This paper was prepared for Stanford University’s Global Food Policy and Food Security Symposium Series, hosted by the Center on Food Security and the Environment, and supported by the Bill and Melinda Gates Foundation.

Food policy makers are increasingly faced with the question of how to adapt to climate change. The increased attention on climate adaptation is partly related to the fact that greenhouse gas emissions and climate change show little sign of slowing, partly because of prospects for large sums of money devoted to adaptation, and partly because of well publicized recent weather events that have affected agricultural regions and rattled global food markets. A common and reasonable reaction from the food policy and agricultural community has been to argue that climate variations have always been a challenge to agriculture, and that climate change just makes addressing these variations more important. A logical conclusion from this perspective is to emphasize activities that help build resilience to unpredictable weather events, as well as to focus on the types of weather variables that exhibit a lot of year-to-year variability and cause the bulk of farmers’ concerns in current climate.

However reasonable as a starting point, this perspective is misguided and risks taking a challenging problem and making it even harder. Anthropogenic global warming (AGW) is fundamentally different from the natural variations driven by internal dynamics in the climate system. Indeed, predicting the course of climate change is less like predicting the weather next week than it is like predicting that summer will be warmer than winter. Progress in climate science has shown that the most indelible hallmarks of AGW will be increased occurrence and severity of high temperature and heavy rainfall extremes in all regions, and increased frequency and severity of drought in sub-tropical regions. Changes in the timing and amount of seasonal rainfall also appear likely in some regions, but at a much smaller pace relative to natural variability. In all of these cases, predictions from climate science are most robust at broader spatial scales, with considerable uncertainty in predicting changes for any single country.

Meanwhile, progress in crop science has shown that most crops show fairly rapid declines in productivity as temperatures rise above critical thresholds, with as much as 10 percent yield loss for +1°C of warming in some locations. Both sub-Saharan Africa and South Asia appear particularly prone to productivity losses from climate change, in part because major staples in these regions are often already grown well above their optimum temperature.

Approaches to climate adaptation should recognize these realities, and should not equate anticipating climate changes with the considerably harder task of predicting next year’s weather. Predicting and building resilience to climate variability still remain important goals for agricultural development, but adaptation efforts should balance these activities with those focused more on the specific threats presented by climate change. Heat tolerant crop varieties and strategies to deal with heavy rainfall provide two examples of important needs. Similarly, balance is needed between the local-scale efforts that attract most of adaptation investment currently, and regional and global networks to develop needed technologies. Given the greater certainty of climate changes at broader scales, as well as the positive track record of international networks for crop breeding, investments in these global systems are very likely to deliver substantial adaptation benefits. Finally, given the downward pressures that climate change will exert on smallholder farm productivity in sub-Saharan Africa, and the critical role productivity gains play in catalyzing an escape from poverty, speeding the pace of investment in African agriculture can also be viewed as a good bet for climate adaptation.

Promotion of smallholder irrigation is cited as a strategy for enhancing income generation and food security for sub-Saharan Africa’s poor farmers, but what makes this technology a successful poverty alleviation tool? In the short run, the technology should pave the way for increased consumption, asset accumulation, and reduced persistent poverty among users. Over the longer run, it should lead to institutional feedbacks that support sustained economic development and nutritional improvements. Our conceptual model and review of case studies reveal the importance of three sub-components of irrigation technology—access, distribution, and use—and the ways in which the design of the technology itself can either bridge, or succumb to, institutional gaps. These critical features are illustrated in an experimental evaluation of a solar-powered drip irrigation project in rural northern Benin, which provides a controlled study of technology impacts in the Sudano-Sahel. The combined evidence highlights the technical and institutional requirements for project success and points to two important areas of research in the scale-up of any small-scale irrigation strategy: the risk behavior of water users, and the evolution of institutions that either support or obstruct project replication over space and time.

This paper was prepared for Stanford University’s Global Food Policy and Food Security Symposium Series, hosted by the Center on Food Security and the Environment, and supported by the Bill and Melinda Gates Foundation.