Pooled time series and cross-sectional data are used to investigate the behavior of rice farmers when a modern rice variety was introduced in a region of the Philippines in the 1970s. The data come from the Farm Record Keeping project in Iloilo, Philippines, conducted by the International Rice Research Institute Department of Agricultural Economics from the crop years 1975-1976 to 1979-1980. Labor inputs generally were found to enhance mean productivity and reduce production risk. Nitrogen fertilizer was not found to be necessarily risk increasing, as it was risk reducing with modern varieties under favorable conditions. Lower production risk was found to be linked with appropriate combinations of management and inputs, but not necessarily with any one component of the technology. During the first 2 years of modern-variety adoption, the modern varieties were less technically efficient than the traditional varieties. However, 4-5 years after introduction, the modern varieties were more efficient than the traditional-variety plots.