5. From Energy Efficiency to Social Utility: Lessons from Cookstove Design, Dissemination, and Use

DANIEL M. KAMMEN¹

With roughly half the world's population cooking daily with the traditional biomass fuels of dung, crop residues, wood, and charcoal, efforts to disseminate improved, more efficient cookstoves are an ideal way to address a wide range of socio-economic and environmental goals. These goals include conserving energy, reducing the time spent collecting wood, expanding economic opportunities for both rural and urban families, empowering women, reducing harmful household smoke exposure, reducing forest clearing and ecological alteration, and mitigating global atmospheric pollution. Widespread dissemination and use of improved woodstoves has the potential to impact each of these objectives, and thus, has been a focal point of household development and quality of life efforts for several decades.

The statistics on the role and impact of biomass use on the energy-health-environment cycle are striking. Half the more than 3,000 million tons of wood harvested annually worldwide is used as fuel. Wood and other biomass fuels comprise 40 to 60 per cent of all energy resources, both industrial and domestic, in many Asian, Latin American, and African countries. In sub-Saharan Africa, on average, domestic cooking accounts for over 60 per cent of total national energy use; in some countries, it exceeds 80 per cent.² Some poor families spend 20 per cent or more of disposable income to purchase wood and charcoal fuels, or devote 25 per cent or more of total household labour to wood collecting.³ Biomass fuels are generally utilized at a low thermodynamic efficiency. Reliance on them thus entails a high opportunity cost for poor households, and has serious health and environmental costs as well.

Biomass is used not only for domestic purposes. Other uses include construction materials, industrial energy, and simply as cleared land for agriculture. However, the need for cooking fuel is constant; in some developing countries, it exceeds one ton per capita per year.⁴ The environmental impact of all forms of wood use - industrial, agricultural, and domestic - varies widely, ranging from areas with ecologically sustainable harvest levels⁵ to regions where population density and woodfuel demand alter the type of forest cover and biodiversity. In the most extreme cases, dramatic deforestation and erosion result.⁶

Biomass cooking fuels are often combusted inefficiently. Open "three stone" fires and some traditional stove designs generate large quantities of smoke and particulate matter, while directing only a small fraction of the resulting heat to the cooking pots. Sometimes, as little as 5 to 15 per cent of the total energy content of the fuel is utilized to heat the food.⁷

High levels of indoor air pollution create serious health problems. Air quality studies in developing countries have shown that woodsmoke exposure can often be twenty times or more the exposure limits recommended by the World Health Organization. The pollution level in homes and cooking huts can exceed those in industrial cities. High indoor air pollution, in turn, is linked to acute respiratory infection (ARI), particularly pneumonia, and other ailments. Those most continuously exposed to indoor air pollution are women - who perform over 90 per cent of domestic chores, including cooking - and children. In fact, ARI is the leading health hazard to children in developing countries, and results in an estimated 4.3 million deaths per year among the overall population.⁸ Among all endemic diseases, including diarrhea, ARI is the most pervasive cause of chronic illness.

Researchers confronted with a continuing dependence on biomass fuels, traditional woodfuel management schemes, and traditional cooking stoves have concluded that the introduction of improved cookstoves could have a dramatic development impact. Specific designs vary widely, but most cookstoves are made to consume less fuel per amount of useful energy delivered and/or to emit less pollution, which benefits the health of the user as well as the environment. Widespread dissemination and adoption of technically and culturally appropriate stoves could reduce the need for fuelwood harvesting (and thus, human impact on forest systems), while reducing human exposure to indoor air pollution.

Reducing the labour and capital required to collect or purchase cooking fuels could also provide new resources and expand economic opportunities to women throughout the developing world.⁹ Indeed, the fact that the management of biomass fuels and cooking techniques is so interrelated with energy, food, health, and the economy in developing countries has led one researcher to coin the phrase "the hearth as central system."¹⁰

Cookstoves and the Technology-for-Development Paradigm

Recent surveys have identified several hundred individual cook-stove research and dissemination projects. The scale of cookstoves programmes ranges from entirely local (often consisting of nongovernmental advocacy and implementation activities) to national initiatives reaching over 100 million homes,¹¹ to regional programmes sponsored by multinational development agencies.¹² These programmes have used a wide range of stove designs, from massive sand and clay models to a variety of portable metal and ceramic stoves. They have tested and utilized various technology transfer paradigms; these paradigms have ranged from promoting commercial cookstove mass-production and sales to training the eventual users of cookstoves, primarily women, to adapt and construct stoves for their home use.

The quality and efficiency of stoves has varied greatly, as has the success of individual projects in reaching the intended audience or developing a self-sustaining market or cook-stove industry.¹³ Because cookstove projects have such a long, diverse history, they can serve as a crucial test of the kind of small-scale, household-oriented paradigm of development long promoted by appropriate technology advocates.¹⁴ They can also serve as a model for design and dissemination efforts of such other renewable energy technologies as biogas digesters, solar ovens and food dryers, household photovoltaic systems, wind-pumps, and micro-hydro stations.

The evolving tools and continuing pitfalls of technology-transfer and implementation projects can be critically examined by analyzing the changing goals and implementation strategies of cookstove projects. This paper summarizes the changes in: a) the technology of improved cookstoves, and b) the resulting economic, health, environmental conservation, and social empowerment opportunities. Particularly notable are the following: the expanded communication and cooperation between project implementors and end-user groups, the growth of market-based technology transfer mechanisms (even in peri-urban settings, where markets did not seem initially to exist), the analysis of rural vs. urban market potential, and the design of multi-objective efforts that integrate disparate actors in the development process. The evolving understanding and interaction of these factors in the cookstove design, development, dissemination, and adoption chain provide the clearest lessons for the role of household energy in managing social change, empowerment, and development generally.

Improved Stoves: Design Engineering and Energy Efficiency

Improved cookstove designs fall into two broad categories: fixed location and portable models. The immobile stoves are commonly made from a combination of metal, clay, ceramic, or cement. These designs generally achieve energy conservation through insulation and are often "complete" stoves, with an enclosed burning chamber and multiple openings for pots (burners). Massive stove designs have been extensively tested, refined, and introduced and re-introduced in Latin America and Asia.¹⁵ Portable stoves are generally constructed of metal with clay or ceramic liners, or as formed clay "burners" that support one pot over an enclosed burning box. As is not surprising, there are a great many designs, all of which are geared to meet the same general objective:

· Maximize combustion of the fuel by keeping the temperature high and ensuring the presence of sufficient oxygen;

· Maximize radiative heat transfer from the fire to the pot(s) by keeping the pot as close to the flame as possible;

· Maximize convection from the tire to the pot(s) with a stove design that passes as much of the hot gases over the pot(s) as possible; reduce drafts;

· Maximize conduction to the food pot(s) by using an insulating material for the stove so that the heat is retained and concentrated near the pot(s);

· Maximize user satisfaction by making the stoves convenient to use (with local fuels, cooking pots, and utensils) and able to easily prepare local dishes well.

In summary, only a stove with what might be called robust efficiency will consistently save fuel under conditions of actual use. Stoves must be easy to use and fuel efficient under a variety of conditions: when it is boiling, simmering, baking, or frying food; when it is using only one opening of a large, three-pot stove; and when it is dirty or worn. Stoves evaluated in idealized laboratory conditions, very different from the environmental and practical constraints of real-world kitchens and cooking huts, have too often failed to meet this requirement of robust efficiency. Cookstoves are workhorses, not racehorses, and must be designed accordingly.

Early cookstove projects were heralded as the solutions to a tremendous array of social, economic, and environmental ills - from deforestation to the oppression of women. Although improved cookstove efficiency and household energy security can lead to improvements in all of these areas, evaluations of early projects were generally disappointing.¹⁶ Many of the early projects failed for both technical and social reasons. As one analyst has stated: "Early improved cookstoves were often designed by development workers with a great deal of zeal and enthusiasm, but little technical background. Under the banner of 'appropriate technology,' new designs were quickly labeled 'improved stoves' and construction manuals prepared, without prior serious scientific research."¹⁷

Advocates and researchers involved in early stove projects fell into the trap of equating "appropriate" technology with "simple" technology. The first four design factors in the list above, essentially thermodynamic criteria, have proved difficult to achieve.¹⁸

Many early programmes expected to see efficiency gains of 75 per cent or more.¹⁹ However, these expectations were based on tremendously idealized cooking conditions never realized in the field. Many new stoves were expensive and difficult to use, and degraded rapidly with use. The wood savings realized in actual home settings tend to follow a distribution, as shown below:²⁰

households	wood savings (%)
15%	40 +
25%	25-40
25%	10-25
10%	0
10%	-10 to -25
15%	no longer in use

In aggregate, these data correspond to fuel savings of about 20 per cent per stove, an impressive and important reduction, although considerably less than the early claims of some wood-stove projects.²¹ Construction of massive, in-place design stoves in each house (i.e., built one at a time without standardization or economies of scale) often resulted in uneven quality and efficiency. For all these reasons, many new stove designs and dissemination programmes failed the test of "robust efficiency."

At the same time, claims that traditional stoves were only 5 to 10 per cent efficient neglected the many other benefits they provided: lighting, space heating, ease of use, and versatility. In fact, under conditions of shielded, carefully tended use, three-stone fires can reach efficiencies of 20 per cent.²²

In many early projects, the end-users, generally women, were not involved in initial discussions, feedback, or training programmes; these were inappropriately targeted to men or to extension workers who rarely cooked food themselves.²³ At the same time, many of the basic lessons of commercial enterprises were lost, partly because stove projects were categorized as development aid and assistance. Little market research was initially undertaken to determine the most pressing concerns of the women who would use the stoves, the local perceptions of fuel scarcity or abundance, or even the suggestions of local communities about cookstove design or the kind of programmes that could make them available. The combination of the oversell of improved cookstoves and the under appreciation of the reliability and versatility of traditional methods meant that the fuel savings in some early efforts amounted to little or nothing.

The disappointing results of the early efforts led to re-evaluation of the engineering and design of stoves, to greater end-user participation, and to far more rigorous and realistic measures of actual stove performance. This more pragmatic approach then began to yield broadly useful efficiency and cost comparison data for second-generation stove projects.

A series of practical measurements of actual stove efficiencies conducted in West Africa is particularly instructive. A useful measure of stove performance is the percentage of heat utilized (PHU), a ratio of the energy utilized/energy expended. It is obtained by boiling and then continuing to cook a volume of water while measuring the total fuel combusted, the volume of water boiled, and water boiled off - all in a series of comparable field settings. Table 5.1 shows the PHU for a variety of stoves commonly used in West Africa.

For the stoves listed in Table 5.1, prices range from essentially no cost for a three-stone fire (except for the cost of the pots) to more than $20. At the prevailing cost of wood, these prices (omitting the three-stone fire) correspond to amortized payback times ranging from one month to over a year.²⁴ Fuel savings compared to traditional stoves are as much as 60 per cent for the expensive CATRU-A stove, and as little as 5 per cent for some simple and inexpensive designs.

Stove dissemination programmes should not be undertaken lightly. They involve significant financial investment by both users and donors, and require a long-term commitment of research, training, and support services to be successful. However, one of the most promising aspects of such programmes is the growing realization that reducing woodsmoke exposure has a wide range of important health benefits. The impact of stove programmes on both pressing energy and health problems can be used effectively to broaden the base of support in providing the needed long-term resource commitment necessary to generate successful, self-supporting programmes and industries.

Table 5.1 - Efficiency Comparison of Several West African Improved and Traditional Stoves

Stove	Description	PHU	% increase in PHU over:		% decrease in wood use over:
			3-Stone	Metal	3-Stone	Metal
Three Stone	Pot supported by three stones over open fire	10.2	0	-	0	-
Metal Stove	Simple 1-pot metal stove	14.5	42	-	29	-
SIM	Sand insulation placed around the simple metal stove	16.1	58	11	36	10
Sota	Clay shell stove around a single pot	18.4	80	27	44	21
AIDR	3-pot partially insulated stove, without chimney	10.9	7	-	6	-
GS Chula	Insulated 2-pot lorena-type stove with chimney	15.2	49	5	32	5
Nouna	Brick and cement lorena-type stove	15.3	50	6	33	5
CATRU-B	Lorena-type, aluminum top plate and matched pots	17.2	69	19	41	15
CATRU-A	CATRU-B with improved chimney	25.9	154	79	61	43

Note: The PHU (or percentage of heat utilized) is based on a combination of the initial boiling and sustained cooking (simmering) phase - thus providing a proxy for realistic cooking conditions. The PHU values reported here are averaged over five or more measurements per stove. Each stove is compared here to both a traditional three-stone fire and a metal stove. The Three-Stone, Metal, SIM, and Sota stoves all accommodate one pot only; the AIDR stove has openings for three pots; and the remaining four stoves all have two openings. Wood savings over traditional fires range from minimal (6 per cent) to dramatic (61 per cent).

Source: Data derived from TS. Wood, Laboratory and Field Testing of Improved Stoves in Upper Volta (Mt. Rainier, MD: VITA, 1981); S. Connors, "Wood-Conserving Cookstoves: A Short Primer for the Design and Implementation of Woodstoves and Woodstove Projects," Peace Corps/Benin (1987), mimeo; and S. Baldwin, H. Geller, G. Dutt, and N.H. Ravindranath, " Improved Woodburning Cookstoves: Signs of Success," Ambio 14(1985).

The Cooking Environment, Smoke Exposure, and Health Risks

The connection between wood use, cooking, and the epidemiology of respiratory and other illnesses is a topic of active current research. However, a consistent pattern linking energy, environment, and health has already become alarmingly clear. Biomass fuels provide 90 to 95 per cent of domestic energy in sub-Saharan Africa, most of it for cooking." Combustion of these fuels in confined, often unventilated, indoor areas and at low thermodynamic efficiency leads to high concentrations of smoke and pollution. To evaluate how effective improved cookstoves and household energy management are in mitigating these harmful health conditions, it is necessary to consider the entire food preparation cycle, including energy and environmental management, and household risk and economic decision-making.

The food preparation process is one of the most important health and development issues facing poor countries. Biomass cooking on traditional stoves is a major source of concentrated air pollutants, including respirable particulate matter, carbon monoxide, nitrogen oxides, benzene, formaldehyde, benzo(a)pyrene, and aromatics.²⁶ Particulates seem to be the primary culprit in smoke-related illnesses.²⁷ In some places, the pollutant exposure levels associated with indoor biomass burning in developing countries is many times greater than accepted health standards. U.S. standards, for example, call for maximum particulate concentrations of 250 micrograms per cubic metre (not to be exceeded more than once a year). However, in developing countries, people are routinely exposed to indoor particulate concentrations many times that high (see Table 5.2). These levels rival or even exceed the pollution levels found in the most polluted industrial cities.²⁸ Women and children are particularly affected, since cooking smoke is confined to indoor settings, where they are exposed for extended periods of time.

The living conditions that expose people to high levels of indoor air pollution have been well documented in Africa. The majority of sub-Saharan Africans live in rural areas; Kenya, for example, is only about 20 per cent urban. Family homes generally consist of multi-use buildings, where the same room or few rooms are used for cooking, sleeping, and working. In many cases, the total indoor volume is less than 40 cubic metres; in some (such as the Maasai homes in Kenya), indoor air volumes may be half that. Rural homes often have minimal ventilation; when people cook, they may close the door or, when they exist, plug the windows with cloth.²⁹ Ventilation is further reduced during rainy seasons, cold spells, and at higher elevations.³⁰

Under these circumstances, pollutant concentrations resulting from cooking can easily build to unhealthy levels and remain that way over the course of a day. Compounding the problem is the type of cooking practiced in much of Africa. In many countries, the staple foods are grain and legume combinations that require long cooking times. In Kenya, for example, preparation of the staple maize and bean dish (ugali, a hardened corn meal, and irio, a simmered mixture of several beans) requires several hours of softening and slow cooking that can consume wood at the relatively high rates of 1.5 -3.0 kilograms per hour.

High pollution levels are not limited to rural areas. The close quarters of urban slums, and even the minimal spaces sometimes allocated to household servants in more affluent households, and the heavy use of traditional cooking fuels, notably charcoal, all contribute to urban pollution. Poverty and overcrowding can increase the ambient pollution concentration over entire neighbourhoods, where woodsmoke mixes with photochemical smog.

Table 5.2 - Indoor Particulate Concentrations In Developing Countries, Summary of Selected Studies

Location	Conditions	Number of measurements	micrograms per cubic metre
Africa
Kenya (1972)	Night: highlands	5	2,700-7,900
Kenya (1972)	Night: lowlands	3	300-1,500
Kenya (1987)	24 hour exposure	64	1,200-1,900
Kenya (1993)	Unvented hut: cooking	4	1,346-37,000
Nepal (1988)	24 hour exposure	18	400-2,400
Gambia (1988)	24 hour exposure	36	800-3,400
Zimbabwe (1990)	While cooking	20	100-4,900
Asia
India (1983)		56	6,800
India (1988)		129	4,700
India (1988)		44	3.600
India (1988)		165	3,700
Nepal (1986)		49	2,000
Nepal (1988)	Traditional stoves	20	8,200
Nepal (1988)	Improved stoves	20	3,000
US 24-hour standard (not to be exceeded more than once/year)			250
US Annual Urban Levels			60

Note: These measurements of particulate concentrations, in micrograms per cubic metre, were obtained under a range of conditions - during cooking, as a daily average, etc. They are not directly comparable, but give a feel for the range of concentrations. The measured concentrations are consistently far above the U.S. one-time exposure standard and the annual average. Measurements are for particles having a diametre greater than 10 microns.

Source: Data from P. Young and K. Wafula, "Smoked Maasai," ITDG/KENGO (London, 1993), mimeo; K.R. Smith et al., "Greenhouse Gases from Biomass and Fossil Fuel Stoves in Developing Countries: A Manila Pilot Study," Chemosphere 26 (1992), pp. 479-506; and K.R. Smith, "The Health Impact of Cookstove Smoke in Africa," in African Development Perspectives Yearbook 3 (Muenster: Lit Verlag, 1994).

Health Effects of Biomass Burning

Numerous studies demonstrate a consistent positive correlation between exposure to smoke from indoor biomass burning and acute respiratory infection and chronic lung disease (Table 5.3). Ongoing research is attempting to determine the precise dose response relationship;³¹ in the meantime, however, long-term exposure to smoke from biomass combustion has been observed to elevate the risk of a child developing ARI by 100 to 400 per cent.³² Less well documented studies have linked woodsmoke to an increased incidence of eye infections, low birth weight, and cancer. Health clinics and mobile physician programmes in developing countries routinely treat both children and adults for serious bums resulting from direct contact with cooking fires.

Table 5.3 - Summary of Studies of Household Biomass Combustion and Acute Respiratory Infection (ARI) Among Infants

Study (year)	Case/Control (ARI determination)	Exposure & measurements	AM relative risk (95% confidence interval)
1. Natal, South Africa (1980)	132/18	Cookfire smoke X-Ray and ARI cases	4.8 (1.7-13.6)
2. Basse, The Gambia (1989)	587	Child carried by mother Reported breathlessness	2.8 (1.3-6.1)
3. Maragua, Kenya (1986,1990)	36	Woodsmoke exposure Particulate monitoring	N/S
4. Marondera, Zimbabwe (1990)	244/500	Open fire cooking ARI hospitalizations	2.2 (1.4-3.3)
5. Basse, The Gambia (1991)	587	Child carried by mother Reported breathlessness	Girls: 6.0 (1.1-34.0) Boys: N/S

Notes: The studies looked at ARI among infants, 0 to 59 months of age. The relative risk (incidence ratio to that in the control group) is shown with the uncertainty range at the 95 per cent confidence interval (CI). N/S= not statistically significant.

Source: K.R. Smith, "The Health Impact of Cookstove Smoke in Africa," in African Development Perspectives Yearbook 3 (Muenster: Lit Verlag, 1994).

The extent of the woodsmoke health crisis in developing countries is beginning to gain international attention. The World Health Organization estimates that 1.5 billion people live under conditions of unhealthy air. Four to five million childhood deaths are attributed to acute respiratory infection every year. Regional health reports provide another measure of the pervasive nature of the health hazard. For example, Laikipia District in Kenya is a relatively prosperous mixed agricultural and farming region. However, even here, annual morbidity data from district hospitals and clinics show that respiratory infections head up the list of most commonly reported diseases, accounting for a third of all afflictions reported and diagnosed. Eye infections, also linked to woodsmoke exposure, are on the list of top ten diseases as well.

Although more research is needed to determine the precise relationship between woodsmoke dose and morbidity response, it has become clear that improved, low-emission cookstoves are as important to improving health conditions as to conserving fuel. Additional dose-response information is critically needed to develop guidelines for predicting at-risk groups and designing preventive health care programmes; it could also be used to evaluate the cost/benefit tradeoffs of various types of education and cookstove design and dissemination projects. For example, the preliminary exposure and morbidity data in Tables 5.1-5.4 provide evidence of a hazard, but are insufficient to delineate what degree of smoke reduction will generate significant improvements in human health; this is the classic "how clean is clean enough" problem in environmental science and engineering. Although they have drawbacks, cost/benefit evaluations are particularly useful in making renewable energy and energy efficiency projects mainstream rather than additional activities. All too often, renewables are still considered an unnecessary luxury. However, when the results of renewable energy dissemination efforts can be measured in terms of cost and benefit, resistance becomes significantly less.³³

Cookstove Technology and Indoor Air Pollution

As the health risks associated with biomass cooking become increasingly clear, the case for continued and expanded improved stove projects is strengthened. Improved cookstove and outreach, education, demonstration, and construction efforts can contribute to reducing harmful woodsmoke exposure in a number of ways: a) by improving venting of combustion gases through use of a flue or chimney, b) by improving combustion efficiency through better stove designs, and c) by encouraging use of cleaner burning fuels and more advanced stove designs - a process called moving up the "energy ladder."

By reducing the time needed to collect (or to purchase) cooking fuels and combining health gains with improvements in household fuel, cookstove programmes can meet economic efficiency, environmental conservation, and quality of life objectives. Transition to more advanced stove designs and cleaner fuels can significantly reduce indoor air pollution (see Table 5.4).

Table 5.4 - Concentration of Carbon Monoxide (CO) From Indoor Biofuel Combustion, Various Fuel and Stove Combinations, Kenya

Fuel and Stove Combination	No. of Measurements	[CO] (parts per million by volume)
Dung/Traditional Stove	25	220-760
Wood/Traditional Stove	38	140-550
Charcoal/Traditional Stove	14	230-650
Charcoal/Improved Stove	22	80-200
Kerosene Fuel and Stove	8	20-65
WHO 1-hour exposure standard		46

Notes: The carbon monoxide measurements art for a typical range of concentrations measured one metre above the stove during food preparation. They are instantaneous values, not time averages, collected in 1992 and 1993 from homes in southern and eastern Kenya; the homes are culturally relatively homogenous, but economically stratified. The energy ladder "increases" from the top to the bottom of this list of stove/fuel combinations.

The "energy ladder" concept is fundamental to efforts to understand household energy decision-making.³⁴ It is the household or micro-economic corollary to what happens at the macro-economic level, as countries industrialize and move from traditional biomass to commercial fossil-fuel-based economies. Under this hypothesis, as households become more affluent, they move from simple stoves and inexpensive fuels to more sophisticated, convenient, and costly fuels and stoves. The ladder climbs from dung or crop residues combusted in three-stone fires, to wood or charcoal use in metal or improved stoves, to kerosene wick or pressure systems, and finally to propane, liquid petroleum, and electric appliances.

The health and energy impact of climbing up this ladder can be dramatic. Simple biomass stoves may use six or seven times as much fuel as a modern stove, and release 50 times more pollutants than a gas stove used to prepare the same quantity of food.³⁵ The poorest segments of society, thus, not only are exposed to higher levels of pollution, but must also spend a greater share of household income and resources to cook the same meal.

Some aspects of moving from traditional to improved stoves - or from dung and wood to charcoal, kerosene, and other fuels - may actually introduce new risks. For example, although an improved, charcoal-burning stove may emit less total pollution than a traditional stove, carbon monoxide may still be produced at higher levels than those recommended by the World Health Organization. Since carbon monoxide is odourless and colourless, there may be acute risk of poisoning without the warning signs of coughing and tearing that are associated with hydrocarbons and particulates in woodsmoke.

In many ways, the energy ladder is too simplistic a view of energy decision-making. In Africa, although families acquire new cooking technologies as resources permit, they rarely abandon the more traditional metal stoves, or even three-stone fires. Instead of moving up an energy ladder or emphasizing energy efficiency, they pursue a more eclectic pattern of acquiring energy security. Improved cookstove programmes designed to augment positive social change must work with, rather than against, this process. The focus of programmes to provide small-scale domestic and community-based energy resources must be to give households a diverse set of energy services. End-users of energy services must have the opportunity to evaluate, and choose among a range of alternatives that fit their specific needs, rather than be presented with a single technology in a "take it or leave it" development project.

Improved Cookstove Programmes in East Africa

Except for the large cookstove programmes of India and China, the varied efforts to introduce improved stoves to East Africa have been the most extensive.³⁶ The most widely used model in East Africa is the Kenya Ceramic Jiko (KCJ), of which more than 700,000 are in use today. Proponents of the KCJ report an adoption rate of 56 per cent among urban households, where most efforts have been focused, and an impressive 16 per cent nationwide.³⁷ The problems encountered and the lessons learned in East Africa are now being incorporated into cookstove programmes worldwide.

Improved cookstove design and dissemination programmes in East Africa date from the 1970s. Financial and technical support was provided by a variety of bilateral and multilateral development agencies and non-governmental organizations, with financial contributions ranging from several thousand to over four million dollars. Local non-governmental organizations and community/activist groups, as well as government agencies, were involved in research, development, support, and dissemination.³⁸

Both the strengths and the problems associated with cookstove programmes can be traced to the diversity of actors, methods, and resources brought to bear on the complex interplay among stove performance and cost, user needs and resources, how the method was introduced, and the local stove/fuel economy

Some early efforts in Kenya generally failed the "robust efficiency" criteria outlined earlier; they were poorly designed, had overly optimistic expectations of the amount of wood and charcoal savings, and the stoves were inconvenient to use.³⁹

In addition, they focused on a difficult market - on the rural poor, who do not purchase fuel, rather than on the more affluent urban population, who already regularly purchased both fuel and stoves. After learning from these mistakes, the Kenya Ceramic Jiko Programme is now considered a model programme with the most popular stove. The KCJ is constructed by several hundred distinct commercial producers; over 13,000 KCJ stoves are sold each month in stores and markets throughout urban areas. Critical to Kenya's success are the efforts of many agencies - bilateral, multilateral, non-governmental, and governmental - to promote and popularize the stove.

Today, the various cookstove efforts in Kenya include the KCJ as well as other designs, and are targeted at more than simply the urban population. They include the Kuni Mbili (two-stick) stove, which has a larger burning box than the KCJ to accommodate the primary rural fuel of wood, and the Maendeleo or Upesi (quick) stove, which consists of an inexpensive manufactured liner built into the user's existing hearth (see Table 5.5). These diverse stove projects provide flexibility and meet diverse needs, fuel requirements, and household incomes and preferences.

As discussed by a World Bank research team, the experience of transferring the KCJ to neighbouring Rwanda provides an important example of technology transfer, evolution, and choice. Laboratory testing of a number of stove designs preceded a large-scale field trial in a Kigali neighbourhood. The KCJ was expected to prove popular, but it did not accommodate Rwanda pots or provide close access to the flame for grilling. A modified version of the KCJ, the Rondereza ("to save") was rapidly adopted by over 25 per cent of the population.⁴⁰

In addition to the percentage of heat utilized (Table 5.1), the direct financial and quality of life benefits are the crucial test of success for cookstove programmes. In Nairobi, average household charcoal savings as a result of KCJ adoption are 0.18 kilograms per person per day, or over 600 kilograms per family per year; this is equivalent to 1,170 Kenyan shillings (US $64.70 at the 1991 conversion rate). In Rwanda, charcoal use declined by 390 kilograms per year, equivalent to US $84.⁴¹

Perhaps the most important aspect of the Kenyan cookstove experience is in the institutional capacity developed by indigenous organizations such as the Kenya Energy and Environment Organization (KENGO) and the Foundation for Woodstove Dissemination (FWD). They have become sources of regional expertise in many facets of improved cookstove design, dissemination, popularization, and follow-up. Along with national and international organizations, they have greatly facilitated the further spread of cookstove programmes in Africa (see Table 5.6).

Table 5.5 - Comparison of Kenyan Improved Stove Designs and Programmes

Stove Description	Institutions Involved	Construction	Dissemination	Cost (US$)
Kenya Ceramic Jiko Lightweight metal cladding and ceramic liner: separate firebox and pot rest sections. Primarily for use with	Kengo, FWD a number of and international NGOs; USAID; Ministry of Energy	By over 200 artisans and jua kali copperatives. These groups are manufacturing about 13,600 stoves/month > 700,000 so far	Pre-existing Commercial channels: artisans sell stoves to supermarkets or retailers, or direct to the consumers. 90% sold in urban areas	Varies by stove size, quality, and by vendor: $3 and up)
Kuni Mbili Stove Larger version of KCJ with expanded firebox to more easily accommodate wood fuel	KENGO, the Bellerive Foundation and Rural Technology Enterprises (RTE)	By some of the same artisans as the KCJ; manufacturing about 1,000 stoves/month > 20,000 so far	Programme in the demonstration phase: targeted at rural households that collect or buy wood	Subsidized retail price; true cost of $6-7 & up
Maeneleo, or Upesi Stove standardized liner: baked and then installed with mud and stone surrounding. Variant of Sota stove.	ITDG, GTZ, Ministry of Energy; Home Economics Office of Ministry of Agriculture	Multiple approaches: artisans and women in the informal sector produce liners; installation by extension workers or end-users	Programme in the demonstration phase: transition from extension-based to commercial project underway. Focus on rural households	Liners cost about $1; installed by rural extension teams or end-users

Notes: Jua kali (literally "fierce heat") refers to a loose network of cooperative shops that undertake projects from automotive repair to construction of KCJs. The jua kalis were targeted for early training in cookstove construction and maintenance; it is now a profitable and independent business.

Household Energy Management and Social Change

The fundamental role that energy management plays in family health, nutrition, economic opportunity, and environmental conservation means that improvements in cookstove technology and efforts to disseminate new technologies offer not only opportunities to climb the energy ladder, but also opportunities for positive social change.

The history of cookstove projects shows that energy efficiency is a necessary, but not a sufficient, condition for a new technology to succeed. The part technical, part social criterion of robust efficiency remains a difficult design standard for technologies and for development programmes. However, improved cookstove programmes have benefited greatly from efforts to adhere to this principle by combining intensive interaction and feedback between stove designers and end-users, long project follow-up times, and greater reliance on market and commercial forces.

One important lesson from cookstove efforts to date is that choice and selection - of both technology and implementation methods - are fundamental to meeting the diverse needs of the intended end-users. To some degree, this simply reflects the strength that technology transfer efforts experience when they are designed to support individuals and communities in achieving a diverse set of objectives that are locally determined, and not imposed. An example would be a cookstove training center in which various stove designs and different approaches to managing community workshops are discussed with community leaders, who then select and refine the methods they consider best suited for their particular local conditions.

Funding agencies such as the World Bank, the Global Environment Fund, as well as national and private donors, must tolerate and even encourage the short-term inefficiency of some early cookstove programmes in order to develop a strong indigenous resource base and breadth of experience.⁴² The mainstream development community has much to learn from the "small is beautiful" or "appropriate technology" approach that emphasizes local knowledge and control of projects, as well as the provision of financial and infrastructure resources that households and poor communities can manage and direct. This approach, however, requires a long-term investment in education that may not initially appear to be warranted in simple cost/benefit terms.

One possible approach that could be implemented by virtually any development organization is to provide small-scale decentralized funding to groups of women, households, or communities to experiment with, and evaluate, a range of energy technologies, including improved cookstoves, solar ovens and food dryers, propane and kerosene stoves, windmills, micro-hydro generators, and photovoltaic systems.⁴³ The resulting technology resource centers, perhaps under collaborative management of the funding agencies and local groups dedicated to researching and promoting new technologies, could also provide wide-ranging technical expertise. This approach of providing a broad selection of technical and managerial resources would ensure that end-users can choose what works for them, rather than what worked best in a laboratory, or what was designed in a grand scheme somewhere else. It promotes technology choice, not dependence.

Table 5.6 - Estimated Number of Improved Stoves in Selected Sub-Saharan Countries, 1994

Kenya	780,000
Burkina Faso	200,000
Niger	200,000
Tanzania	54,000
Ethiopia	45,000
Sudan	28,000
Uganda	25,000
Rwanda	20,000
Zimbabwe	10,880
Malawi	3,700
Botswana	1,500

Source: P. Wickramagamage, Improved Cookstove Programs in East and Central Africa, Draft ESMAP Report (Washington: World Bank, 1991); S. Karakezi, "Disseminating Renewable Energy Technologies in Sub-Saharan Africa," Annual Review of Energy Environment 19 (1994), pp. 387-421; and Stephen Karekezi, "Renewable Energy-Technologies in Sub-Saharan Africa: Case Examples from Eastern and Southern Africa," background brief prepared for seminar, Woodrow Wilson School of Public and International Affairs, Princeton University, April 19,1995.

Although an environment that promotes new, and thus, inherently riskier, technologies is important, market forces and commercial strategies are the final test of whether a project is viable. There is no reason that commercial ventures and tests of market success should not be regarded as allies of "appropriate technology" development efforts - particularly when preceded by a sufficient period of scientific and social "research and development." The most successful cookstove dissemination projects, for example, have capitalized on the market's potential contribution to widespread dissemination: they focused initially on areas where fuels and stoves are already purchased, and then moved into less commercial settings through technical assistance and training, with minimal use of incentives or subsidies.

Ironically, the problem of how to most effectively provide energy services exists under both appropriate technology and mainstream approaches to development. Market-oriented, technology-commercialization development programmes offer little realistic "trickle down," and rural extension efforts have yet to move broadly from subsidy to self-sufficiency. An important goal for future programmes will be to integrate the nurturing, capacity-building features of appropriate technology efforts with the market reach and benefits of scale that can be achieved in programmes that spur commercial interest in small-scale and household energy technologies.

NOTES

¹ Daniel M. Kammen is Assistant Professor of Public and International Affairs, as well as Co-Chair of the Science, Technology, and Public Policy Program, at the Woodrow Wilson School of Public and International Affairs, Princeton University.

It is a pleasure to thank Barbara Saatkamp for research assistance, and Stephen Karekezi, Omar Masera, and Kirk R. Smith for project information and invaluable feedback.

² J. Woods and D.O. Hall, Bioenergy for Development: Environmental and Technical Dimensions (Rome: Food and Agriculture Organization, 1994).

³ J. Pasztor and L.A. Kristoferson, Bioenergy and the Environment (Boulder, CO: Westview Press, 1990).

⁴ S.R. Nkonoki and B. Sorensen, "A Rural Energy Study in Tanzania: The Case of Budilya Village," Natural Resources Forum 8 (1984), pp. 51-62; and J. Goldemberg, T.B. Johanssen, A.K.N. Reddy, and R.H. Williams, "Basic Needs and Much More With One Kilowatt Per Capita," Ambio 14 (1985), pp. 190-200.

⁵ N. Bradley, Women, Woodfuel, and Woodlots (London: Macmillan Ltd., 1991).

⁶ B. Agarwal, Cold Hearths and Barren Slopes (New Delhi: Allied/Zed Books, 1986); G. Foley P. Moss, and L. Timberlake, Trees and Stoves: How Much Wood Would a Woodstove Save if a Woodstove Could Save Wood? (Washington: Earthscan/IIED, 1984).

⁷ K. Openshaw, "A Comparison of Metal and Clay Charcoal Cooking Stoves," paper presented at the Conference on Energy and Environment in East Africa, Kenya Energy and Environmental Organization (KENGO), Nairobi (1979), mimeo; K. Openshaw, "The Development of Improved Cooking Stoves for Urban and Rural Households in Kenya," Report of the Beijer Institute (Stockholm: Royal Swedish Academy of Sciences, 1982); S. Connors, "Wood-Conserving Cookstoves: A Short Primer for the Design and Implementation of Woodstoves and Woodstove Projects," Peace Corps/Benin (1987), mimeo; M. Jones, Energy Efficient Stoves in East Africa: An Assessment of the Kenya Ceramic Jiko (Stove) Programme, Report No. 89-01, Office of the Energy Bureau for Science and Technology, and Regional Economic Development Services Office for East and Southern Africa (Washington: U.S. Agency for International Development, 1989); Food and Agriculture Organization (FAO), Guidelines for the Monitoring and Evaluation of Cookstove Programmes (Rome: FAO, 1990); I. Bialy, Evaluation Criteria for Improved Cookstove Programmes: The Assessment of Fuel Savings, Draft Energy Sector Management Assistance (ESMAP) Report (Washington: World Bank, 1991). ESMAP Reports can be obtained from Dr. K.R. Smith, Director, Improved Biomass Cookstove Program, Program on Environment, East-West Center, Honolulu, HI 96848.

⁸ N.M.H. Graham, "The Epidemiology of Acute Respiratory Infections in Children and Adults: Global Perspectives," Epidemiological Review 12 (1990), pp. 149-78; K.R. Smith, "The Health Impact of Cookstove Smoke in Africa," in African Development Perspectives Yearbook 3 (Muenster: Lit Verlag, 1994), pp. 417-34.

⁹ S. Raju, Smokeless Kitchens for the Millions (Madras, India: Christian Literature Society 1953); S. Baldwin, H. Geller, G. Dutt, and N.H. Raindranath, "Improved Woodburning Cookstoves: Signs of Success," Ambio 14 (1985), pp. 280-87; Smith and Ramakrishna, Improved Cookstove Programs: Where Are We Now?, ESMAP Report No. 2 (Washington: World Bank, 1991); and D.F. Barnes, K. Openshaw, K. Smith, and R. van Plas, What Makes People Cook with Improved Biomass Stoves? Technical Paper No. 242, Energy Series (Washington: World Bank, 1994).

¹⁰ K.R. Smith, The Hearth As System Central, Draft ESMAP Report (Washington: World Bank, 1991).

¹¹ K.R. Smith, G. Shuhua, H. Kun, and Q. Daxiong, "100 Million Biomass Stoves in China: How Was It Done?" World Development 18 (1993), pp. 941-61.

¹² World Bank, World Development Report: Development and the Environment (New York: Oxford University Press, 1992).

¹³ H. Krugmann, Review of Issues and Research Relating to Improved Cookstoves, IDRC-MR152e (Ottawa: International Development Research Centre, 1987).

¹⁴ E. Boserup, Women's Role in Economic Development (London: Allen and Unwin, 1970); F. Schumacher, Small is Beautiful: Economics As If People Mattered (New York: Harper and Row, 1973); B. Agarwal, "Diffusion of Rural Innovations: Some Analytical Issues in the Case of Wood-Burning Stoves," World Development 11 (1983), pp. 359-76; B. Agarwal, Cold Hearths; M.B. Anderson, "Technology Transfer: Implications for Women," in C. Overholt et al. (eds.), Gender Roles in Development Projects (Kumarian Press, 1985), pp. 57-78; L. Fortman and D. Rocheleau, "Women and Agroforestry: Four Myths and Three Case Studies," Agroforestry Systems 2 (1985), pp. 253-72; K.R. Smith, "Biomass, Combustion, and Indoor Air Pollution: The Bright and Dark Sides of Small Is Beautiful," Environmental Management 10 (1986), pp. 61-74; and V. Shiva, Staying Alive: Women, Ecology and Development (London: Zed Books, 1989).

¹⁵ S.F. Baldwin, Biomass Stoves, Engineering Design, Development, and Dissemination (Arlington, VA: Volunteers in Technical Assistance, 1987).

¹⁶ The demand for domestic fuelwood leads to local shortages and often long transportation distances, but it is not a major contributor to deforestation. Commercial logging and agricultural land conversion and alteration are the primary causes of deforestation. See O. Davidson and S. Karekezi, A New Environmentally Sound Energy Strategy for the Development of Sub-Saharan Africa (Nairobi: Africa Energy Policy Research Network, 1992); S. Karekezi, "Disseminating Renewable Energy Technologies in Sub-Saharan Africa," Annual Review Energy Environment 19 (1994), pp. 387-421; and D.R. Ahuja, "Research Needs for Improving Biofuel Burning Cookstove Technologies: Incorporating Environmental Concerns," Natural Resources Forum 14 (1990), pp. 125-34.

¹⁷ H. Krugmann, Review of Issues.

¹⁸ For an excellent analysis of the design and evaluation of cook-stoves based rigorously on the principles of heat transfer and materials science, see S.F. Baldwin, Biomass Stoves, Engineering Design, Development, and Dissemination.

¹⁹ D.F. Barnes et al., What Makes People Cook.

²⁰ These data are an aggregate of several follow-up studies conducted 12-18 months after the various dissemination programmes concluded oven construction or sales sessions. I thank Kirk R. Smith (private communication) for this summary.

²¹ H. Krugmann, Review of Issues.

²² G. Foley et al., Stoves and Trees.

²³ P. Stamp, Technology, Gender, and Power in Africa (Ottawa: International Development Research Center, 1989); I. Bialy, "Evaluation Criteria for Improved Cookstove Programmes"; and P.N. Bradley, Women, Woodfuel, and Woodlots.

²⁴ Based on urban wood or charcoal costs of U.S. $0.30 to $0.40 per family per day. See S. Baldwin et al., "Improved Woodburning Cookstoves."

²⁵ P. Wickramagamage, Improved Cookstove Programmes in East and Central Africa, Draft ESMAP Report (Washington: World Bank, 1991).

²⁶ K.S. Smith, Biofuels, Air Pollution, and Health (New York: Plenum, 1987); D.F. Barnes et al., What Makes People Cook.

²⁷ N.M.H. Graham, "The Epidemiology of Acute Respiratory Infections"; and D.F. Barnes et al., What Makes People Cook.

²⁸ FAO, Guidelines for Monitoring.

²⁹ D.M. Kammen and B. Fayemi Kammen, "Energy, Food Preparation and Health in Africa: The Roles of Technology, Education and Resource Management," African Technology Forum 6(1) (1992), pp. 11-14; P. Young and K. Wafula, "Smoked Maasai," ITDG/KENGO (London, 1993), mimeo; and K.R. Smith, "The Health Impact of Cookstove Smoke in Africa."

³⁰ K.R. Smith, Biofuels, Air Pollution, and Health.

³¹ C. Barnes, J. Ensminger, and R O'Keefe (eds.), Wood, Energy, and Households: Perspectives on Rural Kenya (Stockholm: Beijer Institute, 1984); G.J. Wells, X. Xu, and T.J. Johnson, Valuing the Health Effects of Air Pollution: Application to Industrial Energy Efficiency Projects in China, Chinese Government/UNDP/World Bank Study (Washington: World Bank, 1994).

³² K.R. Smith, "The Health Impact of Cookstove Smoke in Africa."

³³ An important example is the concept of cost per "disability adjusted life year (DALY)," a somewhat problematic measure of the impact of various health and development infrastructure interventions used by the World Bank. The consequences of ARI are felt not only when the disease strikes, but may cause morbidity and mortality decades after the exposure period, causing the loss of many DALYs per person afflicted. The combination of the low cost of stoves and the large DALY impact of woodsmoke exposure makes improved cookstove programmes an excellent investment when measured in cost/benefit terms. For a discussion of the DALY concept, see World Bank, World Development Report, 1993: Investing in Health (New York: Oxford University Press, 1993).

³⁴ R. Hosier and J. Dowd, "Household Fuel Choice in Zimbabwe: An Empirical Test of the Energy Ladder Hypothesis," Resources and Energy 9 (1987), pp. 347-61.

³⁵ S. Connors, "Wood-Conserving Cookstoves."

³⁶ In India, 8 to 10 million improved cookstoves have been disseminated; in China, over 120 million. Both provide lessons for projects elsewhere. A recent World Bank report comparing the two programmes concludes that the Indian programme is problematic; it is characterized by central administration, government involvement in stove production, and high oven costs. The more successful Chinese programme is characterized by government involvement only in dissemination and promotion, leaving stove manufacture and sale generally unsubsidized. (Editor's Note; For a detailed discussion of China's stove programme, see chapter 7, this volume).

³⁷ F.M. Njorge, "An Overview of Improved Stove Dissemination Programmemes in Kenya," Second International Workshop on Stove Dissemination, Antigua, Guatemala, October 1987 (unpublished mimeo); Barnes et al., Wood, Energy, and Households; S. Karakezi and D. Walubengo, Household Stoves in Kenya: The Case of the Kenya Ceramic Jingo (Nairobi: KENGO, 1987); P Wickramagamage, Improved Cookstove Programmes in East and Central Africa; S. Pandey, Criteria and Indicators for Monitoring and Evaluation of the Social and Administrative Aspects of Improved Cookstove Programmes, Draft ESMAP Report (Washington: World Bank, 1991); and D. Walubengo, "Cooking Stoves for Commercial, Sustainable Production and Dissemination in Africa?" Boiling Point 30 (1993), pp. 16-19. Boiling Point is available from the Intermediate Technology Development Group (ITDG), London.

³⁸ P Wickramagamage, Improved Cookstove Programmes in East and Central Africa.

³⁹ E. Hyman, "The Strategy of Production and Distribution of Improved Charcoal Stoves in Kenya," World Development 15 (1987), pp. 375-86; P. Wickramagamage, Improved Cookstove Programmes in East and Central Africa, D.M. Kammen, "Cookstoves for the Developing World," Scientific American, vol.273, pp.72-75.

⁴⁰ D.F. Barnes et al., What Makes People Cook.

⁴¹ F. Hitzhusen, The Economics of Improved Cookstove Programmes, Draft ESMAP Report (Washington: World Bank, 1991); and P Wickramagamage, Improved Cookstove Programmes in East and Central Africa.

⁴² K. King, The Incremental Costs of Global Environmental Benefits, Global Environment Facility (GEF) Working Paper No. 5 (Washington: World Bank, 1993); D. Anderson and R.H. Williams, The Cost Effectiveness of GEF Projects, GEF Working Paper No. 6 (Washington: World Bank, 1993).

⁴³ M. Dulansey and J.E. Austin, "Small-Scale Enterprise and Women," in C. Overholt et al. (eds.), Gender Roles in Development Projects (West Hartford, CT: Kumarian Press, 1985), pp. 79-134.