Unfortunately, as a result of the restrictions arising from the CoviD-19 pandemic, it is not currently possible to update the KonSULT website. It is being maintained as a teaching resource and for practitioners wishing to use its Measure and Package Option Generators and its Policy Guidebook. Practitioners wishing to use it, should do so on the clear understanding that recent experience on existing and new policy measures has not been incorporated.

Fuel Taxes

Fuel taxes are levied on the purchase of fuel in most countries. Levying a tax on fuel consumption not only raises revenues, it is also a relatively unselective means of charging for road use. However, there is some differentiation by fuel type and mode. Fuels that are considered to be the most polluting are often taxed at a higher level. Fuel tax escalators have also been applied to raise fuel tax on an annual basis regardless of need to generate income. This differentiation and annual incrementation are both designed to influence purchasing decisions in favour of more environmentally friendly choices. Some consumers, e.g. bus operators, are offered rebates or tax-free purchase, constituting a subsidy. Fuel tax income is rarely hypothecated.

Demand responses to fuel taxes will not necessarily have a dramatic effect on vehicle kilometres, especially in the short term. Costs may initially be absorbed, and it may take time to find alternatives. Consequently, impacts on policy objectives may follow the same pattern. It should also be noted that fuel taxes may not make a positive contribution to all policy objectives, as such there may be a number of looses as a result of taxation. Consequently, it may be difficult to raise taxes.

Terminology

Fuel taxes are levied on the purchase of fuel in most countries. Levying a tax on fuel consumption not only raises revenues (although this is generally not hypothecated), it is also a relatively unselective means of charging for road use. However, there is some differentiation by fuel type and mode. Fuels that are considered to be the most polluting are often taxed at a higher level. In the UK, four-star petrol is taxed at the highest rate, whilst unleaded and diesel are both cheaper. Fuel tax escalators have also been applied to raise fuel tax on an annual basis regardless of need to generate income. This differentiation and annual incrementation are both designed to influence purchasing decisions in favour of more environmentally friendly choices.

Some consumers are offered rebates or tax-free purchase, constituting a subsidy. In the UK, bus operators receive an 80% fuel tax rebate, train operators and airlines are not taxed on their fuel consumption and diesel purchased for agricultural uses is tax-free. Subsidies are usually provided to support socially necessary services or low profit industries. Some would argue that aviation fuel should be taxed. However, there are two counter arguments. Firstly, aviation fuel does not attract tax anywhere in the world, thus such a tax would need to be introduced and set on an international basis to maintain consistency. Secondly, if air travel were considered public transport, especially with regard to internal flights, then to levy fuel tax on airlines and not other forms of public transport would create an uneven playing field.

Fuel tax income is rarely hypothecated. Revenue for fuel tax and vehicle ownership taxation are both key income generators for many national governments. 

Why introduce fuel taxes?

Fuel taxes exist in three forms, as an indirect tax purely to raise revenue, as a charge for the use of roads, and as a pollution tax (DfT, 2002). The use of indirect taxation to raise revenue has been in existence for many years; the level can be interpreted as a historical accident or a reflection of the relative political marketability of each tax (DfT, 2002). Some economists recommend an increase in fuel tax matched by a decrease in direct taxes seen as harmful to the economy, e.g. income and investment taxes. That is to say an increase in fuel tax is part of a revenue-neutral tax shift (Litman, 2002).

Fuel tax as a charge for road space

When fuel tax is used to charge for road use, and hence to influence the amount of driving (to reduce congestion, pollution, severance and other negative consequences of car use), an excess over and above the economy wide rate of indirect taxation is levied (DfT, 2002). The excess is an amount, which will take the total fuel tax to a level that will influence fuel consumption – i.e. personal decisions about whether to drive or not, and/or driving style. The excess can be increased annually by a specified amount, or reviewed periodically.

Fuel tax as a pollution tax

Fuel tax can also be viewed as a pollution tax, based on the polluter pays principle (i.e. a Pigou tax). “If this tax has been set optimally, the tax payment per unit of fuel is equal to the social cost of the externalities generated by its consumption” (DfT, 2002). When the tax level is set in this way, it is not designed to influence consumption levels. However, if the current rate of tax means that the market prices are below marginal social cost, then raising the levy to equal marginal social cost may influence consumption levels in the first instance.

Fuel tax as a rationing device

Fuel tax can also be used as a rationing device. To meet commitments to the Rio Summit and other such treaties to limit fuel consumption, supply is treated as completely inelastic (DfT, 2002). Thus, an increase in consumption one industry sector must be matched by a reduction in another. In this case, the excess over and above the basic indirect tax would be increased more in sectors where a reduction in consumption is desired.

Demand impacts

Litman (2002) reports a number of price elasticities for fuel consumption. All the elasticities indicate that substantial increases in price are needed to achieve noticeable reductions in fuel consumption in the short term. Nevertheless, the effect appears to be cumulative over time. The following price elasticities are reported:

Estimate by

Short term

Medium term

Long term

Goodwin (1992)

-0.27% over two to three years

 

-0.7% over five to ten

Dahl (1991)

-0.18%

 

-1.0%

Hagler Bailly (1999)

-0.15%

 

-0.6%

DeCicco and Gordon (1993)

 

-0.3 to -0.5% in the US

 


Taking Goodwin’s estimates, a 10% price increase would reduce fuel consumption by 2.7% over two to three years, and 7% over five to ten years (Litman, 2002).

Reductions in fuel consumption can be achieved by reducing the number of journeys made or driving more fuel efficient vehicles.

“The elasticity of vehicle travel with respect to fuel price is typically found to be -0.20 to –0.3 (Harvey, 1994; Schimek, 1997; Johansson and Schipper, 1997), with values of about –0.1 in the short run, and up to –0.50 over the very long run” (Litman, 2002).

Joseph (2000) in DETR (2001) observes that traffic levels in the UK “have been relatively low, despite levels of economic growth that have previously stimulated significant traffic growth. Rail travel is at record levels and even cycle and motorcycle traffic is rising. Joseph considers the key difference between economic growth generating high car traffic growth in the past and not doing so now is high fuel prices.”

Responses and situations
Response Reduction in road traffic Expected in situations
There may be a small effect where changing departure time avoids congestion and thus reduces fuel consumption.
Increased fuel prices may encourage drivers to take the most direct route (assuming it is not overly congested), although it is unlikely that they will be significantly deviating from this anyway.
Destinations closer to home, work or other places that individuals cannot avoid travelling to will become more attractive.
The effect will depend on the size of the price increase, but those on low incomes will be affected first, and make the greatest reductions.
The effect will depend on the size of the price increase, but those on low incomes will be affected first, and make the greatest reductions. Public transport will become more attractive when the fuel cost for a journey becomes more than the public transport fare.
Again the effect will depend on the size of price increases, but those barely able to afford a car will be affected more. Over all drivers the more likely response would be to switch to a more fuel efficient vehicle.
Where the journey to work becomes prohibitively expensive, and cannot be undertaken by an alternative, cheaper mode, some individuals may relocate. In extreme cases, individuals may relocate both home and job, to be nearer to other family members for example. Again, those on low incomes will be affected first.
= Weakest possible response = Strongest possible positive response
= Weakest possible negative response = Strongest possible negative response
= No response

Short and long run demand responses

These estimates are based on fuel tax levels that at least keep pace with inflation.

Demand responses
Response - 1st year 2-4 years 5 years 10+ years
-
  -
  Change job location
- Shop elsewhere
  Compress working week
- Trip chain
- Work from home
- Shop from home
  Ride share
- Public transport
- Walk/cycle
  -
  -
= Weakest possible response = Strongest possible positive response
= Weakest possible negative response = Strongest possible negative response
= No response

Supply impacts

There are no supply impacts.

Financing requirements

There are no financial requirements assuming mechanisms for collecting tax are already in place. Conversely, fuel tax is a major source of income.

Expected impact on key policy objectives

Contribution to objectives

Objective

Scale of contribution

Comment

  By reducing congestion, efficiency will be increased.
  By reducing congestion, severance, pollution, accidents and other negative impacts will be reduced, thus, improving liveability.
  By decreasing car use, negative environmental impacts will be reduced.
  As fuel tax is indiscriminate and affects those on low incomes most, it may have negative effects in terms of equity and inclusion. Those for whom car use is essential due to the nature of the journey or lack of alternative will also be unduly penalised. Where tax levels differentiate between types of fuel, cheaper options may be available if those concerned can afford to change the vehicle they drive.
  Reductions in car use will improve safety.
 
If increased expenditure on fuel reduces spending in other sectors of the economy, economic growth may be stifled. Although, where individuals change their travel behaviour to avoid increased costs, there will be no negative effect on economic growth. Indeed, where the changes in behaviour result in reduced congestion, growth may occur.
  Increased revenue from taxes will benefit finances.
= Weakest possible positive contribution = Strongest possible positive contribution
= Weakest possible negative contribution = Strongest possible negative contribution
= No contribution

Expected impact on problems

Contribution to alleviation of key problems

Problem

Scale of contribution

Comment

Congestion-related delay

Reductions in car use will reduce congestion-related delay.

Congestion-related unreliability

Reductions in car use will reduce congestion-related unreliabilty.

Community severence

Reductions in car use will reduce severance.

Visual intrusion

Reductions in car use will reduce visual intrusion.

Lack of amenity

Increased demand for local goods and services may result in increase amenity.

Global warming

Reductions in car use will reduce pollution contributing to global warming.

Local air pollution

Reductions in car use will reduce local air pollution.

Noise

Reductions in car use will reduce noise.

Reduction of green space

/
Reduced demand for car use will reduce demand for increased road infrastructure and may therefore prevent reductions in green space. However, increased demand for public transport and associated infrastructure may negate this effect.

Damage to environmentally sensitive sites

Reduced demand for car use will reduce demand for increased road infrastructure and may therefore prevent damage to environmentally sensitive sites. However, increased demand for public transport and associated infrastructure may negate this effect.

Poor accessibility for those without a car and those with mobility impairments

/
Those with mobility impairments who rely on a car to get around will be penalised unless exemptions or rebates are offered. Those without a car may benefit if increased demand for alternatives improves provision.

Disproportionate disadvantaging of particular social or geographic groups

/
Those for whom car use is essential, e.g. some rural residents, or shift workers for whom there is no adequate alternative, will be penalised unduly. The effect will be greatest on those with low incomes. Those without a car may benefit if increased demand for alternatives improves provision.

Number, severity and risk of accidents

Reductions in car use will reduce the number, risk and severity of accidents.

Suppression of the potential for economic activity in the area

/
Reduced congestion will improve the attractiveness of an area. Although, increased spending on fuel that reduces spending in other sectors of the economy may stifle economic growth.
= Weakest possible positive contribution = Strongest possible positive contribution
= Weakest possible negative contribution = Strongest possible negative contribution
= No contribution

Expected winners and losers

One would not expect everybody to benefit equally from any transport measures. However, a PJP campaign does not force anybody to change their travel habits, thus there is more potential for winners than losers.

Winners and losers

Group

Winners/Losers

Comment

Large scale freight and commercial traffic

If these users are subject to tax increases (i.e. no exemptions or rebates) they will lose if alternatives are not available. However, if congestion is reduced enough to make substantial cost savings, the impact on these groups may be neutral.

Small businesses

If these users are subject to tax increases (i.e. no exemptions or rebates) they will lose if alternatives are not available. However, if congestion is reduced enough to make substantial cost savings, the impact on these groups may be neutral.

High income car-users

The impact on high income car users may not be significant and not sufficient to prevent car use, but it is nonetheless a negative impact unless reductions in congestion are sufficient to negate the fuel tax cost.
People with a low income / Those dependent on a car will lose, but those who use alternatives may benefit from improved provision if increases in demand are sufficient to cause improvements.

People with poor access to public transport

Where increased demand results in improved provision, these people will benefit. Otherwise the effect will be neutral or negative if public transport is more congested.

All existing public transport users

/ If operators are subject to fuel tax increases, price rises may be passed onto customers who will loose. Where rebates are offered, this negative impact will be minimal. Although, where increased demand results in improved provision, these people will benefit. Otherwise the effect will be neutral or negative if public transport is more congested.

People living adjacent to the area targeted

Fuel tax is not area based.

People making high value, important journeys

These journeys may still be made as solo drivers, but reduced congestion will result in valuable time savings.
The average car user The average car driver will be penalised unless they are able to use an alternative.
= Weakest possible benefit = Strongest possible positive benefit
= Weakest possible negative benefit = Strongest possible negative benefit
= Neither wins nor loses

Barriers to implementation

Scale of barriers
Barrier Scale Comment
Legal Possible within current legislation.
Finance Raises revenue.
Governance Complex coordination needed between national and local government and between adjacent authorities, particularly to avoid spill over effects.
Political acceptability Highly contentious, given expected serious opposition by car owners and suppliers.
Public and stakeholder acceptability Any increase in the costs of motoring is seen as very unpopular.
Technical feasibility No technical constraints.
= Minimal barrier = Most significant barrier

Case study 1: Tromso, Norway
Case study 2: California, USA

 

Tromso, Norway

The material presented here comes from Milne, Niskanen and Verhoef (2000), Operationalisation of Marginal Cost Pricing within Urban Transport, AFFORD Deliverable 1, Government Institute for Economic Research, Helsinki, plus private communications with Dr David Milne, Institute for Transport Studies, University of Leeds, and Terje Tretvik, SINTEF, Norway.

Context

Charging schemes in Norway are implemented on a temporary basis to fund infrastructure construction. Once costs have been recovered, charging schemes are removed. One such charging scheme is the implementation of a fuel tax increment in Tromso. Whilst fuel tax cannot usually be varied between regions due to the high potential for avoidance behaviour, Tromso is so remote that it would not be possible to drive to an area with lower fuel tax to purchase fuel.

The additional fuel tax - 0.50 NOK per litre of petrol and auto diesel - was introduced in 1990, and increased to 0.65 NOK in 1996. The tax will be in place until the end of 2003. Net income from the tax was 18.7 million NOK in 2001. The tax was introduced to part finance an investment package of main highway infrastructure (including a subsea and other tunnels) for Tromso.

Impacts on demand

The tax was expected to reduce car travel by two to two and half percent. However, this does not appear to have happened. It is likely that the demand impacts of the infrastructure funded by the tax have counteracted some impacts of the tax itself. The subsea tunnel is known to have increased car and decreased bus travel for certain origin-destinations. Further to this, total transport costs have decreased despite the tax, and cyclists and buses have obtained improved operating conditions.

Contribution to objectives

Contribution to objectives
Objective Comment
  The additional infrastructure funded by the tax will have improved efficiency.
  Possible increases in car use as a result of additional infrastructure may have had negative impacts on liveability.
  Construction of new infrastructure and possible increases in car use as a result may have had negative impacts on the environment.
  If the new infrastructure has improved accessibility, then equity and inclusion benefits are possible. However, the tax itself may have had negative impacts, especially on those with a low income.
  Increased car use resulting from availability of new infrastructure may have had negative safety impacts.
  Improved infrastructure may have attracted investment and hence, generated economic growth.
  The financial impacts of the tax are neutral, as it is implemented to recover costs and removed once this has been achieved.
= Weakest possible positive contribution = Strongest possible positive contribution
= Weakest possible negative contribution = Strongest possible negative contribution
= No contribution

California, USA

Context

The material cited here is taken from the Online Travel Demand Management Encyclopedia, compiled by Litman.


Whilst the Tromso example cited above is an exception, fuel tax is usually set at a national level, and is not generally hypothecated. This means that fuel tax is generally not implemented in same way as other charging instruments such as urban road charging. Fuel tax is implemented as a general revenue raiser and to influence consumption of fuel to meet environmental objectives. Impacts are therefore national and virtually impossible to disentangle from other national policy instruments. However, Litman (2002) cites modelling of the effect of fuel tax increases in California, USA.

Table 1: Impacts of Fuel Tax Increase, Year 2010 (Harvey and Deakin (1997) Table B.8 in Litman (2002))

Region

Additional fuel taxes applied in addition to current taxes

Change in total vehicle mileage

Change in total vehicle trips

Change in congestion-related delay

Change in fuel consumption

Annual revenue in millions of 1991 US$

 

$0.50

-3.6%

-3.4%

-8.5%

-8.8%

$1,332

Bay Area

$2.00

-11.7%

-11.3%

-25.5%

-30.6%

$4,053

 

$0.50

-4.1%

-3.9%

-7.0%

-9.3%

$414

Sacramento

$2.00

-13.2%

-12.7%

-22.0%

-31.8%

$1,245

 

$0.50

-3.9%

-3.5%

-8.0%

-9.1%

$747

San Diego

$2.00

-12.5%

-12.0%

-23.0%

-31.1%

$2,257

 

$0.50

-4.2%

-3.5%

-9.5%

-9.3%

$3,724

South Coast

$2.00

-13.0%

-12.5%

-28.5%

-31.6%

$11,235

Deakin and Harvey (1997) model the effect of a fuel tax increase on transportation impacts in four major urban regions in California. Table 1 summarizes their results for the year 2010. It indicates, for example, that in the South Coast (Los Angeles) region, an additional $50 per gallon tax would reduce total vehicle trips by only about 3.5%, but congestion delay would decline by 9.5%, and fuel consumption would decline by 9.3%. Another study finds that a $0.40 increase in fuel prices would reduce regional vehicle trips by 1.2% and vehicle mileage by 1.4%, while a $2.00 increase would reduce trips by 6.7%, and mileage by 7.2% (PSRC, 1994) (Litman, 2002)

Contribution to objectives

Contribution to objectives
Objective Comment
  The decrease in congestion-related delay will make a significant contribution to efficiency.
  The reductions in vehilce trips and mileage, and fuel consumption will contribute to liveability improvements.
  The reductions in vehilce trips and mileage, and fuel consumption will contribute to protection of the environment.
  No evidence is presented on this topic.
  The reductions in vehilce trips and mileage should contribute to a reduction in the number of accidents.
  The decrease in congestion-related delay will make a significant contribution to economic growth.
  Substantial incomes can be generated from fuel tax increments.
= Weakest possible positive contribution = Strongest possible positive contribution
= Weakest possible negative contribution = Strongest possible negative contribution
= No contribution

Gaps and weaknesses

As indicated, there is little evidence on the effect of fuel taxes as a policy instrument because they are generally implemented on a national basis to meet a wide variety of objectives, not all of which are transport related.

Contribution to objectives and alleviation of problems

Contribution to objectives and alleviation of problems
Objective Tromso* California
 
 
 
 
 
 
 
= Weakest possible positive contribution = Strongest possible positive contribution
= Weakest possible negative contribution = Strongest possible negative contribution
= No contribution
*The assessment here does not separate the contribution of the tax from that of the infrastructure funded by the tax.

Contribution to alleviation of key problems

Contribution to alleviation of key problems
Objective Tromso* California
Congestion-related delay
Congestion-related unreliability
Community severance **
Visual intrusion
Lack of amenity
Global warming
Local air pollution
Noise
Reduction of green space
Damage to environmentally sensitive sites
Poor accessibility for those without a car and those with mobility impairments
Disproportionate disadvantaging of particular social or geographic groups
Number, severity and risk of accidents
Suppression of the potential for economic activity in the area
= Weakest possible positive contribution = Strongest possible positive contribution
= Weakest possible negative contribution = Strongest possible negative contribution
= No contribution

*The assessment here does not separate the contribution of the tax from that of the infrastructure funded by the tax.
**Whilst car use may have increased, the severance caused by fijords has been significantly reduced.  Appropriate contexts

Appropriate area types is not relevant here, although it is worth noting that those in areas with few alternatives to the car, and/or long journey distances due to low density development may suffer a greater tax burdeon than others.

Adverse side-effects

There is potential for a number of adverse side effects from the application of fuel taxes. Those who depend on a car for accessibility, be it due to the nature of their work (e.g. shift work), residential location (e.g. some rural areas), or mobility impairment may be unfairly penalised if adequate alternatives are not available and rebates/exemptions are not offered. The effects will be most severe for those on low incomes. Existing public transport users will also be penalised if rebates/exemptions are not offered to the operators and price increases are passed on to the customer. Alternatively, if savings are made through cut backs, public transport users will suffer in other ways.

There is also the possibility of wider adverse economic effects if freight and commercial traffic is significantly penalised and no alternatives are available. Whilst costs can be passed on to customers this is likely to have an inflationary effect and reduce spending across the economy. Although, if reductions in congestion are sufficient to make valuable time savings, these negative effects on the economy could be negated.

Department for the Environment, Transport and the Regions (DETR) (2001) ‘The Potential for Further changes to the Personal Taxation Regime to Encourage Modal Shift’, Final Report by The Open University, WS Atkins and Napier University, http://www.dft.gov.uk/itwp/modalshift/index.htm (27/11/02).

Department for Transport (DFT) (2002) ‘Treatment of Taxation in the Cost-Benefit Appraisal of Transport Appraisal’. http://www.roads.dft.gov.uk/roadnetwork/heta/resresult/treatment/09.htm (27/11/02).

Litman T (2002), Online TDM Encyclopaedia, Victoria Transport Policy Institute, http://www.vtpi.org/tdm/tdm17.htm (11/12/02).

Milne, Niskanen and Verhoef (2000), Operationalisation of Marginal Cost Pricing within Urban Transport, AFFORD Deliverable 1, Government Institute for Economic Research, Helsinki