- About ATAI
- Our Approach
- Contact ATAI
A farmerâ€™s choice to adopt a new technology requires several types of information. The farmer must know that the technology exists; she must know that the technology is beneficial; and she must know how to use it effectively. These types of information may come from different sources:
Sources of Information
External sources of information, such as extension workers, may be particularly important for the adoption of new technologies. Though many farmers cite extension workers as an important source of information about agricultural technologies, not all extension workers are motivated to do their job well. Many lack incentives to perform well because they are not monitored and are not rewarded for good work nor held accountable for shirking their responsibilities. In addition, extension workers may under serve disadvantaged groups such as women and minorities.
Other farmers are also an important source of information about new technologies. Studies on technology adoption in fields other than agriculture show that individuals learn from others within their social network. The results are mixed, however: adoption by oneâ€™s peers can make adoption more likely,1 less likely2 or have no effect3. Learning from others can result in a less rapid spread of technology if social networks are small or if the benefits of a technology are hard to observe. Examples include technologies for slow growing crops that take multiple seasons to mature or technologies that require considerable customization for a farmerâ€™s particular growing conditions.
Learning from Personal Experience
One study of fertilizer adoption in Kenya showed that intensive information provision by extension workers had a bigger effect on adoption than did information spread among peers.4 In this setting, farmers learn how to use the technology rather than learning about whether or not it is beneficial. The authors find that learning by trial and error, in this case, did increase adoption significantly over the short run, though not as much as extension services. Even more puzzling is that farmers stopped using the beneficial fertilizer even after experiencing its benefits.
Given that the relative importance of different sources of information is likely to vary across technologies and contexts, understanding this difference can help improve the effectiveness of interventions which seek to provide farmers with information to promote technology adoption.
Research suggests that the way information is presented (who provides the information, how much information is given and in what form) can be as important as the content of the information itself. For example, presenting information in different ways (i.e. framing) can have large effects on decision making. While framing appears to have large effects in some settings, research on the adoption of weather insurance in India shows that framing is much less important than other factors, such as the price of the insurance.5 Information about a technology is, of course, only one of many factors that affect whether or not a farmer will adopt the technology.
1 Foster, A. D. and M. R. Rosenzweig (1995). "Learning by Doing and Learning from Others: Human Capital and Technical Change in Agriculture." The Journal of Political Economy 103(6): 1176-1209.; Oster, E. F. and R. L. Thornton (2008). "Determinants of Technology Adoption: Peer Effects in Menstrual Cup Take-Up." NBER Working Paper.
2 Kremer, M. and E. Miguel (2007). "The Illusion of Sustainability." The Quarterly Journal of Economics 122(3): 1007-1065.
3 Duflo, E., M. Kremer, et al. (2009). Nudging Farmers to Use Fertilizer:
Theory and Experimental Evidence from Kenya, Harvard University
Department of Economics working paper.
5 Cole, S., X. Gine, et al. (2009). Barriers to Household Risk Management: Evidence from India, Harvard Business School Working Paper.
Approaches for improving farmersâ€™ access to information must first consider why the information is not already available. Does the farmer have access to sources of information? Do information providers have an incentive to deliver information to the farmer?
As with the distribution of other goods and services, those who hold information need some incentive (monetary or otherwise) to distribute itâ€”particularly if the process of distribution costs effort, time, or money. Payment for information, however, does not need to be direct. For example, a buyer of agricultural output who faces no competition for local farmersâ€™ output (i.e. a â€śmonopsony buyerâ€ť) would benefit both from increased yieldsâ€”more product to turn over and sellâ€”and possibly lower prices. Such a buyer therefore has an incentive to tell farmers about a new technology that could increase production, since he will be able to capture much of the benefit. A government agricultural extension worker, on the other hand, often lacks both the incentives and accountability needed for reliable information supply to meet the needs of smallholder farmers.
Evidence from Information Interventions
The availability of information clearly makes a difference in determining adoption outcomes. Evidence from a randomized intervention in which school children were given accurate information about the benefits of staying in school suggests that simply providing information about the average benefits of a technology increases adoption, in this case the decision to stay in school.1 Targeted information, however, is more effective in some settings. In a randomized trial in India, Jalan and Somanathan (2008) provide some households with specific information about their water quality and disease risk, while other households learned only about average risks.2 The households that received specific information about their own returns to technology adoption were significantly more likely to adopt water purification technologies.3 In an experiment that varied the information content of a loan offer, simpler information and fewer examples increased take-up of the loan. These findings suggest that simpler information may be more influential because it is easier for the farmer to incorporate into her adoption decision.
Easing Information Constraints
Research suggests that constraints to information access can be lowered by improving incentives for those delivering information, reducing the cost of acquiring information, and improving the design of information provision (content, source and presentation). Advances in information technology, such as cellular telephones and SMS, offer great potential for lowering the costs of regular information provision. One example of this is the e-Choupal project in India, which provided internet kiosks in villages in order to give farmers access to a wide variety of information on agriculture and markets.4 Similarly, a World Bank project in Ghana used SMS to deliver prices and other market information to farmers. In recently funded ATAI research, Kremer, Mullainathan, and Casaburi will explore whether timely text messages reminding sugar farmers when to apply fertilizer can promote product usage. In an ATAI pilot project, Karlan and coauthors will evaluate a model of using video to disseminate information about agricultural practices. Further evaluation of the cost-effectiveness of these programs and their impact on adoption will help in designing future interventions.
Newly supported research under ATAI focuses on how to make information more available, to increase adoption of technologies that are beneficial to the individual and society. Specifically, Duflo, Kremer, and Robinson will investigate the role of farmer cooperatives as a source of information for their members, and will investigate both the spread of information and the results for technology adoption. In an ATAI supported project in Malawi, Magruder and coauthors are investigating the role of contact farmers in information distribution.
Information delivered through agricultural extension services often conveys findings from demonstration plots, which is not always applicable to other contexts. Recent research in western Kenya documents the inappropriateness of the official recommendations for the average Kenyan farmer. Duflo, Kremer and Robinson (2008) show that the package of fertilizer and seed inputs recommended by the Kenyan Agricultural Research Institute results in negative profits for the farmer, though it successfully increases yields. An alternative, locally appropriate quantity of fertilizer is shown to result in positive returns for the majority of farmers in their study.5 These results clearly document the importance of locally appropriate technologies, which in this case consist of correct recommendations about the quantity of fertilizer to use on a maize plot.
In an ATAI-supported follow up to this study, the authors will explore whether a complementary technology that helps guide farmers in applying the appropriate amount of fertilizer improves adoption and yields. In other settings, where micro-climates or soil types vary from plot to plot, appropriate information may vary at the household level. In that case, providing appropriate information may be more challenging and require further innovation.
Advertising to Promote Technology Adoption
Survey research suggests that simple innovations tend to spread more quickly than complex ones because they are more adaptable to individual needs and preferences. Marketers have long understood the power of advertising and other ways of providing simple, salient information. Preferences for food products or agricultural practices are malleable and may be formed or altered through advertising and use. Strategies that take advantage of this malleability can benefit from the substantial research on advertising that illuminates how and when these types of changes occur, though to date, no research has investigate the role of advertising in agricultural technology adoption in the developing world.
1 Jensen, R. (2008). The Perceived Returns to Education and the Demand for Schooling, Watson Institute for International Studies Working Paper.
2 Jalan, J. and E. Somanathan (2008). "The importance of being informed: Experimental evidence on demand for environmental quality." Journal of Development Economics 87(1): 14-28.
3 Bertrand, M., D. Karlan, et al. (2008). "Whatâ€™s Advertising Content Worth? Evidence from a Consumer Credit Marketing Field Experiment." Quarterly Journal of Economics Revise and Resubmit.
4 Upton, D. and V. Fuller (2005). The E-Choupal: Just Enough Bandwidth in Developing Agriculture. The broadband explosion: leading thinkers on the promise of a truly interactive world. S. P. Bradley and R. D. Austin. Cambridge, Harvard Business School Press.
5 Duflo, E., M. Kremer, et al. (2008). "How High Are Rates of Return to Fertilizer? Evidence from Field Experiments in Kenya." American Economic Review 98(2): 482-488.