Friday, April 01, 2011

Improved Stoves Experience in India - Field Research in Orissa

From Mr. Lalloobhoy Battliwala

Below an excerpt from the report by Esther Duflo (MIT) et al. "Indoor Air Pollution, Health, an Economic Well Being", which suggests that some improved stoves testing was under way in Orissa under Gram Vikas.

I think some of this was reported in the WRI/IFMR Power to the People report. I read in a WRI/IFC (2007) report "The Next 4 Billion": "The energy giant Shell aims to create sustainable market systems to sell 20 million affordable stoves in India by 2010."


According to the 2001 Indian Census, 72.3% of households in India—and 90% of the population in poorer, rural regions—use traditional fuels. In response to perceived health threats from the traditional fuels, both the Indian government and many non-governmental organizations (NGOs) have implemented clean stove programs. During the 1980s and 1990s, the government of India alone subsidized and distributed 32 million improved stoves. However, there is little evidence on whether the stoves improve health, and if effective, how the stoves compare with other possible health interventions.

To this end, Gram Vikas, a rural development organization that works with marginalized communities in rural Orissa, India, began an investigation into the efficacy of its stove program. At the start of the study, Gram Vikas had planned to subsidize improved stoves ("chulhas" in Hindi and Oriya) to roughly 15,000 households over the next three to five years. As the stoves are primarily made of locally available mud, the stoves can be easily constructed in remote, rural areas. These stoves enclose the cooking flame, which in laboratory settings leads to increased efficiency and lower biofuel requirements. Importantly, they also include a chimney that directs the smoke out of the room. Gram Vikas subsidizes the stove cost by contributing stove materials (chimney) and design, but households are responsible for providing mud for the stove base, labor and a payment of Rs. 30 which is used to pay the person who assists in building and maintaining the stoves.

To evaluate the stove program, a randomized evaluation has been implemented. Since installing such a large number of stoves takes considerable time and resources, Gram Vikas will phase in the stove construction over the next few years. In about 40 villages (about 2500 households), the order according to which households will receive the stove was determined randomly. In each of these villages, a lottery was conducted to randomly divide the village households into three groups. Based on the lottery results, the first group was given a chance to get the stove immediately. A year and a half into the project, the second group will be given an opportunity to get a stove. The third group will be given the opportunity at the end of three years. It is important to note that because the assignment into groups was conducted at the household level, rather than the village level, we can be sure that differences in outcomes between households that were and were not offered a stove are attributable to the stove, and not to other characteristics common to households in a particular village.

In collaboration with Gram Vikas, the Center for MicroFinance (CMF) at the Institute for Financial Research and Management, a research institute based in Chennai, India, has undertaken a considerable data collection effort to better understand the impacts of the program. Specifically, in 2006, CMF conducted a detailed baseline in the 40 villages. CMF is currently conducting follow-up surveys to understand the impacts of the improved stoves on cooking practices, fuel usage, CO exposure, health status and health expenditures, and labor supply.

This study will further our understanding of the impacts of a clean stove program in the field. First, this is the largest randomized evaluation of a stove program to date. Second, this study will use data from a program wholly run and operated by an NGO, which has the added advantage of more accurately evaluating the likely impact of a program in the real world. Moreover, this study evaluates a fairly cheap stove design, where stoves are constructed using locally available material. Previous studies evaluated planchas, stoves that are large and costly. The stoves we evaluate, while more likely to break, are potentially more cost-effective and practical for remote areas where there are high transport costs. Next, CMF is collecting a rich series of productivity measures, including child school attendance. This study, thus, takes into account the welfare consequences of poor health due to both illness and reduced productivity. Finally, the study is based on richer data that allows for an examination of households’ behavioral responses upon receiving the option to buy a stove (take up, use, changes in who cooks and when they cook, change in other health behaviors) and on economic consequences of better health.

No comments: