One-third of the world’s population lives in water-stressed areas, a figure expected to climb to two-thirds within 25 years, due in part to climate change. In Asia, 20 percent of the total rice cultivation area is prone to drought and climate change is expected to exacerbate the situation. In the impoverished eastern Indian states of Jharkhand, Orissa, and Chhattisgarh, drought occurs approximately every five years and can result in a 40 percent loss in total rice production.
Rice experts expect that future growth in yield will come primarily from stress-tolerant varieties such as the short-duration Sahbhagi Dhan (SD) seeds used in India, designed to withstand up to two weeks of drought exposure. As an early maturing crop, SD rice also allows farmers to plant an additional crop, often one sold at market to augment household income. Tests show that in drought years, SD rice provides a significant yield advantage over traditional varieties but in non-drought years, SD provides slightly lower expected yields compared to long-duration varieties, due to early maturation. This implies a tradeoff between risk and expected returns with long-term positive gains in expected yield if drought frequency is sufficiently high, such as in the study areas. Since its limited release in 2010 in northern and eastern India, knowledge of SD rice has quickly diffused and adoption rates have rapidly increased. The primary research question is to understand the welfare effects of adopting SD rice in a drought-prone environment, specifically the impact on yield and livelihood.
This study takes places in the Indian states of Orissa and West Bengal among small upland farmers who primarily rely on rainfed irrigation where rice is an important staple crop and risk of drought is high. This study is a partnership with the International Rice Research Institute (IRRI), which works with the Stress-Tolerant Rice for Africa and South Asia (STRASA) project, and is situated within a pilot seed diffusion program in these two states. By focusing on two geographically separated states, the research design increases the likelihood that a portion of the study villages will experience drought during the project period.
This is an experimental study, in which minikits of SD seeds will be distributed to 5 randomly selected farmers in 384 randomly selected villages in Orissa and West Bengal in a three-year phase-in from 2012-2014. 5 other farmers in each village will be surveyed but not receive the seeds. This research design closely aligns with the seed promotion strategy employed by the Indian government and IRRI. Another 128 randomly selected villages that do not receive any intervention will serve as a comparison group. Follow-up surveys would begin one-year after initial distribution until 2015. All villages will be surveyed each year, leading to a 3 year panel survey of 5,120 farmers. Observation of the same households over several years will help reveal how learning affects yield outcomes.
The surveys will collect data on the following outcomes to measure impact: a) crop yields, b) farming practices, including crop and variety choices, inputs, land use, and labor activities by gender, c) household characteristics including income and demographic indicators, d) attitudes toward uncertainty, and e) diffusion of seeds to others. Detailed household characteristic data will allow for evaluation of impacts disaggregated by asset-based wealth, gender, and caste.
Results and Policy Implications
2012 - present
Photo Credit: Thomas Chupein