Policy commentary

How Regressive Are Fuel Taxes? A Comparison of Countries from Around the World

May 2, 2011 | Thomas Sterner

Raising fuel taxes could significantly reduce emissions of greenhouse gases and other pollution from the transportation sector. One of the prime arguments against raising fuel taxes is the perception that they are regressive—that they are more costly to the poor and other socioeconomic groups. But recent research suggests the opposite, particularly for developing countries.

Recognition of climate change as a serious problem has become widespread. Although there is still some uncertainty and disagreement about just how serious the problem is, there is wide support for the broad picture painted by the Intergovernmental Panel on Climate Change (IPCC), that the world must drastically cut its emissions of fossil carbon (and other climate gasses). In contrast, there is no agreement on policies at local, national, and international levels—rather, what some might call paralysis.

Questions of fairness and the inequitable effects of climate policies on the poor are primary concerns for policymakers. Here I will focus on a policy that exemplifies these issues, the use of fuel taxes levied at the national level to address carbon emissions in the transport sector. Global emissions from the transport sector account for a large share—about 25 percent—of global fossil carbon emissions and this share has been growing (for example, in the EU, it has gone from 20 to 30 percent in the last 20 years).

Ultimately, any effective policy has to imply that the costs of emissions rise. One of the prime arguments against fuel taxation is concern for the poor, those in rural areas, and certain groups in society. In this context, fuel taxes are considered “regressive,” because everyone pays the same amount of money per liter or gallon, making the percentage of burden heavier on those with less income, less access to alternative means of transportation, or greater distances to travel.

Where the Differences Exist

The world happens almost to have a natural experiment in this area because fuel taxation policy differs considerably across countries, allowing us to study the effects of this instrument. Fuel tax rates vary widely, from a mere 19 cents per liter in the United States to $1.19 per liter, the average for Western European countries (for more detail, see Table 1). Elsewhere, such as in Japan and Australia, rates fall in the intermediate range. There are two notable trends: taxes within Western European countries have been converging over the last 20 years, whereas the spread between Western Europe and other countries has been increasing over time (since 2005, the average in this table for West European countries went up by 35 percent while the other countries’ average rose by less than 20 percent).


Table 1. Gasoline Taxes in US$ cents/liter in Selected Countries, 2008


Western Europe Gasoline Tax Non European Gasoline Tax
 Italy  116  Japan  59
 UK  123  Australia  43
 Netherlands  139  New Zealand  51
 France  121  Canada  32
 Belgium  125  Mexico  10
 Germany  128  USA  13
 Finland  129  Average  35
 Portugal  122    
 Sweden  120    
 Spain  90    
 Austria  100    
 Average  119    

The main reason why fuel prices vary between countries is differences in taxation. Other reasons include the costs of handling, quality related and refinery costs, product transportation, and distribution costs but these only for a small share of the variation. Source: International Energy Association 2009.


Taxes and Emissions

A particularly interesting comparison can be made between the United States and Europe. The very low fuel taxes in the United States compared even to the lowest tax rates in Europe are related to higher fuel use—although the relationship is mainly apparent in the very long run. The average-per-capita consumption of gasoline in the United States is more than four times higher than in the United Kingdom or several other European countries. Such data suggest that if the European Union had followed a similar tax policy to that in the United States, aggregate carbon emissions would have been substantially higher.

In past research, I estimated the effect of reducing fuel taxes in all OECD countries as low as that of United States. The differences in gasoline consumption (and thus emissions) turned out to be substantial. Transportation emissions from the OECD would have been about 30 percent higher if all fuel taxes had been as low as in the United States. These estimates show that fuel taxes are a very potent instrument for climate policy and that this is one of the few policies that have had a measurable effect on the carbon content of the atmosphere.

Similar comparisons could be made for developing countries with high or moderate fuel taxes compared to those that do not tax, or tax domestic consumption very lightly. Oil-exporting developing countries have typically had very low domestic prices, and often subsidize fuel consumption. Given how effective fuel taxes are, one might wonder why they are not adopted in more countries.

Regressivity? Popular Belief versus Actual Fact

One of the main reasons for resistance to fuel taxes appears to be the popular belief in many, many countries is that fuel taxes hurt the poor. However, my studies with colleagues (see Further Reading) have found neutral or very weak regressive results in richer countries and quite strong progressive evidence in the developing countries, such as China, India, Ethiopia, Indonesian, Ghana, Nairobi, Mali, and several more. The intuition is not surprising; in most developing countries, the very poorest households cannot afford to own a car at all. Fuels have more of a “luxury” character and hence fuel taxes are more progressive. On the other hand, fuel taxes also increase the cost of public transportation (and other goods). Because the poor typically use public transportation more intensively, this effect might mitigate the progressivity of fuel taxes, but we have found that the net effect is still progressive in practically all developing countries.

In fact, it is quite often the middle (or upper-middle) classes that bear the brunt of the taxation. Maybe they help spread the myth that fuel taxes are regressive, for they compare themselves to the richest and not to the poor who are much less “visible.” This tendency is often reinforced by anecdotal evidence concerning some individual motorists—often in the countryside—who are adversely affected by fuel taxes. It is inevitable that commuters who drive long distances will be negatively affected by higher fuel taxes. It would be wrong, however, to say that drivers rural areas are negatively affected in general; in fact, many rural people have very low average mileage compared to drivers in urban or quasi-urban areas.

The United States is one of the very few exceptions to some of these trends. This is at least partly explained by the lack of public transport and the high general income level, which implies that even the poor have cars. Moreover, to the extent that the domestic fuel tax is regressive, it could very easily be made progressive by a suitable use of the tax proceeds, through either lump-sum repayments or environmental fiscal reform so that tax revenues are used to lower some other, more regressive tax.

In closing, the reader is reminded that we urgently need to cut our fossil emissions and that transport fuels are probably less sensitive from the viewpoint of jobs and competitiveness and probably a lot easier to deal with than reducing heating and industrial uses in very poor countries.


Further Reading

This article is adapted from chapters by the author in Fuel Taxes and the Poor, to be published by Taylor & Francis, October 2011.

Sterner, Thomas. 2007. Gasoline Taxes: A useful instrument for climate policy? Energy Policy . 35(6): 3194–3202.

Sterner, T., and D Slunge. 2009. Environmental Fiscal Reform in East and Southern Africa and its Effects on Income Distribution, in Environmental Taxes and Fiscal Reform edited by Laura Castellucci and Anil Markandya. Rivista di Politica Economica VII–IX(July–Sept).