2019/102 Supported by K NKONW A A WELDEGDEG E OL N ONTOET E S ESREI R E ISE S F OFRO R P R&A C T HTEH E NEENREGRYG Y ETX ITCREA C T I V E S G L O B A L P R A C T I C E THE BOTTOM LINE Have Improved Cookstoves Benefitted Rural Kenyans? Meeting the goal of universal access to clean cooking by 2030 Findings from the EnDev Initiative remains a formidable challenge, as the current growth rate in clean-cooking coverage lags far Why promote improved cookstoves? consumption in total energy consumption dropping slowly—from about 10 percent in 1990 to less than 8 percent in 2015 (World Bank behind the rate required to meet Millions continue to use biomass for cooking. 2018). Sub-Saharan Africa, where population growth outpaces the the goal (0.5 percent per year More efficient stoves will protect their health growth of clean cooking adoption by a margin of 4:1, fares worst. vs. 3 percent). An estimated $4 billion in investment is needed and the environment Most of the rural population in developing countries, particularly in Sub-Saharan Africa, will continue to use biomass for cooking for to achieve universal access in 20 Biomass is the main source of energy for cooking in most developing the foreseeable future. In the meantime, use of improved cookstoves high-impact countries—125 times countries. An estimated 3 billion people—40 percent of the world’s (ICS) is a good transitional solution. The idea is, until the dependence current spending of $32 million a population—use inefficient biomass cookstoves, exposing them to on biomass can be completely eliminated, improved cookstoves will year (SE4All 2018). Most of the 20 indoor air pollution, drudgery, inconvenience, and physical endanger- relieve pressure on biomass resources and slow down deforestation high-impact countries are in Africa ment (World Bank 2018). Within households, women and children are and greenhouse gas emissions. and Asia. the most affected. Harvesting of biomass also causes deforestation, This Live Wire summarizes the socioeconomic benefits of ICS loss of ecosystem services, and emission of greenhouse gases. from a study of a cookstove program in rural Kenya administered Clean cooking is a priority theme of the United Nation’s by the EnDev program of GIZ, the German development agency. The Sustainable Development Goals. Clean cooking solutions are defined study is a product of the Status of Energy Access Report (SEAR), as those that are efficient, have low emissions of carbon monoxide an initiative of the Energy Sector Management Assistance Program and fine particulate matter (PM2.5), and are safe to use.1 The Clean at the World Bank. By documenting the effects of energy access Hussain Samad is a Cooking Alliance (formerly the Global Alliance for Clean Cookstoves) programs on beneficiaries’ welfare (for example, income, health, and consultant in the Energy has set a goal of having 100 million households adopt clean cook- education), SEAR enables policy makers and other stakeholders to Sector Management stoves by 2020 (GACC 2012). Clean cookstoves include liquefied assess energy-access benefits and make informed decisions about Assistance Program at the petroleum gas, electric, and ethanol stoves. Fan gasifier stoves that future projects. World Bank, where he acts as use biomass pellets can also be considered clean stoves, as they principal investigator of the Multi-Tier meet the emission criteria. Framework Global Survey activities. Initiatives like these have reduced biomass consumption. But Elisa Portale is a senior the results have been marginal, with the global share of biomass energy specialist in the 1 The Clean Cooking Alliance developed a tiered framework of cookstoves (Tier 0 indicating Energy Sector Management the worst performance and Tier 4 the best) based on the 2012 ISO International Workshop Assistance Program at the Agreement. According to the agreement, stoves meeting at least the Tier 3 standard for indoor World Bank, where she is emissions are considered clean for health impacts; stoves meeting the Tier 3 standard for responsible for coordinating ESMAP’s overall emissions are considered clean for environmental impacts. For details, see GACC (2012). knowledge products. 2 H a v e I m p r o v e d C o o k s t o v es B e n e f i t t e d R u r a l Ke n y a n s ? Fi n d i n g s f r o m t he E n De v I n i t ia t i v e What are the benefits of improved cookstoves? Figure 1. Benefits of improved cookstoves Fuel savings, better health, less pollution, and new Adoption of improved jobs head the list cookstoves Because improved cookstoves are more efficient than traditional or Until the dependence on open-fire stoves, they can yield substantial fuel savings.2 According Less biomass fuel Less time Less time spent used for cooking spent cooking collecting fuel biomass can be completely to a recent study, in Burkina Faso, improved stoves reduce fuelwood eliminated, improved consumption by 28 percent (Bensch, Grimm, and Peters 2015) and Less pollution and More time available for rewarding shorten cooking time.3 Using less fuel also means spending less Less drudgery cookstoves will relieve exposure to disease and productive activities time gathering it. Fuel is often collected by women and children, pressure on biomass who often must walk long distances from the home and return Monetary savings resources and slow carrying loads of firewood. The time women save in cooking and fuel Income/asset Benefits for growth women down deforestation and collection can be used to generate income, care for children, read, or socialize. Children can use the time they save to study. The substan- greenhouse gas emissions. tial savings in fuel can also increase disposable income, which can Source: Barnes and Samad (2018). Note: The benefits shown do not include other related benefits, such as employment generation be spent on productive activities.4 and training and skill development. Improved cookstoves also improve health. Household air pollution from solid fuel consumption is the leading risk factor in the global burden of disease. Indoor air pollution, caused mainly by Thus the accumulated benefits from using ICS for a long period smoke from traditional biomass cookstoves, accounts for an esti- can be high. The benefits are shown in figure 1. mated 3.7 million premature deaths each year, making it the second EnDev Kenya, a division of GIZ (the German development leading cause of disease, after smoking (Smith and others 2014). agency), has been promoting improved cookstoves since 2006, Because of its higher efficiency and more complete combustion, ICS in collaboration with the government of Kenya, nongovernmental can substantially reduce smoke and indoor pollution. organizations, and private firms. It has also helped develop the Finally, improved cookstoves reduce greenhouse gas emissions. cookstove market in Kenya. Its activities span the ICS supply chain, Cooking accounts for an estimated 2 percent of all greenhouse gas including production, marketing, installation, and end-use. EnDev is emissions worldwide, and cooking with three-stone stoves using also involved in results-based financing programs that encourage firewood produces 45 percent of the CO2 emissions attributed to firms to promote modern cookstoves by raising awareness of their cooking (Grupp 2004). A 2015 study finds that emissions from wood benefits and lowering barriers to distribution. fuel account for 1.9–2.3 percent of global CO2 emissions (Bailis and The ICS market has created jobs in rural Kenya in the production, others 2015). marketing, and installation of stoves. Some 4,200 previously unem- ployed people (mostly women and youth) have become self-em- ployed in the ICS market. EnDev was instrumental in the formation of 2 For example, the Tier 3 ICS developed by the Infrastructure Development Company Limited the Improved Stoves Association of Kenya (ISAK). in Bangladesh is claimed to increase thermal efficiency by at least 35 percent over traditional stoves (http://idcol.org/home/ics). EnDev Kenya promotes two types of energy-efficient and 3 There is, however, some evidence to the contrary. A study in Bangladesh finds that cooking improved cookstoves, the Jiko Kisasa and the Rocket Stove, the time was slower with improved stoves than with traditional ones (USAID 2014). 4 Because firewood is collected more than purchased, money saved from reduced fuel con- benefits of which are reported here. The Jiko Kisasa stove uses a sumption is largely imputed rather than actual. ceramic-lined chamber for combustion. It can be adapted for either 3 H a v e I m p r o v e d C o o k s t o v es B e n e f i t t e d R u r a l Ke n y a n s ? Fi n d i n g s f r o m t he E n De v I n i t ia t i v e stationary or mobile use, is up to 40 percent more efficient than variables at the household and community level.7 The control a three-stone stove, and has a lifespan of three years. The Rocket variables chosen are those that are expected to influence ICS Stove has a stationary brick structure. It is 20 percent more efficient adoption. In the second stage, the propensity score is used than the Jiko Kisasa stove and has a lifespan of five years.5 Both to create a weight, which is then used in a regression-based types are produced locally, use firewood, and have no chimneys but estimation to assess impacts.8 Adoption of improved provide good combustion. • Naive estimation is also based on regression estimation. In this cookstoves is a transitional EnDev’s program covers three geographical clusters: Western, method, the treatment group are households that use only an Central, and Trans Mara. As of December 2017, about 9.6 million improved stove; the control group includes households from solution that yields people had benefited from it. According to EnDev, ICS lowered control villages that use only traditional stoves but have shown many benefits, for both firewood consumption by 638,000 tons (corresponding to 38,000 willingness to adopt an improved stove if offered. The idea households and the hectares of forest) and cut CO2 emission by more than 738,000 tons behind this method is that people adopting ICS and people from environment. between 2016 and 2017. control villages showing interest in doing so are statistically comparable. How were the benefits of the EnDev Kenya Of the two methods, inverse probability weighting is more program measured? rigorous; it is thus preferred. Two evaluation techniques were used on the The study controlled for a number of factors—including “stove cross-sectional data collected for this study stacking” (use of multiple stoves in combination), fuel type, stove condition, stove maintenance, and kitchen characteristics—that At the heart of any evaluation lies a comparison of outcomes (such may affect the estimated benefits of ICS.9 Depending on whether as fuel consumption) between a treatment group (those who have stacking involves cleaner or traditional stoves, the measured benefits an ICS but no traditional stoves) and a control group (those have of ICS may over- or underestimate actual benefits. If not controlled traditional stoves but no ICS), called the counterfactual.6 Determining for, the use of fuel for improved and traditional stoves may bias the counterfactual is the most critical task in any impact evaluation, the estimated benefits. Kitchen characteristics also affect the and evaluation techniques must make various assumptions to do so. measured benefits, especially for health. Time spent in the kitchen With a valid counterfactual, the difference in outcomes between the was controlled for in estimating health benefits. The study also took treatment group and control group yields the impact of the inter- into account dealer characteristics, such as experience, training, and vention. Here, two evaluation techniques were used to compare the membership in ISAK. Community characteristics, such as community treatment and control groups—inverse probability weighting (IPW) infrastructure and the prices of alternate cooking fuels, were also and naive estimation. accounted for. • Inverse probability weighting is a two-step process. In the first Because the study was based on recalled data, information step, the probability of ICS adoption by each household (called collected may be subject to response bias, leading to under- or the propensity score) is calculated, based on a range of control 7 Control variables considered included the age, gender, and education of the household 5 A 2013 study finds that the Rocket Stove is 33 percent more efficient than three-stone stoves head; the amount of agricultural land; the condition of housing structure; village characteristics (Ochieng, Tonne, and Vardoulakis 2013). (mostly infrastructure variables); and prices of alternate fuels. 6 “Counterfactual” represents a hypothetical situation of what would have happened to the 8 The weight is the inverse of probability of ICS adoption. That is, it is given by w = 1 for p treatment group had they not received the intervention. Obviously, it is an imagined scenario, owners of an ICS, and w = 1 for nonowners. (1–p) since the treatment group, by definition, always receives the intervention. In reality, a subset of 9 Enumerators were asked to check stove conditions and to rate them as poor, average, or the control group is selected to mimic the counterfactual. good, based on specific criteria. 4 H a v e I m p r o v e d C o o k s t o v es B e n e f i t t e d R u r a l Ke n y a n s ? Fi n d i n g s f r o m t he E n De v I n i t ia t i v e Table 1. Distribution of villages and households in the sample Villages Households Cluster Treatment Control Total ICS Non-ICS Total Central 27 15 42 399 472 871 Trans Mara 32 15 47 499 446 945 Cooking accounts for an Western 40 20 60 600 586 1,186 estimated 2 percent of all Total 99 50 149 1,498 1,504 3,002 greenhouse gas emissions Source: ESMAP household survey, 2015.: worldwide, and cooking with three-stone stoves overestimation. Especially for information about cooking (such as households, which may not necessarily be the distribution in the using firewood produces cooking time and fuel consumption), survey data are less accurate underlying population, household sampling weights were used in the 45 percent of the CO2 than information collected in a controlled field test. Moreover, par- analysis to make the findings representative within the survey areas. emissions attributed to ticulate emissions or exposure to indoor air pollution from cooking The survey collected information on basic demographic charac- can be measured only using specialized instruments. This study, teristics of the household members, such as age, gender, education, cooking. on the other hand, depended solely on survey data, and collected and employment, as well as information on household assets, information on various symptoms of household members (such as income, expenditure, housing, sanitation, incidence of illness, and respiratory problems) that can be linked to exposure to indoor air women’s use of time. Detailed data were collected on households’ pollution.10 biomass collection and consumption, kitchen characteristics, cooking Data for this study came from a household survey conducted patterns, and stove characteristics. ICS owners were asked about in August–October 2015 with funding from the World Bank’s Energy the adoption and maintenance of their units. Opinion questions were Sector Management Assistance Program. The survey sample also asked about various aspects of the stove. included two types of villages: villages in which the EnDev ICS pro- In addition to the household survey, a community survey was gram operated (treatment villages) and villages in which the program conducted in every survey village. This survey collected information did not operate but that had similar socioeconomic characteristics on village infrastructure, development activities, prices of different (comparison villages). In most of the treatment villages, the EnDev fuels, and consumer goods. program had been operating for at least four years (enough time for Finally, a dealer questionnaire was administered to the dealers benefits to have materialized). who supplied or built improved stoves for the sample households. One hundred treatment and 50 comparison villages were This questionnaire covered demographic characteristics of the deal- selected; about 20 households were selected in each. Households ers, as well as their experience, training, and various other aspects selected from the treatment villages included both owners and of the ICS business (time in business, revenue and profit, number of nonowners of an ICS. A total of 3,197 households were interviewed, units sold or produced per month, training, and so on). split evenly between users and nonusers of an ICS. After data clean- The sample consisted of 1,498 ICS and 1,504 non-ICS households ing, 3,002 households were left for analysis. As the ICS households from 149 villages in three clusters.11 Table 1 shows the distribution of are purposely sampled to constitute about 50 percent of all sample wards, villages, and households in the sample. 10 Notwithstanding the limitations, recall-based surveys have certain advantages. They are useful for collecting a wide range of socioeconomic information on the household and its mem- bers, for example, and are feasible when the sample size is as large as in this study (about 3,000 11 Because of administrative changes in Kenya, former clusters were divided into smaller households). administrative units called counties. 5 H a v e I m p r o v e d C o o k s t o v es B e n e f i t t e d R u r a l Ke n y a n s ? Fi n d i n g s f r o m t he E n De v I n i t ia t i v e Table 2. Distribution of sample households by cookstove use Figure 2. Reductions in stove-use time from adoption of (percent) improved cookstoves Traditional Other 60 Cluster ICS stoves cookstoves 51.1 50 47.4 Central 25.5 53.8 20.7 Naive estimates (minutes per day) 37.7 Households saved an Trans Mara 48.6 44.7 5.7 40 Inverse probability Time savings 34.3 weighted estimates average of 34–47 minutes Western 75.5 20.4 4.1 30 a day in cooking and about Total 52.1 39.0 8.9 20 Source: ESMAP household survey, 2015. 10 38–52 minutes a day in 3.4 3.7 0 overall stove use. Adopting Cooking meals Water boiling All use Table 2 shows the distribution of households by cookstove and heating an improved stove reduced type. Households are categorized into three groups by cookstove Source: ESMAP household survey, 2015. the time women spent ownership: collecting biomass fuels by • All ICS-owners, who may or may not have traditional (three- or five-stone) or other stoves fuels, including dung (animal waste), maize cobs, sorghum stalks, 92–105 minutes a week. • Owners of traditional stoves only sawdust, and others, declined by about 35–40 percent.13 • Owners of other stove types, who may or may not have tradi- Improved cookstoves saved time. Households save time using tional stoves but did not have an ICS.12 an ICS in two ways—by cooking faster and reducing the time spent collecting fuel. Stove use time is the daily use in minutes averaged More than 50 percent of households had an ICS, 39 percent had over a seven-day period. Households were asked about the time traditional stoves only, and less than 10 percent had other types of they spent using stoves to cook meals vs. other uses (boiling water stoves. The ICS penetration was highest in the Western cluster (about and heating). Households saved an average of 34–47 minutes a day 75 percent) and lowest in the Central cluster (about 25 percent). on cooking and about 38–52 minutes a day in overall stove use after adopting the ICS (figure 2). What did the survey reveal? ICS adoption reduced the time women spent collecting biomass fuels by 92–105 minutes a week. It saved other household members Adoption of an improved stove had positive effects 43–57 minutes a week (figure 3).14 on the outcomes Improved cookstoves increased the time women spent on Fuel consumption declined; households saved time (which they used income-generating activities, child care, and leisure. ICS adoption, for income-generating activities, child care, and leisure activities); by saving time and money, enables women to use the freed-up time and some of the symptoms associated with exposure to household air pollution abated. Findings that are statistically significant at the 13 The monetary value of the consumed biomass is not necessarily what households actually 10 percent level or better are reported below. spent, because a large share of the biomass consumed is collected from nature. The value of Improved cookstoves reduced fuel consumption. ICS the collected biomass is estimated using community prices of the fuels in question. Monetary savings in the fuel are therefore imputed rather than actual. adoption reduced firewood consumption by about 20–32 kilograms a 14 The estimated time saved in collecting fuel may be overestimated because the reported month (about 18–29 percent of consumption). Use of other biomass time may include time spent on other activities. For example, children may collect fuel on their way back from school but report the entire time as spent on collecting fuel. However, if we as- sume that such overlapping of activities is random and does not depend on the type of stoves 12 Most of these households own ceramic Jiko or metal charcoal stoves. Charcoal stoves have (ICS or traditional), the time savings still can be attributed to the ICS, though the magnitude of been in use in rural Kenya for more than 20 years. the saving may be overestimated. 6 H a v e I m p r o v e d C o o k s t o v es B e n e f i t t e d R u r a l Ke n y a n s ? Fi n d i n g s f r o m t he E n De v I n i t ia t i v e Table 3. Impacts of improved cookstoves on fuel consumption Figure 3. Savings in fuel-collection time from adoption of Inverse improved cookstoves probability weighted Naive 120 105 (minutes per week) Item estimates estimates 100 92 Time savings 80 Naive estimates Change in firewood consumption (kg/month) −31.5 −19.5 The benefits of improved 60 57 Inverse probability Change in monetary value of all consumed −40.3 −35.1 43 weighted estimates stoves depend on biomass (percent) 40 20 sustained use, but not all Source: ESMAP household survey, 2015. 0 Women’s time All family members’ time households continued using theirs. Table 4. Changes in women’s use of time as a result of use of Source: ESMAP household survey, 2015. improved cookstoves (minutes per day, except as otherwise indicated) Figure 4. Reduction in emissions-related symptoms among Inverse probability women associated with adoption of improved cookstoves weighted Outcome estimates Naive estimates symptoms (percentage points) 20 18.7 Reduction in incidence of Engagement in income- 4.4 5.7 14.5 Naive estimates generating activities (percent) 15 Inverse probability Time spent taking care of 19 6 10 9.4 weighted estimates children 6.0 7.1 5 4.3 Time spent resting 23 16 Time spent listening to radio or 15 7 0 watching TV Coughing Chest pain Eye irritation Source: ESMAP household survey, 2015. Source: ESMAP household survey, 2015. in more rewarding activities, including those that generate income. six months preceding the survey, which it considered as proxies for ICS adoption increased the probability of women’s engagement in exposure to indoor air pollution or smoke.16 self-employment by 4.4–5.7 percentage points. It also increased the The self-reported incidence of coughing decreased by about amount of time they spent taking care of their children, resting, and 15–19 percentage points among women as a result of ICS use. The listening to radio or watching TV (table 4). incidence of chest pain and eye irritation fell by about 4–9 per- Improved cookstoves reduced some of the symptoms associated centage points (figure 4). The study found little impact on children’s with exposure to household air pollution. This study did not measure symptoms—among the three ailments, only coughing was found the level of pollution (for example, concentrations of PM2.5 and to have decreased among children as a result of ICS adoption, by CO). Instead, it investigated symptoms of ailments associated with 10–14 percentage points. household air pollution and smoke.15 The survey collected infor- mation on the incidence of coughing, chest pain, and eye irritation among women (over 15) and children (15 and younger) during the 16 Most women were heads of household or the wives of the household head; they were also the main cook in the household. For a small number of households, unmarried daughters over the age 15 were included in this group. Other symptoms, such as shortness of breath, irritation 15 These questions were asked only of women (female heads of household and wives of male of nose and throats, and asthma, were also investigated. However, ICS use had no statistically heads). significant impact on these symptoms. 7 H a v e I m p r o v e d C o o k s t o v es B e n e f i t t e d R u r a l Ke n y a n s ? Fi n d i n g s f r o m t he E n De v I n i t ia t i v e What can be done to improve outcomes? also one of the focus countries of the Clean Cooking Alliance, which has been working with international and local partners to help the The EnDev Kenya initiative achieved positive government prioritize clean cooking by reducing taxes and import results, but not all households reaped full benefits. duties on clean cookstoves. Several further steps could help… ICS is a transitional Ensure that households continue to use their stoves. The benefits of References solution toward cleaner improved stoves depend on sustained use, but not all households Barnes, Douglas F., and Hussain A. Samad. 2018. Measuring the cooking; the ultimate goal continued using theirs. About 12 percent of nonusers had used Benefits of Energy Access: A Handbook for Development an ICS in the past; the most common reason for stopping was the Practitioners, Washington, DC: Inter-American Development is for households to adopt inconvenience of not being able to use large pots or large pieces of Bank. cookstoves that use clean, firewood. Thus, user concerns should be taken into account in stove Bailis, Robert, Rudi Drigo, Adrian Ghilardi, and Omar Masera. 2015. modern fuels design. To ensure sustained use, stakeholders and promoters should “The Carbon Footprint of Traditional Woodfuels.” Nature Climate also include more sensitization activities in their dissemination plans. Change 5: 266–72. Ensure that households use and maintain their stoves properly. Bensch, G., M. Grimm, and J. Peters. 2015. “Why Do Households More than 40 percent of ICS users did not keep their stoves in good Forego High Returns from Technology Adoption? Evidence from condition, and only 60 percent reported having received any training Improved Cooking Stoves in Burkina Faso.” Journal of Economic on stove use and maintenance. ICS promoters must put training Behavior & Organization 116 (2015): 187–205. programs in place to ensure that stoves are kept in good shape and GACC (Global Alliance for Clean Cookstoves). 2012. “Measuring regularly maintained. Customer training should be made mandatory Progress During Phase I: Building on the IWA Interim Guidelines.” and training programs monitored. Discussion Paper, Washington, DC. https://www.cleancookingalli- Before customers can be trained, dealers must have proper ance.org/binary-data/ATTACHMENT/file/000/000/150-1.pdf. training, affiliation, and experience. On average, ICS dealers have Grupp, Michael. 2004. “Domestic Cooking Appliances in Developing been in business for almost six years. Only 45 percent of dealers are Countries Economic and Environmental Aspects.” In Proceedings members of ISAK, the premier organization for dealers, producers, of the Domestic Use of Energy Conference. Cape Town. http:// and marketers of cookstoves in Kenya. About 70 percent of dealers timetable.cput.ac.za/_other_web_files/_cue/DUE/2004/ seem to have received formal training from EnDev, ISAK, or another PDF/26_M_Grupp.pdf. organization. Improvements in these areas would enhance the IDCOL (Infrastructure Development Company Limited). “Improved benefits of ICS. Cook Stove Program.” Dhaka. http://idcol.org/home/ics. Discourage stove stacking. About 10 percent of ICS owners stack Ochieng, Caroline A., Cathryn Tonne, and Sotiris Vardoulakis. 2013. “A their improved stoves with traditional stoves, a practice that can Comparison of Fuel Use between a Low Cost, Improved Wood reduce the benefits of the improved stove. More awareness raising Stove and Traditional Three-Stone Stove in Rural Kenya.” Biomass and sensitization are needed to encourage households to gradually and Bioenergy 58 (November): 258–66. replace their traditional stoves with an improved one. Smith, K.R., N. Bruce, K. Balakrishnan, H. Adair-Rohani, J. Balmes, Z. Encourage the adoption of clean cookstoves. ICS is a transitional Chafe, and others. 2014. “Millions Dead: How Do We Know and solution toward cleaner cooking; the ultimate goal is for households What Does It Mean? Methods Used in the Comparative Risk to adopt cookstoves that use clean, modern fuels. An initiative by Assessment of Household Air Pollution.” Annual Review of Public Netherlands-based SNV disseminates clean and efficient (gasifier) Health 35 (1): 185–206. stoves and fuel pellets in urban and rural areas of Kenya. Kenya is 8 H a v e I m p r o v e d C o o k s t o v es B e n e f i t t e d R u r a l Ke n y a n s ? Fi n d i n g s f r o m t he E n De v I n i t ia t i v e MAKE FURTHER SE4All (Sustainable Energy for All. 2018). “Today Three World Bank. 2018. Tracking SDG7: The Energy Access Report. CONNECTIONS Billion People Still Lack Access to Clean Cooking.” Washington, DC. December 17. https://www.seforall.org/content/ Live Wire 2014/7. “Understanding the today-three-billion-people-still-lack-access-clean-cooking. The authors thank Richard H. Hosier (Senior Energy Specialist, GEE08), Yabei Differences Between Cookstoves,” by Zhang (Senior Energy Specialist, GEEES), and Caroline Adongo Ochieng Koffi Ekouevi, Kate Kennedy Freeman, USAID (U.S. Agency for International Development). 2014. “What and Ruchi Soni. (Energy Specialist, GEEES) for providing valuable comments on drafts of this Do Cooks Want? What Will They Pay? A Study of Improved Live Wire. Live Wire 2015/46. “Results-Based Cookstoves in Bangladesh.” Technical Brief. Washington, Financing to Promote Clean Stoves: DC. http://www.washplus.org/sites/default/files/ics_bangla- Initial Lessons from Pilots in China desh2014_0.pdf. and Indonesia,” by Yabei Zhang and Norma Adams. Live Wire 2016/62. “Toward Universal Access to Clean Cooking and Heating: Lessons from the East Asia and Pacific Clean Stove Initiative,” by Yabei Zhang and Norma Adams. Live Wire 2016/63. “The Lao Cookstove Experience: Redefining Health through Cleaner Energy Solutions,” by Rutu Dave and Rema N. Balasundaram. Live Wire 2016/64. “Contextual Design and Promotion of Clean Biomass Stoves: The Case of the Indonesia Clean Stove Initiative,” by Laurent Durix, Helene Carlsson Rex, and Veronica Mendizabal. Live Wire 2017/74. “Increasing the Use of Liquefied Petroleum Gas in Cooking in Developing Countries,” by Richenda Van Leeuwen, Alex Evans, and Besnik Hyseni. Live Wire 2018/89. “Kenya’s Strategy to Make Liquefied Petroleum Gas the Nation’s Primary Cooking Fuel,” by Inge C. van den Berg. Live Wire 2019/99. “Beyond the Last Mile: Piloting High-Efficiency, Low-Emissions Heating Technologies in Central Asia,” by Yabei Zhang, Norma Adams, and Crispin Pemberton-Pigott. Find these and the entire Live Wire archive at https://openknowledge. worldbank.org/handle/10986/17135.