Page 82

Box 4-1 The Green Revolution in Agriculture as an Analogy to Climate Forecasting

Evidence from India suggests that farmers consider two kinds of information when faced with an innovation such as a new seed variety or production practice: information about how to use the innovation best and about whether the innovation, if used optimally, will be profitable. Farmers learn about both from their own experience—by trying the innovation themselves—and by observing or talking with their neighbors about their experiences with the innovation. Although essentially all farmers ultimately adopt a practice if it is profitable, a number of patterns of innovation adoption are observed. Farmers with larger land holdings adopt new practices faster, partly because the return to a profitable innovation is positively related to farm size, whereas the cost of acquiring the innovation is scale-independent. Thus, larger farmers have the biggest payoffs to an increase in knowledge.

More educated farmers also appear to learn best practice for an innovation faster than others, all else being equal. Such farmers thus adopt new practices faster and gain the benefits earlier. Uneducated farmers benefit later, by learning from their better-educated neighbors. In sum, more educated farmers with large land holdings bear more of the initial costs associated with learning best practice but, at least initially, reap the most from an innovation.

Because the best use of an innovation is not known initially, many early adopters may experience initial losses even if the innovation is profitable under best use. These losses are part of the cost of learning about the innovation. Farmers thus have an incentive to delay adoption when there is less known about how to use the innovation because they will learn from their neighbors' experiences without bearing the costs. This situation creates an economic externality, in that a farmer's decision to adopt an innovation does not take into account the learning benefits that go to his or her neighbors. This situation creates a disincentive for early adoption with the result that, in the absence of interventions, adoption patterns are not efficient.

The green revolution in Mexico yielded some lessons that elaborate on some possible consequences of unequal rates of use of new and valuable information. Although agricultural production increased dramatically, there were also increases in social inequality and environmental degradation. In the Yaqui valley, for example, the less-well-off farmers and indigenous peoples were unable to afford the new seeds or the inputs required for them to produce high yields, whereas wealthier farmers, who obtained bumper harvests with the new seeds, became able to rent or buy the land of the poor, who then became migrant workers or destitute. In some cases, farm labor was mechanized, further increasing unemployment and poverty (Hewitt de Alcantara, 1973, 1976). Regional inequality also increased, as localities with irrigation infrastructure were able to take advantage of the new technologies but rainfed areas were not. Some have argued that the new technologies increased Mexico's international debt by creating dependency on imported fertilizers and pesticides; increased pollution from fertilizer and pesticide runoff, erosion, and salinization; and increased vulnerability of crop monocultures to disease (Wright, 1984). It has also been suggested that the new seeds increased vulnerability to climatic variations because of their extreme vulnerability to drought and other climatic extremes (Michaels, 1979; Liverman, 1990). The majority of farmers, who could not afford insurance, lost heavily in drought years such as 1982-1983, when they went into debt to buy seeds and fertilizers and then suffered harvest failures (Austin and Esteva, 1987). In response to these developments, the Mexican government and research centers such as Centro Internacional de Mejoramento de Maiz y Trigo (the International Center for the Improvement of Corn and Wheat, or CIMMYT) developed new programs to bring benefits to poorer farmers, develop national seed and fertilizer industries, prevent environmental degradation, and subsidize insurance.

If the experience of the green revolution is replicated with climate forecasts, it is reasonable to expect that well-educated farmers and those with large land holdings will be the first to benefit from climate forecasts and will reap the greatest benefits, at least in the early stages of climate prediction. The Mexican experience also suggests that, if a forecast should prove overly optimistic about crop production, the losses may be relatively devastating to small, poorer farmers, with long-term negative effects on social equity.


agencies, which then inform individuals and organizations locally. Sometimes it is advisable to use redundant information sources, with the same message sent through multiple channels to increase the chances that everyone in the intended audience will get the message from a trusted source. It is often important to plan for the possibility that different information sources will distribute conflicting information.


Research on disaster warnings has contributed to a good understanding of the information systems involved in short-term warnings of major negative weather events, such as hurricanes and floods, as well as of aspects of the information systems that can be employed for longer-term warning of such events. These information systems include organizations that detect impending disasters (e.g., weather services) and management organizations that interpret information and decide what and how

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement