Cover Image
close this bookBoiling Point No. 38 : Household Energy in High Cold Regions (ITDG - ITDG, 1997, 40 p.)
close this folderGTZ News: Non-theme articles
View the document(introduction...)
View the documentImproved institutional stoves for Sudan schools
View the documentHaiti: Cooking stoves and domestic energy
View the documentHousehold energy in a recently electrified rural settlement in Mpumalanga, South Africa
View the documentImproved Tunisian domestic bread ovens: Flying saucer lids save 50 per cent fuelwood
View the documentMumu: A traditional method of slow cooking in Papua New Guinea
View the documentReducing the risks of poisonous emissions from stoves
View the documentUpdate on biogas in Nepal
View the documentResearch and development

Household energy in a recently electrified rural settlement in Mpumalanga, South Africa

Bernard T Luvhimbi and Harald H Jawurek, School of Mechanical Engineering, University of the Witwatersrand, Johannesburg, WITS, 2050 South Africa.

Wood is still the predominant source of energy for cooking and water heating, but much of it is now purchased, rather than gathered.

South Africa is in the process of rapid electrification. The number of houses newly connected to the grid was approximately 623,000 for the period 1991 to 1993, 436,000 for 1994 and 478,00() for 1995. It is estimated that this programme will increase household access to electricity for the country from 35 per cent in 1990 to 80 per cent in 2010. The households being electrified are mainly those of low income areas. These include - in descending order of income and present access to electricity - the traditionally black, formal townships attached to cities and towns, informal urban communities (shack settlements) and rural settlements.

The effect of electrification on household energy consumption has been studied for several urban and peri-urban communities in South Africa. Very little comparable post electrification information on rural, mainly wood-buning settlements appears to be available. This study looked at such a settlement.

During August and September 1994, an energy consumption survey in a traditionally wood-burning, recently electrified, remote settlement was carried out. The settlement that was studied was Green Valley (24°36'S, 31°01'E), adjacent to Acomhoek in Mpumalanga (previously Eastern Transvaal). Green Valley is an administrative unit of a low income, densely populated, urban like sprawl stretching for kilometers and set in a remote, semi-arid, rural region. Land area per household is of the order of 1000 m2; there is thus no question of subsistence agriculture, though gardening for food is practised, despite frequent water supply difficulties.

The electrification of Green Valley was started in 1990 and essentially completed in 1993. Very few houses were 'wired' in the conventional sense. The majority were fitted with a 'Ready Board', a simplified distribution board into which lights and appliances are plugged directly. The boards are operated by means of a magnetic card with which the customer repurchases electricity at a central pay-point in the settlement.

Data collection

Data was obtained mainly by means of structured interviews based on a questionnaire. A total of 80 randomly selected households was covered. Interviews were conducted in Pedi and Tsonga, the two local languages. Additionally, numerous informal interviews were held. The method involved determining how many households used each type of energy source. For a sample of this size, measurements to determine the exact quantities of each fuel consumed are problematic as they are both intrusive for the households involved and excessively time-consuming for researchers.


Figure 1 shows the percentage of households using various energy sources. All sample households in Green Valley use electricity and 84 per cent of households use wood. There is a sharp reduction in the use of paraffin when settlements are supplied with mains electricity. In the nearby non-electrified settlement of Cottondale paraffin is used for lighting (96 per cent of households), cooking (53 per cent) and refrigeration (7 per cent); in Green Valley it is used for cooking only - the paraffin lamp has passed completely out of use.

The use of dry cell batteries is considerably lower in Green Valley than in Cottondale. This is largely due to the reduced use of batteries in radios. The reduction would have been larger still had not many radios been of the battery-only type. In Green Valley (where all households use electric lighting) candles serve as backups in the case of power failures, or when the pre-paid card is out of credit. Coal, LPG and dung are not used by the Green Valley sample households; coal is not available and LPG is little known. Some dung possibly may have been used, but this was not admitted.

Table 1 shows the percentage of households which use grid electricity for specific functions The main Uses are: lighting, radio for news and entertainment, ironing, cooking and water heating (for washing and hot beverages). The percentage of households in possession of appliances for the last two activities is as follows: full electric stove (5 per cent); double electric hotplate (34 per cent); electric kettle (33 per cent).

Table 1 Percentage of Green Valley households using grid electricity for particular activities


Percentage of households





Radio/music system






Heating water (kettle)




Deep freezing


Cooling house (fan)


Heating house (heater)


Figure 1. Percentage of household using particular energy sources

Of the households using electricity for lighting only, 52 per cent stated that they had too little money to buy other appliances, 30 per cent that they had too little money to run other appliances, or that they had appliances (radios/music systems) but not the AC/DC adapters that permit operation off the mains. Only 8 per cent prefer the use of a non-electrical energy source - wood for cooking.

Of the 84 per cent of households using wood for cooking and water heating, 34 per cent use wood exclusively, 31 per cent use wood supplemented by paraffin, 10 per cent use wood and electricity, and 9 per cent use all three. Wood thus remains a major source of energy.

Table 2 shows how fuelwood is obtained. Purchased wood is obtained mainly from veld clearing operations for agriculture and from vegetation thinning in game reserves suffering from bush encroachment. All purchased wood was veld wood in this study; exotic plantation woods are, however, known to be used at times.

The transition from gathered to purchased wood has also been observed in non-electrified settlements in the area; it reflects the increasing scarcity of free fuelwood from the veld. Table 2 also shows a significant increase in the number of households that do not use wood at all. This is most likely due to the combined effects of wood scarcity and electrification.

Table 2 Sources of wood

Source of wood

Percentage of households

Green Valley 1994

Cottondale 1990

Gathered in veld






Free, collected by hired van



Households not Using wood




In a recently electrified rural settlement in Mpumalanga, electricity is extensively used for lighting and media applications, but less so for cooking and water heating. For the latter, energy intensive activities, wood remains the major fuel.

Comparisons can be made with results from earlier studies in urban and peri-urban areas where it was found that for cooking, water heating and space heating (high energy consumption activities), the 'old', pre-electrification fuels, predominantly coal (for settlements near the coal fields), paraffin (kerosene), and to a lesser degree LPG, remained in extensive use.

Rural households using wood for cooking were found (with a single exception) to use the traditional open fire built on the ground. There have been several attempts locally, and numerous programmes elsewhere, aimed at the development of low cost, wood-burning stoves that are more efficient and that pollute less than open fires. Woodstove programmes thus remain relevant in the face of rural electrification. With the increasing scarcity of traditional veld wood there is a major energy transition from gathered to purchased wood.