- Email: [email protected]
Use of grid electricity by rural households in South Africa
Articles
Use of grid electricity by rural households in South Africa Cecile Thom Energy & Development Research Centre, PO Box 34178, Rhodes Gift 7707, South Africa
Almost half of rural households in South Africa had been connected to the grid by the end of 1999. This paper describes the utilisation of electricity by rural households in South Africa with reference to quantitative as well as qualitative information on energy use by electrified low-income households. The main findings are the following. Significant percentages of electrified households in rural areas own electrical appliances such as hotplates, kettles, irons, refrigerators, televisions and radios/hi-fis. Not all households that own electrical appliances use them on a regular basis. The reasons for acquiring appliances are of a complex nature -- considerations such as usefulness and cost are only some of the factors that influence this, while the symbolic value of appliances is also of importance. While electric lighting is commonly used by electrified households, a significant percentage of electrified rural households continue to use other fuels, particularly candles, for lighting purposes. A significant percentage of electrified households also continue to use battery-operated radios. Furthermore, a significant percentage of rural households use electricity for cooking, although only higher income households are likely to use electricity as the sole fuel for cooking. The service of cooking with electricity is highly valued by many rural households, even if cooking is not done exclusively with electricity. Most electrified households in rural areas use three fuels in combination to meet their energy needs. It seems that electricity does not replace other fuels, but adds to the fuels used by rural households. Many low-income households in South Africa continue to use paraffin after electrification, particularly for cooking and water heating purposes. One of the factors that contributes to this is the view that paraffin is cheaper to use than electricity for these thermal purposes. Nevertheless, a range of factors contributes to the persistent use of paraffin. 1. Introduction This paper aims to present some of the available information on the use of electricity by ‘‘rural’’[1] households in South Africa. Much of the work reported on in this paper was done as part of the research project The Role of Electricity in the Integrated Provision of Energy to Rural Areas (REIPERA) that was undertaken from 1995 to 1998 at the Energy and Development Research Centre (EDRC)[2]. The main findings of the project have been summarised by Thom [1998]. Some of them, particularly pertaining to household electrification, are presented here in greater detail together with some additional information which has since become available. Although the focus of this paper is household electrification, it needs to be mentioned that the REIPERA project recommended that the priorities of the electrification programme in South Africa -- mainly households, as well as clinics and schools -- should be expanded to include community water supply schemes and small-scale agricultural projects in particular, while small, medium and micro-enterprises (SMMEs) also need greater attention.
households have access to grid electricity. About two-thirds of households in South Africa (66%), and almost half (46%) of rural households, had been connected to the grid by the end of 1999 [Kotzé, 2000]. This can be compared with an estimated 44% of households, and 12% of rural households, which had been electrified at the end of 1994 [Thom et al., 1995]. In the six years from 1994 to 1999 between 400,000 and 500,000 new connections were made annually as part of the accelerated national electrification programme, in accordance with the Reconstruction and Development Programme (RDP) targets agreed upon between the government and the electricity industry (see Table 1). The programme was implemented by the national electricity utility Eskom as well as by municipalities. Eskom has been responsible for most of the connections in rural areas. Electrification projects in these areas typically involve the electrification of an entire settlement, or at least a substantial part of it. All households in a settlement who pay the connection fee (75 rands (R) in 1998, increased to R150 in 2000, for the standard 20A supply) and can be electrified at a reasonable cost are provided with an electricity supply, including a pre-payment meter and a ready board (see below). This is sometimes referred to as a ‘‘blanket’’ electrification approach. In spite of this approach substantial numbers of
2. Access to grid electricity The first matter to consider when discussing the use of electricity by rural households is to what extent these 36
Energy for Sustainable Development
l
Volume IV No. 4
l
December 2000
Articles
Table 1. The national electrification programme 1994-1999 (number of households) 1994
1995
1996
1997
1998
1999
Eskom
254,383
313,179
307,047
274,345
280,977
293,006
Local government
164,535
150,454
137,534
213,768
136,074
144,043
16,838
15,134
9,414
11,198
10,375
6,241
Total
435,756
478,767
453,995
499,311
427,426
443,290
RDP targets
350,000
400,000
450,000
450,000
450,000
450,000
Farm workers
Source: Kotzé, 2000
Table 2. Electrical appliance ownership for different income groups (all areas) % of households with monthly income of: R1-499
R500-799
16
25
39
67
98
Geyser
1
3
9
43
93
Heater
3
7
14
33
70
Stove
3
10
19
54
94
Hot plate
6
9
14
14
10
Kettle
5
12
22
55
94
Iron
6
14
26
61
96
Fridge
6
13
25
61
96
Television
8
17
32
63
96
Radio/hi-fi
9
16
28
58
92
Access to grid electricity
R800-1399
R1400-3999
R4000+
Electrical appliances
Source: Eskom, 1996
Table 3. Appliance ownership among electrified households in rural areas % of households with per capita[1] monthly income of:
% of all households
R130-270 [2]
< R130 Geyser
> R270
5
0
2
8
Stove/hot plate
37
10
27
53
Kettle
32
12
25
45
Fridge
43
12
28
65
Television
47
19
33
67
Radio/hi-fi
87
79
85
91
Source: Davis and Ward, 1995 Notes 1. Average household size was about 5.5. 2. Estimated minimum monthly expenditure to satisfy basic needs was about R270 per capita.
households in areas connected to the grid do not have access to electricity. In 1998 one of Eskom’s regional electrification managers indicated that only 65-85% of households in areas connected to the grid generally paid the connection fee [Kinners, 1998]. In an attempt to address this problem Eskom has been piloting an approach since the beginning of 2000 whereby households that are unable to pay a connection fee are provided with a 2.5A supply [Sokopo, 2000]. This enables them to use electricEnergy for Sustainable Development
ity for lighting, radios, television and refrigeration. The tariff is the same as the 20A tariff, comprising a simple unit energy charge with no fixed monthly charges. Users can upgrade to a higher level of supply by paying the required connection fee. 3. Appliance ownership Table 2 provides an overview of electrical appliance ownership in South Africa. It is presented in relation to l
Volume IV No. 4
l
December 2000
37
Articles
Table 4. Ownership and use of electrical appliances by households in Loskop electrified for 2 years or more[1] Service
Households using it frequently (%)
Households using it occasionally (%)
Households using it (total) (%)
Households owning appliance (%)
Lighting
94
0
94
-
Radio/hi-fi
77
0
77
84[2]
Refrigeration
48
0
48
48
Television
42
3
45
52[2]
61
10
71
65 (electric iron)
32
16
48
45 (electric kettle)
Cooking (hot plate/stove)
29
13
42
3 (stove) 45 (hot plate)
Space heating
3
3
6
16 (heater)
Ironing (iron/hot plate)[3] Water heating (kettle/hot plate)
[3]
Source: Compiled from Hansmann et al., 1996, A86. Notes 1. A sample of 30 was used, chosen randomly as far as possible within practical constraints. 2. These figures may include some radios and televisions operated with batteries -- for example, 3% of the sample (one household) indicated that they frequently used car batteries to operate a television. 3. The percentage of households using electricity for ironing and water heating probably includes households using a flat-plate metal iron and simple stove-top kettle respectively, heated on a hot plate.
household income to illustrate some of the trends, particularly since large income disparities exist in South Africa which correspond to disparities in access to services. At first glance the table gives the impression that the ownership of electrical appliances is highly dependent on household income levels. However, as shown in the first row, access to grid electricity itself was highly dependent on income at the time the study was conducted (1995). Generally, ownership of electrical appliances increases in relation to access to electricity as income increases, confirming that there is a relationship between household income and appliance ownership among electrified households, with hot plates being the main exception. The data suggests that more than one-third of the electrified households in the lowest income group own hot plates, irons and refrigerators, while more than half own radios/hi-fis and televisions which operate on grid electricity. The table shows that electric stoves with ovens are more common than electric hot plates in all but the lowest income group. Some recent energy studies in specific localities have indicated, however, that hot plates are far more common among lower-income electrified households than stoves [James and Ntutela, 1997; Mehlwana and Qase, 1999; Palmer, 1999; White, 2000]. 3.1. Appliance ownership in rural areas Data on the ownership of appliances in rural areas is presented in Table 3, which has been drawn from the analysis done by Davis and Ward [1995] on data collected in the Project for Statistics on Living Standard and Development (PSLSD). Davis and Ward [1995] observe that the ownership of electrical appliances appears to be closely related to income -- with the exception of radios, electrical appliance penetration in the lowest income group is between 10% and 20%. 3.2. Limitations of an income-based analysis Although there is undoubtedly a relationship between 38
Energy for Sustainable Development
appliance ownership and income, it nevertheless seems that income is not the main determinant of electrical appliance purchases. Recent anthropological studies in metropolitan areas have shown that ‘‘the interplay of multiple (social and economic) factors determines appliance acquisition and use’’ [Mehlwana, 1999]. These are further discussed below. 3.3. Electrical appliance ownership versus use Some information on electrical appliance ownership and use was obtained in a post-electrification study in Loskop (KwaZulu-Natal), conducted as part of the REIPERA project [Hansmann et al., 1996]. A small quantitative survey was undertaken in an area where households had been electrified for two years or more. The information pertaining to electrical appliances is presented in Table 4. It is particularly interesting to compare the percentage of households that use appliances frequently with the percentage that own the appliances concerned. A significant percentage of households that own an electric stove or hot plate do not use these frequently for cooking purposes. The difference is even more striking in the case of electric heaters. Some of the reasons for this are discussed under ‘‘Multiple fuel use’’ (Section 5) below. Research in metropolitan areas has confirmed that the fact that households own appliances does not necessarily mean that they use them [Mehlwana, 1999]. In fact, anthropological studies have indicated that some of the factors that contribute to the purchasing of appliances are not related to use at all. According to White [2000], utilitarian considerations such as usefulness, cost, and the availability of space are only one set of factors that influence decisions to buy, use or keep appliances. Coupled with this are considerations of symbolic value. Examples of such considerations are the wish to conceal poverty, and to ‘‘embody’’ a particular lifestyle: The fashioning of the ‘‘public’’ spaces within the home l
Volume IV No. 4
l
December 2000
Articles
is one of the primary mechanisms used to ward off the suspicion that one is ‘‘struggling’’. In an attempt to create an image of style and well-being for outsiders, any electric appliances that can add to that image are prominently displayed. ... In some instances, appliances have symbolic value only, as evidenced by the fact that broken appliances are kept on prominent display in many homes [White, 2000]. Electric appliances are infused with positive symbolic associations that are quite absent from most non-electric appliances. Electricity is a marker of modernity and progress. ... In East London, the drive to accumulate appliances for the purposes of marking style and social position is particularly prevalent among single women with good permanent jobs. ... Electrical appliances ... serve to distance them from the way of life of paraffin, the fuel associated with dependent and subordinated women, and enhance their image as independent and self-sufficient [White, 2000].
and/or bear the cost of extending the supply to other rooms, and are not provided any assistance with this. Extension cords are often used to ‘‘wire’’ houses informally. A survey conducted at Tambo found that about 76% of households had electric lighting in one room only, while about 14% had lights in two rooms, and only about 10% had lights in more than two rooms [James and Ntutela, 1997, p. 34]. This apparently resulted from the cost of materials for informal wiring, as well as the cost of installation and the availability of people with the necessary skills. In very poor households at both Tambo and Mafefe no wiring of additional rooms had occurred [James, 1997]. The placement of the ready board in the home clearly becomes an important consideration under these circumstances. In some cases electrification contractors consult households about the placement of ready boards, but not always. Since the research in Tambo and Mafefe was conducted only about six months after electrification, the findings need to be compared with studies done a longer time after electrification. In a longitudinal study on energy use in Kameelrivier B -- a peri-urban area in Mpumalanga province -- Palmer [1999] found that the situation changed significantly over a period of two years after electrification. Immediately after electrification the most common fuels for lighting were a combination of electricity and other fuels, while almost 40% of households were using no electric lighting at all. After two years the situation had changed completely, with 79% of households relying solely on electricity, and 21% using it in conjunction with other fuels. The main reason for continuing to use other fuels, and candles in particular, seemed to be that not all rooms had been provided with electric lighting, thus confirming the findings in the Tambo and Mafefe studies. As shown in Table 4, about 6% of households in Loskop (KwaZulu-Natal) that had been electrified for two years or more showed that they did not use electric lighting at all. The majority of the sample (94%) all used electric lighting frequently. More information on households’ use of electric lighting emerged from qualitative interviews undertaken in Loskop. The qualitative sample was smaller and differed substantially from the quantitative sample[3]. Electric lights were generally available in all rooms of the main dwellings (as opposed to additional ‘‘outbuildings’’ that form part of the homestead) of the households included in the qualitative sample. Most of these households also had an outside electric light. Candles, and occasionally paraffin lamps, were used in outbuildings and when power failed [Annecke, 1996]. The wiring of the houses was in most cases done by family members and sometimes by neighbours, and was therefore presumably not particularly costly to households. Some other factors also influence whether electric lighting is used -- for example, cases have been found where people prefer to use other forms of lighting because they fear electricity, or because the electric light switch cannot be reached once they are in bed [James, 1997]. 4.2. Radios A large percentage of rural households own radios. For
4. Use of electricity for particular services Specific attention is given here to the use of electricity for lighting and cooking, and to operate radios. The aim is to illustrate that the utilisation of electricity for these purposes is much more complex than the general perception that rural households use electricity for lighting and to operate radios, but not for cooking. 4.1. Lighting Electric lighting is commonly used by electrified households. However, research done in the REIPERA project indicates that a significant percentage of electrified households continue to use other fuels -- particularly candles, and to a lesser extent paraffin lamps -- for lighting purposes. Davis and Ward [1995] found that as much as 60% of the electrified sample in rural areas use candles and paraffin for lighting in addition to electricity, while about 20% use electricity only, and about 18% use candles and/or paraffin only (no electric lighting). The fact that not all households connected to the electricity grid are able to use electricity, as discussed under ‘‘Access to grid electricity’’ (Section 2) above, probably contributes significantly to this situation. Other factors nevertheless also have an impact on the use of electric lighting, as indicated by some qualitative post-electrification studies conducted during the REIPERA project. At both Tambo (Eastern Cape province) and Mafefe (Northern Province) where the 2.5A supply was originally piloted, candles and paraffin were still being used substantially for lighting about six months after electrification [Wentzel et al., 1997; James and Ntutela 1997]. The main reason for this was that electric lighting was not available in all the rooms in houses [James, 1997]. Eskom generally provides households with ready boards comprising a light and a row of sockets, which allows households immediate access to electricity without the expense of formal house wiring, while reducing the reticulation cost to the utility. The disadvantage of this approach is that the light and sockets are located in a specific room. Households have to get access to the skills Energy for Sustainable Development
l
Volume IV No. 4
l
December 2000
39
Articles
Table 5. Use of electricity by different income groups % of households with monthly income of: R1-499
R500-799
R800-1399
R1400-3999
R4000+
Access to grid electricity
16
25
39
67
98
Use of electricity as sole energy source
3
12
21
51
82
Use of electricity in combination with other fuels
12
13
17
16
16
Use of electricity for cooking
8
18
29
63
96
Use of electricity for lighting
15
25
37
67
98
Source: Eskom, 1996, Table E8:1
example, Davis and Ward [1995] found that 74% of unelectrified households and 87% of electrified households in the PSLSD sample owned radios. In the absence of grid electricity radios are mostly powered with dry-cell batteries. It seems that unelectrified households generally buy radios that can be operated on batteries only, without the option to use grid electricity. At both Tambo and Mafefe none of the households interviewed in the qualitative research were able to power their small radios with electricity [James, 1997]. For example, in Tambo eight households included in the qualitative sample (a total of seventeen households) owned radios in working condition, none of which could be operated with grid electricity [James and Ntutela, 1997, p. 35]. Only hi-fis could be operated from the grid. Findings in metropolitan areas in the Western Cape have been similar, with significant percentages of electrified low-income households using dry-cell batteries to power radios: The use of dry-cell batteries in these electrified households occurs because ownership of battery-operated radios dates back to the pre-electrification period [Mehlwana and Qase, 1999, p. 55]. In Loskop the majority of households in the quantitative sample (77%), which had all been electrified for more than two years, were using radios and/or hi-fis operated on grid electricity. Nevertheless, a significant percentage of households (32%) still used battery-operated radios, with 19% using them frequently, and the rest using them only sometimes (see Table 4). About 10% of households indicated that they did not own a radio or hi-fi. Clearly there were people who used both batteries and electricity to operate their radios -- one reason is probably that batteries are needed if radios are used a distance away from available sockets (outside of houses, for example). Market research conducted by Eskom has confirmed these findings. They found that 25% of grid-electrified households (in all areas, including rural areas) continue to use battery-operated radios [Eskom, 1996]. The cost of dry-cell batteries is extremely high in relation to the energy delivered, and some studies have found that dry-cell batteries form a significant component of households’ energy expenditure (up to 30-40% in some unelectrified villages) [Griffin et al., 1992]. Furthermore, James and Ntutela [1997, p. 40] found that the energy expenditure of households in Tambo that got a 2.5A elec40
Energy for Sustainable Development
tricity supply (for which they had to pay a fixed charge of R15 per month)[4] had increased significantly in cases where electricity had not displaced other fuels for lighting, or batteries for radio and television. Dry-cell batteries for radios seemed to have the greatest impact on household budgets under these conditions, because of their expense. Electrification had least impact on budgets of 2.5A households without radios, which were therefore not spending at least R8.20 per month on batteries [James and Ntutela, 1997, p. 39]. Research among farm workers in the Free State found that in addition to not having mechanisms to run their radios on electricity, farm workers did not realise that they might be able to power their radios with electricity, nor did they have the money to buy new radios which could run on electricity [Hofmeyr, 1996, cited in James, 1997]. The expenditure on batteries was found to be high, especially when considered as a proportion of wages. 4.3. Cooking Qualitative studies done during the REIPERA project indicate that cooking with electricity is important to many rural households, even if cooking is not done exclusively with electricity. The study in Tambo indicated that many people value the use of electricity for cooking. For example, some women in Tambo showed a lot of determination in saving money from their earnings in order to purchase cooking appliances. Furthermore, 75% of the households in the quantitative sample in Tambo that chose a 20A supply did this because they wanted to be able to use electricity for services not available with a 2.5A supply, such as cooking [James and Ntutela, 1997, p. 16]. They therefore prioritised the expenditure on the connection fee (R200 in Tambo village in 1996) for the 20A supply. Similar cases were found in Mafefe in spite of the fact that households had to pay a connection fee of R450 and a monthly service charge of R37 in order to obtain a level of supply that would enable them to use electricity for cooking [James, 1997, p. 7]. These households did not necessarily belong to a high-income group. In fact, in Tambo, households reliant on one pension or less per month also prioritised expenditure on hot plates/stoves so that electricity could be used for cooking [James, 1997, p. 31]. James and Ntutela [1997, p. 17] point out that, even in cases where a hot plate is used only for food which is cooked speedily, and total household expenditure on electricity is only R10 per month (corresponding to l
Volume IV No. 4
l
December 2000
Articles
about 37kWh used), the service of cooking with electricity can be highly valued. The importance attached to cooking with electricity was also evident in Loskop, where only two of the thirteen households included in the qualitative sample had no electric stove or hot plate [Hannsman et al., 1996]. One household had a large electric stove with an oven, two had two-plate mini-stoves combined with small ovens, and the rest had two-plate hot plates. A significant number of these households (seven) had bought electrical cooking appliances before any other appliances (except lights), and in two cases hot plates were the first and only appliances (except lights) bought by households. Households with a variety of demographic and gender characteristics -- for example, male and female household heads and/or income-earners -- as well as incomes were included in this sample. Nevertheless, while many households regard cooking with electricity as a priority, it is true that those who purchase electric hot plates and stoves often do not use these for all or even most of their cooking activities. Davis and Ward [1995] found that a third of electrified households in the rural PSLSD sample used electricity for cooking (also see Table 3 on appliance ownership). Only higherincome households were likely to use electricity as the sole fuel for cooking, however (26% of these households). Most of the middle- and lower-income households used electricity in combination with other fuels such as firewood and paraffin. Furthermore, the quantitative survey in Loskop found that 29% of households used electricity frequently for cooking, while an additional 13% occasionally used electricity to cook (see Table 4). Although most households in the qualitative sample in Loskop owned electric cooking appliances, there did not seem to be much evidence of the displacement of other fuels by electricity [Annecke, 1996, cited in James, 1997]. A variety of appliances and fuel combinations were used for cooking. It was found that stoves used before electrification (utilising wood, paraffin or gas) were still in use in all the households. Although it was difficult to quantify, the research suggested that these other fuel and appliance combinations were used as often as the hot plates were. Similar results were found at both Tambo and Mafefe in households which had acquired hot plates and stoves. None of the households had switched entirely to electricity, and other fuels were still used in conjunction with electricity [James, 1997]. In Tambo it was found that those households with hot plates and stoves all used paraffin as the main cooking fuel, while wood and dung fires were also used. These matters are discussed further under ‘‘Multiple fuel use’’ (Section 5) below. It should be pointed out that not all households want to cook with electricity. In Loskop the households which consisted of men only tended not to have electric stoves, as well as households which considered themselves to be ‘‘too poor’’ or did not like electricity for cooking [Annecke, 1996, cited in James, 1997]. Some people in Tambo also indicated that they were not interested in Energy for Sustainable Development
cooking with electricity, but were happy with wood and paraffin [James and Ntutela 1997, p. 38]. This attitude was prevalent among older women pensioners and households which consisted of men only. 5. Multiple fuel use Multiple fuel use -- the use of more than one fuel for household purposes -- is a key feature of energy use among lower-income households in South Africa, including electrified households. The National Domestic Energy Use Database (NDEUD) indicates that about 68% of all households (electrified and unelectrified) use combinations of two or more fuels, with many households (about 30%) using three fuels [Afrane-Okese, 1999]. Some energy studies in low-income urban areas have found that the majority of electrified households use electricity in combination with other fuels [Simmonds and Mammon, 1996, p. 50]. This is confirmed by Eskom [1996] which has found that the use of electricity in combination with other fuels is particularly prevalent among lower-income groups (see Table 5) and among newly electrified households. Davis and Ward [1995] found that most electrified households (39%) in rural areas use three fuels in combination to meet their energy needs (mostly wood, paraffin and electricity), while using four or more fuels is also very common (29%). As most unelectrified rural households use two or three fuels (34% and 56% respectively), it would seem that electricity does not replace other fuels, but adds to the fuels used by rural households. Regarding the reasons for combining electricity with other fuels, White [2000] points out that most lower-income households in metropolitan areas with access to electricity, and particularly recently electrified ones, perceive it as being too expensive for meeting all their energy needs. Nevertheless, cost is not the only reason for multiple fuel use: The choices involved are driven not by economics alone but by factors as ‘‘tradition’’, ‘‘culture’’, sociability, taste, and preferences which some researchers have chosen to call ‘‘fuel biographies’’ -- in other words, people’s personal histories of using fuel [White, 2000]. As a result, multiple fuel use is seen as a permanent phenomenon among low-income electrified households in urban areas [Mehlwana and Qase, 1999, p. 40]. It is also likely that most electrified rural households will remain heavily dependent on fuels other than electricity (particularly fuelwood and paraffin) for thermal purposes in the foreseeable future. 5.1. Persistence of paraffin One of the most interesting phenomena in post-electrification energy use among low-income households is the persistence of paraffin as a fuel, particularly for purposes other than lighting. For example, reporting on time trends in appliance acquisition over a three-year period in a periurban area which had been electrified, Golding [1993, p. 45] observed that paraffin stoves were the only appliances that were not affected by access to electricity. One of the factors that contributes to the continued use l
Volume IV No. 4
l
December 2000
41
Articles
of paraffin is the view that it is cheaper than electricity for cooking and water heating. In some areas this seems to be related to people’s ability to control the appliances involved. For example, a number of women in Loskop commented that it was more expensive to use an electric hot plate than a paraffin stove [Hansmann et al., 1996, A135]. This seemed to be related to their expertise in using paraffin stoves in an economical way -- that is, their ability to control the paraffin stoves accurately to perform certain tasks, and their knowledge of the time taken for particular tasks using the paraffin stoves, the consumption rates of their paraffin stoves and the costs of paraffin. A few women could estimate the costs of particular tasks using the paraffin stoves, which they saw as ‘‘cheap’’. They could not do the same for electricity, however, one of the reasons being that the cheapest prepayment electricity tokens cost R10, while the smallest unit registered by the electricity dispenser used in the area was also R10 [Hansmann et al., 1996, A126]. One woman expressed the wish to operate the electricity dispenser with coins, so that she could regulate their daily expenditure on electricity more closely. A general trend towards using paraffin for time-consuming cooking processes has been found in areas such as Loskop and Tambo. Cooking processes such as simmering were seen as expensive by households in Loskop, and paraffin was preferred because it was regarded as cheaper than electricity [Hansmann et al., 1996, A126]. Paraffin stoves were also seen as easy to regulate in order to simmer food [Annecke, 1996, p. 23]. Households therefore tended to prepare the main meal (in the evenings) on paraffin stoves. By comparison, the speed of cooking food was regarded as one of the main advantages of cooking with electricity -- it was thus sometimes used exclusively for dishes that require quick, intense heat inputs such as frying [Hansmann et al., 1996, A149]. Coupled with this were some forms of behaviour that probably added to the experience that electricity is expensive for cooking purposes. For example, some people tended to use electric hot plates on the highest setting all the time, without reducing the heat once the food is boiling. One of the reasons given for this was that they thought it was more expensive to use all the settings on the hot plate [Annecke, 1996, p. 23]. In Tambo electric hot plates were also mainly used to cook food with short preparation times [James and Ntutela, 1997, p. 38]. The main uses of electric hot plates were: boiling water and making porridge for breakfast in the mornings, making tea and preparing fast cooking vegetables (such as cabbage), soups, and occasionally fried meat in the evenings. Paraffin, on the other hand, was used specifically to prepare slow-cooking foods such as samp (maize) and beans. These use patterns were observed not only among lowincome households, but also among relatively affluent households in both Loskop and Tambo. This points to the complexity of factors that contribute to the persistent use of paraffin. For example, on the basis of research in a metropolitan area in the Eastern Cape province, Bank 42
Energy for Sustainable Development
[1999] argues that fuel use strategies and consumption patterns cannot simply be viewed as a function of affordability: ... [The] persistence of paraffin use in household units in Duncan Village was much more strongly correlated to variables such as gender, generation and diet, than it was to income. ... [Paraffin] was viewed as particularly well-suited to the preparation of the staple township diet of samp, beans and stiff porridge. As a commodity paraffin also bore a particularly strong gender identity. ... [By] virtue of its location in the domestic division of labour it had become synonymous with motherhood and household provisioning. ... [Women] and paraffin had come to represent each other, and together they symbolised domesticity [Bank, 1999]. These values were not as prevalent among younger people, however: The desire of the urban youth to differentiate themselves from the older generation by shirking conventional practices and attitudes affected their attitudes toward fuel consumption [Bank, 1999]. 6. Conclusions In this paper an attempt has been made to describe the utilisation of electricity by rural households in South Africa, with reference to quantitative as well as qualitative information on energy use by electrified low-income households in both urban and rural areas. The main findings can be summarised as follows: • In spite of the ‘‘blanket’’ electrification approach followed by Eskom, a significant percentage of rural households in areas connected to the national electricity grid during the last five years are not able to utilise electricity because they cannot afford the connection fee. Since the beginning of 2000 this problem has been addressed in some pilot areas by providing currentlimited connections to these households at no connection fee. • Significant percentages of low-income electrified households and electrified households in rural areas own electrical appliances such as stoves/hot plates, kettles, irons, refrigerators, televisions and radios/hifis. Generally, ownership of these appliances appears to increase as household income increases. Nevertheless, anthropological studies have shown that household income is not the main determinant of household appliance purchases. • Not all households that own electrical appliances use them on a regular basis. The reasons for acquiring appliances are of a complex nature -- ‘‘utilitarian’’ considerations such as usefulness and cost are only some of the factors that influence this, while the symbolic value of appliances is also important. • The common perception that rural households use electricity for lighting and to operate radios, but not for cooking, is too simplistic: -- While electric lighting is commonly used by l
Volume IV No. 4
l
December 2000
Articles
electrified households, a significant percentage of electrified rural households continue to use other fuels, particularly candles, for lighting purposes. One of the reasons for this is that many households find it difficult to extend electric wiring from the ready board provided by Eskom to all the rooms in their houses. -- A significant percentage of electrified households continue to use battery-operated radios. One reason that seems to contribute to this is that unelectrified households tend to buy radios that can be operated on batteries only, without the option to use grid electricity. The cost of dry-cell batteries is extremely high in relation to the energy delivered. -- The service of cooking with electricity is highly valued by many rural households, even if cooking is not done exclusively with electricity. A significant percentage of rural households use electricity for cooking, but only higher-income households are likely to use electricity as the sole fuel for cooking. Nevertheless, cooking with electricity is also highly valued by many low-income households and households that only use electricity for very limited cooking purposes. • Most electrified households in rural areas use three fuels in combination to meet their energy needs (mostly wood, paraffin and electricity). It seems that electricity does not replace other fuels, but adds to the fuels used by rural households. • The persistent use of paraffin, particularly for cooking and water heating, is one of the most interesting phenomena in energy use among low-income electrified households in South Africa. One of the factors that contributes to this is the view that paraffin is cheaper to use than electricity for these thermal purposes. Nevertheless, a range of factors contributes to the persistent use of paraffin. For example, in one metropolitan area paraffin use was found to be much more strongly correlated to variables such as gender, generation and diet, than it was to income.
3. For example, 85% of the households interviewed qualitatively owned electric hot plates compared with 45% of the quantitative sample. Furthermore, the qualitative sample (comprising thirteen households) included three households that had been electrified for less than one year. 4. This tariff was rejected by Eskom after the negative responses at the pilot sites such as Tambo village. References Afrane-Okese, Y., 1999. National Domestic Energy Use Database System as a Tool for Integrated Energy Planning, Pretoria, Department of Minerals and Energy. Annecke, W., 1996. Post-electrification Study of Loskop -- Synthesis Report, Cape Town, Energy and Development Research Centre. Bank, L., 1999. ‘‘Basic energy needs and multiple fuel-use: some reflections on energy policy and social theory’’, Proceedings of the Domestic Use of Electrical Energy Conference, Cape Town, Cape Technikon, 30 March to 1 April. Davis, M., and Ward, S., 1995. Household Energy-use Patterns in Rural Areas: the Effects of Access to Electricity, Cape Town, Energy and Development Research Centre. Eskom, 1996. SA to Z: The Decision Maker’s Encyclopaedia of the South African Consumer Market, Sandton, Eskom. Golding, A.P., 1993. The Socio-economic Impact of Electrification on a Peri-urban Community in Bophuthatswana, Pretoria, Department of Minerals and Energy. Griffin, N.J., Banks, D., Mavrandonis, J., Shackleton, S.E., and Shackleton, C.M., 1992. Household Energy and Wood Use in a Peripheral Rural Area of the Eastern Transvaal Lowveld, Pretoria, Department of Minerals and Energy. Hansmann, C., Van Gass, M., Annecke, W., Despins, P.M., and Kargas, S., 1996. Postelectrification Study of Loskop -- Appendices, Cape Town, Energy and Development Research Centre. Hofmeyr, I.-M., 1996. A Comparative Study of the Access to and Use of Electricity by Farmworker Households in the Free State, Cape Town, Energy and Development Research Centre. James, B., and Ntutela, P., 1997. Rural Households’ Response to the 2.5A Electricity Supply Option in the Tambo Village Pilot Project, Cape Town, Energy and Development Research Centre. James, B., 1997. Current-limited Supplies of Electricity in the Context of South African Rural Areas, Cape Town, Energy and Development Research Centre. Kinners, L., (Eskom), 1998. Personal communication. Kotzé, I.A., 2000. ‘‘The South African National Electrification Programme: past lessons and future prospects’’, Proceedings of ISES 2000 Utility Initiative for Africa: Rural Electrification in Africa, 17-18 April, Midrand, South Africa. Mehlwana, M., 1999. The Economics of Energy for the Poor: Fuel and Appliance Purchase in Low-income Urban Households, Cape Town, Energy and Development Research Centre. Mehlwana, M., and Qase, N., 1999. The Contours of Domesticity, Energy Consumption and Poverty: the Social Determinants of Energy Use in Low-income Urban Households, Cape Town, Energy and Development Research Centre. Palmer, R., 1999. The Impact of Domestic Electrification on Household Hydrocarbon Fuel Consumption, Pretoria, Department of Minerals and Energy. Sokopo, I., (Eskom), 2000. Personal communication. Simmonds, G., and Mammon, N., 1996. Energy Services in Low-income Urban South Africa: a Quantitative Assessment, Cape Town, Energy and Development Research Centre.
Notes
Thom, C., Davis, M., and Borchers, M., 1995. Review of South African Experience in Rural Electrification, Cape Town, Energy and Development Research Centre.
1. Definitions of rural areas in South Africa are generally both inadequate and inconsistent. These generally only exclude towns and cities which had some form of local authority under the apartheid government, and include many of the informal settlements bordering on such towns and cities. In addition, most areas in the former homelands are classified as rural areas, irrespective of the size or density of the settlements.
Thom, C., 1998. Rural Electrification Policy: Some Key Recommendations, Cape Town, Energy and Development Research Centre. Wentzel, M., Manzini, M., Mudlaudzi, C., Sehlapelo, D., and Wood, C., 1997. A post-electrification Study of the Mafefe Electrification Project, Cape Town, Energy and Development Research Centre.
2. The project was jointly funded by the South African Department of Minerals and Energy (DME), Eskom’s Technology Research & Investigations (TRI), the Norwegian Agency for International Development Cooperation (NORAD), and to a lesser extent the Development Bank of Southern Africa (DBSA). It aimed to develop policy proposals and implementation guidelines for rural electrification in South Africa.
Energy for Sustainable Development
White, C., 2000. Synthesis Report on the Social Determinants of Energy Use in Low-income Households in Four Metropolitan Areas of South Africa, Pretoria, Department of Minerals and Energy.
l
Volume IV No. 4
l
December 2000
43
Use of grid electricity by rural households in South Africa Cecile Thom Energy & Development Research Centre, PO Box 34178, Rhodes Gift 7707, South Africa
Almost half of rural households in South Africa had been connected to the grid by the end of 1999. This paper describes the utilisation of electricity by rural households in South Africa with reference to quantitative as well as qualitative information on energy use by electrified low-income households. The main findings are the following. Significant percentages of electrified households in rural areas own electrical appliances such as hotplates, kettles, irons, refrigerators, televisions and radios/hi-fis. Not all households that own electrical appliances use them on a regular basis. The reasons for acquiring appliances are of a complex nature -- considerations such as usefulness and cost are only some of the factors that influence this, while the symbolic value of appliances is also of importance. While electric lighting is commonly used by electrified households, a significant percentage of electrified rural households continue to use other fuels, particularly candles, for lighting purposes. A significant percentage of electrified households also continue to use battery-operated radios. Furthermore, a significant percentage of rural households use electricity for cooking, although only higher income households are likely to use electricity as the sole fuel for cooking. The service of cooking with electricity is highly valued by many rural households, even if cooking is not done exclusively with electricity. Most electrified households in rural areas use three fuels in combination to meet their energy needs. It seems that electricity does not replace other fuels, but adds to the fuels used by rural households. Many low-income households in South Africa continue to use paraffin after electrification, particularly for cooking and water heating purposes. One of the factors that contributes to this is the view that paraffin is cheaper to use than electricity for these thermal purposes. Nevertheless, a range of factors contributes to the persistent use of paraffin. 1. Introduction This paper aims to present some of the available information on the use of electricity by ‘‘rural’’[1] households in South Africa. Much of the work reported on in this paper was done as part of the research project The Role of Electricity in the Integrated Provision of Energy to Rural Areas (REIPERA) that was undertaken from 1995 to 1998 at the Energy and Development Research Centre (EDRC)[2]. The main findings of the project have been summarised by Thom [1998]. Some of them, particularly pertaining to household electrification, are presented here in greater detail together with some additional information which has since become available. Although the focus of this paper is household electrification, it needs to be mentioned that the REIPERA project recommended that the priorities of the electrification programme in South Africa -- mainly households, as well as clinics and schools -- should be expanded to include community water supply schemes and small-scale agricultural projects in particular, while small, medium and micro-enterprises (SMMEs) also need greater attention.
households have access to grid electricity. About two-thirds of households in South Africa (66%), and almost half (46%) of rural households, had been connected to the grid by the end of 1999 [Kotzé, 2000]. This can be compared with an estimated 44% of households, and 12% of rural households, which had been electrified at the end of 1994 [Thom et al., 1995]. In the six years from 1994 to 1999 between 400,000 and 500,000 new connections were made annually as part of the accelerated national electrification programme, in accordance with the Reconstruction and Development Programme (RDP) targets agreed upon between the government and the electricity industry (see Table 1). The programme was implemented by the national electricity utility Eskom as well as by municipalities. Eskom has been responsible for most of the connections in rural areas. Electrification projects in these areas typically involve the electrification of an entire settlement, or at least a substantial part of it. All households in a settlement who pay the connection fee (75 rands (R) in 1998, increased to R150 in 2000, for the standard 20A supply) and can be electrified at a reasonable cost are provided with an electricity supply, including a pre-payment meter and a ready board (see below). This is sometimes referred to as a ‘‘blanket’’ electrification approach. In spite of this approach substantial numbers of
2. Access to grid electricity The first matter to consider when discussing the use of electricity by rural households is to what extent these 36
Energy for Sustainable Development
l
Volume IV No. 4
l
December 2000
Articles
Table 1. The national electrification programme 1994-1999 (number of households) 1994
1995
1996
1997
1998
1999
Eskom
254,383
313,179
307,047
274,345
280,977
293,006
Local government
164,535
150,454
137,534
213,768
136,074
144,043
16,838
15,134
9,414
11,198
10,375
6,241
Total
435,756
478,767
453,995
499,311
427,426
443,290
RDP targets
350,000
400,000
450,000
450,000
450,000
450,000
Farm workers
Source: Kotzé, 2000
Table 2. Electrical appliance ownership for different income groups (all areas) % of households with monthly income of: R1-499
R500-799
16
25
39
67
98
Geyser
1
3
9
43
93
Heater
3
7
14
33
70
Stove
3
10
19
54
94
Hot plate
6
9
14
14
10
Kettle
5
12
22
55
94
Iron
6
14
26
61
96
Fridge
6
13
25
61
96
Television
8
17
32
63
96
Radio/hi-fi
9
16
28
58
92
Access to grid electricity
R800-1399
R1400-3999
R4000+
Electrical appliances
Source: Eskom, 1996
Table 3. Appliance ownership among electrified households in rural areas % of households with per capita[1] monthly income of:
% of all households
R130-270 [2]
< R130 Geyser
> R270
5
0
2
8
Stove/hot plate
37
10
27
53
Kettle
32
12
25
45
Fridge
43
12
28
65
Television
47
19
33
67
Radio/hi-fi
87
79
85
91
Source: Davis and Ward, 1995 Notes 1. Average household size was about 5.5. 2. Estimated minimum monthly expenditure to satisfy basic needs was about R270 per capita.
households in areas connected to the grid do not have access to electricity. In 1998 one of Eskom’s regional electrification managers indicated that only 65-85% of households in areas connected to the grid generally paid the connection fee [Kinners, 1998]. In an attempt to address this problem Eskom has been piloting an approach since the beginning of 2000 whereby households that are unable to pay a connection fee are provided with a 2.5A supply [Sokopo, 2000]. This enables them to use electricEnergy for Sustainable Development
ity for lighting, radios, television and refrigeration. The tariff is the same as the 20A tariff, comprising a simple unit energy charge with no fixed monthly charges. Users can upgrade to a higher level of supply by paying the required connection fee. 3. Appliance ownership Table 2 provides an overview of electrical appliance ownership in South Africa. It is presented in relation to l
Volume IV No. 4
l
December 2000
37
Articles
Table 4. Ownership and use of electrical appliances by households in Loskop electrified for 2 years or more[1] Service
Households using it frequently (%)
Households using it occasionally (%)
Households using it (total) (%)
Households owning appliance (%)
Lighting
94
0
94
-
Radio/hi-fi
77
0
77
84[2]
Refrigeration
48
0
48
48
Television
42
3
45
52[2]
61
10
71
65 (electric iron)
32
16
48
45 (electric kettle)
Cooking (hot plate/stove)
29
13
42
3 (stove) 45 (hot plate)
Space heating
3
3
6
16 (heater)
Ironing (iron/hot plate)[3] Water heating (kettle/hot plate)
[3]
Source: Compiled from Hansmann et al., 1996, A86. Notes 1. A sample of 30 was used, chosen randomly as far as possible within practical constraints. 2. These figures may include some radios and televisions operated with batteries -- for example, 3% of the sample (one household) indicated that they frequently used car batteries to operate a television. 3. The percentage of households using electricity for ironing and water heating probably includes households using a flat-plate metal iron and simple stove-top kettle respectively, heated on a hot plate.
household income to illustrate some of the trends, particularly since large income disparities exist in South Africa which correspond to disparities in access to services. At first glance the table gives the impression that the ownership of electrical appliances is highly dependent on household income levels. However, as shown in the first row, access to grid electricity itself was highly dependent on income at the time the study was conducted (1995). Generally, ownership of electrical appliances increases in relation to access to electricity as income increases, confirming that there is a relationship between household income and appliance ownership among electrified households, with hot plates being the main exception. The data suggests that more than one-third of the electrified households in the lowest income group own hot plates, irons and refrigerators, while more than half own radios/hi-fis and televisions which operate on grid electricity. The table shows that electric stoves with ovens are more common than electric hot plates in all but the lowest income group. Some recent energy studies in specific localities have indicated, however, that hot plates are far more common among lower-income electrified households than stoves [James and Ntutela, 1997; Mehlwana and Qase, 1999; Palmer, 1999; White, 2000]. 3.1. Appliance ownership in rural areas Data on the ownership of appliances in rural areas is presented in Table 3, which has been drawn from the analysis done by Davis and Ward [1995] on data collected in the Project for Statistics on Living Standard and Development (PSLSD). Davis and Ward [1995] observe that the ownership of electrical appliances appears to be closely related to income -- with the exception of radios, electrical appliance penetration in the lowest income group is between 10% and 20%. 3.2. Limitations of an income-based analysis Although there is undoubtedly a relationship between 38
Energy for Sustainable Development
appliance ownership and income, it nevertheless seems that income is not the main determinant of electrical appliance purchases. Recent anthropological studies in metropolitan areas have shown that ‘‘the interplay of multiple (social and economic) factors determines appliance acquisition and use’’ [Mehlwana, 1999]. These are further discussed below. 3.3. Electrical appliance ownership versus use Some information on electrical appliance ownership and use was obtained in a post-electrification study in Loskop (KwaZulu-Natal), conducted as part of the REIPERA project [Hansmann et al., 1996]. A small quantitative survey was undertaken in an area where households had been electrified for two years or more. The information pertaining to electrical appliances is presented in Table 4. It is particularly interesting to compare the percentage of households that use appliances frequently with the percentage that own the appliances concerned. A significant percentage of households that own an electric stove or hot plate do not use these frequently for cooking purposes. The difference is even more striking in the case of electric heaters. Some of the reasons for this are discussed under ‘‘Multiple fuel use’’ (Section 5) below. Research in metropolitan areas has confirmed that the fact that households own appliances does not necessarily mean that they use them [Mehlwana, 1999]. In fact, anthropological studies have indicated that some of the factors that contribute to the purchasing of appliances are not related to use at all. According to White [2000], utilitarian considerations such as usefulness, cost, and the availability of space are only one set of factors that influence decisions to buy, use or keep appliances. Coupled with this are considerations of symbolic value. Examples of such considerations are the wish to conceal poverty, and to ‘‘embody’’ a particular lifestyle: The fashioning of the ‘‘public’’ spaces within the home l
Volume IV No. 4
l
December 2000
Articles
is one of the primary mechanisms used to ward off the suspicion that one is ‘‘struggling’’. In an attempt to create an image of style and well-being for outsiders, any electric appliances that can add to that image are prominently displayed. ... In some instances, appliances have symbolic value only, as evidenced by the fact that broken appliances are kept on prominent display in many homes [White, 2000]. Electric appliances are infused with positive symbolic associations that are quite absent from most non-electric appliances. Electricity is a marker of modernity and progress. ... In East London, the drive to accumulate appliances for the purposes of marking style and social position is particularly prevalent among single women with good permanent jobs. ... Electrical appliances ... serve to distance them from the way of life of paraffin, the fuel associated with dependent and subordinated women, and enhance their image as independent and self-sufficient [White, 2000].
and/or bear the cost of extending the supply to other rooms, and are not provided any assistance with this. Extension cords are often used to ‘‘wire’’ houses informally. A survey conducted at Tambo found that about 76% of households had electric lighting in one room only, while about 14% had lights in two rooms, and only about 10% had lights in more than two rooms [James and Ntutela, 1997, p. 34]. This apparently resulted from the cost of materials for informal wiring, as well as the cost of installation and the availability of people with the necessary skills. In very poor households at both Tambo and Mafefe no wiring of additional rooms had occurred [James, 1997]. The placement of the ready board in the home clearly becomes an important consideration under these circumstances. In some cases electrification contractors consult households about the placement of ready boards, but not always. Since the research in Tambo and Mafefe was conducted only about six months after electrification, the findings need to be compared with studies done a longer time after electrification. In a longitudinal study on energy use in Kameelrivier B -- a peri-urban area in Mpumalanga province -- Palmer [1999] found that the situation changed significantly over a period of two years after electrification. Immediately after electrification the most common fuels for lighting were a combination of electricity and other fuels, while almost 40% of households were using no electric lighting at all. After two years the situation had changed completely, with 79% of households relying solely on electricity, and 21% using it in conjunction with other fuels. The main reason for continuing to use other fuels, and candles in particular, seemed to be that not all rooms had been provided with electric lighting, thus confirming the findings in the Tambo and Mafefe studies. As shown in Table 4, about 6% of households in Loskop (KwaZulu-Natal) that had been electrified for two years or more showed that they did not use electric lighting at all. The majority of the sample (94%) all used electric lighting frequently. More information on households’ use of electric lighting emerged from qualitative interviews undertaken in Loskop. The qualitative sample was smaller and differed substantially from the quantitative sample[3]. Electric lights were generally available in all rooms of the main dwellings (as opposed to additional ‘‘outbuildings’’ that form part of the homestead) of the households included in the qualitative sample. Most of these households also had an outside electric light. Candles, and occasionally paraffin lamps, were used in outbuildings and when power failed [Annecke, 1996]. The wiring of the houses was in most cases done by family members and sometimes by neighbours, and was therefore presumably not particularly costly to households. Some other factors also influence whether electric lighting is used -- for example, cases have been found where people prefer to use other forms of lighting because they fear electricity, or because the electric light switch cannot be reached once they are in bed [James, 1997]. 4.2. Radios A large percentage of rural households own radios. For
4. Use of electricity for particular services Specific attention is given here to the use of electricity for lighting and cooking, and to operate radios. The aim is to illustrate that the utilisation of electricity for these purposes is much more complex than the general perception that rural households use electricity for lighting and to operate radios, but not for cooking. 4.1. Lighting Electric lighting is commonly used by electrified households. However, research done in the REIPERA project indicates that a significant percentage of electrified households continue to use other fuels -- particularly candles, and to a lesser extent paraffin lamps -- for lighting purposes. Davis and Ward [1995] found that as much as 60% of the electrified sample in rural areas use candles and paraffin for lighting in addition to electricity, while about 20% use electricity only, and about 18% use candles and/or paraffin only (no electric lighting). The fact that not all households connected to the electricity grid are able to use electricity, as discussed under ‘‘Access to grid electricity’’ (Section 2) above, probably contributes significantly to this situation. Other factors nevertheless also have an impact on the use of electric lighting, as indicated by some qualitative post-electrification studies conducted during the REIPERA project. At both Tambo (Eastern Cape province) and Mafefe (Northern Province) where the 2.5A supply was originally piloted, candles and paraffin were still being used substantially for lighting about six months after electrification [Wentzel et al., 1997; James and Ntutela 1997]. The main reason for this was that electric lighting was not available in all the rooms in houses [James, 1997]. Eskom generally provides households with ready boards comprising a light and a row of sockets, which allows households immediate access to electricity without the expense of formal house wiring, while reducing the reticulation cost to the utility. The disadvantage of this approach is that the light and sockets are located in a specific room. Households have to get access to the skills Energy for Sustainable Development
l
Volume IV No. 4
l
December 2000
39
Articles
Table 5. Use of electricity by different income groups % of households with monthly income of: R1-499
R500-799
R800-1399
R1400-3999
R4000+
Access to grid electricity
16
25
39
67
98
Use of electricity as sole energy source
3
12
21
51
82
Use of electricity in combination with other fuels
12
13
17
16
16
Use of electricity for cooking
8
18
29
63
96
Use of electricity for lighting
15
25
37
67
98
Source: Eskom, 1996, Table E8:1
example, Davis and Ward [1995] found that 74% of unelectrified households and 87% of electrified households in the PSLSD sample owned radios. In the absence of grid electricity radios are mostly powered with dry-cell batteries. It seems that unelectrified households generally buy radios that can be operated on batteries only, without the option to use grid electricity. At both Tambo and Mafefe none of the households interviewed in the qualitative research were able to power their small radios with electricity [James, 1997]. For example, in Tambo eight households included in the qualitative sample (a total of seventeen households) owned radios in working condition, none of which could be operated with grid electricity [James and Ntutela, 1997, p. 35]. Only hi-fis could be operated from the grid. Findings in metropolitan areas in the Western Cape have been similar, with significant percentages of electrified low-income households using dry-cell batteries to power radios: The use of dry-cell batteries in these electrified households occurs because ownership of battery-operated radios dates back to the pre-electrification period [Mehlwana and Qase, 1999, p. 55]. In Loskop the majority of households in the quantitative sample (77%), which had all been electrified for more than two years, were using radios and/or hi-fis operated on grid electricity. Nevertheless, a significant percentage of households (32%) still used battery-operated radios, with 19% using them frequently, and the rest using them only sometimes (see Table 4). About 10% of households indicated that they did not own a radio or hi-fi. Clearly there were people who used both batteries and electricity to operate their radios -- one reason is probably that batteries are needed if radios are used a distance away from available sockets (outside of houses, for example). Market research conducted by Eskom has confirmed these findings. They found that 25% of grid-electrified households (in all areas, including rural areas) continue to use battery-operated radios [Eskom, 1996]. The cost of dry-cell batteries is extremely high in relation to the energy delivered, and some studies have found that dry-cell batteries form a significant component of households’ energy expenditure (up to 30-40% in some unelectrified villages) [Griffin et al., 1992]. Furthermore, James and Ntutela [1997, p. 40] found that the energy expenditure of households in Tambo that got a 2.5A elec40
Energy for Sustainable Development
tricity supply (for which they had to pay a fixed charge of R15 per month)[4] had increased significantly in cases where electricity had not displaced other fuels for lighting, or batteries for radio and television. Dry-cell batteries for radios seemed to have the greatest impact on household budgets under these conditions, because of their expense. Electrification had least impact on budgets of 2.5A households without radios, which were therefore not spending at least R8.20 per month on batteries [James and Ntutela, 1997, p. 39]. Research among farm workers in the Free State found that in addition to not having mechanisms to run their radios on electricity, farm workers did not realise that they might be able to power their radios with electricity, nor did they have the money to buy new radios which could run on electricity [Hofmeyr, 1996, cited in James, 1997]. The expenditure on batteries was found to be high, especially when considered as a proportion of wages. 4.3. Cooking Qualitative studies done during the REIPERA project indicate that cooking with electricity is important to many rural households, even if cooking is not done exclusively with electricity. The study in Tambo indicated that many people value the use of electricity for cooking. For example, some women in Tambo showed a lot of determination in saving money from their earnings in order to purchase cooking appliances. Furthermore, 75% of the households in the quantitative sample in Tambo that chose a 20A supply did this because they wanted to be able to use electricity for services not available with a 2.5A supply, such as cooking [James and Ntutela, 1997, p. 16]. They therefore prioritised the expenditure on the connection fee (R200 in Tambo village in 1996) for the 20A supply. Similar cases were found in Mafefe in spite of the fact that households had to pay a connection fee of R450 and a monthly service charge of R37 in order to obtain a level of supply that would enable them to use electricity for cooking [James, 1997, p. 7]. These households did not necessarily belong to a high-income group. In fact, in Tambo, households reliant on one pension or less per month also prioritised expenditure on hot plates/stoves so that electricity could be used for cooking [James, 1997, p. 31]. James and Ntutela [1997, p. 17] point out that, even in cases where a hot plate is used only for food which is cooked speedily, and total household expenditure on electricity is only R10 per month (corresponding to l
Volume IV No. 4
l
December 2000
Articles
about 37kWh used), the service of cooking with electricity can be highly valued. The importance attached to cooking with electricity was also evident in Loskop, where only two of the thirteen households included in the qualitative sample had no electric stove or hot plate [Hannsman et al., 1996]. One household had a large electric stove with an oven, two had two-plate mini-stoves combined with small ovens, and the rest had two-plate hot plates. A significant number of these households (seven) had bought electrical cooking appliances before any other appliances (except lights), and in two cases hot plates were the first and only appliances (except lights) bought by households. Households with a variety of demographic and gender characteristics -- for example, male and female household heads and/or income-earners -- as well as incomes were included in this sample. Nevertheless, while many households regard cooking with electricity as a priority, it is true that those who purchase electric hot plates and stoves often do not use these for all or even most of their cooking activities. Davis and Ward [1995] found that a third of electrified households in the rural PSLSD sample used electricity for cooking (also see Table 3 on appliance ownership). Only higherincome households were likely to use electricity as the sole fuel for cooking, however (26% of these households). Most of the middle- and lower-income households used electricity in combination with other fuels such as firewood and paraffin. Furthermore, the quantitative survey in Loskop found that 29% of households used electricity frequently for cooking, while an additional 13% occasionally used electricity to cook (see Table 4). Although most households in the qualitative sample in Loskop owned electric cooking appliances, there did not seem to be much evidence of the displacement of other fuels by electricity [Annecke, 1996, cited in James, 1997]. A variety of appliances and fuel combinations were used for cooking. It was found that stoves used before electrification (utilising wood, paraffin or gas) were still in use in all the households. Although it was difficult to quantify, the research suggested that these other fuel and appliance combinations were used as often as the hot plates were. Similar results were found at both Tambo and Mafefe in households which had acquired hot plates and stoves. None of the households had switched entirely to electricity, and other fuels were still used in conjunction with electricity [James, 1997]. In Tambo it was found that those households with hot plates and stoves all used paraffin as the main cooking fuel, while wood and dung fires were also used. These matters are discussed further under ‘‘Multiple fuel use’’ (Section 5) below. It should be pointed out that not all households want to cook with electricity. In Loskop the households which consisted of men only tended not to have electric stoves, as well as households which considered themselves to be ‘‘too poor’’ or did not like electricity for cooking [Annecke, 1996, cited in James, 1997]. Some people in Tambo also indicated that they were not interested in Energy for Sustainable Development
cooking with electricity, but were happy with wood and paraffin [James and Ntutela 1997, p. 38]. This attitude was prevalent among older women pensioners and households which consisted of men only. 5. Multiple fuel use Multiple fuel use -- the use of more than one fuel for household purposes -- is a key feature of energy use among lower-income households in South Africa, including electrified households. The National Domestic Energy Use Database (NDEUD) indicates that about 68% of all households (electrified and unelectrified) use combinations of two or more fuels, with many households (about 30%) using three fuels [Afrane-Okese, 1999]. Some energy studies in low-income urban areas have found that the majority of electrified households use electricity in combination with other fuels [Simmonds and Mammon, 1996, p. 50]. This is confirmed by Eskom [1996] which has found that the use of electricity in combination with other fuels is particularly prevalent among lower-income groups (see Table 5) and among newly electrified households. Davis and Ward [1995] found that most electrified households (39%) in rural areas use three fuels in combination to meet their energy needs (mostly wood, paraffin and electricity), while using four or more fuels is also very common (29%). As most unelectrified rural households use two or three fuels (34% and 56% respectively), it would seem that electricity does not replace other fuels, but adds to the fuels used by rural households. Regarding the reasons for combining electricity with other fuels, White [2000] points out that most lower-income households in metropolitan areas with access to electricity, and particularly recently electrified ones, perceive it as being too expensive for meeting all their energy needs. Nevertheless, cost is not the only reason for multiple fuel use: The choices involved are driven not by economics alone but by factors as ‘‘tradition’’, ‘‘culture’’, sociability, taste, and preferences which some researchers have chosen to call ‘‘fuel biographies’’ -- in other words, people’s personal histories of using fuel [White, 2000]. As a result, multiple fuel use is seen as a permanent phenomenon among low-income electrified households in urban areas [Mehlwana and Qase, 1999, p. 40]. It is also likely that most electrified rural households will remain heavily dependent on fuels other than electricity (particularly fuelwood and paraffin) for thermal purposes in the foreseeable future. 5.1. Persistence of paraffin One of the most interesting phenomena in post-electrification energy use among low-income households is the persistence of paraffin as a fuel, particularly for purposes other than lighting. For example, reporting on time trends in appliance acquisition over a three-year period in a periurban area which had been electrified, Golding [1993, p. 45] observed that paraffin stoves were the only appliances that were not affected by access to electricity. One of the factors that contributes to the continued use l
Volume IV No. 4
l
December 2000
41
Articles
of paraffin is the view that it is cheaper than electricity for cooking and water heating. In some areas this seems to be related to people’s ability to control the appliances involved. For example, a number of women in Loskop commented that it was more expensive to use an electric hot plate than a paraffin stove [Hansmann et al., 1996, A135]. This seemed to be related to their expertise in using paraffin stoves in an economical way -- that is, their ability to control the paraffin stoves accurately to perform certain tasks, and their knowledge of the time taken for particular tasks using the paraffin stoves, the consumption rates of their paraffin stoves and the costs of paraffin. A few women could estimate the costs of particular tasks using the paraffin stoves, which they saw as ‘‘cheap’’. They could not do the same for electricity, however, one of the reasons being that the cheapest prepayment electricity tokens cost R10, while the smallest unit registered by the electricity dispenser used in the area was also R10 [Hansmann et al., 1996, A126]. One woman expressed the wish to operate the electricity dispenser with coins, so that she could regulate their daily expenditure on electricity more closely. A general trend towards using paraffin for time-consuming cooking processes has been found in areas such as Loskop and Tambo. Cooking processes such as simmering were seen as expensive by households in Loskop, and paraffin was preferred because it was regarded as cheaper than electricity [Hansmann et al., 1996, A126]. Paraffin stoves were also seen as easy to regulate in order to simmer food [Annecke, 1996, p. 23]. Households therefore tended to prepare the main meal (in the evenings) on paraffin stoves. By comparison, the speed of cooking food was regarded as one of the main advantages of cooking with electricity -- it was thus sometimes used exclusively for dishes that require quick, intense heat inputs such as frying [Hansmann et al., 1996, A149]. Coupled with this were some forms of behaviour that probably added to the experience that electricity is expensive for cooking purposes. For example, some people tended to use electric hot plates on the highest setting all the time, without reducing the heat once the food is boiling. One of the reasons given for this was that they thought it was more expensive to use all the settings on the hot plate [Annecke, 1996, p. 23]. In Tambo electric hot plates were also mainly used to cook food with short preparation times [James and Ntutela, 1997, p. 38]. The main uses of electric hot plates were: boiling water and making porridge for breakfast in the mornings, making tea and preparing fast cooking vegetables (such as cabbage), soups, and occasionally fried meat in the evenings. Paraffin, on the other hand, was used specifically to prepare slow-cooking foods such as samp (maize) and beans. These use patterns were observed not only among lowincome households, but also among relatively affluent households in both Loskop and Tambo. This points to the complexity of factors that contribute to the persistent use of paraffin. For example, on the basis of research in a metropolitan area in the Eastern Cape province, Bank 42
Energy for Sustainable Development
[1999] argues that fuel use strategies and consumption patterns cannot simply be viewed as a function of affordability: ... [The] persistence of paraffin use in household units in Duncan Village was much more strongly correlated to variables such as gender, generation and diet, than it was to income. ... [Paraffin] was viewed as particularly well-suited to the preparation of the staple township diet of samp, beans and stiff porridge. As a commodity paraffin also bore a particularly strong gender identity. ... [By] virtue of its location in the domestic division of labour it had become synonymous with motherhood and household provisioning. ... [Women] and paraffin had come to represent each other, and together they symbolised domesticity [Bank, 1999]. These values were not as prevalent among younger people, however: The desire of the urban youth to differentiate themselves from the older generation by shirking conventional practices and attitudes affected their attitudes toward fuel consumption [Bank, 1999]. 6. Conclusions In this paper an attempt has been made to describe the utilisation of electricity by rural households in South Africa, with reference to quantitative as well as qualitative information on energy use by electrified low-income households in both urban and rural areas. The main findings can be summarised as follows: • In spite of the ‘‘blanket’’ electrification approach followed by Eskom, a significant percentage of rural households in areas connected to the national electricity grid during the last five years are not able to utilise electricity because they cannot afford the connection fee. Since the beginning of 2000 this problem has been addressed in some pilot areas by providing currentlimited connections to these households at no connection fee. • Significant percentages of low-income electrified households and electrified households in rural areas own electrical appliances such as stoves/hot plates, kettles, irons, refrigerators, televisions and radios/hifis. Generally, ownership of these appliances appears to increase as household income increases. Nevertheless, anthropological studies have shown that household income is not the main determinant of household appliance purchases. • Not all households that own electrical appliances use them on a regular basis. The reasons for acquiring appliances are of a complex nature -- ‘‘utilitarian’’ considerations such as usefulness and cost are only some of the factors that influence this, while the symbolic value of appliances is also important. • The common perception that rural households use electricity for lighting and to operate radios, but not for cooking, is too simplistic: -- While electric lighting is commonly used by l
Volume IV No. 4
l
December 2000
Articles
electrified households, a significant percentage of electrified rural households continue to use other fuels, particularly candles, for lighting purposes. One of the reasons for this is that many households find it difficult to extend electric wiring from the ready board provided by Eskom to all the rooms in their houses. -- A significant percentage of electrified households continue to use battery-operated radios. One reason that seems to contribute to this is that unelectrified households tend to buy radios that can be operated on batteries only, without the option to use grid electricity. The cost of dry-cell batteries is extremely high in relation to the energy delivered. -- The service of cooking with electricity is highly valued by many rural households, even if cooking is not done exclusively with electricity. A significant percentage of rural households use electricity for cooking, but only higher-income households are likely to use electricity as the sole fuel for cooking. Nevertheless, cooking with electricity is also highly valued by many low-income households and households that only use electricity for very limited cooking purposes. • Most electrified households in rural areas use three fuels in combination to meet their energy needs (mostly wood, paraffin and electricity). It seems that electricity does not replace other fuels, but adds to the fuels used by rural households. • The persistent use of paraffin, particularly for cooking and water heating, is one of the most interesting phenomena in energy use among low-income electrified households in South Africa. One of the factors that contributes to this is the view that paraffin is cheaper to use than electricity for these thermal purposes. Nevertheless, a range of factors contributes to the persistent use of paraffin. For example, in one metropolitan area paraffin use was found to be much more strongly correlated to variables such as gender, generation and diet, than it was to income.
3. For example, 85% of the households interviewed qualitatively owned electric hot plates compared with 45% of the quantitative sample. Furthermore, the qualitative sample (comprising thirteen households) included three households that had been electrified for less than one year. 4. This tariff was rejected by Eskom after the negative responses at the pilot sites such as Tambo village. References Afrane-Okese, Y., 1999. National Domestic Energy Use Database System as a Tool for Integrated Energy Planning, Pretoria, Department of Minerals and Energy. Annecke, W., 1996. Post-electrification Study of Loskop -- Synthesis Report, Cape Town, Energy and Development Research Centre. Bank, L., 1999. ‘‘Basic energy needs and multiple fuel-use: some reflections on energy policy and social theory’’, Proceedings of the Domestic Use of Electrical Energy Conference, Cape Town, Cape Technikon, 30 March to 1 April. Davis, M., and Ward, S., 1995. Household Energy-use Patterns in Rural Areas: the Effects of Access to Electricity, Cape Town, Energy and Development Research Centre. Eskom, 1996. SA to Z: The Decision Maker’s Encyclopaedia of the South African Consumer Market, Sandton, Eskom. Golding, A.P., 1993. The Socio-economic Impact of Electrification on a Peri-urban Community in Bophuthatswana, Pretoria, Department of Minerals and Energy. Griffin, N.J., Banks, D., Mavrandonis, J., Shackleton, S.E., and Shackleton, C.M., 1992. Household Energy and Wood Use in a Peripheral Rural Area of the Eastern Transvaal Lowveld, Pretoria, Department of Minerals and Energy. Hansmann, C., Van Gass, M., Annecke, W., Despins, P.M., and Kargas, S., 1996. Postelectrification Study of Loskop -- Appendices, Cape Town, Energy and Development Research Centre. Hofmeyr, I.-M., 1996. A Comparative Study of the Access to and Use of Electricity by Farmworker Households in the Free State, Cape Town, Energy and Development Research Centre. James, B., and Ntutela, P., 1997. Rural Households’ Response to the 2.5A Electricity Supply Option in the Tambo Village Pilot Project, Cape Town, Energy and Development Research Centre. James, B., 1997. Current-limited Supplies of Electricity in the Context of South African Rural Areas, Cape Town, Energy and Development Research Centre. Kinners, L., (Eskom), 1998. Personal communication. Kotzé, I.A., 2000. ‘‘The South African National Electrification Programme: past lessons and future prospects’’, Proceedings of ISES 2000 Utility Initiative for Africa: Rural Electrification in Africa, 17-18 April, Midrand, South Africa. Mehlwana, M., 1999. The Economics of Energy for the Poor: Fuel and Appliance Purchase in Low-income Urban Households, Cape Town, Energy and Development Research Centre. Mehlwana, M., and Qase, N., 1999. The Contours of Domesticity, Energy Consumption and Poverty: the Social Determinants of Energy Use in Low-income Urban Households, Cape Town, Energy and Development Research Centre. Palmer, R., 1999. The Impact of Domestic Electrification on Household Hydrocarbon Fuel Consumption, Pretoria, Department of Minerals and Energy. Sokopo, I., (Eskom), 2000. Personal communication. Simmonds, G., and Mammon, N., 1996. Energy Services in Low-income Urban South Africa: a Quantitative Assessment, Cape Town, Energy and Development Research Centre.
Notes
Thom, C., Davis, M., and Borchers, M., 1995. Review of South African Experience in Rural Electrification, Cape Town, Energy and Development Research Centre.
1. Definitions of rural areas in South Africa are generally both inadequate and inconsistent. These generally only exclude towns and cities which had some form of local authority under the apartheid government, and include many of the informal settlements bordering on such towns and cities. In addition, most areas in the former homelands are classified as rural areas, irrespective of the size or density of the settlements.
Thom, C., 1998. Rural Electrification Policy: Some Key Recommendations, Cape Town, Energy and Development Research Centre. Wentzel, M., Manzini, M., Mudlaudzi, C., Sehlapelo, D., and Wood, C., 1997. A post-electrification Study of the Mafefe Electrification Project, Cape Town, Energy and Development Research Centre.
2. The project was jointly funded by the South African Department of Minerals and Energy (DME), Eskom’s Technology Research & Investigations (TRI), the Norwegian Agency for International Development Cooperation (NORAD), and to a lesser extent the Development Bank of Southern Africa (DBSA). It aimed to develop policy proposals and implementation guidelines for rural electrification in South Africa.
Energy for Sustainable Development
White, C., 2000. Synthesis Report on the Social Determinants of Energy Use in Low-income Households in Four Metropolitan Areas of South Africa, Pretoria, Department of Minerals and Energy.
l
Volume IV No. 4
l
December 2000
43