Background and Motivation
China's national carbon market has grown rapidly in recent years, emerging as one of the world's largest Emissions Trading Systems (ETS). Carbon price volatility not only affects market stability and pricing credibility but also influences corporate investment and emissions strategies. While prior research has identified various factors affecting carbon price fluctuations, most studies focus on a narrow set of variables and rarely compare broader potential drivers across regions. This leaves a gap in understanding which factors are truly critical in explaining volatility dynamics in China's ETS markets, especially given the frequency mismatch between daily carbon prices and monthly or quarterly macroeconomic indicators.
Methodology and Scope
To address these challenges, researchers from the University of Science and Technology of China and Southwest University of Science and Technology developed an integrated GARCH-MIDAS-Adaptive-Lasso (GM-AL) model. This framework combines the ability to handle mixed-frequency data with advanced variable selection techniques, enabling the identification of the most influential predictors from a large set of macroeconomic, financial, energy, and environmental variables. The study focuses on three major regional ETS pilots in China: Hubei, Guangdong, and Shenzhen, using daily carbon price data from 2014 to 2023. A structured pool of low-frequency variables—including energy prices, financial indices, policy uncertainty indicators, and environmental factors—was analysed to uncover region-specific drivers of carbon price volatility.
Key Findings and Contributions
The study reveals clear regional differences in what drives carbon price volatility:
- In Hubei, the electricity and energy sectors (measured by the CSI 300 Electricity and Energy Indices) are the primary drivers.
- In Guangdong and Shenzhen, crude oil prices and the energy index play a more dominant role.
- Overall, energy-related factors exert the strongest influence on China's carbon market volatility, while policy and environmental variables show limited impact.
The proposed GM-AL model significantly outperforms benchmark models in both forecast accuracy and economic value. It also demonstrates robustness across different weighting schemes and alternative dimensionality reduction methods. The research contributes to the literature by systematically integrating multidimensional factors into volatility modelling and providing a scalable framework for other developing countries seeking to establish or enhance their own ETS mechanisms.
Why It Matters
Understanding the drivers of carbon price volatility is crucial for policymakers, regulators, and market participants. The findings highlight the importance of energy market signals and regional industrial structures in shaping carbon price dynamics. By identifying key predictors, this research supports the development of early warning systems and more responsive regulatory frameworks. It also offers investors improved tools for risk management and decision-making in carbon markets.
Practical Applications
The GM-AL model can be used by:
- Regulators to design differentiated, region-specific policies and stabilise carbon markets through proactive monitoring.
- Investors and financial institutions need to enhance volatility forecasting, optimise portfolio strategies, and assess carbon market risks.
- Energy and industrial firms need to anticipate better compliance costs and adjust emissions strategies.
- International researchers and policymakers in other emerging economies can use it as a reference for building robust carbon pricing systems.
Discover high-quality academic insights in finance from this article published in China Finance Review International . Click the DOI below to read the full-text!
