Climate change is predicted to affect global rainfall patterns, but there is mixed evidence with regard to the effect of rainfall on civil conflict. Even among researchers who argue that rainfall reduces civil conflict, there is disagreement as to the underlying mechanism. Using data from the Philippines for the period 2001-2009, we exploit seasonal variation in the relationship between rainfall and agricultural production to explore the connection between rainfall and civil conflict. In the Philippines, above-average rainfall during the wet season is harmful to agricultural production, while above-average rainfall during the dry season is beneficial. We show that the relationship between rainfall and civil conflict also exhibits seasonality, but in the opposite direction and with a one-year lag. Consistent with the hypothesis that rebel groups gain strength after a bad harvest, there is evidence that lagged rainfall affects the number of violent incidents initiated by insurgents but not the number of incidents initiated by government forces. Our results suggest that policies aimed at mitigating the effect of climate change on agricultural production could weaken the link between climate change and civil conflict.