Ed Dolan's Econ Blog

How Fuel Subsidies Around the World Burden the Rich and the Poor Alike, with Lessons for the US

I have posted frequently (most recently in a three-part series that starts here) on the topic of underpricing of energy in the United States, but we are not the only offender. Many countries around the world subsidize consumer energy prices in ways that bring them to levels even lower than what U.S. consumers pay. These policies burden the rich and the poor alike—rich countries like Saudi Arabia and poor ones like Egypt, and within each country, both rich and poor citizens.

Who subsidizes fuel prices and why?

Many countries around the world subsidize fuel prices. A recent IMF report divides the subsidies into two types. Pre-tax subsidies are those that reduce the cost of energy to consumers to a level below the world price, adjusted for transportation costs. Government budgets sometimes record such subsidies explicitly and sometimes hide them, for example as losses of state-owned enterprises, accumulation of payment arrears, or in other ways. Post-tax subsidies include all policies that hold the after-tax price of energy below the level consistent with efficient taxation. The IMF defines efficient taxation as a system that applies uniform rates of consumer taxes like VAT across all goods, and also includes compensatory taxes to reflect externalities of energy use. Except for a note at the end, this post is concerned primarily with pre-tax subsidies. Also, although the prices often extend to electricity and industrial energy, what follows will focus on consumer prices for petroleum products.

The following table, excerpted from Appendix Table 2 in the IMF study, shows pre-tax subsidies for a representative sample of countries. Subsidizers include wealthy countries like Saudi Arabia, poor ones like Yemen, and many in between. Oil exporters are especially prominent on the list, but there are importers there as well. Some, like Indonesia and Egypt, are former exporters turned importers. At the same time, some major exporters engage in little or no subsidization, Russia and the United Arab Emirates being notable examples.

Subsidies in oil exporters seem to be motivated primarily by “spread the wealth” considerations. For net importers, especially former exporters, subsidies often stem from a desire to protect people from oil price shocks. Since the long-term trend of world prices has been upward, even price controls and subsidies that originally aim only to smooth volatility easily evolve into permanent subsidies, as they have in Indonesia and Egypt.

How subsidies hurt the poor

Fuel subsidies both help and hurt consumers. The trouble is that poor consumers get a disproportionately small portion of the help and a disproportionately larger share of the hurt.

The help comes because subsidies make fuels more affordable. That not only reduces direct costs for cooking and lighting, but also indirectly holds other prices down, for example, by reducing transportation costs for food. For individual families, the price reductions can be most welcome. For example, a study by Arze del Granado and others, cited by the IMF study, found that an increase of $.25 per liter in the price of fuel would reduce the real purchasing power of a poor household by more than 5 percent.

Still, in the aggregate, poor households account for only a small part of total fuel use. As a result, on average, consumers in the richest 20 percent of the population get six times as much total benefit from fuel subsidies as do those in the poorest 20 percent. The specific amount varies by fuel. For example, vehicle ownership is low among poor households in poor countries, so they get little direct benefit from a reduction in gasoline prices. On the other hand, since poor households are less likely to be connected to the electric grid, they account for a larger share of kerosene consumption and get more benefit from subsidies of that fuel. The following chart from the IMF study provides estimates of the distribution of subsidy benefits for four important fuels.

The harm to the poor arises from the way that fuel subsidies drain government budgets of funds that could benefit the poor in a more targeted way. In Indonesia and Saudi Arabia, fuel subsidies amount to some 14 percent of the government budget. For Yemen, the figure is 20 percent, and for Egypt, it is an alarming 30 percent. The same funds now squandered on fuel subsidies could be used to bolster pubic spending on education, health, or other programs more efficiently targeted at the poor. Freeing up public funds for such needs should be one of the major objectives of energy price reform.

How fuel subsidies hurt the rich

Since upper-income consumers get the bulk of the benefit from policies that hold fuel prices below global market levels, we might at first think that they would be net beneficiaries of subsidy programs, but that is not necessarily true. There are several channels through which subsidies harm the rich as well as the poor.

Taxes are one channel. In some exporting countries, including Saudi Arabia, individual income taxes and VAT are insignificant. Elsewhere, including Egypt and Indonesia, wealthy citizens have to cover the budgetary costs of fuel subsidies through income taxes and consumption taxes. The progressivity of the tax system in such countries more than offsets the regressive nature of the fuel subsidies themselves. Where that is true, fuel subsidy policies constitute a net drain on the budgets of the rich.

Second, fuel subsidies have side effects that directly reduce the quality of life for people of all income levels. Traffic congestion and air pollution are two of the most striking. For example, The New York Times recently carried a horror story on traffic conditions in Jakarta, reported to be the largest city in the world without a rapid transit system. Lower-income commuters get to work aboard slow and crowded city busses or rickety private minibuses. Things are really no better for those able to afford cars, however. The nine-mile, 90-minute drive to work that one office manager reports is apparently typical. Because cheap gasoline encourages tens of thousands of extra cars and motorbikes to join the fray, traffic spills over everywhere, including into supposedly designated bus lanes. Higher fuel prices would free up funds that now go to subsidies so that they could be used for investment in modern transit facilities. At the same time, they would reduce the number of cars on the road. Instead, the main thing that fuel subsidies seem to buy is transportation hell for the rich and poor alike, not to mention intolerable levels of air pollution.

Slower economic growth is the third channel through which energy subsidies harm the rich as well as the poor. In countries with entrenched subsidies, businesses often express concern that higher fuel prices would undermine their competitiveness on world markets. In the short run, that may be true. However, several studies suggest that the opposite is the case in the long run. (See Box 4 of the IMF paper for citations.) One way that fuel subsidies inhibit growth is by tying up budget resources that could otherwise go to investment in human capital or infrastructure. Another way is by distorting the structure of trade. Rather than exporting goods in which it has a true comparative advantage, countries export goods that have high opportunity costs but have low out-of-pocket costs because of artificially low prices for energy inputs. When the fuel in question comes from domestic sources, it would be better to export it directly than use it in inefficient domestic industries. Subsidizing imported fuel in order to convert it into exports in which the country does not have a true comparative advantage is even more absurd.

How to end subsidies

Unfortunately, once they become established policy, fuel subsidies are hard to end. A major reason is that the benefits of fuel subsidies, unevenly spread though they are, are highly visible to those who receive them. The benefits of ending them—improvement in trade and budget balances allowing potential spending on worthy public projects—are neither immediate, nor plainly seen, nor certain. As a result, removal of subsidies draws impassioned protests from people on the street but more cerebral and limited backing from policy makers, academics, and development advisers.

The attempt by President Goodluck Jonathan to lift Nigeria’s gasoline subsidies in early 2012 is a case in point. Nigeria is a major exporter of crude oil, but because of corruption and mismanagement, it has little domestic refining capacity. As a result, it must import some 70 percent of its gasoline. As of the end of 2011, gasoline was selling for 65 naira per liter, or about $1.50 per gallon at the then-prevailing exchange rate. The withdrawal of subsidies roughly doubled the price to $3.00 per gallon.

Nigerians by the millions immediately set aside their many differences to protest the increases. They saw the subsidies as one of the few tangible benefits they receive from a corrupt and inefficient government. Christians and Muslims stood side by side to block highways, urged on by their respective religious leaders. Trade unions joined the protest movements and declared a general strike.

The government objected that it needed the money raised by price reforms to fund human development and infrastructure projects that would benefit all Nigerians, but few believed them. Protest leaders noted that the government could long since have been using the country’s vast oil export revenues to fund the construction of schools, highways, and power stations, not to mention efficient domestic refineries. Instead, the oil money had been stolen corrupt officials.

After just two weeks of protests, the government largely gave in. The price increase was cut in half, leaving gasoline costing about $2.25 per gallon, which the government argued still required a significant subsidy.

The Nigerian fiasco is by no means unique, but some countries, including Brazil, South Africa, and Turkey, have been more successful in phasing out their subsidies. Based on a comparison of successful and unsuccessful reforms, the IMF has identified several elements that improve the chances for success in removal of fuel subsidies. These include:

  • Better transparency, planning, communication, and consultation prior to implementation of price increases.
  • Replacing subsidies with appropriately targeted policies to aid the poor.
  • Measures to depoliticize future energy prices, for example, by establishing automatic formulas that link domestic prices to world prices.
  • Sound macroeconomic and monetary policies to prevent fuel price increases from feeding through into self-sustaining inflation.

Unfortunately, many of the countries with the largest subsidies are deeply lacking in the transparency, trust, and technical competence needed to follow the IMF recommendations.

Implications for the United States

As the first chart in this post showed, the United States provides only minor pre-tax subsidies for petroleum products and other fossil fuels, as measured by the IMF. At first glance, that might suggest that the issue of fuel subsidies has little relevance to U.S. policy, but for two reasons, that is not really the case.

The first reason is that fuel subsidies, wherever they are applied, affect global energy prices through their impact on demand. Sources cited in the IMF study estimate that ending pre-tax fuel subsidies in all non-OECD countries would reduce world oil prices by 8 percent and natural gas prices by 13 percent. Both the price reductions and the accompanying reductions in pollution would benefit the United States. For that reason, if for no other, the United States, through all economic and diplomatic channels available, should support efforts to eliminate fuel subsidies everywhere.

Second, although the United States does not have large pre-tax subsidies, it does have substantial post-tax subsidies. The IMF calculates that for petroleum products alone, U.S. post-tax subsidies come to 2.42 percent of GDP. That is much higher than any other advanced economy—higher, for example, than 0.35 percent in Japan, 0.09 percent in Sweden, and zero in the UK and Germany. These post-tax subsidies occur because the U.S. government, unlike those of other countries, taxes gasoline, natural gas, and coal at rates that are well below the levels needed to compensate for negative externalities.

For the benefit of climate-change skeptics, it is worth noting that the IMF study puts a relatively low weight on carbon-related externalities when calculating post-tax subsidies. The study assigns a value of just $25 per ton of CO2 to climate change externalities, a figure that falls in the low to middle range of estimates used in discussions of optimal carbon pricing. As the following chart shows, the bulk of negative externalities of petroleum use come instead from local pollution, road congestion, and increased traffic accidents.

If the federal government is ever to succeed in implementing appropriate corrective taxes, as I have often urged, it will need to heed the same guidelines that the IMF identifies for the removal of pre-tax subsidies in developing countries. Any such policy changes will require planning, transparency, and a sound communication strategy. They will also require appropriate policies to compensate low-income households, while maintaining incentives for energy compensation. In addition, it would also be worth considering policies that depoliticize energy pricing, such as the price-smoothing oil tax discussed in this earlier post.

The bottom line is that underpricing of energy burdens both the rich and the poor. As the IMF puts it,

Subsidies aggravate fiscal imbalances, crowd-out priority public spending, and depress private investment, including in the energy sector. Subsidies also distort resource allocation by encouraging excessive energy consumption, artificially promoting capital-intensive industries, reducing incentives for investment in renewable energy, and accelerating the depletion of natural resources. Most subsidy benefits are captured by higher-income households, reinforcing inequality.

These words apply as much to the United States as they do to Saudi Arabia, Egypt, or Indonesia.


13 Responses to “How Fuel Subsidies Around the World Burden the Rich and the Poor Alike, with Lessons for the US”

RikAugust 12th, 2013 at 4:25 am

An important issue you forget completely to take into consideration is cooking. Without subsidized fuel that will largely be done by wood. A country like India would not have a tree left if so. And half its population would have astma because of the smoke/polution. Next to the huge drain on financial resources herefor for the very poor.
You need to find a solution herefor before you can even think about abolishing fuel subsidies in the 3rd world.
This have been brought up already several times, better do your homework.

Ed Dolan EdDolanAugust 12th, 2013 at 9:41 am

Thank you for bringing this up. Yes, it would have been good to use providing poor households with environmentally sound cooking alternatives as an example of why removal of fuel subsidies needs to proceed in a planned, phased manner, as part of a comprehensive energy reform, and accompanied by targeted programs to help poor households cope with the consequences of price reform.

Ed Dolan EdDolanAugust 12th, 2013 at 12:00 pm

Good question.

If we allow for environmental benefits of reducing fossil fuel use, it is theoretically possible that subsidies for wind and solar, considered in isolation, could benefit both the rich and the poor. However, as I have frequently argued, even greater benefits would come from a policy that raises the price of fossil fuels to cover negative externalities (that is, in IMF terminology, that eliminates "post-tax subsidies").

If we consider the three policies jointly–(1) do nothing, (2) subsidize renewables, or (3) charge fossil fuels for externalities, I think we can show that (3) is superior for the economy as a whole. If accompanied by appropriate target policies to compensate low-income households for higher energy prices, (3) can provide greater benefits to both the rich and the poor than either (1) or (2)

Jardinero1August 12th, 2013 at 12:50 pm

Thanks. I really respect your reasoned approach to these issues. I hope that in a future post you will explore the components that make up the costs of negative CO2 externalities.

In this essay you cite a value of 25 dollars a ton of negative externalities. I struggle and struggle with what those negative externalities really are. I also struggle with whether that is a net value, i.e. the economic benefit of CO2 emissions less the economic cost of CO2. Or, is that simply the value of the negative effects with no consideration of the positive effects?

My interest in this has nothing to do with climate change skepticism, I resolutely believe that the climate is changing. No, my interest stems from skepticism that man made CO2 emissions impose any real economic costs on society.

Ed Dolan EdDolanAugust 12th, 2013 at 11:52 pm

The short answer is that the $25 is supposed to be net effects, i.e., coastal cities flooded less gains from cheaper trans-Arctic shipping.

The longer answer is that you are not alone in struggling with this issue. The correct value of externalities is subject to great uncertainty. The uncertainty has two major sources.

First, there is uncertainty inherent in the climate science. I am not a climate scientist, but I find it useful to follow the latest on "Climate, Etc." ( This blog, written by a real climate scientist, is sometimes characterized as a "denier" blog, but to my point of view, it gets that label simply from the fact that it openly and honestly discusses the state of knowledge within the science community. (Disclosure: I also like it because Judith is kind enough to reprint some of my posts on the economics of climate change.)

The second biggest source of uncertainty is the appropriate discount rate to apply to climate effects. For any given scientific finding of harm, the dollar amount is very high if you use near-zero discount rates, as in the Stern report, and very modest if you use market discount rates, say, the 10-year Treasury rate. It is a conundrum. I guess it is about time to blog about it.

ThomasGrennesAugust 14th, 2013 at 9:28 am

Subsidies to oil use in Indonesia have also transformed that country from a net exporter
of oil to a net importer.

ThomasGrennesAugust 15th, 2013 at 12:03 pm

Jardinero1's question raises an economic issue in addition to time and the discount rate. To translate units of temperature change into units of dollars worth of goods and services one must know how responsive producers are to climate change. For example, the loss from warming will be greater if farmers stubbornly plant the same crops in the same locations in spite of changes in temperature, rainfall, wind patterns, etc. There is evidence that farmers did respond to climate change in the past, and they are responding in the present. The corn belt is already moving north from Iowa to North Dakota to Manitoba, Canada. Damage can be mitigated by responding to climate change. Taking a longer view, humans have greatly mitigated the damage that would have occurred after the Great Ice Age by moving all around the world and producing very different goods and services.

ThomasGrennesAugust 15th, 2013 at 3:31 pm

A good complement to Ed Dolan's stimulating posting on this topic is the Aug 7th posting by John Cochrane in his blog, the Grumpy Economist: He takes a financial economics perspective on environmental issues, including dealing with greenhouse gases. He observes that in low income countries cleaner water and air are much more important issues than climate change, and he asks whether a carbon tax is the most beneficial environmental investment.

Bryan WillmanAugust 15th, 2013 at 11:33 pm

I think there are two issues with this line of argument, both of which are hinted at but not really made clear. They suggest the issue will not go away in the forseeable future.

1. In spite of what people say, redistribution of resources to "the poor" is not actually a goal. It's a pretend goal, a feel good goal, perhaps a public relations goal. But it's not the real goal. The real goals are about redistributing resources to the well connected, or the powerful by numbers. Hence, plans which have low distortion or other such merits are politically doomed. (This similar to the argument that the truly arcane US tax law is a *feature* from the perspective of politicians.)

Bryan WillmanAugust 15th, 2013 at 11:33 pm

part 2:
2. The 'externalities' argument is based on one judgement about the value of environmental quality. People are forever saying they care about that – until they're asked to actually pay for it. If we were to have a clear transparent direct referendum on taxes and 'externalities' in the US tomorrow, I wager there would overwhelming support for… the status quo. As in "don' t make the air any dirtier, but don't make my life any more difficult"

And so, once again, it's easy to see a pretty clear plan that would be better, that can never happen in the real world.