Journal Issue

Petroleum taxes: how high and why?: A survey of 1973 and 1974 taxes in 64 countries showing the government’s share in the price rise

International Monetary Fund. External Relations Dept.
Published Date:
December 1975
  • ShareShare
Show Summary Details

Katrine W. Saito

The upsurge in the price of crude oil during the second half of 1973 and early 1974 has generated substantial discussion over its impact on the economies of developing and developed countries alike. Attention has tended to be concentrated on the crude price increases, and one aspect which has been relatively ignored is the magnitude of the changes in the retail prices of petroleum products. Such changes reflect not only the increase in the value of crude oil but also changes in the taxation of petroleum products.

The purpose of this article is to examine how different governments responded with the taxation of petroleum products and how much of the increase in prices of oil products is, in fact, attributable to the changes in taxation. Petroleum products have traditionally been a favorite target of excise taxation, and the tax component of petroleum products is frequently quite sizable so that minor adjustments in the tax per gallon can significantly affect the retail price. But fuel taxes vary according to product, and these variations and their implications can be examined in terms of equity, efficiency, and revenue.

Price and tax data collected from 64 developing and developed countries for the months of July 1973 and July 1974 give a composite picture of the situation. Considering that the major increases in crude prices occurred between October 1973 and January 1974, these two points of observation should adequately reflect the impact of the changes in crude prices on the retail prices of refined products, as well as the corresponding change in taxation. The prices chosen for the survey were those in the capital cities and the taxes were those levied on domestically refined products, with the exception of those few countries surveyed which have no refining capacity. In such cases, import duties were taken into account, in addition to taxes levied on domestic sales.

Modes of taxation

In both developing and the more developed countries petroleum products are regarded as a reasonably equitable and administratively easy way of raising revenue. They provide an obvious and rather simple tax device since they are usually manufactured by large enterprises and reach a wide market. They are generally taxed by a specific levy on the quantity sold. Such specific fuel taxes have the merit of administrative simplicity, and in most of the countries surveyed they were preferred to an ad valorem tax, which involves problems of valuation of the tax base. When the value of the tax base is highly volatile, however, as in the case of petroleum products over the last few years, an ad valorem tax does have the advantage of automatically adjusting to changes in the tax base. In many European countries, while the main levy is specific, a value-added tax is also commonly levied on petroleum products; although such a tax is essentially ad valorem in its method of assessment, it would not be automatically elastic to the ex-refinery price and would vary only in proportion to the value added at the stage of refinement (or at whatever stage the tax is levied).

Administration of petroleum product taxation is made easier by the availability of a few easily identifiable points of collection. Those countries which levy specific taxes generally collect them at the refinery gate or through a few wholesalers. Ad valorem taxes are collected at the point of importation (as in the French-speaking African countries) or at the refinery gate (as in Korea, where the tax is assessed as a percentage of the ex-refinery price). In Brazil, the tax on petroleum products is assessed by multiplying the c.i.f. price of the imported crude oil by the relevant tax rate. In Mexico, an oil producing country, the state-owned oil company is obliged to pay a fixed percentage of the total value of its sales to the Government. Clearly, when the importing or refining company also has interests in distribution, there is additional incentive for underassessment of the taxable value.

In practice, taxes on petroleum products are important revenue raisers in most countries, as shown in Sijbren Cnossen’s study on “Revenue Aspects of Excise Systems” (unpublished Fund document), where the revenue from petroleum products as a percentage of total tax revenue for selected countries was found to range between 10 and 16 per cent. Considering how far reaching the effects of such taxes can be, it is essential to minimize their distortionary effects on relative prices.

However, taxes on petroleum products, whether specific or ad valorem, do introduce some measure of distortion into the market. The situation is complicated by the fact that pricing policies of oil marketing agencies deviate significantly from the optimal so that there is a likelihood that taxes might lessen distortions in the price structure instead of increasing them. This is not the only problem in determining an optimal system of tariffs for petroleum products. In the transport sector the demand for fuels is a derived one and the question arises as to what extent taxes should reflect such costs as the maintenance of transport facilities or congestion costs. Further, the tax structure affects the utilization of different transport modes, and taxes on petroleum products need to be graduated so as to bring about the “best” distribution of traffic across the various modes. Finally, there is the question of incidence. Social objectives may favor progressive taxes, and this must be kept in mind when differentiating fuel taxes according to product.

Fuel tax differentiation in practice

All 64 countries surveyed differentiated fuel tax rates according to product. With the exception of Nigeria, Switzerland, and the United Kingdom, where diesel oil was taxed at the same specific rate as gasoline, the tax per gallon of gasoline was higher than that of other products, with rates frequently differing according to octane value. Diesel oil, kerosine, and heavy fuel oil were generally taxed at lower rates; heavy fuel oil was usually taxed the least and was exempt from tax in one fourth of the countries surveyed. While such a gradation of tax rates corresponds in rank to the costs of producing the different products, it is impossible to say that the price-cost ratios would be maintained with the application of the tax. Marginal costs vary, among other things, according to type of crude import, state of technology in the refinery, volume, and composition of output.

An interesting distinction can be made between the fuel tax levels in developing and developed countries. In July 1973, the tax per gallon of gasoline was, on average, much less in developing than in developed countries, averaging 26 and 28 U. S. cents a gallon of regular and premium grades respectively in the developing countries, and 51 and 53 cents respectively in the developed countries (see Table 1). Retail prices of gasoline were consequently higher, on average, in developed countries, though the prices net of tax did not differ substantially between the two groups.

Table 1Average retail prices and taxes on petroleum products and their percentage change, July 1973–July 1974

(U.S. cents per gallon)

GasolineHouseholdAutomotiveHeavy fuel
RegularPremiumkerosinedlesel oiloil
All countries
July 197363.234.375.336.833.07.439.817.212.71.3
July 197497.141.9108.445.649.58.157.617.624.41.9
Percentage change53.622.144.123.749.99.944.82.691.843.3
Developed countries
July 197383.250.891.152.841.98.264.836.414.31.2
July 1974113.655.0123.557.760.010.481.935.827.11.7
Percentage change36.68.335.69.343.226.926.4−1.788.840.5
Developing countries
July 197352.525.566.327.928.
July 197488.234.999.938.743.96.947.710.322.82.1
Percentage change68.036.850.639.155.2−0.661.19.394.144.9
Of which:
Major oil producing countries
July 197324.18.731.29.714.92.517.96.08.2−0.31
July 197426.17.634.49.915.21.418.53.79.6−0.71
Percentage change8.1−12.810.41.81.7−45.13.3−38.517.0−110.7
Non-oil Producing countries
July 197368.938.181.940.935.
July 1974107.446.9119.651.
Percentage change55.923.246.024.553.012.547.56.899.245.7



Household kerosine was generally subject to similar taxes per gallon on both developed and developing countries. In many developing countries kerosine is the main source of energy for lower-income groups, whereas in more developed countries consumption of kerosine is relatively unimportant compared with other refined products. In Bangladesh, for example, kerosine comprised 32 per cent of the total volume of consumption of refined products in 1973, compared with the Federal Republic of Germany where it was less than 2 per cent. While tax per gallon differed little, on average, between the two groups of countries, retail prices averaged almost 50 per cent higher in developed countries. The explanation, presumably, lies in the higher distribution costs in the former group, where per capita consumption is much lower compared with many developing countries, and also in the tendency for the price of kerosine to become frequently an issue of social and political concern in many developing countries and thus to be subject to greater control.

Taxes per gallon of automotive diesel oil were relatively high in developed countries, averaging 39 U. S. cents per gallon (four times the average tax in developing countries). This can be attributed to two factors: first, the tendency for less emphasis to be placed on maintaining low cost public and truck transportation in many developed countries; and second, diesel oil is used for certain passenger cars in Western Europe. The before tax price did not differ significantly between the two groups. Heavy fuel oil prices and taxes were similar in both groups.

Changes in prices and taxes

Between July 1973 and July 1974, there was an approximately fourfold increase in crude prices. Retail prices of refined products reacted variously to these changes, both among products and among countries. On average, prices of gasoline, kerosine and diesel oil rose between 44 and 54 per cent, while, curiously, the price of heavy fuel oil rose by more than 90 per cent to an average price of 24 U. S. cents per gallon ($10.08 a barrel). Generally, both developing countries and developed countries experienced this huge jump in the price of fuel oil; prices of other products, however, rose less substantially on average in the developed countries than in the developing countries. Prices of gasoline increased by less than 40 per cent in the developed countries compared with increases of 68 and 50 per cent for regular and premium grades respectively in the developing countries. The price of diesel oil rose by almost twice as much in percentage terms in developing countries as in the developed ones, and the average increases in the price of kerosine were 55 per cent in developing countries and 36 per cent in developed countries. In spite of these larger percentage increases, the prices of gasoline, kerosine, diesel oil, and heavy fuel oil in developing countries in July 1974, were on average still substantially below those in the more developed countries.

During these 12 months, very few countries left the taxation of petroleum products unchanged; most of them increased the tax per gallon on the products surveyed, though they raised it more on some products than on others. Taxes on heavy fuel oil, while substantially increasing in percentage terms (on average, for all countries, 43 per cent), remained less than 20 cents a gallon (84 U. S. cents a barrel). Thus, less than 5 per cent of the 92 per cent average price increase of heavy fuel oil can be attributed to the change in taxation.

Gasoline tax increases

Changes in the taxation of gasoline, kerosine and diesel oil differed markedly between the developing and the developed countries. Gasoline was already subject to high levels of taxation in the developed countries in 1973, and during the 12 months from July 1973 to July 1974 the tax per gallon of gasoline was raised on average by less than 10 per cent while prices increased by 37 per cent. Thus, less than 15 per cent of the average retail price increase during this period can be explained through taxation. In developing countries, however, gasoline taxes were increased by an average of almost 40 per cent, with taxes on the premium grade increasing slightly more than those on the regular grade. Prices rose by 68 per cent in the same period. As a result, 26 and 32 per cent of the price changes of regular and premium grades, respectively, can be attributed to the tax increases.

Rather surprisingly, those developing countries which more than doubled the gasoline tax tended to be those with relatively high tax ratios of total tax revenues to gross national product. The average tax ratio of such countries was 16 per cent, compared with an overall average of 14 per cent. By contrast, no consistent pattern can be detected in the more developed countries. One possible explanation for this phenomenon in developing countries is institutional. Those countries with relatively high tax ratios tend to be those with a more effective tax administration and with less resistance from taxpayers. The increase in crude prices provided roles this commodity has in the consumption patterns of developed and developing countries. In many of the latter, kerosine expenditures are a significant item in the budget of lower-income groups, since kerosine is the principal source of their energy supply. Its price, therefore, takes on a social as well as political importance, and any increase can meet with strong resistance. This is especially true in Southeast Asian countries; in Indonesia, for example, social unrest in 1968 was price rise can be attributed to taxation.

While these generalizations hold true for these two groups of countries as a whole, within each group divergent responses to the upsurge in crude prices can be identified in terms of tax modifications and policy motivations. Almost one fourth of the countries surveyed increased the tax rate on all the products surveyed. These include both developed countries (such as Austria, Italy, and the Netherlands) and developing countries (such as both the justification as well as the excuse for substantially increasing gasoline taxes.

Taxes on automotive diesel in developed countries, which were already an average of four times the level prevailing in developing countries in July 1973, declined slightly, on average, during the next 12 months. The 26 per cent price increase in the group, therefore, is attributable entirely to nonfiscal elements such as increases in the crude input cost, refinery costs, profit margins, and the general inflationary impact. Developing countries, on the other hand, raised the tax on diesel by an average of 9 per cent; however, less than 5 per cent of the substantial increase of 61 per cent in diesel prices was due to increased taxation.

Role of consumption patterns

The different tax treatment of kerosine during the 12 months between these two groups of countries reflects the different partly attributable to the proposed increase in the price of kerosine. In many South and Central American countries, however, liquefied petroleum gas is becoming an increasingly important source of energy for lower-income groups. This concern over the price of kerosine is reflected in the slight decline in the average tax per gallon of this product in developing countries during these 12 months. The 55 per cent price increase in this group, therefore, reflects the impact of the rise in crude and other production costs, as well as the inflationary trend. In developed countries kerosine is generally regarded as a luxury item, used for heating or cooking at vacation resorts, camping, and so on. In terms of total refined products, its consumption is relatively unimportant. Tax per gallon of kerosine in this group increased by almost 30 per cent during these 12 months while the price went up 43 per cent so that almost 13 per cent of the Argentina, Burma, Korea, and Paraguay) and they were presumably motivated by the need to constrain demand, as well as by revenue considerations.

The oil producing countries

The oil producing nations surveyed (Bolivia, Colombia, Ecuador, Indonesia, Nigeria, Saudi Arabia, and Venezuela) had relatively low levels of taxes on petroleum products, and generally reduced these tax rates during the 12 months between July 1973 and July 1974. Saudi Arabia, for example, abolished all taxes on petroleum products in August 1974, and prices fell by the full amount of the tax. In Indonesia the fiscal monopoly profits of the state-owned oil company were reduced on all the products surveyed in the first quarter of 1974. In the sense that the official retail prices were less than the stated production cost of the oil company, kerosine, industrial diesel oil, and heavy fuel oil were all subsidized in 1973, and in the first half of 1974 these subsidies were increased substantially. The oil producing countries of Central and South America which either reduced or left unchanged the taxes on petroleum products in 1974 are Bolivia, Colombia, and Ecuador. Each of these countries was a net exporter of crude oil in 1973, and each obtained its domestic supply of crude at concessionary prices through a variety of agreements with foreign oil companies. In common with other oil producing countries, the combination of cheap supplies of crude, minimal taxes, and (at least in the case of Indonesia) retail prices below cost, have made prices in these countries among the lowest in the world.

Oil producing countries which sell their petroleum products below the market value, subsidize the petroleum users to the full extent of the price differential between the free market prices and the domestic prices. The petroleum industries could withstand such an arrangement only because the prices in the export sector were those prevailing in the world market. Thus export sales, in effect, supported the lower-priced domestic sales.

The only non-oil producing country surveyed which attempted to absorb some of the increase in crude costs through reducing the tax per gallon of the five products considered was Ghana. It is, in fact, subsidizing both kerosine and diesel oil. Despite tax reductions, however, the prices of gasoline, kerosine, and diesel oil increased by about 70 per cent in Ghana.

A “compromise” system of taxation

Most of the countries surveyed opted for the “compromise” choice regarding its taxation policy, increasing the tax rates on certain products, notably gasoline, while leaving the rates unchanged or reducing them on others, such as kerosine, diesel oil, and fuel oil. In this way, some attempt was made to restrict demand for crude oil imports through limiting gasoline consumption, yet price increases affecting the industrial and agricultural sectors, as well as the lower-income groups, were lessened. Panama, for example, doubled the specific tax on gasoline between July 1973 and July 1974, while introducing subsidies on kerosine and diesel. Bangladesh, Paraguay, Tunisia, and Uruguay, all raised the tax on gasoline and reduced the tax on kerosine. Bangladesh, in fact, substantially reduced the tax on both kerosine and diesel to 9 and 15 U. S. cents a gallon, respectively, compared with a gasoline tax of 55 to 60 U. S. cents a gallon. El Salvador, Jamaica, and Kenya continued exempting kerosine from excise tax. However, Jamaica and Kenya increased the tax on gasoline. In Korea, the fuel tax differentials were among the largest of any country surveyed; the taxes are levied ad valorem, so that all taxes increased on a per gallon basis. While the tax per gallon of less refined distillates in Korea remained below 15 U. S. cents a gallon, the doubling of the gasoline tax rate resulted in a tax of $1.13 to $1.34 a gallon—the highest gasoline tax in any of the developing countries surveyed in July 1974 and rivaled only by Italy among the developed countries.

This exaggeration of the fuel tax differentials not only widened the price-cost relationship for individual products but also had a distorting, effect since it increased the difference between these ratios for products with some degree of substitutability. In Korea, for example, in July 1973 the retail price of regular gasoline was four times the ex-refinery price (net of taxes and distribution costs) and increased to almost five times by July 1974. In 1973, the retail price for diesel, however, was less than twice the ex-refinery price, and the ratio declined slightly during the next 12 months. This distortion of the price-cost relationship was most pronounced in Korea; in Brazil, Nicaragua, and Thailand the retail prices of gasoline and diesel oil were approximately double the ex-refinery price in July 1974, declining somewhat over the previous 12-month level.

Taxes on ad valorem basis

Taxes on petroleum products estimated on an ad valorem basis (see Table 2) declined on average during these 12 months in both developed and developing countries for all products surveyed. For the developed countries, the ratios of taxes to retail prices were higher on average than those in the developing countries in both years, despite a larger percentage decline during the 12-month interval (a 20 per cent decline, compared with decreases of 18 and 8 per cent in developing countries for regular and premium grades, respectively). On less refined distillates, these ratios declined rather more—an average of between 20 and 30 per cent in both groups of countries. Of special interest is the oil producing group of countries, where for both kerosine and diesel oil, the ratio of taxes to retail prices declined on average more than 40 per cent (reflecting the decline in the specific tax rates for these products). For heavy fuel oil, the decline was even greater (more than 80 per cent).

Table 2Average taxes on petroleum products estimated on an ad valorem basis1 and their percentage change(Per cent)
GasolineHouseholdAutomotiveHeavy fuel
RegularPremiumkerosinediesel oiloil
All countries
July 197354.348.922.443.210.2
July 197443.242.116.430.67.8
Percentage change−20.5−13.9−27.0−29.3−23.9
Developed countries
July 197361.158.019.656.28.4
July 197448.446.717.343.76.3
Percentage change−20.4−19.4−11.4−22.2−25.3
Developing countries
July 197348.942.124.731.812.0
July 197439.638.715.721.69.2
Percentage change−8.5−7.9−36.5−32.0−23.0
Of which:
Major oil producing countries
July 197336.131.116.833.5−4.32
July 197429.−7.72
Percentage change−19.3−7.4−45.1−40.3−80.6
Non-oil producing countries
July 197349.843.225.631.612.5
July 197440.339.616.222.29.0
Percentage change−19.1−8.5−36.6−29.7−28.1

Ratio of taxes to retail price.


Ratio of taxes to retail price.


These declines in the ratios of taxes to retail prices reflect the overwhelmingly specific nature of petroleum taxation in the countries surveyed, rather than ad valorem taxes, and show that despite an overall increase in the average specific tax rate, the retail price increased even more on average. Two implications can be drawn from this: first by maintaining specific taxes on petroleum products, potential revenue was, on the average, being lost, and second, product prices were lower on average than they would have been if the taxes had been ad valorem.

Diversity in price rise

During the second half of 1973 and the early part of 1974 crude prices increased approximately fourfold. The World Bank survey of changes in the prices and taxes of five petroleum products in 64 countries between July 1973 and July 1974 shows the diversity in the reactions of the retail prices of petroleum products to the increase in crude prices. On average, prices of gasoline, kerosine, and diesel oil rose between 44 and 54 per cent; the price of heavy fuel oil, however rose by over 90 per cent. While both developed and developing countries experienced this huge jump in the price of heavy fuel oil, prices of other products rose less substantially, on average, in the developed than in the developing group of countries. Despite these larger percentage increases in the developing countries, the prices of these products in this group in July 1974 still averaged well below those in the more developed countries.

The explanation for these diverse movements lies primarily in the tax response to the increase in crude prices. Taxation of petroleum products was found to be overwhelmingly specific in nature, rather than ad valorem. The former method has the advantage of administrative simplicity and, given some degree of rate differentiation according to product, is a reasonably equitable way of raising revenues. From an efficiency viewpoint, differentiation of rates by product is also necessary if the taxation is to reflect not only the costs of production but also the social costs of using the product.

While revenue raising and constraints on the demand for crude imports may have been the motivation for certain tax changes in non-oil producing countries, these motives were tempered by concern over the impact the increased petroleum product prices would have on the industrial and agricultural sectors, as well as on the lower-income groups. This is evident from the substantial decline in the ratios of taxes to retail prices for the less refined distillates; gasoline bore the brunt of the tax increases in most countries. This was particularly true in developing countries, where gasoline taxes increased on average by almost 40 per cent. In more developed countries, where taxes per gallon of petroleum products were already at a much higher level, the average tax increases were much less, though again gasoline taxes were increased the most. For all the countries surveyed, the tax changes were responsible for, on average, 25 per cent of the gasoline price rise, compared with a rise of 4 per cent for kerosine, 2 per cent for diesel oil, and 5 per cent for heavy fuel oil.

Other Resources Citing This Publication