Demand response is a key solution in smart grid to address the ever-increasing peak energy consumption. With multiple utility companies, users will decide from which utility company to buy electricity and how much to buy. Consequently, how to devise distributed real-time demand response in the multiseller-multibuyer environment emerges as a critical problem in future smart grid. In this paper, we focus on the real-time interactions among multiple utility companies and multiple users. We propose a distributed real-time demand response algorithm to determine each user's demand and each utility company's supply simultaneously. By applying dual decomposition, the original problem is firstly decoupled into single-seller-multibuyer subsystems; then, the demand response problem in each subsystem can be distributively solved. The major advantage of this approach is that each utility company and user locally solve subproblems to perform energy allocation, instead of requiring a central controller or any third party. Therefore, privacy is guaranteed because no entity needs to reveal or exchange private information. Numerical results are presented to verify efficiency and effectiveness of the proposed approach. © 2014 IEEE.