This paper seeks to understand the main sources of uncertainty in assessing the impacts of climate change on agricultural output, international trade, and poverty. We incorporate biophysical uncertainty by sampling from a distribution of productivity shocks reflecting the impacts of climate on agricultural yields in 2030. These shocks, in turn, affect the global economy. The response of economic agents to climate change is the second source of uncertainty in our estimates. We find that, even though it is difficult to predict where in the world agricultural crops will be favorably affected by climate change, we find that the responses of output and exports can be far more robust. This is due to the fact that supply and demand decisions depend on relative prices, and relative prices depend on productivity changes relative to other crops in a given region, or relative to similar crops in other parts of the world. We also find that uncertainty in poverty impacts of climate change appears to be almost entirely driven by biophysical uncertainty. Further reducing this uncertainty will require improving climate models and the crop models used to interpret the impacts of likely future climate on crop productivity.