To clarify the impacts of El Nino events on production of major food crops in Asia and the Pacific, log linear regression, incorporating a El Nino year dummy variable was conducted. Production variability was measured by percentage deviation from a five-year moving average as trend. ENSO (El Nino Southern Oscillation) sensitivity was measured by correlation coefficient between percentage deviation from the five-year moving average and monthly average of SOI (June to September in the year concerned). Estimated period was 1961 to 2000. Six crops (rice, maize, soybean, groundnut, sweet potato and cassava) and eleven countries (Myanmar, Thailand, Laos, Cambodia, Viet Nam, Malaysia, Indonesia, the Philippines, Australia, New Zealand, and Fiji) were covered. The El Nino event is becoming more frequent as it occurred once in the 1960s, twice in the 1970-1980s and three times in the 1990s. Compared with production loss in non-El Nino years, that in El Nino years was characterized as: 1. Absolute magnitude (7 per cent loss during El Nino vs. 3 per cent loss during non-El Nino); 2. Simultaneity of both yield and area decline (production loss during non-El Nino was mostly caused by yield decline); 3. Single year phenomenon with recovery of production in the following year. Maize was most variable and sensitive to ENSO for both the area harvested and yield, followed by soybean which was significantly sensitive to ENSO in terms of area but not yield. A unique feature of maize is its positive correlation between area harvested and yield. Rice is most stable, while its area harvested is moderately affected by ENSO. Cassava, sweet potato and groundnut were not significantly affected by ENSO, although their production variability was larger than that of rice. Enhancement of the production stability of maize and soybean and diversification of farming systems through incorporating root crops are expected to improve the stability of food crop production in the region.