Cooking Stoves, Indoor Air Pollution, and Respiratory Health in IndiaPDF version

Esther Duflo
Michael Greenstone
Rema Hanna
Gram Vikas
Fieldwork implemented by: 
Centre for Micro Finance (CMF)
Rural Orissa, India
2,600 households
2006 - 2010
Policy Issue: 
Policy Issue: 

One half of the world’s households, and up to 95 percent of people in poor countries, burn wood, dung, peat and other biomass fuels, as well as coal, for energy. As women are primarily responsible for cooking, and children often spend time with their mothers, women and young children are disproportionately affected by the indoor air pollution caused by the use of solid fuels and traditional stoves. The WHO states that indoor air pollution from cookstoves “contributes to nearly 2 million deaths annually.”  In response, NGOs and governments have distributed tens of millions of “improved” or “clean” stoves that aim to reduce the concentration of air pollutants indoors by burning cleaner fuels and/or removing the smoke from houses through chimneys. However, there is surprising little evidence on the benefits of these stoves to health or fuel use and greenhouse gas emissions, and whether or not people are willing to use and maintain them. This study comprehensively examines how clean stoves perform in a large scale, real world distribution program.

Context of the Evaluation: 

According to the Indian Census, in 2001, over 70 percent of all households in India and 90 percent of households in the country’s poorer, rural areas used traditional solid fuels, such as crop residue, cow-dung and firewood, to meet their cooking energy needs. Burning solid fuels indoors results in high levels of toxic pollutants, a major risk factor for lung cancer, and cardiovascular and respiratory disease. Indoor air pollution in participating houses was high: health tests to measure smoke levels in the lungs found that women had an average Carbon Monoxide (CO) reading of 7.55ppm , while children had an average reading of 6.48ppm. This suggests that children have CO levels similar to those that would result from smoking about 7 cigarettes per day.

Details of the Intervention: 

A local NGO, Gram Vikas, obtained funding to subsidize the construction of 15,000 improved stoves in rural Orissa. The improved stoves (“chulas” in Hindi) were developed and tested by a local NGO, the Appropriate Rural Technology Institute. The chulas are made of mud, and have two pots and an enclosed cooking flame, which in lab settings decreased cooking times and fuel use; a chimney directs smoke out of the room.  Gram Vikas subsidized the stove cost by contributing the materials, the design, and the mason, but households were responsible for providing the mud for the base, labor, and a payment of Rs. 30 (about US$0.75) used to pay the person who assisted in building and maintaining the stoves. The total cost of the stove was approximately $12.50. In addition to building the stoves, Gram Vikas provided information on and encouraged use of the new stoves. For example, training sessions on proper use and maintenance were held during the construction. Good users were also identified within each village and hired to help promote the use of the stoves and alert Gram Vikas if any stove was in need of repair.

The researchers studied the impact of the stoves in a sample of 2,600 households.  At a community meeting, Gram Vikas explained that the stoves were being built in three waves, and households would be randomly assigned to each wave. A public lottery was then held to choose the third of the households who would be offered a stove in the first wave. When the second wave of stove building took place, Gram Vikas also made a big push to repair and rebuild stoves that had been destroyed or broken since the first wave.

Results and Policy Lessons: 

Stove use: Initial household take-up of the (almost free) new stoves was high, but over time households generally chose to let the stoves fall into disrepair and used them less frequently. In the first two years, households that received stoves in the first wave were 65 percent more likely to have a Gram Vikas improved stove than comparison houses. However, by year three, this number had fallen to 44 percent. Further, households that obtained new stoves continued to use the traditional one in addition. On average, households that received stoves cooked only about 3 more of their meals per week using improved stoves than the comparison group.

Smoke exposure: Households that were eligible for stoves showed little improvement in smoke exposure. Household members were tested for the levels of Carbon Monoxide (CO) in exhaled breath to measure for smoke inhalation. For the household’s primary cook, CO exposure fell by 7.5 percent during the first year but there was little change for other women or children. However, this effect for the primary cooks disappeared after the first year.  

Health status: Given that there was little to no change in smoke exposure, any health effects are unlikely to be due to decreased smoke inhalation. Respiratory tests of lung functioning, as well as a battery of health measures, both observed and self-reported, such as blood pressure, cough, cold and infant health outcomes, confirms that being offered a stove had no impact on health outcomes.

Fuel Usage and Cost of stoves: The clean stoves were meant to decrease fuel usage, and therefore fuel costs, and decrease cooking time. They do not appear to have done either, although interestingly, households report that the new stoves performed well on both measures. The new stoves did require substantially more repair than traditional stoves.

Policy Lessons:  In the context of a real, large-scale program, improved stoves were not used regularly, recipients did not make the necessary investments to maintain them properly, and proper use declined further over time.  There were no observable effects on smoke inhalation, health, or fuel usage. More research needs to be done to identify stoves that people value enough to use regularly and are willing to devote the resources necessary to maintain them.  And, it is vital that these technologies be rigorously tested in real-world conditions over long time horizons, since household behavior will determine their ultimate effectiveness.

Related Papers Citations: 
Duflo, Esther, Michael Greenstone, and Rema Hanna. “Up in Smoke: The Influence of Household Behavior on the Long-Run Impact of Improved Cook Stoves.” Working Paper, MIT, April 16, 2012.
Duflo, Esther, Michael Greenstone, and Rema Hanna. "Indoor Air Pollution, Health, and Economic Wellbeing." Working Paper, MIT, February 2008.
Duflo, Esther, Michael Greenstone, and Rema Hanna. 2008. "Cooking Stoves, Indoor Air Pollution, and Respiratory Health in Rural Orissa." Economic & Political Weekly 43(32): 71-76.