Internet fraud has been on the rise in online consumer-to-consumer (C2C) auction markets, posing serious challenges to people's trust in electronic markets. Among various remedies to promote trust and reduce trader's risk, online escrow service has been proposed as a trusted third party to protect online transactions from Internet fraud. However, whether an escrow service constitutes a viable business model for a trusted third party to effectively block Internet fraud remains an open question. This research proposes a dynamic game model for online traders and a profit maximization model for the escrow service provider. Through the investigation of the optimal strategies of online traders, we explore the relationships among traders' decision making, escrow service fee rates, and adoption rates. We reveal the demand for escrow services and establish the optimal pricing rule for the escrow service provider. A numerical study based on the theoretical analysis is conducted to provide detailed guidelines of the model application for an escrow service provider and to explore if the escrow service is a viable business model in C2C auction markets.