Market-based pollution control mechanisms such as pollution levy and cap and trade have received increasing attention from both academics and practitioners. A good understanding of the optimal pollution price under these mechanisms is a premise for regulators to make sound pollution control policies. In this paper, we propose a method for deriving the optimal pollution price for a given pollution target. This method consists of two steps that integrate cost function estimation and market equilibrium analysis: First, historical data is used to estimate the pollution abatement cost functions of the polluters; second, market models are used to solve the equilibrium pollution price under each control mechanism. For illustration, we apply the method to investigate SO2 emission control policies in China, using a dataset of SO2 emissions and abatement costs from three major industry sectors (Electricity, Gas, and Water Supply; Manufacturing; and Mining). Our analysis shows that the optimal levy rate is significantly higher than the actual rate adopted by the Chinese government. For example, the optimal levy rate should be 4.92RMB/kg, while the actual rate is 1.26RMB/kg in 2010. As a result, the actual emission structure is much less efficient: The overall cost savings would be 49.7% for all three sectors during 2006–2010 if the optimal emission structure is achieved. These findings have useful policy implications for the Chinese government. In addition, the method may be applied to analyze control policies at different aggregate levels (for example, provincial economies) or for other pollutants (for example, CO2 and chemical oxygen demand).