Columnists: Evan Mills, US Lawrence Berkeley National Laboratory

Published on: 4 Jul 2017

Beware kerosene subsidies: An obstacle to energy efficiency and development

The International Monetary Fund estimates that global energy subsidies topped $5 trillion in 2015, or a whopping 6.5% of GDP.   This represents an immense amount of money trimmed off the true price of energy and financed, instead, in a cloaked way through other parts of the economy.   We all still pay for this energy, but through general taxes that governments must collect in order to finance their support to energy providers, and through externalities such as health costs and roadway congestion caused by energy use.

It is widely held that energy subsidies impede the efficient functioning of markets.  In particular, this takes the form of boosting energy use and creating artificial bias against efforts to use energy more wisely.  This is blatantly at odds with policies to raise energy efficiency and reduce the cost of energy services and associated externalities, most notably the growing impacts of climate change.

The outcry against subsidies is spearheaded by such respected institutions as the IMF, World Bank, and the G-20, not to mention scores of NGOs.  The focus is almost exclusively on mainstream energy sources like gasoline and electricity.  It never sat well with me that energy sources even more important to some populations, like kerosene in the developing world, are largely overlooked in the discussion.  Hence the just released article in the journal World Development (online, in-press here or download here).

Kerosene prices vary widely by country, ranging from just a few pennies per litre in oil-producing countries such as Iran and Venezuela, to around $2.00/litre in countries like Israel and Italy.  Subsidies are largely limited to developing and emerging economies.  Nominal prices from the IMF database.

The key findings are that kerosene is used in 173 countries, at a cost to consumers of $43.4 billion/year as of 2013, or $60.3 billion/y including direct economic subsidies, and $77.2 billion/y including certain externalities. Despite low world oil prices, direct economic subsidies for kerosene are $18.4 billion, and $34.7 billion including environmental externalities. These values correspond to 72% and 56% of total kerosene costs being passed through to consumers, respectively. When excluding advanced economies, the pass-through values fall to 40% and 35% -- i.e., the price paid is well less than half the true cost. Approximately 52% of the global kerosene supply receives direct subsidy, or 63% when externality costs are considered.  It is worth noting that the IMF’s estimates don’t include many of the specific hazards of kerosene, including indoor air pollution and house fires.  Depending on relative prices, subsidized kerosene is often mixed with motor fuels, creating additional air pollution and damage to engines.

An innovation in the analysis was to quantify subsidies by energy end use.  Traditional top-down economists don’t tend to look at expenditures in this fashion.  I found that the cooking end use receives $2.0B/y in direct kerosene subsidies, lighting $7.1B/y, and heating and other residual uses $9.3B/y, or $76/y averaged over all subsidized households.  Defining subsidies at this level of granularity helps pinpoint policy issues and opportunities.

Kerosene subsidies are concentrated in the developing world.  They are greatest for heating in the Middle East, and for lighting in Sub-Saharan Africa and Developing Asia.

Some surprises emerged from the results.  These include that large amounts of kerosene (about one-half of the world’s total) are used in industrialized countries.  Japan alone consumes one-third of the world’s total, predominantly for space heating, particularly in rural areas where electricity is the only alternative source of energy.  There are also important cultural drivers, including the custom of conditioning rooms individually rather than relying on centralized systems. Fortunately Japan taxes kerosene above the market price, although far less so than other fossil fuels, and their price still remains about 30% below that reflecting externalities.

Unvented kerosene heaters for sale in Japan. Source: Japan Times.

Promoting a transition to energy efficient off-grid energy services is one of the most cost-effective ways of reducing dependency on subsidies.  This can be seen clearly in the case of solar-LED lanterns, which could be paid for by the avoided kerosene subsidy in just a few months to a few years, depending on type of lantern and level of subsidy.

Annual per-lantern kerosene subsidy by lantern type and direct subsidy.

However, the very presence of subsidies undercuts this process by diluting market price signals and rendering energy efficiency investments less cost-effective, while competing with other social and development-focused budgetary needs. Kerosene subsidies are additionally counterproductive because the emerging technologies they impede (e.g. improved lighting and cook stoves) also improve productivity, safety, and quality of life.

Forty-five countries—mostly in the developing world—have used taxes of various sorts to price kerosene such that there are no direct subsidies.  For twenty-two of these, the taxes are high enough to reflect environmental externalities.  This suggests that subsidy reform is economically and politically feasible.


Evan Mills is a Senior Scientist at Lawrence Berkeley National Laboratory, and leads the Lumina Project, which cultivates technologies for low-carbon off-grid lighting in the developing world.

The views expressed in this column are those of the columnist and do not necessarily reflect the views of eceee or any of its members.

Other columns by Evan Mills