Monitoring is one way to alleviate the moral hazard problem rampant in most crop insurance programs. Lacking explicit data on monitoring, we test the effectiveness of monitoring on reducing moral hazard behavior indirectly. We first propose a theoretical model that takes into account several features of the Philippines crop insurance program – the empirical setting of interest. Our model predicts that if monitoring is effective, then crop insurance should have a positive effect on the use of certain inputs. Our empirical analysis of a survey dataset of corn farmers in the Philippines confirms this theoretical prediction and lends empirical support to the hypothesis that monitoring is effective in reducing moral hazard behavior by the farmers.