Providing privacy guarantees has been one of the primary motivations of Federated Learning (FL). However, to guarantee the client-level differential privacy (DP) in FL algorithms, the clients' transmitted model updates have to be clipped before adding privacy noise. Such clipping operation is substantially different from its counterpart in the centralized differentially private SGD and has not been well-understood. In this paper, we first empirically demonstrate that the clipped FedAvg can perform surprisingly well even with substantial data heterogeneity when training neural networks, which is partly because the clients' updates become similar for several popular deep architectures. Based on this key observation, we provide the convergence analysis of a DP FedAvg algorithm and highlight the relationship between clipping bias and the distribution of the clients' updates. To the best of our knowledge, this is the first work that rigorously investigates theoretical and empirical issues regarding the clipping operation in FL algorithms.