existing policy scenario; sensitive to assumptions on marginal external cost of driving related externalities, price responsiveness of fuel use and driving, and EV prices.
- Combined efficient road user charges and energy taxes result in net welfare benefits of around 0.4 percent of GDP per year, equivalent to over USD 15 billion or 7 percent of annual transportation costs.
- Efficient RUC increases welfare by slightly less due to below-optimal pricing of pollution.
- Partial reform of RUC and energy taxes generates around two-thirds of the welfare benefits of full pricing, with less impact on energy prices and driving costs.
- Maintaining the domestic carbon price slightly increases welfare and generates revenue around 0.1 percent of GDP.
- Feebate results in a very slight decline in welfare due to a slight increase in the incentive to reduce pollution relative to existing policies (existing incentive slightly above efficient levels).
- Increased EV subsidies generate welfare losses (text ends mid-sentence in source).

*Source: IMF Working Papers — Balancing Environmental, Fiscal, and Welfare Impacts of Transportation Decarbonization in France (figures and text excerpts from provided PDF).*

### 0.10 percent of GDP (~2 percent of annual transport costs) since economic distortions caused by the higher

### wpiea2024145-print-pdf - 0.10 percent of GDP (~2 percent of annual transport costs) since economic distortions caused by the higher

### Welfare impacts of evaluated transportation policies (2030)
- Environmental benefits of efficient road user and congestion charges and energy taxes exceed economic costs by 0.4 percent of GDP and 7 percent of annual transportation costs.
- Road transportation emissions decline to 80 Mt by 2030 under efficient RUC and energy taxes, meeting the SGPE 2023 target and achieving the provisional medium-term carbon budget.
- Road user and congestion charges internalizing accidents, congestion, and road damage are estimated to generate annual reductions in:
  - Driving-related externalities of €33 billion by 2030.
  - Pollution-related externalities of €7 billion by 2030.
  - Economic costs of €28 billion by 2030 (net welfare improvement when compared to externality reductions).
- Efficient road pricing policies are estimated to generate revenue of 2.5 percent of GDP annually when applied at efficient levels.

### Distributional and political economy considerations
- Welfare impacts reported are economy-wide; some households and firms will be better off while others worse off (e.g., households who drive more and cannot easily reduce driving).
- Redistribution of (a portion of) gains to vulnerable households, careful communication, and phased implementation can support political acceptability and equity.
- Detailed, comprehensive distributional analysis is needed, potentially requiring novel locational data on distance travelled by household and firm categories.
- Evidence from congestion and distance-charge literature:
  - Gothenburg and Stockholm congestion charges are (slightly) regressive before revenue use, with large within-income-group variation.
  - Ex-ante studies find incidence of distance-based vs. fuel taxes is mixed; congestion charges are generally regressive before revenue recycling (Denne et al 2020).
  - Revenue use is likely the main determinant of incidence; means-tested public transit subsidies, cash transfers, and targeting by exposure to congestion can help alleviate regressivity.
- Practical interim revenue-recycling and compensation options suggested (means-tested subsidies, targeted transfers, promoting public transport accessibility).

### Electricity demand and grid implications (2030)
- Under current policies, electricity demand from road transportation:
  - Grows from 4 TWh in 2023 to 31 TWh in 2030.
- Under efficient RUC and energy tax policies:
  - Road transport electricity demand reaches 20 TWh in 2030.
- With increased EV subsidy:
  - Road transport electricity demand increases to nearly 40 TWh in 2030.
- Total national electricity use was around 450 TWh in 2023.
- Complementary measures recommended to support grid resilience: smart metering, real-time electricity pricing, and demand-side management.

### Feebate reform and near-term fiscal impacts
- Transitioning the existing bonus/malus to a linear feebate provides practical short-term reform with fiscal and environmental benefits:
  - Linear feebate provides uniform marginal incentives across vehicle purchases versus current scheme that targets above 118 g per km non-linearly.
  - A feebate of €250 per ton with a pivot point of 97 grams of CO2 per km is modelled (Figure 15 reference).
  - Compared to phasing out EV subsidies between 2025 and 2035, a feebate of €250 per ton would:
    - Result in slightly lower emissions in 2030.
    - Generate an estimated fiscal savings of €3.2 billion (0.1 percent of GDP).
- Feebate design considerations:
  - Heavy goods vehicles may require conversion of the feebate into an annual circulation tax/subsidy scheme due to complex value chains.
  - Feebate could be combined with means-tested EV subsidies or general budget funding to promote equity.
  - Additional charge by Euro class could account for diesel vehicle local pollutant differences.

### Carbon pricing
- Gradually increasing the domestic carbon price to €90 per ton by 2030 and implementing it as a price floor for EU ETS II would:
  - Reduce road transportation emissions by 2.5 percent by 2030 and increase revenue by 0.1 percent of GDP by 2030 (earlier statement).
  - Raise revenue of 0.1 percent of GDP, generate net welfare improvements of €1 billion, and reduce emissions by an additional 3 percent in 2030 (fine-tuning statement).
- Half of EU ETS II revenues are earmarked for a Social Climate Fund; the other half to national budgets earmarked for climate measures.
- Carbon-price coordination across neighboring EU countries could reduce aggregate abatement costs; options include removing domestic carbon prices from EU ETS II sectors, keeping taxes and receiving an exemption under Article 30e.3., or keeping both levels with equivalent price floors.

### EV subsidies, production subsidies, and distortion risks
- Increasing broad-based EV purchase subsidies is costly and distortionary:
  - Doubling broad-based subsidies leads to revenue loss of 0.4 percent of GDP by 2030 and increased economic efficiency costs and driving-related externalities.
- Well-designed production-side subsidies can be justified to correct network externalities and spillovers but must avoid funding activities that would occur without subsidies.
  - Example: Netherlands SDE++ auctions subsidies to investments with lowest decarbonization costs to reduce deadweight loss, but requires substantial administration and compliance.

### Policy package and implementation sequencing
- Market-based measures (road user/congestion charges, fuel taxes) should form the centerpiece for efficiency and fiscal resilience.
  - These policies apply to both EVs and ICEVs and help stabilize the revenue base as fleet electrifies.
- Complementary policies to reduce driving levels include promoting teleworking, modal shifts, and expanding public transportation.
- Interim proxy measures for road user charging while building administrative capacity:
  - Annual registration tax scaled by prior-year distance driven (odometer-based).
  - Regular pre-payments for additional kilometers.
  - High frequency tolling on regularly congested roads and peak-hour rate increases (example: Santiago, Chile).
- Administrative and public-sensitization steps recommended prior to real-time GPS-based charges (including time, location, and driver/vehicle specific per-km fee estimates).
- Exemptions can support political acceptability but should be designed carefully (e.g., EV exemptions may shift vehicle mix but not reduce congestion).

### Fiscal and emission baselines, and final conclusions
- Emissions projections and targets:
  - Under current policies, road transportation emissions projected to reach 94 Mt by 2030 (a 22 percent reduction relative to 2023 levels).
  - Provisional carbon budget: 75 Mt by 2030.
  - SGPE target: 81 Mt by 2030.
  - In 2030 projections:
    - Passenger and light commercial vehicles: 68 Mt (target 63 Mt).
    - Heavy goods: 26 Mt (target 18 Mt).
- Projected fiscal impacts from electrification:
  - Fiscal revenue from road transportation estimated to fall by about 0.6 percent of GDP over the next decade under current policies.
  - Road user and congestion charges and higher fuel taxes estimated to generate revenue of 2.5 percent of GDP annually when applied at efficient levels.
- Policy trade-offs and recommended direction:
  - Achieving carbon budgets requires policies that also reduce driving levels, not only improved emissions-intensity of vehicles.
  - A combination of higher fuel and road user charges produce significant emissions reductions and environmental co-benefits while raising revenue.
  - Phasing, consultation, and distributional analysis are necessary prior to implementation; a portion of revenue should be used to support vulnerable households and firms.
  - Fine-tuning measures with near-term benefits: transition to a linear feebate (0.1 percent of GDP fiscal savings), gradual carbon price increase to €90 per ton (0.1 percent of GDP revenue, net welfare of €1 billion, additional emissions reduction), and refinement/targeting of EV subsidies.

*Source: IMF Working Paper excerpt (wpiea2024145-print-pdf).*

### Annex I. Model Description

### Annex I. Model Description

### Model overview
- A “technoeconomic” model forecasts vehicle and driving demand, road transportation energy use, and emissions under various policies.
- Projections are performed separately for light-duty vehicles (LDVs, cars and commercial vehicles with weight below 3.5 tons) and heavy-duty vehicles (HDVs, medium- and heavy-duty freight).
- New and used vehicles are distinguished given differences in purchase prices, fuel economy, and reliance on used vehicle markets.
- Historical data on fuel demand, fuel economy, vehicle ownership, and fuel prices are used to estimate total cost of ownership (TCO) for EVs and ICEVs and as model inputs.

### Key historical inputs (from Table 3)
- EV purchase share: 25% (passenger), 7.6% (light), and 6.2% (freight) in 2023.
- Vehicle ownership (vehicles per 1,000 people): 559 (passenger), 88 (light), and 9.3 (freight) in 2021.
- Fuel economy, average (Liters per 100 km): 5.4 (passenger), 14.24 (light), and 33.72 (freight) in 2019.
- Fuel economy, new ICEV: Percent 20 percent below average (passenger, light) and 10 percent below average (freight).
- Fuel economy, new EV (kWh per 100 km): 14.24 (passenger) in 2019; 76 (light) and 150 (freight) in 2020 with future decline.
- Share of first-time used vehicles: 46 percent of vehicles (all vehicle types) in 2021.
- Useful life: 15 years (passenger and light), 8 years (freight).
- Fuel consumption, ICEVs (Billion liters) in 2022: 11 (gasoline, passenger), 18 (diesel, passenger), 4 (diesel, light), and 11 bln liters (diesel, freight).
- Fuel consumption, EVs (TWh) in 2022: 2.3 (passenger) and 0.006 (freight).
- VKT per vehicle (kilometers): 13,500 (passenger and light) and 53,000 (freight) in 2022.
- Energy prices (including VAT) in 2023 (USD per liter or kWh): 2.03 (gasoline, passenger), 1.94 (diesel, passenger), 1.41 (diesel, freight), 0.22 (electricity, passenger), and 0.14 (electricity, freight).

### Total Cost of Ownership (TCO) structure and assumptions
- TCO is calculated as a net present value (NPV) per-kilometer cost over vehicle lifetime and is separated into:
  - purchase price (adjusted for depreciation),
  - insurance,
  - maintenance,
  - fuel costs,
  - taxes on driving.
- Purchase price determinants:
  - Retail prices: ICE retail prices are projected to remain constant in the baseline for a given income level but increase with rising fuel economy in policy scenarios and as income rises. Retail prices for EVs fall steeply with declining battery costs for both light- and heavy-duty EVs but increase as income rises.
  - Depreciation: faster for freight vehicles due to higher annual mileage and faster in the first year. Depreciation rates assumed identical for EVs and ICEVs.
  - Salvage/resale value: equal to depreciated value with adjustment for changes in retail price of new vehicles relative to vehicles entering used-car market; adjustment reduces resale value of EVs as newer vehicles become relatively cheap.
  - Discount rates: reflect households’ and firms’ cost of capital. The discount rate for LDVs and HDVs reflect households’ and firms’ cost of capital, respectively; the household rate is assumed to be the risk-free discount rate.
  - Taxes on vehicle purchases: include modelled subsidies, feebates, and existing country-specific VAT and customs for passenger vehicles. Fuel efficiency standards are modelled as a shadow carbon price.
  - Range anxiety: an additional cost for EVs that declines over time and as charging station availability grows; calibrated so new vehicle sales in 2023 match actual values.
- Maintenance: varies with labor costs and is lower for EVs due to fewer moving parts.
- Insurance: assumed the same for EVs and ICEVs.
- Fuel costs: function of VKT, fuel economy, and fuel prices. Calculated separately for gasoline and diesel LDVs and then weighted by ownership shares.
  - An additional mark-up is applied to electricity prices to cover charging infrastructure costs, which decline with technological improvements and EV deployment expansion.

### Fuel economy, VKT, and elasticity assumptions
- Exogenous fuel economy improvement:
  - Light- and heavy-duty ICEVs: fuel economy improves 1 and 2 percent a year respectively from gradual autonomous improvement.
  - Heavy-duty EVs: improve from 0.7 km per kWh in 2020 to 1.5 km per kWh in 2050.
- Price-induced fuel economy responses:
  - Elasticities of -0.4 and -0.3 are assumed for light- and heavy-duty ICEVs respectively (fuel economy increases with higher fuel prices).
  - Fuel economy of EVs is assumed to decline exogenously but not in response to changing prices.
- VKT adjustments:
  - Driving elasticities to post-tax marginal cost of driving: -0.3 (light-duty) and -0.2 (heavy-duty).
  - Income elasticity of VKT per vehicle: 0.05.
  - VKT per vehicle is capped using a cross-country regression with urbanization rates and population density; VKT per vehicle assumed identical for ICEVs and EVs for new vehicles.
- Range anxiety and charging supply elasticities:
  - A 10 percent increase in charging availability increases EV uptake by 3.7 percent at relatively low levels of charging availability.
  - The elasticity of charging station supply to charging station costs is assumed to be 0.67.
  - For simplicity, only fast charging stations are modelled.

### TCO parameter table highlights (from Table 4)
- Depreciation:
  - Passenger: 30% in first year and then linearly depreciated over 14 years, sold at 7 years for new cars; used vehicles are held until retirement.
  - Freight: 30% in first year and then linearly depreciated over 8 years, sold at year 5; used vehicles are held until retirement.
- Range anxiety:
  - EV – passenger: EUR 6,000 in 2023.
  - EV – freight: EUR 240,000 in 2023.
- Fuel/energy efficiency:
  - ICE – passenger: 1% annual exogenous improvement.
  - EV – passenger: Country-specific for around 30 countries (IEA), average for others, 1% annual exogenous improvement.
  - ICE – freight: exogenous improvement of 2.5% per year (ICCT).
  - EV – freight: 150 to 100 kWh per 100 km from 2020 to 2030 (ICCT) and continued, linear improvement after 2030.
- Maintenance:
  - ICE – passenger: $0.06 per km (Argonne).
  - EV – passenger: 1/3 of ICE (Argonne).
  - ICE – freight: $0.125 per km (ICCT).
  - EV – freight: 1/3 of ICE (ICCT).
- Insurance: 1.5% of vehicle value (Argonne) and $0.04 per km (ICCT) noted.
- Discount rate: 2.5% (passengers, Argonne) and 5% (freight).

### Vehicle demand, stock dynamics, and allocation
- Total vehicle demand is estimated using the Gompertz function with population and GDP projections from the IMF WEO.
- Vehicle additions equal the difference between total demand and ownership after depreciation and used cars sold domestically.
- A logit function based on relative TCO of ICE and EV determines the ICE vs. EV share of vehicle demand, separately for new vs. used purchases.
  - The logit function is fitted to minimize difference between modelled and actual EV sales between 2015 and 2023 for select economies.
  - The k-parameter for the logit function is 3.25 and increases to 4.75 as EV uptake crosses 50 percent.
- Availability of used EVs is constrained based on new vehicles reaching the end of the initial ownership period and is split between imported used vehicles and domestically previously owned used vehicles.
  - Limited supply of both imported and second-hand domestic used vehicles is determined based on global and domestic number of EV purchases from 10 years ago for passenger vehicles and 4 years ago for freight.
- For used vehicles, purchase price equals salvage value of a newly purchased vehicle and fuel economy is the lagged values for new vehicles.
- After vehicle stock estimation, driving distances adjust according to changes in post-tax marginal cost of driving, driving elasticities, and income per capita to estimate fuel use, fiscal revenue, and emissions.

### Congestion charging assumptions
- Elasticity of congestion to congestion charges: -0.6 overall, with -0.75 for passenger vehicles and -0.1 for heavy goods.
- Driving increases by half of the reduction in congestion since some vehicles re-route and congestion levels increase exponentially with traffic.
- Congestion charging is assumed to apply to 12 percent of driving (the portion of driving that occurs during peak, working hours in France according to TomTom).

*Source: Annex I. Model Description.*

### 2021. https://ec.europa.eu/commission/presscorner/detail/en/qanda_21_3662

### Balancing Environmental, Fiscal, and Welfare Impacts of Transportation Decarbonization in France

### Document identity and scope
- Working Paper No. WP/2024/145
- Theme: Transportation decarbonization in France; balancing environmental, fiscal, and welfare impacts.

### Bibliographic and data sources cited (selected items as presented)
- European Commission. “Taxes in Europe Database V4.” European Commission, 2024a. https://ec.europa.eu/taxation_customs/tedb/#/simple-search
- EEA (European Environment Agency). “Member States’ Greenhouse Gas (GHG) Emission Projections,” 2023. https://www.eea.europa.eu/en/datahub/datahubitem-view/4b8d94a4-aed7-4e67-a54c-0623a50f48e8
- Fell, Harrison; Ian MacKensie; William A. Pizer. “Prices versus Quantities versus Bankable Quantities.” Resource and Energy Economics 34(4) (2012): 607–23.
- French Strategie. “The Value for Climate Action: A Shadow Price of Carbon for Evaluation of Investments and Public Policies.” French Strategie, 2019.
- GoF (Government of France) publications, including:
  - “2024 CO2 Malus Tax on the Most Polluting Vehicles,” 2024a.
  - “Déduction Exceptionnelle : Véhicules Lourds Utilisant Des Énergies Propres,” 2024e.
  - “Green Bonus, Aid for the Acquisition of Low-Emission Vehicles: What Changes in 2023?” 2024c.
  - “Heavy Goods Vehicle Toll,” 2014.
  - “Key Figures on Climate France, Europe and Worldwide 2023 Edition,” 2023b.
  - “NATIONAL ENERGY CLIMATE-PLAN OF FRANCE (DRAFT UPDATE),” 2023a.
  - “Prime à La Conversion Des Véhicules et Bonus Écologique 2023,” 2024d.
  - “Rapport Sur l’impact Environnemental Du Budget de l’Etat,” 2023c.
  - “Taxes Sur l’affectation Des Véhicules à Des Fins Économiques (Ex-TVS),” 2024b.
- Harberger, Arnold C. “The Measurement of Waste.” American Economic Review 54 (1964): 8–76.
- ICCT. “European Vehicle Market Statistics: Pocketbook 2021/22.” The International Council for Clean Transportation, 2021a.
- IEA. “Global Energy and Climate Model.” Paris: International Energy Agency, 2023.
- IEA. “Global EV Outlook 2024.” International Energy Agency, 2024.
- IMF, FM. “Climate Crossroads: Fiscal Policies in a Warming World.” Washington, D.C.: International Monetary Fund, 2023.
- IPCC. “Climate Change 2022: Mitigation of Climate Change. Contribution of Working Group III to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change.” 2022.
- Lindsey, Robin; Ioannis Tikoudis; Katherine Hassett. “Distributional Effects of Urban Transport Policies to Discourage Car Use: A Literature Review.” OECD Environment Working Papers. Vol. 211. March 3, 2023.
- Parry, Ian; Dirk Heine; Eliza Lis; Shanjun Li. Getting Energy Prices Right: From Principle to Practice. International Monetary Fund, 2014.
- Pyddoke, Roger; Jan-Erik Swärdh; et al. “Distributional Effects from Policies for Reduced CO2-Emissions from Car Use in 2030.” Transportation Research Part D: Transport and Environment 101 (December 2021): 103077.
- Qorbani, Davood; Hubert P. L. M. Korzilius; Stein-Erik Fleten. “Ownership of Battery Electric Vehicles Is Uneven in Norwegian Households.” Communications Earth & Environment 5, no. 1 (April 1, 2024): 170.
- Rennert, Kevin; Frank Errickson; Brian Prest; Lisa Rennels. “Comprehensive Evidence Implies a Higher Social Cost of CO2.” Nature 610 (2022): 687–92.
- Rystad Energy. “Mystery Solved: Norway’s Persistent Road Fuel Demand Won’t Last amid Rapid EV Adoption,” 2023.
- SGPE. “La Planification Écologique Dans Les Transports.” Secrétariat général à la planification écologique, 2023.
- Sharpe, Ben; Hussein Basma. “A Meta-Study of Purchase Costs for Zero-Emission Trucks.” The International Council for Clean Transportation, 2022.
- Stiglitz, Joseph E.; Nicholas Stern; Maosheng Duan; Ottmar Edenhofer; Gaël Giraud; Geoffrey M. Heal; Emilio Lèbre La Rovere; et al. “Report of the High-Level Commission on Carbon Prices,” 2017.
- T&E. “Tolling: The Highway to Green Trucking.” Transport & Environment, 2024.
- US EPA. “Multi-Pollutant Emissions Standards for Model Years 2027 and Later Light-Duty and Medium-Duty Vehicles.” 2023b.
- US EPA. “Supplementary Material for the Regulatory Impact Analysis for the Final Rulemaking, ‘Standards of Performance for New, Reconstructed, and Modified Sources and Emissions Guidelines for Existing Sources: Oil and Natural Gas Sector Climate Review.’” 2023.
- Vailles, Charlotte; Sirine Ousaci; Louise Kessler. “La Transition Est-Elle Accessible à Tous Les Ménages ?” I4CE, 2023.
- Zhao, Lujin; Elizabeth R. Ottinger; Arthur Hong Chun Yip; John Paul Helveston. “Quantifying Electric Vehicle Mileage in the United States.” Joule 7, no. 11 (November 2023): 2537–51.

### Thematic emphasis of cited literature (as presented)
- Policy instruments and carbon pricing (shadow prices, malus/bonus vehicle policies).
- Distributional and welfare impacts of transport decarbonization.
- Vehicle market statistics and EV adoption trends.
- Emissions projections and climate mitigation modeling.
- Regulatory standards for vehicle emissions and energy sector performance.
- Fiscal implications and environmental impact of public budgets.

*Source: Balancing Environmental, Fiscal, and Welfare Impacts of Transportation Decarbonization in France. Working Paper No. WP/2024/145*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2024/english/wpiea2024145-print-pdf.pdf_