Typical street scene in Santa Ana, El Salvador. (Photo: iStock)

Typical street scene in Santa Ana, El Salvador. (Photo: iStock)

IMF Survey: Problems Follow When Officials Set Fuel Prices

January 15, 2007

  • Fuel subsidies may direct public spending away from more productive uses
  • Low fuel prices encourage inefficient use of energy
  • Well targeted social assistance must accompany subsidy removal

Developing country governments commonly resort to controlling the prices of petroleum products, at least in part to protect low-income residents. But how effective are these subsidies?

Traffic in Accra, Ghana—where the welfare effect of fuel subsidies amounts to 8.5 percent of total household consumption (photo: Kambou Sia/AFP)


A new IMF Working Paper—drawing on poverty and social impact analyses in Bolivia, Ghana, Jordan, Mali, and Sri Lanka—underscores how expensive and poorly targeted energy subsidies are. There are more effective ways, it argues, for governments to help shield the poor from high energy prices.

Developing countries intervene in energy markets in a variety of ways. In oil-importing countries, some governments directly control quantity, distribution, and prices. Others allow the private sector to import and distribute petroleum products freely, but set price ceilings and compensate private sector distributors to cover any losses. In oil-exporting countries, governments often set domestic prices below world levels, imposing an opportunity cost on its suppliers.

Although politically popular, subsidized fuel prices have significant downsides. For governments, the subsidies may direct public expenditures away from more productive uses, reduce revenues from domestic production, or contribute to unsustainable budget deficits. At the household level, low fuel prices also encourage inefficiency in the use of energy. And, as a means of shielding the poor from rising petroleum costs, universal energy subsidies are simply not cost-effective, because they inevitably entail a substantial leakage of benefits to higher-income groups.

Gauging the cost of subsidies

Because most countries are either net exporters or net importers of petroleum products, the appropriate reference price in measuring the extent of subsidy is the relevant border price—that is, the world price adjusted for trade and transport costs to the country's border. For an exporting country, the border f.o.b. (free on board) price minus trade and transport margins to the border represents the revenue forgone in consuming rather than exporting petroleum. For an importing country, the border c.i.f. (cost, insurance, freight) price plus trade and transport margins from the border represents the cost of domestic consumption. These reference prices are thus efficient because they maximize the sum of consumer and producer surpluses.

The difference between the actual consumer price and the reference price represents the unit subsidy (actual less than reference) or tax (actual greater than reference) for that product. It is not uncommon for some products to be subsidized and for others to be taxed. Multiplying this difference by annual product consumption and summing across products gives the total cost of the fuel subsidy. Comparing the actual to the reference price also provides the basis for identifying how much actual prices need to increase to eliminate the subsidy. That said, of course, not all subsidies are recorded in the budget. Some cost can be passed on, at least initially, to public enterprises or, through regulations, to private sector market participants.

To gauge the magnitude of these consumer subsidies, the paper compared actual consumer prices with reference prices that capture the true opportunity cost of domestic consumption (see box). In the five countries analyzed, the aggregate budget subsidy was 2.0-3.2 percent of GDP in 2004 (go to table on page 15 of IMF Survey in pdf format). Extrabudgetary subsidy costs were also incurred. For instance, in Bolivia, the private sector shouldered production and distribution costs of more than 1½ percent of GDP. Based on the prices at the time each country was analyzed (between October 2004 and September 2005), subsidized fuel products would need to rise by one-third to two-thirds to reach world market levels.

Who benefits?

Fuel subsidies affect households in two ways. First, they have a direct effect on the prices of the petroleum products that households consume. Second, they have an indirect effect on the prices of other goods and services that use petroleum products as inputs consumed by households.

For the five countries, the total welfare effect—that is, the sum of the direct and indirect effects—ranged from 1.7 percent of total household consumption in Mali to 8.5 percent of total consumption in Ghana. The subsidies were progressive in that they represented a (slightly) higher proportion of total consumption for poorer households, but they were badly targeted. Richer households received a disproportionate share of the benefits, with the bottom 40 percent receiving between 15 percent (Bolivia) and 25 percent (Sri Lanka). At this level of leakage, for every unit of resources transferred to the poorest households, three to five or more units are transferred to better-off households.

Better options

To protect the poor and gain political support, the removal of subsidies must be accompanied by well-targeted social assistance measures. Budgetary savings from reduced fuel subsidies should be directed to higher priorities, such as increasing access to or improving the quality of education and health care services, improving physical infrastructure, or reducing taxes.

Ideally, an existing, well-designed social protection program—such as Oportunidades in Mexico or Bolsa Familia in Brazil—can be used to safeguard the real incomes of the poorest households. Such a system helps generate efficiency gains by directly addressing any possible adverse affects on low-income households. In the event of price increases—for instance, for petroleum products—the desired transfer to low-income households can be maintained in real terms by inflation-indexing the transfer.

A government's ability to protect the poor over the short term is curtailed if such a system is absent or ineffective. However, a thoughtful plan for eliminating subsidies along with ad hoc mitigation measures can reduce the adverse effects on the poor. In this regard, the paper notes the positive steps that several countries have taken. Ghana eliminated fuel subsidies in February 2005 and used some of the fiscal savings to eliminate fees for primary and junior secondary school, increase funding for primary health care in the poorest areas, and expand investments in urban mass transport.

Jordan is phasing out subsidies over four years while strengthening its primary social assistance program. Ad hoc measures—including a small increase in the minimum wage—are also shielding the poor during the phase-out of subsidies. Sri Lanka is working on better targeting its primary safety net program. Bolivia and Mali, on the other hand, have been slower to take aggressive action to reduce subsidies and more effectively target social assistance.