Electrification of the World

The world’s 2 billion users of fuel-based lighting collectively emit substantial amounts of greenhouse gases, even compared to households served with luxurious electric lighting. Thanks to low lamp efficiencies, lighting expenditures rival those seen by affluent households who enjoy the vastly higher levels of quality, safety, and services provided by electric light.

Thomas Edison’s seemingly forward-looking statement that “we will make electricity so cheap that only the rich will burn candles” was true enough for the industrialized world. But it did not anticipate the plight of 2 billion people—more than the world’s population in Edison’s time—who 100 years later still have no access to electricity. According to the World Bank, 24% of the urban population and 67% of the rural population in developing countries are without electricity today.

Unlike heating or cooking, lighting is one of the energy end uses that is often associated exclusively with electricity. But the reality is different: In fact, about a third of the world’s population uses fuel-based lighting. The exact number of people who lack direct access to electric lighting is not known. Barozzi and Guidi put the value at 2.2 billion, Efforsat and Farcot at 2.13 billion, and World Bank at just under 2 billion.

The extent of rural electrification varies widely from country to country, e.g. about 90% of the population in Africa is not served by grid electricity, versus 20% in Mexico. Some countries (e.g. Burundi and Rwanda) have barely passed the 1% electrification threshold. While the levels of illumination provided by flame-based lamps are far lower than with modern electric lighting, the efficiency of fuel-based light production is also low. The result is a substantial amount of primary energy use with little service received in return.

In some instances, the rate of electrification is high and one could argue that fuel-based lighting energy use is a temporary problem. Yet, in Sub-Saharan Africa the rate of per-capita electrification has been only 25% of the birth rate over the past 20 years (i.e. 55 out of 220 million people). An estimate for Kenya projected rural growth of 65000 to 85000 households in 1996, of which only 4000 to 8000 would have electric grid connections. In Southeast Asia the net effect of new electrification and population growth was an increase of 250 million people without electricity during the two-decade period of 1970 to 1990. However, in China, the opposite effect was seen (electrification exceeding population growth). At the best, population growth counteracts many efforts at electrification. At worst, the tug-of-war between population growth and electrification may be resulting in an increase in the number of people without electric light. This was not the lighting future imagined by Edison.

The state of affairs concerning fuel-based lighting is worrisome. Oil import dependency is generally high in developing countries, and it drains valuable hard currency. By virtue of its inefficiency, fuel-based light is hard to work or read by, imposes a high cost on very poor households, and seriously compromises indoor air quality. Further complicating the picture, subsidized kerosene intended for domestic lighting sometimes finds its way into vehicles, which creates additional environmental consequences. Meanwhile, electrification (for the sake of lighting and other energy services) has its own problems, not the least of which is the extraordinary cost of electric transmission and distribution costs coupled with the high capital costs and low system efficiencies associated with providing centralized power generation in such conditions.

THE WORLD OF FUEL-BASED LIGHTING

There are a wide variety of fuel-based light sources, including candles, oil lamps, ordinary kerosene lamps, pressurized kerosene lamps, biogas lamps, and propane lamps. According to most studies, ordinary kerosene lamps are the most common type of fuel-based lighting in developing countries. Ironically, more efficient kerosene lamps tend to increase both light output and fuel consumption, whereas an efficient electric compact fluorescent lamp provides an eight-fold reduction in primary energy consumption compared to standard incandescent light sources.

According to a 1995 study by van der Plas and Floor, typical household kerosene lamp use is 3 to 4 hours per day, with weekly fuel consumption of about 1 liter. Typical light outputs are 10 to 15 lumens for locally made lamps and 40 to 50 lumens for store-bought models. Placed in perspective, the lower end of this range corresponds to about 1% of the light produced by a typical 100-watt incandescent lamp.

A study conducted by the joint UNDP/World Bank Energy Sector management Assistance Programme (ESMAP) found rural households spending as much as US$10 per month on lighting from candles, kerosene and drycell batteries. This operating cost is not dissimilar from that paid by industrialized households with two dozen bright light sources throughout their home.

Many suppliers of energy-efficient lighting equipment have not found the rural markets in developing countries worth exploring. However, the large amounts of money spent on lighting fuel indicates that there is a considerable potential for spending money on alternatives, for instance PV-based lighting solutions. This was verified in a field test by the World Bank (See IAEEL Newsletter 2/98).

HOW MUCH ENERGY?

We have found no prior estimate of the global lighting energy use associated with fuel-based lighting. A very approximate one is developed here, attempting to capture the uncertainties by considering a range of values for important factors that are not well known. We assume an unelectrified population of 2 billion and take the kerosene lamp as the reference light source, the penetration of lamps at one per six people to one per two people, the fuel consumption at 0.04 to 0.06 litres per hour, and the daily usage at three to four hours

We have estimated only the household contribution to fuel-based lighting, lacking sufficient basis for assumptions necessary to evaluate the service and industrial sectors. The pressurized kerosene lamps used in businesses have a much higher hourly fuel-use rate. The energy requirements for households who use fuel-based lighting as an alternate light source (e.g. during blackouts) have also not been estimated.

The main findings, including ranges of uncertainty, are:

Between 15 and 88 billion litres of kerosene (or 550 and 3300 petajoules, PJ, 1015 J) are consumed each year to provide residential fuel-based lighting in the developing world. For comparison, the total energy use (all sectors and fuels) in Austria is 1200 PJ, in Sweden 2200 PJ, and in the UK 10000 PJ). Our central estimate is 38 billion liters (1440 PJ).
The primary energy consumed for this fuel-based residential lighting is between 13% and 78% of that used to provide the approx. 400 TWh of electricity consumed for residential electric lighting globally. Our central estimate is 34%.
The cost of this energy ranges from $15 to $88 billion/year (assuming a kerosene price of $1/ liter), or $44 to $175 per household.
Fuel-based lighting results in between 37 and 223 million metric tonnes of carbon dioxide emissions to the atmosphere each year, or 18% to 110% of the emissions from residential electric lighting globally. Our central estimate is 98 million metric tonnes (48%).
The energy services provided are 1/80th of the level if the light sources were electric (same number of sources; more efficient sources).
The actual energy services in lumen hours provided are approximately 1/1000th that enjoyed by households in the industrialized world (more sources; more efficient sources).
Within the developing countries, national fuel-based lighting energy use can be on a par with that for electric lighting, and large even compared to total electricity used for all purposes. One study noted that kerosene accounted for nearly 60% of the total energy requirement for lighting in India’s household sector in 1986. This is generally consistent with our own findings. According to our “central” estimates, fuel-based lighting in Brazil consumes 40% as much energy as that required to produce the electricity used for lighting in the country.

TOWARDS BETTER LIGHTING SERVICES

Among the more startling implications of these findings is that users of fuel-based lighting in the developing world spend as much or more money on household lighting as do households in the industrialized world, but receive a vastly smaller level of services. On a percentage-of-income basis, households in developing countries spend many times more for lighting than their counterparts in the North.

Some argue that the problem of fuel-based lighting is not a priority given the environmental impacts and costs of other end uses, such as cooking. While that zero-sum analytical perspective is certainly debatable, few would dispute that improving the quality and quantity of light available to households in the developing world would yield dramatic social and health benefits.

Evan Mills

This article summarizes some of the preliminary findings from a study for the International Energy Agency on global lighting energy use.