Abstract

This study evaluates the value of perfect forecasts of El Niño phases for selected rain-fed agricultural locations of Chile. The analysis framework incorporates a soil-crop-atmosphere system model and employs an expected utility decision algorithm that reflects a wide range of possible risk attitudes. The value of perfect forecasts is generally greater than zero indicating that real El Niño forecasts could potentially have economic value. Forecast value depends upon crop and location. The value of forecasts increases as the agricultural system becomes more susceptible to climatic variability. Among the regions under study, Temuco (38.5°S) and Valdivia (39.4°S) are likely to see the largest gains from long-term sea surface temperature forecasts. © 2003 Elsevier Science B.V. All rights reserved.