The study analyzes the effect of risk aversion on the optimal level of fertilizer by employing a pooled time-series cross-sectional survey data collected from 46 rainfed rice farmers in Tarlac, Central Luzon, Philippines. Based on a heteroscedastic specification of production function, fertilizer was found to be a risk-increasing input, but the effect of risk aversion on the optimal fertilizer use was estimated to be moderate. Explanations for the low average rate of application of fertilizer under rainfed conditions thus need to be found elsewhere.