This paper proposes a fully distributed social welfare optimization solution that solves the economic dispatch and demand response problems in an integrated way. Compared with sequentially implementing these two operations one after another, the integrated solution can efficiently maximize the benefits of customers and minimize the generation cost of generators simultaneously. By adjusting both generations and dispatchable loads, line flow constraints and generation bounds can be satisfied easier. The proposed solution has two layers of operations for consensus-based information discovery and gradient-based generation or demand adjustment, respectively. It is fully distributed in the sense that there is no need for a specialized/central controller to coordinate the operations of the autonomous local controllers (agents). Compared with centralized solutions, the multiagent system-based distributed solution is more reliable against single-point failures and can better accommodate customer participation. The proposed solution has been tested with a 5-bus system and the IEEE 30-bus system under light- and heavy-load conditions. Both static optimization and dynamic simulation results are provided to demonstrate the performance of the proposed solution.