An agent-based model for an air emissions cap and trade program: A case study in Taiwan

To determine the actual status of individuals in a system and the trading interaction between polluters, this study uses an agent-based model to set up a virtual world that represents the Kaohsiung and Pingtung regions in Taiwan, which are under the country's air emissions cap and trade program. The model can simulate each controlled industry's dynamic behavioral condition with the bottom-up method and can investigate the impact of the program and determine the industry's emissions reduction and trading condition. This model can be used elastically to predict the impact of the trading market through adjusting different settings of the program rules or combining the settings with other measures. The simulation results show that the emissions trading market has an oversupply, but we find that the market trading amounts are low. Additionally, we find that increasing the air pollution fee and offset rate restrains the agents' trading decision, according to the simulation results of each scenario. In particular, NO_x and SO_x trading amounts are easily impacted by the pollution fee, reduction rate, and offset rate. Also, the more transparent the market, the more it can help polluters trade. Therefore, if authorities want to intervene in the emissions trading market, they must be careful in adjusting the air pollution fee and program rules; otherwise, the trading market system cannot work effectively. We also suggest setting up a trading platform to help the dealers negotiate successfully.