The prevalence of P2P technology contributes the ease of file-sharing, but it also brings some problems such as free-riding and copyright infringement. To solve these problems, many researchers have proposed incentive mechanisms. At the same time, some applications and business models based on P2P technology are developed. However, those new business models would not allow users redistribute files which they have downloaded. This study proposes a new business model which adopts a reselling mechanism to distribute contents on P2P networks. Users who have downloaded contents from peers other than the author peer can resell them by paying a portion of reselling gains as the royalty to the author. Content providers and consumers may re-price shared contents dynamically via incremental learning. This study aims to verify that authors can gain more through P2P reselling model than through client/server architecture, and we will investigate into the factors which influence this possibility. Keywords: Peer-to-Peer, Reselling mechanism, Multi-Agent Simulation, Dynamic pricing.