Clustering Analysis of Keywords and Hot Topics in Carbon Finance and Carbon Market Research

3.2 Clustering Analysis of Keywords

Clustering analysis of keywords is based on the co-occurrence analysis of keywords, and it classifies dispersed keyword nodes into closely related clusters using specific algorithms (Martinez et al., 2019). In order to further explore the themes of carbon finance research, we conducted keyword clustering analysis in Citespace. Since it's difficult to find specific information in keywords solely from the clustering names, it's necessary to analyze the specific research content included in each cluster by combining the main keywords in the clusters in a more in-depth manner (Huang et al., 2020). Based on the network structure and the clarity of the clusters, CiteSpace provides module values and average silhouette values to evaluate the effectiveness of the network visualization. Figure 6 illustrates the keyword clustering clusters of carbon finance and carbon market, with a Q value of 0.4787 (greater than 0.3) and an S value of 0.7534 (greater than 0.5), indicating that the clustering is significant and the reference value is relatively high. Based on the size of the nodes in the clusters, we selected the top 8 clusters for presentation, namely #0 supply chain, #1 emissions trading, #2 carbon sequestration, #3 taxation, #4 economic development, #5 carbon market, #6 power market, and #7 forecasting. For the convenience of analysis, the main keywords included in these 8 clusters are listed in Table 3.

Figure 6. Clustering diagram of keywords in carbon finance and carbon market research. Table 3. Clustering information of keywords in carbon finance and carbon market research.

3.3 Summary of Hot Topics

In order to analyze the topics related to carbon finance and carbon market more reasonably and accurately, this study, based on the co-occurrence and clustering diagrams generated by the software, read literature nodes and conducted manual screening, summarizing them into the following categories:

• Carbon market policies and carbon trading mechanism research
• Carbon finance and carbon market risk research
• Carbon market spillover effect research
• Carbon finance tool design, pricing, and price forecasting research

1. Design of Carbon Trading Market Mechanisms

In order to reduce the total cost of achieving emission targets, the Kyoto Protocol stipulates three mechanisms: Joint Implementation (JI), Clean Development Mechanism (CDM), and Emission Trading (ET). Since the implementation of this agreement in 2005, governments around the world have been taking measures to reduce greenhouse gas emissions. Scholars at home and abroad have conducted extensive research on carbon trading market mechanisms, mainly focusing on carbon quota supply and demand mechanisms (Ji et al., 2017), price mechanisms (Ji et al., 2018a), competition mechanisms (Liu et al., 2022), and risk mechanisms and regulatory mechanisms (Hepburn, 2007). The pricing of carbon emission rights under the total control and emission trading mechanisms is the core issue in the field of carbon finance for solving the problem of fund integration. For example, based on price theory, Ji et al. (2018b) analyzed the theoretical foundation and transmission mechanism of carbon price formation from the perspective of the main influencing subjects; Du et al. (2020) discussed the design of emission quota allocation mechanism from an operational perspective by establishing a Stackelberg model.

2. Measurement and Prevention of Carbon Finance and Carbon Market Risks

With the continuous increase in the scale of carbon finance transactions worldwide, the corresponding risks it faces have also increased, making carbon finance risk management and prevention an important challenge. Therefore, many researchers have conducted studies on the measurement and prevention of carbon finance and carbon market risks, mainly including price risk (Fehr and Hinz, 2006; Dutta, 2018), policy risk (Song et al., 2018; Lin and Jia, 2019), and market risk (Jiao et al., 2018; Jin et al., 2020). Wang and Yan (2022) used the Copula-EVT-VaR model to measure the comprehensive risk of the Chinese carbon market, and the results showed that compared with other methods that do not consider the interdependence between risk factors, using the Copula function to measure the integrated risk of the carbon market is more effective. Tang et al. (2015) used the Capital Asset Pricing Model to analyze the market risk of the European Union Emission Trading System (EU ETS) and the Clean Development Mechanism (CDM), and the results showed that carbon prices are influenced by market mechanisms and external factors (economic crises and environmental policies) under low return expectations. However, under high return expectations, the carbon price changes in the EU ETS market are less stable and carry higher risks compared to the CDM market.

3. Research on Carbon Market Spillover Effects

The carbon finance market is widely recognized as an effective means to reduce carbon emissions and develop a low-carbon economy, and therefore, it is closely linked to energy markets and financial markets. In recent years, the spillover effects between carbon markets, as well as between carbon markets and energy markets or financial markets, have become a hot topic of concern for many researchers. The main research methods include Granger causality test (Yu et al., 2015; Peng et al., 2020), wavelet analysis (Khalfaoui et al., 2015; Ren et al., 2022), vector autoregression (VAR) (Zhang and Sun, 2016; Wang and Guo, 2018), GARCH model (Balcılar et al., 2016; Zhao et al., 2023), and DY spillover index (Zhang and Sun, 2016; Tan et al., 2020), among others.

4. Research on Design, Pricing, and Price Forecasting of Carbon Finance Tools

Firstly, carbon finance tools are similar to other financial tools and can be divided into primary products and derivatives. The basic trading product of carbon finance is carbon emission rights, and there are many types of derivatives, such as carbon futures (Chevallier, 2009; Evans and Karvonen, 2014; Creutzig, 2016), carbon forwards (Andrews, 2021), carbon options (Viteva et al., 2014; Liu and Huang, 2021), and carbon swaps (Kanamura, 2016). The healthy and sustainable development of carbon finance requires the support of carbon finance tools, and the continuous innovation of carbon financial products contributes to a more perfect carbon finance system. Secondly, the carbon pricing mechanism is an important means to influence the supply-demand balance in the carbon finance market and is a key part of the carbon market mechanism. The main carbon pricing mechanisms include carbon tax (Klenert et al., 2018; Muhammad, 2022), carbon emission trading system (ETS) (Narassimhan et al., 2018; Rontard and Hernandez, 2022), carbon credit mechanism (Kale et al., 2009), outcome-based climate finance (RBCF) (Group et al., 2017), and internal carbon pricing (Zhu et al., 2022b). Finally, carbon emissions are a major contributor to environmental pollution and climate change, and the efficient operation of carbon emission trading markets effectively promotes the process of carbon reduction. The accurate prediction of carbon prices is of great importance for carbon trading market management, the formulation of relevant policies, and investment decisions. In carbon price prediction, researchers focus on discussing how to improve the accuracy and stability of carbon price prediction. They have proposed many carbon price prediction model methods, such as the deep neural network model TCN-Seq2Seq (Zhang and Wen), the carbon price prediction hybrid model based on quadratic decomposition and improved extreme learning machine (ELM) (Zhou and Wang, 2021a), the GARCH-LSTM hybrid model (Huang et al., 2021), the quadratic decomposition carbon price prediction model based on Sparrow search algorithm optimized kernel extreme learning machine (Zhou and Wang, 2021b), and other related carbon price prediction models.

4. Evolutionary Path of Carbon Finance and Carbon Markets

4.1 Timeline Diagram Analysis

To analyze the evolutionary path of carbon finance and carbon markets, this paper uses a timeline diagram to explore the development of carbon finance and carbon market research hotspots over time. The keyword timeline diagram shows the time when hotspots emerge on the horizontal axis, the automatically calculated cluster modules by the software on the vertical axis, each node represents a research topic, the circle layer and color represent the richness of the topic, and the connection lines represent the connection between different topics. From the figure, we can see that there are two periods of relatively intensive keywords in carbon finance and carbon markets. The reason is that the United Nations Framework Convention on Climate Change and the Kyoto Protocol were signed in 1992 and 1997, respectively, after which the academic community began to pay widespread attention to carbon finance and carbon markets. Secondly, a new wave of research related to carbon finance and carbon markets emerged after the Kyoto Protocol entered into force in 2005.