A simulation model was built to assess the impacts of environmental tax reform in road transport on revenues, environmental outcomes (i.e. GHG and NOX emissions) and distributional effects. The analysis evaluates two scenarios against a baseline from 2026 to 2030: first, the planned reform of the excise and motor vehicle tax at the national level, and second the implementation of the EU ETS2 and another excise tax and motor vehicle tax reforms that aligns with policy principles discussed above.
In the absence of additional policy measures (baseline), GHG emissions from cars are expected to grow due to a likely increase in car ownership and low uptake of EVs in the short to medium term. NOx emissions could stabilise over time as second-hand vehicle imports gradually comply with stricter Euro emission standards, but low vehicle turnover slows the pace of this decline.
Fiscal impacts: While the ETS2 is expected to raise substantial revenues (one third of the increase under Scenario 2) these will not be available to the general budget of Romania. Revenues from excise tax reform are expected to rise under the baseline by 24% in 2030 compared to 2025 due to increasing car ownership, by 35% under Scenario 1 and by 31% under Scenario 2. The increase from motor vehicle taxes is expected to be more limited under the baseline (16%), but higher under Scenario 1 (68%) and under Scenario 2 (189%).
Environmental impacts: Under both policy scenarios, GHG and NOX emissions decline compared to the baseline but fall short of the respective policy targets under current modelling assumptions.
Distributional impacts: An increase of taxation on automotive fuels will affect high income households relatively more. Large households and households living in rural areas are also more likely to be impacted compared to rural households under current assumptions.
No air pollution tax at the national level is included in the scenarios. Because time and location matter for air pollution, instruments at the local level (e.g. urban tolls, low emission zone in dense areas) could be considered to further improve outcomes.
Environmental Tax Policy Review of Romania
Simulation models were built to evaluate short- and medium-term impacts of selected tax reforms related to cars in Romania. Two simulation tools are developed, one to evaluate the impact of tax reform on revenue and emissions based on fleet structure and traffic evolution data and another one to analyse the distributional impacts related to households’ energy expenditures. Tax policy scenarios incorporate selected reforms for energy and car taxation, which are either already in force at the end of 2025 (baseline), were planned to start as of 2026 (Scenario 1) and were planned to start later or could be implemented to follow suggestions from Chapter 4 (Scenario 2). This chronology is indicative. Most of the instruments discussed here are already in place or planned, and the suggestions is about modulating them, which makes the timeline coherent. This chapter begins by a presentation of the models and their assumptions, followed by a presentation of the simulated scenarios. Results are then sequentially presented for the fiscal, environmental and distributional impact.
The impact analysis combines two different modelling approaches. The environmental and fiscal impact analysis relies on a car fleet turnover model, combining emissions data from representative cars with car fleet data in a dynamic approach. The distributional outcomes evaluated focusses on households and instruments related to energy in a static perspective.
5.1.1. Vehicle turnover model for fiscal and environmental impact
The first part of the model aims to assess how tax reform in road transport may influence energy use, emissions, and fiscal revenues in a partial equilibrium setting. The model relies on a fleet turnover model based on the current Romania fleet structure, related energy consumption and distances driven. The model captures some behavioural changes to tax reform reflected in vehicle technologies, distance driven, and energy use and their implications on public revenue and environmental outcomes, i.e. GHG and NOX emissions. The model focuses on cars. The structure and different channels of impact are summarised in Figure 5.1. Additional information and main hypotheses are outlined in Box 5.1 for revenues.
Source: Author’s elaboration.
Fleet turnover is modelled through the ageing of existing vehicles, annual vehicle retirements and additions, and fleet growth consistent with past trends. Vehicle retirement follows a survival curve1 that is calibrated on the 2018-2023 Romanian fleet data. Vehicles enter the fleet both by replacing retired vehicles, and through continued growth in Romania’s motorisation rate. This is a business-as-usual assumption, assuming that the car fleet will continue to grow according to the average trend of the past three years, i.e. a 3% car’s fleet increase, translating into around 0.5 million newly registered cars per year. This growth balances between the projected GDP growth and declining population (IMF, n.d.[1]), both key determinants of road transport activity. If translated into energy demand, this baseline is more conservative than the one from the Romania long term strategy for decarbonisation (Government of Romania & PwC, 2023[2]), which assumes a stable energy demand in the road transport sector, and of GHG emissions for 2030 – instead of a stable fleet growth as in the present analysis (see Annex M for details).
The characteristics of newly added cars per year evolve according to past trends in the baseline. Inserted vehicles are split in categories with four powertrains (ICE diesel, ICE gasoline, PHEV gasoline, BEV), eight Euro emission standard (Euro 0 to Euro 6d) and two to three sizes for diesel and gasoline-powered cars.2 In the baseline scenario, each category of newly added cars evolves exogenously in line with trends observed over the past three years. For instance, battery electric vehicles accounted for 3% of new registrations in Romania in 2024 and are projected to reach 9% in 2030. Euro emission standards of newly added cars are also changing exogenously towards more recent ones, as a result of the uptake of second-hand cars with similar age over time. For instance, the share of Euro 6d diesel or gasoline-powered cars evolve from 40% in 2025 to 54% in 2030.
In both tax reform scenarios, fleet structure adjusts following behavioural reactions to energy cost. The adoption of powertrain through newly added cars is modelled using cross-elasticities of the car’s uptake by powertrain depending on the energy costs (Fridstrøm and Østli, 2021[3]). For instance, an increase of 10% of the diesel prices and the gasoline prices will increase the demand for BEVs by respectively by 2.3% and 3.8% respectively.
Kilometres driven and emissions per kilometre are assumed to vary by car category. Mileages and emissions per car category are modelled based on the EU Copert model (see Box 5.1). Annual mileages vary by car category (differentiated by powertrain and Euro emission standard), with diesel-powered and younger cars assumed to drive more. CO2 and NOX emissions in gram per kilometre are also associated to each powertrain and Euro emission standard. Emissions are fixed over time for each category of cars.
Fleet data are drawn from two sources, covering the existing stock and new vehicle entries. The vehicle stock as of 2023 is taken from the methodological annex of Romania’s GHG inventory submitted to UNFCCC, which relies on the EU COPERT model. Data on new car registrations are obtained from the General Directorate for Driving Licenses and Registrations of the Ministry of internal affairs. These detailed, vehicle-level datasets provide information on powertrains and age, from which EURO emission standard are inferred.
Activity data (kilometres driven by vehicle) are combined with default emissions factors based on vehicle characteristics. An ad hoc simulation model was developed, primarily using publicly available COPERT data. COPERT is a fleet model used by EU member states to calculate real-world GHG and air pollutant emissions for reporting and monitoring targets. Activity data were constructed by merging mileage by vehicle type, powertrain and Euro emission standards with average CO2 and NOX emissions for the same categories from the EEA. For cars, the datasets were merged using assumptions on the distribution of cars sizes within each segment (i.e. 50% of gasoline cars under 1.6L are classified as medium cars).
Overall, the impact of reform affecting energy cost happens through two channels with different time horizons: reduced mileage and fleet structure. In the short term, increases in energy prices, e.g. through fuel excise tax, ETS or carbon tax reforms, cause a one-off shock to domestic energy demand. For instance, the introduction of a carbon tax in 2028 raising diesel and gasoline prices by 10% is assumed to reduce their consumption by 3% in 2028, with consumption remaining at that level thereafter. In the medium term, drivers respond by replacing vehicles with alternative technologies. In the same example, a 10% increase in diesel and gasoline prices in 2028, is assumed to increase the share of electric vehicles newly registered by 6.1% in 2029 and keep it at the same level for the following years (see above).
Under the two tax reform scenarios, driver’s reaction to the increased cost of energy is modelled based on own price elasticities from the literature. A single price elasticity of -0.3 is assumed across all fuels and vehicle types reflecting a medium-to-upper-bound estimate for gasoline (see Box 4.2 and (Labandeira, Labeaga and López-Otero, 2017[8]). This implies that a 10% increase in fuel prices leads to a 3% reduction in fuel consumption, and thus of kilometre driven. Although diesel demand is often found to be less elastic than gasoline (-0.1, see Box 4.2), tax differentials between Romania and neighbouring countries may induce fuel tourism; therefore, a higher-than-usual estimate is used in the model.
In both tax reform scenarios, fleet structure adjusts following behavioural reactions to energy cost. The adoption of powertrain through newly added cars is modelled using cross-elasticities of the car’s uptake by powertrain depending on the energy costs (Fridstrøm and Østli, 2021[3]). For instance, an increase of 10% of the diesel prices and the gasoline prices will increase the demand for BEVs by respectively by 2.3% and 3.8% respectively.
Fiscal impact is modelled by applying tax rates to their dedicated tax base. In the context of Romania, two different tax bases are considered: the car stock per year for the motor vehicle tax and the vignette, and yearly energy consumption for the excise duty, the carbon tax and the ETS2 (see Box 5.2). Modelling the motor vehicle tax required additional splitting of the car fleet according to the size bracket of the tax schedule.3 The estimated fiscal impact on motor vehicle tax can also vary depending on the effective tax schedule decided at the municipal level.
Modelling the fiscal impact of tax reform relies on bottom-up models of road transport. Vehicle types included are cars with four different powertrains (ICE diesel, ICE gasoline, PHEV gasoline and BEV).
Tax framework
A driver may be liable to pay taxes or fees on four tax bases: new vehicles (NV), vehicle stock (V), energy consumption (E) and road use (M), which can be summarised in the following country-wide revenue function:
Evolution of the different tax revenue components
Evolution of tax revenues:
New vehicle tax revenues (RNV): Calculated by multiplying the number of new vehicles in each category (j) by the tax rate for that category (differentiated by powertrain, Euro emission standard and size). No tax is modelled in this category for Romania.
Vehicle tax revenue (RV): Calculated by multiplying the number of vehicles in each category (k) by the tax rate for that category (differentiated by powertrain, Euro emission standard and size). It encompasses the annual motor vehicle tax (tax on the mean of transport) and the duration-based charge (vignette).
Energy tax revenue (RE): Calculated by multiplying the tax rate for each energy type (i) by total consumption of that energy type. Total consumption is derived by summing across all car categories the product of their average fuel efficiency, number of vehicles and average kilometre driven. It encompasses the excise tax, carbon tax and ETS2.
Road use revenue (RM): Calculated by multiplying the tax rate per kilometre by total kilometre driven per vehicle category (l). Total kilometres are obtained by multiplying the number of vehicles in each category by their average distance driven. No tax is modelled in this category
Source: Author’s elaboration based on OECD/ITF (2019[9])
Although the model includes numerous effects, some impacts are excluded because of data constraints or uncertainty. First, the impact of the motor vehicle tax increase on vehicle size, powertrain and Euro emission standard is not modelled. However, this effect should be limited, as the tax differential is relatively small with a maximum of RON 100 (roughly EUR 25) for large cars with the oldest Euro emission standard. Second, energy costs increases may affect fuel efficiency (through size) of newly added vehicles but is not modelled. Romania has only a minority of cars in the large segment (8.9% of cars have a size larger than 2L (Eurostat, 2025[10])). Finally, HEVs are not part of the model but considered as ICE, which result in a slight overestimation in emissions.
5.1.2. Distributional impact model
An additional model was developed to simulate the distributional impacts of tax reform using data from Household Budget Surveys (HBS). The most recent HBS data available at the required level of detail are from 2015 focusing only on households’ expenditures (see Box 5.3). To ensure precision, the model focuses on household energy taxation and does not include vehicle taxes.4 As car ownership has increased significantly in Romania since 2015, the structure of current energy expenditures may be different from HBS.
Eurostat Household Budget Survey database
The Eurostat Household Budget Survey (HBS) is a harmonised survey carried out approximately every five years in all EU Member States by the national statistical offices. The latest available editions are 2015 and 2020. The anonymised sample provides observations for representative households, each with associated weight. Household expenditures are recorded at a granular level, including by energy type in road transport (diesel, gasoline and LPG) and the buildings sector (electricity, natural gas). However, detailed data at the level of individual energy types – which is critical for this analysis – is only available for 2015.
Other variables of interest relate to the location of the households (NUTS and density of the area), the number and age of household members, and total net income.
Simulations of the reforms
The tax reforms are simulated based on 2015 HBS data and 2025-2030 price variation, applied to 2015 expenditures. Indeed, no granular data is available for the 2020 edition.
Price and tax data come from the IEA Energy Prices database 2015 and own research for 2025.
This impact is differentiated by socio-economic variables: income, location and household size. Income deciles are expressed in terms of net monetary income. Households with no income or with expenditures four time superior to their income are excluded following the methodology used by Flues and Thomas (2015[11]).
Source: Eurostat (2025[12]); Flues and Thomas (2015[11]) (2015[11]); IEA (2025[13]).
In addition to the baseline, two tax policy reform scenarios are assessed over the 2025-2030 period. Scenarios are summarised in Table 5.1, with detailed tax rates provided in Annex K and Annex L. The model does not incorporate potential inflation (i.e. rise of pretax energy cost). They cover selected energy and vehicle-related policy instruments regarding cars.
Note: Changes with respect to the previous column are marked in bold. Recent increases in VAT, excise tax and vignette implemented as of September 2025 are included in the baseline.
Source: Author’s elaboration.
The baseline refers to the policy landscape already existing as of the 1st September 2025. Rates for excise taxes and the vignette are the rates that entered into force in August 2025 and September 2025 respectively. Tax rates used for the motor vehicle tax are those from Bucharest ins 2025, although they are set at the local level and may vary by municipality. Detailed rates are available in Annex K. No reform is added. Impacts derived for the baseline will thus only depend on the (exogenous) evolution of the fleet structure (e.g. growth of the car fleet, changes in characteristics of newly added cars relative to their Euro emission standard and powertrain).
Scenario 1 models reforms that are planned to happen starting from 2026. Reforms included in Scenario 1 are those enacted in 2025 scheduled to happen in 2026: in particular the rise of the motor vehicle tax for cars and an increase of the fuel excise taxes. The tax schedule for the first two instruments is already legislated for 2026 and assumed to stay identical until 2030, while the ETS2 price will depend on the market. The excise tax for diesel and gasoline increased by 10% in January 2026. A reform on the motor vehicle tax is planned, with a component for Euro emission standard. Tax rates modelled are those legislated at the national level and implemented without variation in Bucharest. Rates and sources are indicated in Annex L. Energy costs are assumed to have increased respectively by 4% for diesel and 4.5% for gasoline compared to 2025.
Scenario 2 models reforms in addition to those in Scenario 1, introducing a carbon price on transport emissions starting from 2028 with the ETS2, alongside a reform of energy taxation. The ETS2 price is assumed to be EUR 48 per tCO2e starting from 2028, in line with the European Commission’s projections (European Commission, 2021[14]). Simultaneously, a harmonisation of the diesel and gasoline excise tax is implemented: the excise tax increases for diesel (+3.4%) and decreases (-5.2%) for gasoline to reach the same level of taxation at RON 2.9 per litre. This rate is the mid-level between the current diesel and gasoline excise tax rates). rates, a similar reform being currently implemented in Italy for environmental reasons (see Box 4.5). The motor vehicle tax rate is increased by 150% for Euro 0 to 3 cars, by 100% for Euro 4 cars and by 50% for Euro 5 cars. Even with such reform, the level of taxation stays below. Other reforms such as the increase of excise tax and the reform of the motor vehicle tax in 2026 are kept until 2027. Energy costs are thus assumed to increase respectively by 14.5% for diesel and 10% for gasoline compared to 2025.
This section presents the results of the modelling exercise, assessing the fiscal, environmental, and distributional impacts of the two policy scenarios. The analysis focuses first on the presentation of cost variation for representative cars, before presenting the results to two different cars. Before presenting overall results, the analysis highlights channels through which tax reform affects different car categories
3.1 Effects of tax reform on tax payments for different cars
Overall, Scenario 2 provides incentives that align road transport prices more closely with environmental costs than the current reforms (Scenario 1). Figure 5.2 shows the total tax payment for hypothetical medium-sized cars under different scenarios. To highlight the different channels of tax reform all car characteristics are kept constant (size, fuel efficiency, kilometre driven) except for type of fuel used (diesel vs gasoline) and the Euro emission standards (Euro 2, 4 and 6d). Gasoline-powered cars pay more taxes in the baseline and in Scenario 1, due to the higher excise tax rate on gasoline, and the limited variation of taxation by Euro emission standard (EUR 20). In Scenario 2, cars with stronger environmental impact are taxed more strongly: Euro 2 diesel-powered cars pay the most (EUR 514), while Euro 6d gasoline-powered pay the least (EUR 443). This difference is driven by the newly introduced ETS2 and the stronger differentiation of the motor vehicle tax with Euro emission standards (+ EUR 59 for the oldest Euro emission standard).
Note: Representative vehicles are a medium-sized ICE diesel car (1 900cc, 6.6 litre/100km), a medium-sized ICE gasoline car (1 900cc, 9 litre/100km) and a medium-sized battery electric car (20 kWh/100 km).
Source: Author’s elaboration.
In all scenarios, excise taxes make up the largest share of taxes paid each year. In the baseline, the motor vehicle tax amount is small in absolute amount (EUR 103) and when compared to the annual payments (6%). While the motor vehicle tax increased in both scenarios, its level stays lower than excise taxes and is still a small share of total annual taxes (8-10% and 7-18% of the yearly paid tax under Scenario 1 and 2).
In practice, cars differ by mileage and fuel efficiency, which is reflected in the model, but not on this figure showing hypothetical tax payments under constant assumptions. For instance, at a similar car size, diesel cars in Romania are more fuel-efficient than gasoline cars. Additionally, diesel-powered and newer cars are typically driven more.5
5.3.1. Fiscal impact
Baseline projections indicate rising fiscal revenues from higher car ownership and traffic, with both reform scenarios adding a further positive fiscal impact in the short to medium term. Figure 5.3 presents the revenues and their composition by tax increase under the different scenarios. Revenues are projected to increase by 23% in 2030 compared to 2025 in the baseline due to the higher number of vehicles and fuel consumed. Both tax reform scenarios are projected to generate additional revenues, with a 37% increase for Scenario 1 and a 68% increase for Scenario 2 under current modelling assumptions.
Note: Scenario 1 models an increase of the excise tax and the reform of the motor vehicle tax in 2026. Scenario 2 introduces an excise tax additional reform in 2028, with the harmonisation of diesel a gasoline tax rates alongside the introduction of the ETS2 and a reform of the motor vehicle tax with higher tax rates for old Euro emission standard. All impacts are reported relative to the baseline 2025 fiscal revenue. ETS2 revenues are not fully available to the Romania budget.
Source: Author’s elaboration.
Most of additional revenues under Scenario 1 come from the excise tax. Though the highest share of revenues is generated by the excise tax, most of it comes from the increase of fuel consumption expected from growing motor vehicle ownership. Compared to the baseline, the extra increase is 10 percentage-point for the excise tax and 3.8 percentage point for the motor vehicle tax.
In Scenario 2, the rise in revenue above the baseline is driven primarily by the ETS2, followed by the motor vehicle tax. More than one third of the increase of tax revenues is explained by the ETS2 under Scenario 2 and one fourth by the motor vehicle tax. This assumes no behavioural adjustment of the tax to the car fleet structure in terms of Euro emission standard and size and should be read with caution, as the motor vehicle tax adjustments are voted at the local level, which could not be depicted. While indicated as revenues, the monetary outcomes of the auctions of ETS2 are earmarked for climate action and social measures (European Commission, 2023[15]). Part of the revenue will be pooled at the EU level for the Social Climate Fund, and Romania should receive more than its contribution, as being one of the top beneficiaries (see Section 4.3).
Fleet structure changes towards electric vehicles are expected to have little impact on short-term revenue. Both in the baseline and the scenarios, the EV penetration rate is low (9% to 10% of newly inserted cars, so only 2% of the total fleet in all scenarios). This is partly due to the assumed low elasticity of EV uptake related to rising energy cost, as well as other assumptions (e.g. no behavioural adjustment in terms of fuel of newly inserted cars efficiency and no HEV modelled). The slight increase of taxation for PHEVs and BEVs contributes a modest increase in tax revenues (3% of the total tax increase in 2030 compared to 2025 in scenario 2). As a result, revenue foregone from fleet electrification is limited by 2030.
5.3.2. Environmental impact
While GHG emissions are projected to continue growing, air pollutant emissions are likely to stabilise in the baseline scenario. GHG emissions growth in the baseline is driven by increasing car ownership and the slow adoption of clean technologies under current assumption. The stabilisation of air emission result from gradual penetration of vehicles with stricter Euro emission standards through fleet renewal, including second-hand vehicles.6
Note: On panel A, GHG emissions are expressed as a share of 2005 GHG emissions from cars. On Panel B NOx emissions are expressed as a share of the 2025 level (no precise 2005 level available).
Source: Author’s elaboration.
Under Scenario 1 and Scenario 2, emissions are expected to decrease by 4 to 7% in 2030 compared to the baseline. GHG emissions are estimated decline by respectively 5% and 6% in 2030 under Scenario 1 and 2 compared to the baseline in 2030 (see Figure 5.4). The NOX emissions reduction is more varied, with a decrease of 1% and 4% respectively compared to the baseline in 2030, almost fully driven by reduced domestic fuel consumption in reaction to the increased energy cost and not changes in the fleet. Indeed, the effect of the reforms on the structure of newly purchased cars is limited: the share of new BEVs rises by only one percentage points in response to changes in energy cost and no changes in fuel efficiency (smaller vehicle’s size) or Euro emission standard is modelled.
To reach emission reduction targets in road transport additional policy measures will be required. In both scenarios, carbon pricing, fuel taxes and vehicle tax reform are insufficient to achieve Romania’s emissions reduction target in road transport. Further measures could help bridge the gap (see 4.1). For example, urban tolls or low-emission zones can effectively reduce local air pollution, particularly in densely populated urban areas where external costs are higher.
5.3.3. Distributional impact
Tax reforms that raise energy costs are expected to affect high-income households relatively more, reflecting higher rates of car ownership. shows the impact of Scenario 2 tax reforms on households’ total energy expenditures by income decile based on HBS information. On average, net expenditures are expected to increase by 0.4 percentage points of income for the top decile, against 0.04 percentage points for the bottom decile. These results are weighted averages, i.e. the small effect at the lowest decile reflect the lower car ownership rates and driving among low-income households. Results do not include behavioural adjustments to tax reform.
Note: The tax reform includes an equalisation of excise tax rates for diesel and gasoline and the introduction of the ETS (i.e. Scenario 2 in 2030). The motor vehicle tax reform is not modelled due to data limitation. The simulation is presented without elasticities.
Source: Author’s elaboration based on Eurostat 2015 HBS data.
The stronger effect at higher income levels mainly reflects the higher share of income that richer households spend on automotive fuels. In 2015, 28% of households reported a positive consumption of energy in road transport, with important variations across deciles and highest levels at the highest income level. On average, the richest households spent over twice the amount road transport fuels compared with the poorest. As richer households consume more in absolute terms, they will be more impacted by the price increase.
Note: The tax reform includes an equalisation of excise tax rates for diesel and gasoline and the introduction of the ETS (i.e. Scenario 2 in 2030). The motor vehicle tax reform is not modelled due to data limitation. The simulation is presented without elasticities. “Towns” refer to towns and suburbs.
Source: Author’s elaboration based on Eurostat 2015 HBS data.
Larger households would also be slightly more impacted by tax reforms that increase energy costs due to higher energy consumptions. Figure 5.6 shows expected impacts of tax reform on households’ total energy expenditures. Larger households are expected to be more affected by the tax reform (Panel A). For example, households with three or at least four members face an average increase of 0.3 percentage points of income compared with 0.1 and 0.2 percentage points for one- and two-person households respectively. This can be explained by a higher energy consumption, as 94% of households reporting an expense have more than one member, probably driven by higher number of cars owned.
Energy price increases from tax reforms have a greater impact on larger households because they consume more energy. Figure 5.6, Panel A shows expected impacts of tax reform on households’ total energy expenditures differentiating by household size. Larger households are expected to be more affected by the tax reform. For example, households with three or at least four members face an average increase of 0.3 percentage points of income compared with 0.1 and 0.2 percentage points for one- and two-person households respectively. This effect is driven by higher energy use, as 94% of households reporting positive energy expense consist of more than one member.
Households in moderately dense areas are slightly more affected, reflecting greater diesel consumption. Figure 5.6, Panel B shows expected impacts of tax reform differentiating by household location. Households in towns and suburbs are slightly more affected, with expenses rising by 0.32 percentage points of income, compared with 0.26 and 0.24 percentage points in cities and rural areas. This effect is related to slightly higher diesel use in rural areas compared to cities. Diesel is experiencing the highest variation in the reform simulated (+14.5%) compared to gasoline (+10%). The difference in terms of total energy consumption is less important. While households living cities tends to consume slightly more in average (11% of households in cities reporting an expense for automotive fuels, more than in rural areas (9%) and suburbs (7%)),7 the average amount spent is overall similar depending on the location.
The effective distributional impact of tax reform could differ reflecting changes in vehicle ownership since 2015. Due to data limitation, the model applies tax reform scenarios to 2015 data although the car fleet structure has changed since then. First, more households in Romania own a car today, thus a higher share of the population will be impacted. In 2015, 28% of households reported a positive expense for road transport fuel use, but car ownership rose by 70% between 2015 and 2024 (see Section 4.1). Second, the expenses reported for gasoline were higher than for diesel in the 2015 data. Though, vehicles entering the fleet over the past decades have mainly been second-hand diesel cars, which could reverse this picture, and thus more households could be impacted by the relatively strong tax increase on diesel.
References
[5] Copert (2025), , https://copert.emisia.com/copert-data/.
[6] EEA (2025), EMEP/EEA air pollutant emission inventory guidebook, https://www.eea.europa.eu/en/analysis/publications/emep-eea-guidebook-2023/part-b-sectoral-guidance-chapters/1-energy/1-a-combustion/1-a-3-b-i.
[15] European Commission (2023), ETS2: buildings, road transport and additional sectors, https://climate.ec.europa.eu/eu-action/carbon-markets/ets2-buildings-road-transport-and-additional-sectors_en.
[14] European Commission (2021), Impact assessment report accompanying the document directive of the european parlement and of the council amending Directive 2003/87/EC, https://eur-lex.europa.eu/resource.html?uri=cellar:7b89687a-eec6-11eb-a71c-01aa75ed71a1.0001.01/DOC_1&format=PDF#page=122.
[12] Eurostat (2025), Household budget surveys, https://ec.europa.eu/eurostat/web/household-budget-surveys.
[10] Eurostat (2025), Passenger cars in the EU, https://ec.europa.eu/eurostat/statistics-explained/index.php?title=Passenger_cars_in_the_EU&oldid=626230.
[11] Flues, F. and A. Thomas (2015), “The distributional effects of energy taxes”, OECD Taxation Working Papers, No. 23, OECD Publishing, Paris, https://doi.org/10.1787/5js1qwkqqrbv-en.
[3] Fridstrøm, L. and V. Østli (2021), “Direct and cross price elasticities of demand for gasoline, diesel, hybrid and battery electric cars: the case of Norway”, European Transport Research Review, Vol. 13/1, https://doi.org/10.1186/s12544-020-00454-2.
[2] Government of Romania & PwC (2023), Long term strategy of Romania, https://www.mmediu.ro/app/webroot/uploads/files/Long%20Term%20Strategy%20of%20Romania.pdf.
[13] IEA (2025), End-use Energy Prices, https://www.iea.org/data-and-statistics/data-product/end-use-energy-prices.
[1] IMF (n.d.), Romania: Economic Data, https://www.imf.org/external/datamapper/profile/ROU (accessed on 10 August 2026).
[8] Labandeira, X., J. Labeaga and X. López-Otero (2017), “A meta-analysis on the price elasticity of energy demand”, Energy Policy, Vol. 102, pp. 549-568, https://doi.org/10.1016/j.enpol.2017.01.002.
[7] Ministry of internal affairs (2025), Statistica, https://dgpci.mai.gov.ro/news-and-media/statistica.
[4] National Environmental Protection Agency (2025), Romania. 2025 National Inventory Document (NID). Annex, https://unfccc.int/documents/646458.
[9] OECD/ITF (2019), Tax Revenue Implications of Decarbonising Road Transport: Scenarios for Slovenia, OECD Publishing, https://doi.org/10.1787/87b39a2f-en.
← 1. The survival curve shows the probably that each age class remain in the fleet the following year.
← 2. For diesel, the sizes are below and above 2 L, while for gasoline they are split in three groups (up to 1.4 L, from 1.4 to 2 L and above 2 L).
← 3. Tax schedule by size in terms of cylinder capacity (cc) are up to 1 600 cc, from 1 600 to 2 000 cc, from 2 000 to 2 600 cc, from 2 600 to 3 000 cc, and over 3 000 cc.
← 4. While the HBS includes amounts spent on car purchase and maintenance over the year, no details are provided on the number of cars per households neither on their characteristics (e.g. Euro emission standard) which would be necessary to evaluate tax reform. Only the powertrain can be inferred from fuel consumption.
← 5. For instance, representative medium-size cars of the model based on Copert representative cars and Romania data have the following characteristics: a medium diesel-powered car (1 950 cc; 6.6, 6.3, 6.3 litre/100km, 8 811, 10 150, 13 082 km driven for Euro 2, 4 and 6d respectively) and a gasoline-powered car (1 900 cc; 8.6, 9.3 and 9.3 litre/100km, 5692, 4 804, 9 360 km driven respectively for Euro 2, 4 and 6d respectively).
← 6. If the average age of second-hand vehicles remains around 14 years, newer vehicles entering the fleet will comply with progressively stricter Euro emission standard. For instance, a 14-year-old car imported in 2025 is Euro 5, while in 2030 it is assumed to be Euro 6, reducing NOx emissions for diesel cars by 56%.
← 7. This finding is different from other EU member states where there is a strong difference in total energy use between rural areas (with strong reliance on personal cars) and urban areas (using less cars).