Does green direct financing work in reducing carbon risk?

doi:10.1016/j.econmod.2023.106495

Economic Modelling 经济建模

Volume 128, November 2023, 106495
卷 128， November 2023， 106495

https://doi.org/10.1016/j.econmod.2023.106495 Get rights and content 获取权限和内容

Highlights 突出

•
We examine the relation between green bond issuance (GBI) and firms' carbon risk.
我们研究了绿色债券发行（GBI）与企业碳风险之间的关系。
•
We find that GBI reduces firms' carbon risk.
我们发现GBI降低了企业的碳风险。
•
GBI reduces firms' carbon risk by improving their energy efficiency.
GBI通过提高能源效率来降低企业的碳风险。
•
GBI reduces firms' carbon risk by improving their energy consumption structure.
GBI通过改善企业的能源消费结构来降低企业的碳风险。
•
Green direct financing plays a positive role in reducing firms' carbon risk.
绿色直接融资在降低企业碳风险方面发挥着积极作用。

Abstract 抽象

How green direct financing represented by green bonds affects carbon risk is a topic of great significance. The impact of green indirect financing represented by green credit on carbon risk has been widely discussed, while the relationship between green bonds and carbon risk is yet to be explored. We examine the effect of green bond issuance (GBI) on the carbon risk of firms using the data of Chinese listed firms from 2009 to 2019. We find that GBI reduces firms' carbon risk. We verify the channels of energy efficiency and energy consumption structure through which GBI reduces firms' carbon risk. Furthermore, we also find that the negative effect of GBI on the carbon risk of firms is more pronounced for firms with lower GBI costs and a higher level of digital transformation. We are the first to prove the positive role of green direct financing in reducing carbon risk.
以绿色债券为代表的绿色直接融资如何影响碳风险，是一个具有重要意义的话题。以绿色信贷为代表的绿色间接融资对碳风险的影响已被广泛讨论，而绿色债券与碳风险的关系尚待探讨。本文利用2009—2019年中国上市公司的数据，研究了绿色债券发行（GBI）对企业碳风险的影响。我们发现GBI降低了企业的碳风险。我们验证了GBI降低企业碳风险的能源效率和能源消耗结构的渠道。此外，我们还发现，GBI对企业碳风险的负面影响在GBI成本较低、数字化转型水平较高的企业中更为明显。我们率先证明了绿色直接融资在降低碳风险方面的积极作用。

Keywords 关键字

Green direct financing

Green bonds

Carbon risk

Staggered difference-in-differences model

绿色直接融资绿色债券碳风险交错双重差分模型

1. Introduction 1. 引言

For a long time, China has maintained a development model characterized by high levels of pollution and energy use, which has brought serious environmental pollution problems (Wang et al., 2023). The Chinese government actively encourages firms to become more environmentally friendly in order to achieve low-carbon economic growth. It is more important for firms to actively boost their investment in the adoption of green technologies if they want to undergo a green transformation (Kabir et al., 2021). The long investment period and low return of green technology transformation significantly reduce the internal motivation for green transformation in firms. After the proposal of the double carbon target in 2020, China accelerated the pace of green transformation and introduced a series of policy measures to promote the green transformation of firms, leading to a significant increase in the financial vulnerability of firms in the process of transitioning from a fossil fuel-based to a low-carbon economy, and this financial vulnerability is known as carbon risk (Nguyen and Phan, 2020; Shu and Tan, 2023). Carbon risk has significant adverse effects on the real economy and financial system, which has attracted increasing attention (Nguyen and Phan, 2020; Bolton and Kacperczyk, 2021; Bose et al., 2021; Zhu and Hou, 2022; Shu and Tan, 2023).
长期以来，我国一直保持着高污染和高能耗的发展模式，这带来了严重的环境污染问题（Wang et al.， 2023）。中国政府积极鼓励企业更加环保，以实现低碳经济增长。对于企业来说，如果想进行绿色转型，积极增加对采用绿色技术的投资更为重要（Kabir et al.， 2021）。绿色技术改造投资周期长、回报低，显著降低了企业绿色转型的内在动力。2020年双碳目标提出后，中国加快绿色转型步伐，出台一系列政策措施促进企业绿色转型，导致企业在从化石燃料向低碳经济转型过程中的财务脆弱性显著增加，这种金融脆弱性被称为碳风险（Nguyen and Phan， 2020;Shu 和 Tan，2023 年）。碳风险对实体经济和金融体系具有显著的不利影响，越来越受到关注（Nguyen and Phan， 2020;Bolton 和 Kacperczyk，2021 年;Bose 等人，2021 年;Zhu 和 Hou，2022 年;Shu 和 Tan，2023 年）。

Green bonds refer to negotiable securities that are issued to support green industries (Bhutta et al., 2022; Reboredo et al., 2022; Wang and Jiang, 2023). The Guidelines on GBI issued on December 31, 2015 aim to actively play the role of corporate bond financing in promoting green development, solving prominent environmental problems, and coping with climate change. Fig. 1, Fig. 2 show the number and amount of green bonds issued between 2016 and 2021. Varieties, including carbon neutral bonds, green rural revitalization bonds, and green sustainable development bonds, have been launched. China's regulators have released guidelines to standardize the requirements and application of green bonds in order to promote the growth of green bond markets.
绿色债券是指为支持绿色产业而发行的可转让证券（Bhutta et al.， 2022;Reboredo 等人，2022 年;王和江，2023）。2015年12月31日发布的《全球投资指南》旨在积极发挥公司债券融资在促进绿色发展、解决突出环境问题、应对气候变化等方面的作用。图1、图2显示了2016年至2021年间发行的绿色债券的数量和数量。碳中和债券、绿色乡村振兴债券、绿色可持续发展债券等品种陆续推出。中国监管机构发布了规范绿色债券要求和应用的指南，以促进绿色债券市场的发展。

The R&D of green and low-carbon technology has the characteristics of a long investment period and high risk, which also weakens the motivation of firms to engage in green and low-carbon technology R&D. Green bonds are an important green direct financing tool and a driving force in green development. Green bonds have the characteristics of a long issuance period and low financing costs, which are matched by the long investment period and high risk of green and low-carbon technology R&D. The long use period and low use cost of the funds raised by green bonds encourage firms to be more willing to engage in green and low-carbon technology R&D (Tang and Zhang, 2020). The issuance of green bonds by firms also sends a signal to the market that they actively fulfill their social responsibilities, thereby winning them a good social reputation and recognition from stakeholders (Flammer, 2021), and is in line with the green preference of bank loans, which helps alleviate their financing constraints and reduce financing costs (Del Gaudio et al., 2022), and encourages them to increase their investment in green and low-carbon technology R&D (Wang et al., 2022). The improvement of green and low-carbon technologies can improve the energy efficiency of firms, thereby reducing their carbon emissions (Sun et al., 2019; Zhang et al., 2018; Li et al., 2022). Stakeholder support can help reduce the financial vulnerability of firms in low-carbon transformation, thereby reducing their carbon risk. The funds raised by firms through GBI are earmarked for designated green projects, including the efficient and clean utilization of coal and oil, as well as the development and utilization of natural gas and new energy, with a focus on supporting the replacement of coal and oil with natural gas and new energy. Therefore, GBI may change the energy consumption structure of firms, reduce their consumption of coal and oil, increase their consumption of natural gas and new energy, and further reduce firms' carbon emissions and their financial vulnerability in low-carbon transformation, so as to reduce their carbon risk. However, it remains unclear whether GBI will have an impact on firms' carbon risk and how it affects firms' carbon risk.
绿色低碳技术研发具有投资周期长、风险高等特点，也削弱了企业从事绿色低碳技术研发的积极性，绿色债券是重要的绿色直接融资工具，是绿色发展的驱动力。绿色债券具有发行周期长、融资成本低等特点，与绿色低碳技术研发投资周期长、风险高等特点相匹配。绿色债券募集资金使用周期长、使用成本低，促使企业更愿意从事绿色低碳技术研发（Tang and Zhang，2020）。企业发行绿色债券也向市场发出了积极履行社会责任的信号，从而赢得了良好的社会声誉和利益相关者的认可（Flammer，2021），符合银行贷款的绿色偏好，有助于缓解融资限制，降低融资成本（Del Gaudio et al.， 2022），并鼓励他们加大对绿色低碳技术研发的投入（Wang et al.， 2022）。绿色低碳技术的改进可以提高企业的能源效率，从而减少碳排放（Sun et al.， 2019;Zhang等人，2018;Li 等人，2022 年）。利益相关者的支持有助于降低低碳转型企业的财务脆弱性，从而降低其碳风险。企业通过GBI筹集的资金专门用于指定的绿色项目，包括煤炭和石油的高效清洁利用，以及天然气和新能源的开发和利用，重点是支持以天然气和新能源替代煤炭和石油。因此，GBI可以改变企业的能源消费结构，减少煤炭和石油的消费，增加天然气和新能源的消费，进一步降低企业在低碳转型中的碳排放和财务脆弱性，从而降低企业的碳风险。然而，目前尚不清楚GBI是否会对企业的碳风险产生影响，以及它如何影响企业的碳风险。

In order to verify the above argument that GBI affects firms' carbon risk, we use data from Chinese listed firms to examine the impact of GBI on firms' carbon risk. We find that GBI significantly reduces carbon risk. After robustness tests such as the parallel trend test, endogenous test, and change of carbon risk measurement method, the above findings remain correct. We further explore the potential mechanisms by which GBI affects firms' carbon risk, taking into account two factors: energy efficiency and energy consumption structure. Our results show that GBI improves energy efficiency and improves energy consumption structure by reducing the proportion of coal and oil consumption, thereby reducing firms' carbon risk. We further analyze the heterogeneity of the impact of GBI on firms' carbon risk. We find that GBI has a greater negative impact on the carbon risk of firms with lower GBI costs and a higher level of digital transformation.
为了验证上述GBI影响企业碳风险的论点，本文利用中国上市公司的数据研究了GBI对企业碳风险的影响。我们发现GBI显著降低了碳风险。经过平行趋势检验、内生检验、碳风险测度方法变化等稳健性检验，上述结果仍然正确。本文进一步探讨了全球生物多样性指数影响企业碳风险的潜在机制，同时考虑了能源效率和能源消费结构两个因素。结果表明，GBI通过降低煤油消费比例，提高能源效率，改善能源消费结构，从而降低企业碳风险。进一步分析了GBI对企业碳风险影响的异质性。我们发现，GBI对GBI成本较低、数字化转型水平较高的企业的碳风险有较大的负面影响。

We choose China as the research sample for the following reasons. First, China's carbon emissions rank first among countries in the world and show an increasing trend. China is currently promoting low-carbon transformation of the economy to achieve high-quality development. Especially after China proposed the double carbon goal. China has introduced a large number of carbon emission reduction planning and policy measures to accelerate the smooth realization of the low-carbon transformation of the economy and the double carbon goal. In this process, the carbon risks faced by firms will significantly increase, which will have a negative impact on the real economy and financial system, which is not conducive to the low-carbon transformation of the economy and the smooth realization of the double carbon goal and thus affect the sustainable development of the global economy. Therefore, using China as a research sample is of great significance for exploring the sustainable development of the global economy. Second, China's GBI scale ranks among the top in the world, and the GBI scale of emerging market countries represented by China is rapidly growing and taking up a rising proportion of the total global GBI scale. Emerging market countries are in a stage of rapid economic development. Compared to developed countries, the development of green bonds in these countries is relatively late. The green bond market generally has the characteristics of imperfect green bond standard systems and regulatory systems, and these countries generally face significant carbon emission reduction pressure. Therefore, our results using China as a sample can provide a reference for other countries to better develop the green bond market and play the role of green bonds in reducing carbon risk.
我们选择中国作为研究样本，原因如下。首先，中国的碳排放量在世界各国中排名第一，并呈上升趋势。当前，中国正在推进经济低碳转型，实现高质量发展。尤其是在中国提出双碳目标之后。中国出台了大量碳减排规划和政策措施，加快顺利实现经济低碳转型和双碳目标。在这个过程中，企业面临的碳风险将显著增加，对实体经济和金融体系产生负面影响，不利于经济低碳转型和双碳目标的顺利实现，从而影响全球经济的可持续发展。因此，以中国为研究样本，对于探索全球经济的可持续发展具有重要意义。其次，中国GBI规模位居世界前列，以中国为代表的新兴市场国家的GBI规模增长迅速，占全球GBI总规模的比重不断上升。新兴市场国家正处于经济快速发展阶段。与发达国家相比，这些国家绿色债券的发展相对较晚。绿色债券市场普遍存在绿色债券标准体系和监管体系不完善的特点，这些国家普遍面临较大的碳减排压力。因此，以中国为样本的研究结果可为其他国家更好地发展绿色债券市场，发挥绿色债券降低碳风险的作用提供参考。

The contributions of this paper are as follows. First, we contribute to the growing literature on green bonds. Existing literature mainly focuses on the impact of green bonds on stock returns (Baulkaran, 2019; Wang et al., 2020), stock liquidity (Tang and Zhang, 2020), environmental performance (Flammer, 2021), capital cost (Zhang et al., 2021), country value (Dell’ Atti et al., 2022), etc. Sartzetakis (2021) theoretically analyze the important role of green bonds in low-carbon transformation financing and propose some suggestions for developing the green bond market. Unlike existing literature, we empirically investigate the impact of GBI on firms' carbon risk and find that GBI can significantly reduce firms' carbon risk. We verify that green direct financing helps reduce firms' carbon risk, providing new evidence for the positive role of green direct financing in low-carbon transformation.
本文的贡献如下。首先，我们为越来越多的绿色债券文献做出了贡献。现有文献主要关注绿色债券对股票回报的影响（Baulkaran，2019;Wang et al.， 2020）、股票流动性（Tang and Zhang， 2020）、环境绩效（Flammer， 2021）、资本成本（Zhang et al.， 2021）、国家价值（Dell' Atti et al.， 2022）等。Sartzetakis（2021）从理论上分析了绿色债券在低碳转型融资中的重要作用，并提出了一些发展绿色债券市场的建议。与现有文献不同，我们实证研究了GBI对企业碳风险的影响，发现GBI可以显著降低企业的碳风险。验证了绿色直接融资有助于降低企业碳风险，为绿色直接融资在低碳转型中的积极作用提供了新的证据。

Second, we contribute to the literature on carbon risk. Existing literature mainly focuses on the impact of carbon risk on equity capital costs (Kim et al., 2015), stock returns (Oestreich and Tsiakas, 2015; Bolton and Kacperczyk, 2021), dividend policy (Balachandran and Nguyen, 2018), debt financing costs (Jung et al., 2018), capital structure (Nguyen and Phan, 2020), acquisition decisions (Bose et al., 2021), etc. Less attention has been paid to the influencing factors of firms' carbon risk and how to reduce it, while how to effectively reduce firms' carbon risk is undoubtedly a topic of great significance. We explore the role of GBI in reducing firms' carbon risk from the perspective of green direct financing and further analyze the mechanisms by which GBI affects firms' carbon risk from two aspects: energy efficiency and energy consumption structure. We find that GBI improves energy efficiency and energy consumption structure and further reduces the carbon risk of firms. We provide a new approach to reduce firms' carbon risk and supplement and expand the literature in the field of carbon risk.

The rest of this paper is as follows. Section 2 is the theoretical analysis and assumptions. Section 3 shows the research design. Section 4 presents our empirical findings and examines their reliability. Section 5 shows mechanism tests. Section 6 introduces heterogeneity tests. Section 7 provides conclusion and policy implications.
本文的其余部分如下。第二部分是理论分析和假设。第 3 节展示了研究设计。第4节介绍了我们的实证发现并检验了其可靠性。第 5 节显示了机理测试。第 6 节介绍了异质性测试。第7节提供了结论和政策影响。

2. Related literature and hypothesis development
2. 相关文献和假设发展

Firms mainly face carbon risks during the low-carbon transformation process, and those with higher carbon emissions are more susceptible to sanctions or carbon-related risks, leading to higher financial vulnerability during the low-carbon transformation process (Bose et al., 2021; Shu and Tan, 2023). Chinese firms mainly rely on banks for financing. Banks usually prefer to provide loans to large state-owned firms with an implicit government guarantee (Dong et al., 2021). Banks also prefer short-term loans to long-term loans to reduce credit risks (Custódio et al., 2013). The long investment period and high-risk characteristics of green and low-carbon technology R&D increase the difficulty for firms to obtain loans from banks, which limits their investment in green and low-carbon technology R&D. By issuing green bonds, firms can save intermediary fees paid to third-party financial institutions for financing, and the financing cost of green bonds is lower than that of conventional bonds (Zhang et al., 2021), thereby promoting firms to increase investment in green and low-carbon technology R&D. The funds raised through green bonds are earmarked for designated green projects, and clean energy projects are heavily supported by green bonds. The use of funds and project progress need to be monitored and evaluated by third-party certification agencies to ensure that the raised funds are used for designated green projects (Tang and Zhang, 2020). The innovation of green and low-carbon technologies and the use of clean energy can help reduce firms' carbon emissions and improve their competitiveness, thereby reducing the financial vulnerability of firms in low-carbon transformation and their carbon risk.
企业在低碳转型过程中主要面临碳风险，碳排放量较高的企业更容易受到制裁或碳相关风险的影响，导致低碳转型过程中的财务脆弱性更高（Bose et al.， 2021;Shu 和 Tan，2023 年）。中国企业主要依靠银行进行融资。银行通常更愿意向大型国有企业提供有隐性政府担保的贷款（Dong et al.， 2021）。银行也更喜欢短期贷款而不是长期贷款，以降低信用风险（Custódio et al.， 2013）。绿色低碳技术研发的投资周期长、风险高，增加了企业从银行获得贷款的难度，限制了企业对绿色低碳技术研发的投入。通过发行绿色债券，企业可以节省向第三方金融机构支付的融资中介费用，绿色债券的融资成本低于传统债券（Zhang et al.， 2021），从而推动企业加大对绿色低碳技术研发的投入。通过绿色债券募集的资金专门用于指定的绿色项目，清洁能源项目则由绿色债券大力支持。资金的使用和项目进度需要由第三方认证机构进行监控和评估，以确保筹集的资金用于指定的绿色项目（Tang and Zhang，2020）。绿色低碳技术创新和清洁能源的使用有助于企业减少碳排放，提高竞争力，从而降低低碳转型企业的财务脆弱性和碳风险。

The issuance of green bonds by firms will attract more investors' attention, especially long-term and green investors who attach importance to the environment (Flammer, 2021). These investors, by playing a supervisory role, urge firms to improve information disclosure and engage in green transformation. The relationship between firms and stakeholders determines whether they can obtain sufficient resources to maintain their own development and resist risk shocks (Choi and Wang, 2009; Blake and Moschieri, 2017). The issuance of green bonds by firms also sends a signal to the market that they actively fulfill their social responsibilities, thereby winning good social reputation and recognition from stakeholders (Tang and Zhang, 2020; Flammer, 2021). The green image of firms helps to establish good relationships with the government, thereby obtaining more financial support and preferential policies. Banks are more inclined to invest in green projects to reduce credit risk, and consumers also prefer products produced by green firms under the government's green and low-carbon guidance, which helps alleviate financing constraints and enhance the competitiveness of firms. Therefore, firms establish good relationships with stakeholders by issuing green bonds to enhance their ability to resist risks and reduce their financial vulnerability during the low-carbon transformation process, thereby reducing their carbon risk. The arguments mentioned above lead to the following hypothesis.

Hypothesis 1

GBI reduces the carbon risk of firms.

Chinese firms mainly obtain funds through indirect financing. Indirect financing often has significant uncertainty and may have adverse effects on the normal business activities of firms. Therefore, firms will prioritize the use of funds for business activities that can generate profits in the short term, while their willingness to develop green and low-carbon technologies is relatively low. Green bonds, as an important direct financing tool, are characterized by a long issuance period and a lower financing cost. The issuance period of green bonds is usually more than three years, and the financing cost of green bonds is lower than that of conventional bonds. Besides, intermediary fees and other costs of indirect financing can be saved (Tang and Zhang, 2020; Zhang et al., 2021). The long issuance period of green bonds perfectly matches the long investment period of green and low-carbon technology R&D (Wang et al., 2022). The lower financing cost of green bonds also reduces the failure cost of green and low-carbon technology R&D, thereby enhancing the willingness of firms to engage in green and low-carbon technology R&D. Firms need to regularly disclose the use of funds raised through green bonds, the progress of green projects, environmental benefits, etc., and third-party certification agencies need to evaluate the green performance of firms (Tang and Zhang, 2020). These measures reduce information asymmetry between firms and investors, which is beneficial for investors to accurately evaluate the environmental performance of firms (El Ghoul et al., 2018), thereby enhancing the motivation of firms to obtain environmental competitive advantages through green and low-carbon technology R&D. Green and low-carbon technological innovation can effectively improve the production efficiency and resource utilization efficiency of firms, thereby improving their energy efficiency (Sun et al., 2019; Huang et al., 2019). The improvement of energy efficiency is one of the key factors in reducing firm carbon emissions and improving firm competitiveness (Trianni et al., 2016; Zhang et al., 2018; Li et al., 2022). The improvement of energy efficiency helps to reduce the financial vulnerability of firms in low-carbon transformation, thereby reducing their carbon risk. The arguments mentioned above lead to the following hypothesis.

Hypothesis 2

GBI reduces the carbon risk of firms by improving their energy efficiency.

The inherent resource conditions rich in coal in China make the cost of coal relatively low, leading to the dominance of coal consumption in China's energy consumption (Chen et al., 2020). The massive demand for oil in industrial production and transportation has led to China being the world's largest oil consumer. The carbon emissions of coal and oil are significantly higher than those of natural gas and other non-fossil fuels, so China's carbon emissions mainly come from coal and oil consumption. The carbon emissions generated by coal and oil consumption account for over 80% of China's total carbon emissions from energy consumption (Chen et al., 2020). The funds raised by firms through green bonds are earmarked for designated green projects, including the efficient and clean utilization of coal and oil, as well as the development and utilization of natural gas and new energy, with a focus on supporting the replacement of coal and oil with natural gas and new energy. In the Chinese green bond market, the scale of green bonds used for the development and utilization of new energy usually accounts for a large proportion of the total scale of green bonds. This may be due to the high investment and long payback period of new energy projects, which increase the difficulty for new energy firms to obtain loans from financial institutions. This further prompts them to obtain funds through the direct financing method of issuing green bonds (Liu et al., 2022). Therefore, GBI may change the energy consumption structure of firms, reduce their consumption of coal and oil, increase their consumption of natural gas and new energy, and further reduce their carbon emissions and financial vulnerability in low-carbon transformation, thereby reducing their carbon risk. The arguments mentioned above lead to the following hypothesis.

Hypothesis 3

GBI reduces the carbon risk of firms by improving their energy consumption structure.

3. Research design

3.1. Sample selection and data sources

China's listed firms and GBI data during the period from 2009 to 2019 are used in this paper. The GBI data of firms comes from the Wind database, and 85 firms that have issued green bonds are obtained. The operating cost of the industry comes from the China Industrial Statistical Yearbook, and the data on various industries' energy use is taken from the China Energy Statistics Yearbook. The carbon emission coefficients for different energy sources are derived from the IPCC National Greenhouse Gas Inventory Guidelines issued in 2006. The balance sheet, financial statements, and energy consumption are from the Wind database, the CSMAR database, and the China Statistical Yearbook. The economic policy uncertainty index released by Baker et al. (2016) is used to measure economic policy uncertainty. We winsorize all variables at 1% and 99% to alleviate the effect of extreme values.

3.2. Measures of carbon risk

Following Bose et al. (2021), Bolton and Kacperczyk (2021), and Zhu and Hou (2022), carbon emissions and carbon emission intensity are used to represent the carbon risk of firms, where carbon emissions are taken as a natural logarithm and carbon intensity is calculated by dividing a firm's annual carbon emissions by its operating income. Since the carbon emission information of firms is disclosed voluntarily, data on carbon emissions disclosed by Chinese listed firms is limited. In order to get as much information as possible about firms' carbon emissions, referring to previous studies (Clarkson et al., 2011; Chapple et al., 2013; Shu and Tan, 2023), we approximately calculate firms' carbon emissions using the ratio of the operating costs of firms to the operating costs of their industries. Operating costs are the costs generated by the production and operation activities of firms. Generally, higher production and operation costs mean that firms need to consume more energy in their production and operation activities, leading to higher carbon emissions. Energy is the main raw material for the production and operation activities of firms, and energy costs are also an important component of firm operating costs. Therefore, the ratio of firm operating costs to industry operating costs is used to approximate the carbon emissions of firms. Different energy consumption and associated carbon emission factors are used to calculate the carbon emissions of various industries.

3.3. Empirical model

Following previous literature (Zhang et al., 2021; Flammer, 2021), the relationship between GBI and the carbon risk of firms is investigated using the staggered difference-in-differences model. The specification we estimate is as follows:(1) $C R_{i, t} = α_{0} + α_{1} I S S U A N C E_{i, t} + β X_{i, t} + θ_{i} + η_{t} + ε_{i, t}$ where $C R_{i, t}$ is the carbon risk for firm $i$ in year $t$ . $I S S U A N C E_{i, t}$ is a dummy variable that equals 1 in the years after firm $i$ issues green bonds and 0 otherwise. The coefficient for $I S S U A N C E_{i, t}$ reflects the relationship between GBI and the carbon risk of firms. We also include the firm fixed effects ( $θ_{i}$ ) and the year fixed effects ( $η_{t}$ ). $ε_{i t}$ is the error term. We control the following factors that may affect firms' carbon risk: firm size ( $S I Z E$ ), firms on a larger scale have more resources available to invest in green and low-carbon technology R&D to reduce their carbon emissions; firm age ( $A G E$ ), older firms may have stronger risk awareness and are more willing to take the initiative to conduct green transformation to reduce their carbon risk; financial leverage ( $L E V$ ), highly leveraged firms may face high financing constraints (Cai and Zhang, 2011), thereby inhibiting their investment in green and low-carbon technology R&D; the number of board members ( $B O A R D$ ), the board of directors can play a supervisory role in urging firms to take measures to address carbon risk; the percentage of annual growth in a firm's revenue ( $G R O W T H$ ), higher operating revenue may encourages firms to focus on the development of their main businesses rather than undertake green transformation; return on assets ( $R O A$ ), firms with higher returns may have more spare funds to support their green transformation; the nature of property rights ( $S O E$ ), state owned firms need to respond to national policies for green transformation of the economy, making it more likely to invest in green and low-carbon technology R&D to reduce their carbon emissions; economic policy uncertainty ( $E P U$ ), higher economic policy uncertainty increases the uncertainty of future financing and operational environment risks for firms, and firms may tend to reduce risk investment to cope with future risks, thereby inhibiting their green transformation; environmental regulation ( $E R$ ) is measured by the pollutant discharge fees as in Xie et al. (2017). Strict environmental regulation may encourage firms to reduce carbon emissions to avoid penalties for failing to meet environmental standards. In addition, we also control the total bond issuance amount ( $T B A$ ) to test whether firms' carbon risk will be affected by their issuance of bonds other than green bonds.

3.4. Descriptive statistics

Table 1 shows the descriptive statistics before 2016 and after 2016. The average value of the two carbon risk indicators after 2016 is lower than that before 2016, indicating that the average carbon risk of firms has decreased since the launch of the green bond market. The average values of energy efficiency before and after 2016 are 0.648 and 0.946, respectively, indicating that the average level of energy efficiency among firms has increased since the launch of the green bond market. The mean values of the number of directors are greater than 2, suggesting that firms generally have multiple directors. The average growth rates of operating revenue are less than 0.3, reflecting the low development rates of firms. The average values of the nature of property rights are 0.328 and 0.357, respectively, suggesting that most of the firms are private firms.

Table 1. Descriptive statistics.

Before 2016
Variable	Mean	Std	Min	Max
$C R 1$	6.451	2.221	2.862	11.354
$C R 2$	0.662	1.086	0.007	3.514
$S I Z E$	20.841	1.175	18.874	24.845
$A G E$	2.241	0.312	1.231	3.175
$L E V$	0.628	0.161	0.035	1.671
$B O A R D$	2.026	0.176	1.624	2.227
$G R O W T H$	0.287	1.757	−0.971	8.684
$R O A$	0.041	0.065	−0.254	0.241
$S O E$	0.328	0.475	0.000	1.000
$E P U$	3.276	2.341	1.205	5.415
$T B A$	0.531	1.146	0.000	5.523
$E R$	10.541	1.141	6.622	12.209
$E E$	0.648	1.622	0.043	5.267
$E C S$	0.387	0.316	0.065	0.846
$D T$	0.241	0.182	0.000	1.000
After 2016
Variable	Mean	Std	Min	Max
$C R 1$	4.514	1.746	2.281	9.443
$C R 2$	0.461	0.943	0.005	2.273
$S I Z E$	22.237	1.321	20.625	26.276
$A G E$	2.574	0.321	2.362	3.622
$L E V$	0.481	0.153	0.020	1.347
$B O A R D$	2.267	0.202	1.715	2.614
$G R O W T H$	0.240	1.318	−0.628	7.628
$R O A$	0.039	0.094	−0.423	0.364
$S O E$	0.357	0.496	0.000	1.000
$E P U$	3.955	2.669	1.485	6.882
$T B A$	0.622	1.391	0.000	5.983
$E R$	10.627	1.037	7.442	12.791
$E E$	0.946	1.825	0.064	7.258
$E C S$	0.288	0.256	0.044	0.691
$G B I C$	0.041	0.013	0.018	0.072
$D T$	0.375	0.214	0.000	1.000

4. Empirical results

4.1. Baseline results

The coefficients for $I S S U A N C E$ in Columns (1)–(4) in Table 2 are significantly negative. The results in different situations prove that GBI has a negative correlation with the carbon risk of firms, thus verifying Hypothesis 1. The regression coefficients for $S I Z E$ are significantly negative. Firms on a larger scale have more resources available to invest in green technologies and equipment upgrades. The coefficients for $G R O W T H$ are significantly positive. Higher operating revenue may encourage firms to focus on the development of their main businesses rather than undertake a green transformation. The coefficients for $R O A$ are significantly negative. Firms with higher returns have more funds for carbon reduction. The coefficients for $T B A$ are not significant, reflecting that the total amount of bond issuance will not affect the carbon risk of firms. The coefficients for $E R$ indicates that strict environmental regulation may encourage firms to reduce carbon emissions to avoid penalties for failing to meet environmental standards.

Table 2. GBI and the carbon risk of firms.

Variable	$C R 1$		$C R 2$
Variable	(1)	(2)	(3)	(4)
$I S S U A N C E$	−0.325*** (−3.05)	−0.292*** (−3.25)	−0.221*** (−2.78)	−0.206*** (−3.11)
$S I Z E$		−0.263*** (−2.82)		−0.171*** (−2.70)
$A G E$		−0.152 (−0.86)		−0.103 (−0.77)
$L E V$		0.076 (0.95)		0.047 (1.12)
$B O A R D$		−0.049** (−2.34)		−0.029** (−2.18)
$G R O W T H$		0.368*** (2.77)		0.251*** (3.12)
$R O A$		−0.008*** (−3.54)		−0.005*** (−3.38)
$S O E$		−0.043 (−0.58)		−0.031 (−1.22)
$E P U$		0.063 (1.05)		0.046 (1.13)
$T B A$		0.007 (0.75)		0.004 (0.85)
$E R$		−0.342*** (−3.81)		−0.228*** (−3.62)
Firm FE	Yes	Yes	Yes	Yes
Year FE	Yes	Yes	Yes	Yes
Obs	7538	7538	7538	7538
Adj R2	0.265	0.288	0.321	0.343

This table presents the regression results of GBI and the carbon risk of firms. We use outcome regression, inverse probability weighting, or doubly-robust estimands of Callaway and Sant'Anna (2021) to mitigate potential estimation bias. ***, **, and * indicate statistical significance at the 1%, 5%, and 10% levels, respectively. The t-statistics based on standard errors cluster at the firm level. The other parts of the paper are the same.

4.2. Robustness tests

4.2.1. Parallel trend test

If the carbon risk of the treatment group and the control group present different trends before GBI, the reduction of firms' carbon risk may be caused by other factors, thus affecting the robustness of our results. Following Beck et al. (2010) and Dang et al. (2022), we examine the dynamic impact of GBI on carbon risk by replacing the dummy variable $I S S U A N C E$ in Eq. (1) with a series of dummy variables corresponding to pre-treatment lags (up to 6 years) and post-treatment leads (up to 3 years):(2) $C R_{i, t} = α_{0} + \sum_{k = 1}^{k = 6} α_{- k} I S S U A N C E_{i, t}^{- k} + \sum_{k = 1}^{k = 3} α_{+ k} I S S U A N C E_{i, t}^{+ k} + β X_{i, t} + θ_{i} + η_{t} + ε_{i, t}$ where $I S S U A N C E_{i, t}^{- k} (k = 1,2,3,4,5,6)$ equals 1 for firms in the $k$ th year before GBI and $I S S U A N C E_{i, t}^{+ k} (k = 1,2,3)$ equals 1 for firms in the $k$ th year after GBI. The year each firm issues green bonds is specified as the baseline year and is excluded from the regression. Fig. 3 plots the estimated coefficients and the corresponding 95% confidence intervals. In Fig. 3, we can see that the coefficients on $I S S U A N C E^{- k} (k = 1,2,3,4,5,6)$ are insignificantly and the coefficients on $I S S U A N C E^{k} (k = 1,2,3)$ are significantly negative, suggesting that changes in carbon risk do not precede GBI and that GBI has a negative effect on carbon risk. The results above verify the parallel trend assumption and the robustness of our findings.

4.2.2. Endogeneity tests

Green bonds reduce carbon risk, which in turn encourages firms to continue issuing green bonds, omitting some important variables that may have an impact on both GBI and carbon risk and leading to endogeneity problems. We use the instrumental variable method to address the above concerns. The greater support from local governments for GBI can help reduce the cost of GBI and improve its efficiency, thereby enhancing the motivation of firms to issue green bonds. However, the support provided by local governments for GBI is unlikely to have a direct impact on the carbon risk of firms. Therefore, we choose the support level of local governments for GBI ( $S U P P O R T$ ) as the instrumental variable, and the support level of local governments for GBI is expressed by the number of policy regulations on green bonds issued by local governments. The coefficient on $S U P P O R T$ in Table 3 indicates that local governments' support for green bonds is positively correlated with the issuance of green bonds by firms. The results of the Cragg-Donald Wald F statistic, the Kleibergen-Paap rk LM statistic, and the Hansen J statistic verify the validity of the instrumental variable. The coefficients on $I S S U A N C E$ in the second stage demonstrate that the baseline results are still right after considering endogeneity problems.

Table 3. Endogeneity tests.

Variable	First Stage	Second Stage
	$T R E A T \times P O S T$	$C R 1$	$C R 2$
	(1)	(2)	(3)
$I S S U A N C E$		−0.284*** (−3.35)	−0.183*** (−3.48)
$S U P P O R T$	0.053*** (2.79)
$S I Z E$	0.273** (2.10)	−0.244*** (−2.72)	−0.163*** (−2.91)
$A G E$	0.075 (0.79)	−0.127 (−1.25)	−0.082 (−0.71)
$L E V$	0.134 (1.46)	0.063 (1.44)	0.041 (1.28)
$B O A R D$	0.061** (2.46)	−0.143 (−0.76)	−0.083 (−1.20)
$G R O W T H$	0.023 (1.15)	0.324*** (2.73)	0.255*** (3.28)
$R O A$	0.377 (1.28)	−0.006** (−2.46)	−0.004*** (−2.81)
$S O E$	0.552 (1.12)	−0.035 (−1.15)	−0.026 (−1.31)
$E P U$	0.008 (1.41)	0.052 (0.82)	0.038 (0.61)
$T B A$	0.037 (1.18)	0.005 (1.29)	0.004 (0.83)
$E R$	0.025** (2.32)	−0.328*** (−2.77)	−0.214*** (−3.16)
Cragg-Donald Wald F statistic	57.54
Kleibergen-Paap rk LM	97.18
P-value	0.000
Hansen J	0.374
Firm FE	Yes	Yes	Yes
Year FE	Yes	Yes	Yes
Obs	7538	7538	7538
Adj R2	0.453	0.282	0.339

The P-value corresponds to the Kleibergen-Paap rk LM statistic.

4.2.3. Alternative staggered difference-in-differences specification

Green bonds have an implementation period, and after the expiration of green bonds, firms need to repay the principal and interest and cannot continue to obtain financing from the bonds. We evaluate the impact of GBI on firms' carbon risk by only considering the changes in carbon risk during the implementation of green bonds. The estimated model is as follows:(3) $C R_{i, t} = α_{0} + α_{1} G R E E N_{i, t} + β X_{i, t} + θ_{i} + η_{t} + ε_{i, t}$ where $G R E E N_{i, t}$ is a dummy variable that equals 1 during the implementation years of green bonds issued by firm $i$ and 0 otherwise. The above specification allow us to more accurately assess the changes in carbon risk during the implementation of green bonds. The coefficients for $G R E E N$ in Table 4 suggest that GBI reduces the carbon risk of firms using the alternative staggered difference-in-differences specification, which are in line with the baseline findings.

Table 4. Alternative staggered difference-in-differences specification.

Variable	$C R 1$		$C R 2$
Variable	(1)	(2)	(3)	(4)
$G R E E N$	−0.305*** (−2.72)	−0.284*** (−3.29)	−0.207*** (−3.40)	−0.182*** (−2.88)
$S I Z E$		−0.234*** (−2.75)		−0.144** (−2.25)
$A G E$		−0.121 (−1.14)		−0.094 (−0.87)
$L E V$		0.062* (1.65)		0.034 (1.12)
$B O A R D$		−0.051** (−2.55)		−0.037 (−0.44)
$G R O W T H$		0.415*** (3.35)		0.287*** (4.52)
$R O A$		−0.008** (−2.65)		−0.005*** (−2.78)
$S O E$		−0.049 (−1.22)		−0.034 (−1.15)
$E P U$		0.075 (1.48)		0.052 (1.13)
$T B A$		0.006 (1.02)		0.004 (0.73)
$E R$		−0.361*** (−2.76)		−0.249*** (−3.06)
Firm FE	Yes	Yes	Yes	Yes
Year FE	Yes	Yes	Yes	Yes
Obs	7538	7538	7538	7538
Adj R2	0.258	0.266	0.313	0.337

4.2.4. Alternative measure of carbon risk

We use carbon emissions as a proxy variable for firms' carbon risk in the baseline regression, which may not accurately measure the carbon risk of firms, thereby affecting the accuracy of our conclusion. Following Nguyen and Phan (2020), we classify firms into heavy emitters and light emitters based on the carbon emission characteristics of their respective industries. Firms in industries with high carbon risks are heavy emitters, and firms in industries with low carbon risks are light emitters. Industries with higher carbon risks include those that emit more greenhouse gases and consume more energy. Based on the actual situation of carbon emissions and energy consumption in China, the Guidelines for the Industry Classification of Listed Companies, and the Environmental Information Disclosure Guidelines for Listed Firms, we classify 16 industries, including thermal power, chemical, steel, cement, coal, metallurgy, petrochemicals, mining, etc., as industries with high carbon risk. These industries belong to heavily polluting industries with high emissions and energy consumption. We set a dummy variable for carbon risk ( $C R_D U M M Y$ ) to be 1 when firms belong to the sixteen industries mentioned above and 0 otherwise. We replace the dependent variable in Eq. (1) with $C R_D U M M Y$ and perform regression again. The coefficients for $I S S U A N C E$ in Table 5 verify our finding that GBI reduces the carbon risk of firms.

Table 5. Alternative measure of carbon risk.

Variable	$C R_D U M M Y$
Variable	(1)	(2)
$I S S U A N C E$	−0.055*** (−2.92)	−0.052*** (−3.13)
$S I Z E$		−0.354*** (−2.81)
$A G E$		−0.081 (−1.30)
$L E V$		0.022 (1.42)
$B O A R D$		−0.142*** (−2.72)
$G R O W T H$		0.251 (1.08)
$R O A$		−0.033** (−2.40)
$S O E$		−0.007 (−1.34)
$E P U$		0.235 (1.22)
$T B A$		0.034 (0.28)
$E R$		−0.064** (−2.58)
Firm FE	Yes	Yes
Year FE	Yes	Yes
Obs	7538	7538
Adj R2	0.341	0.3622

4.2.5. Excluding the sample of the financial crisis

The financial crisis will significantly increase firm risk and financing difficulty. The primary consideration for firms is to survive in times of economic turmoil, while green technology upgrades will not be a priority for firms, which may lead to a high level of carbon risk for firms before GBI and affect the accuracy of our results. For this reason, we do not include data from that period in our re-estimation of Eq. (1) and instead use a fresh data set that does not include the 2008–2009 recession. The coefficients for $I S S U A N C E$ in Table 6 are the same as the baseline results, which verify the negative correlation between GBI and the carbon risk of firms.

Table 6. Excluding the sample of the financial crisis.

Variable	$C R 1$		$C R 2$
Variable	(1)	(2)	(3)	(4)
$I S S U A N C E$	−0.271*** (−2.78)	−0.285*** (−3.11)	−0.217*** (−3.28)	−0.194** (−2.43)
$S I Z E$		−0.304*** (−3.05)		−0.214*** (−2.78)
$A G E$		−0.134 (−1.25)		−0.078 (−0.68)
$L E V$		0.105 (1.15)		0.071 (1.23)
$B O A R D$		−0.068*** (−2.78)		−0.042** (−2.66)
$G R O W T H$		0.426*** (4.10)		0.374*** (3.67)
$R O A$		−0.012** (−2.22)		−0.008** (−2.41)
$S O E$		−0.037 (−1.18)		−0.025 (−0.75)
$E P U$		0.242 (1.21)		0.162 (1.05)
$T B A$		0.005 (0.75)		0.004 (0.85)
$E R$		−0.341*** (−3.81)		−0.227*** (−3.62)
Firm FE	Yes	Yes	Yes	Yes
Year FE	Yes	Yes	Yes	Yes
Obs	7164	7164	7164	7164
Adj R2	0.278	0.290	0.335	0.346

4.2.6. PSM-DID

Because the initial sample's treatment and control groups may not have been chosen at random, this may affect the effectiveness of our findings. We match firms that haven't issued green bonds but have issued conventional bonds with the treatment firms using one-to-two nearest-neighbor propensity score matching (PSM). We compare the matching results in Table 7. After matching, these differences between these two groups are not significant, indicating that the two groups are comparable after the PSM processing. We estimate Eq. (1) again using the new sample screened by the PSM approach. In Table 8, the regression results demonstrate that GBI can greatly lower firms' carbon risk (see Table 8).

Table 7. Balance tests.

Panel A: pre-matching					Panel B: post-matching
Variable	Mean		t-test		Variable	Mean		t-test
Variable	Treated	Control	t-value	p-value	Variable	Treated	Control	t-value	p-value
$S I Z E$	23.547	22.125	21.38	0.000	$S I Z E$	23.524	23.614	0.54	0.798
$A G E$	2.715	2.545	5.14	0.000	$A G E$	2.774	2.751	1.22	0.134
$L E V$	0.574	0.741	7.85	0.000	$L E V$	0.572	0.581	1.34	0.180
$B O A R D$	2.455	2.130	4.88	0.000	$B O A R D$	2.452	2.460	0.82	0.514
$G R O W T H$	0.244	0.283	−0.77	0.425	$G R O W T H$	0.2351	0.242	1.05	0.746
$R O A$	0.035	0.042	3.58	0.000	$R O A$	0.0358	0.036	0.42	0.274
$S O E$	0.874	0.341	14.28	0.000	$S O E$	0.872	0.876	1.28	0.175
$E P U$	3.728	3.687	0.42	0.756	$E P U$	3.764	3.745	1.47	0.144
$T B A$	1.257	0.428	18.42	0.000	$T B A$	1.254	1.2715	0.77	0.645
$E R$	11.057	9.147	8.74	0.000	$E R$	11.047	11.072	1.52	0.415

Table 8. PSM-DID.

Variable	$C R 1$	$C R 2$
Variable	(1)	(2)
$I S S U A N C E$	−0.315*** (−3.27)	−0.216*** (−2.84)
$S I Z E$	−0.284*** (−2.72)	−0.195*** (−3.25)
$A G E$	−0.171 (−1.02)	−0.134 (−0.48)
$L E V$	0.084 (0.95)	0.054 (1.05)
$B O A R D$	−0.054*** (−2.82)	−0.030* (−1.87)
$G R O W T H$	0.383*** (3.25)	0.284*** (2.87)
$R O A$	−0.010*** (−3.17)	−0.007** (−2.38)
$S O E$	−0.054 (−0.81)	−0.036 (−1.42)
$E P U$	0.084 (1.23)	0.052 (1.41)
$T B A$	0.008 (0.45)	0.005 (0.62)
$E R$	−0.375** (−2.50)	−0.251*** (−3.12)
Firm FE	Yes	Yes
Year FE	Yes	Yes
Obs	2487	2487
Adj R2	0.287	0.343

4.2.7. Placebo test

The reduction of firms' carbon risk may be influenced by other random factors rather than driven by GBI. We use a placebo test to address the concern above. Following Bae et al. (2021), we randomly select the same number of firms that have issued green bonds as in the original sample and randomly assign the years of GBI to these firms from 2009 to 2019. We set the dummy variable of $P O S T_P S E U D O$ that equals 1 for the year and after GBI and 0 otherwise based on the new sample. We calculate the average carbon risk of firms until the year before GBI ( $C R_B E F O R E_G B I$ ) and the average carbon risk of firms in the years after GBI ( $C R_A F T E R_G B I$ ). By subtracting the difference between $C R_A F T E R_G B I$ and $C R_B E F O R E_G B I$ from the carbon risk of firms in different years after GBI, we obtain the pseudo-carbon risk series, which excludes the effect of GBI. Based on the pseudo-carbon risk series, we re-estimate Eq. (1) by replacing $P O S T$ with $P O S T_P S E U D O$ and repeat the above process 1000 times, and the distribution of the coefficients and t-statistics is shown in Table 9. The results show that the actual coefficients for both the dependent variables of $C R 1$ and $C R 2$ are less than the coefficients of $P O S T_P S E U D O$ under 5% level, and the actual t-statistic values for both the dependent variables of $C R 1$ and $C R 2$ are less than the t-statistic values of $P O S T_P S E U D O$ under 1% level. The above results indicate that the reduction of carbon risk in firms is unlikely to be influenced by other random factors, thus further verifying the robustness of the conclusion.

Table 9. Placebo test.

pseudo- $C R 1$	Actual	Mean	1%	5%	10%	90%	95%	99%
Coefficients of $P O S T_P S E U D O$	−0.292	−0.001	−0.315	−0.271	−0.215	0.201	0.267	0.336
t-statistic of $P O S T_P S E U D O$	−3.25	−0.04	−3.21	−2.64	−1.97	2.12	2.84	3.45
pseudo- $C R 2$	Actual	Mean	1%	5%	10%	90%	95%	99%
Coefficients of $P O S T_P S E U D O$	−0.206	−0.001	−0.231	−0.195	−0.125	0.114	0.184	0.241
t-statistic of $P O S T_P S E U D O$	−3.11	−0.03	−3.07	−2.53	−1.85	2.06	2.74	3.28

5. Mechanism tests

5.1. Energy efficiency

Firms can use the funds they get from green bonds to reduce financing constraints, expand their investment in green R&D, and improve green innovation (Wang et al., 2022). Technological innovation can improve the production efficiency of firms, while green and low-carbon technological innovation can help improve the resource utilization efficiency of firms in clean production, thus improving their energy efficiency (Sun et al., 2019; Huang et al., 2019). The improvement of energy efficiency is one of the key factors in reducing firms' carbon emissions, improving their competitiveness (Trianni et al., 2016; Zhang et al., 2018; Li et al., 2022), and further reducing their financial vulnerability in low-carbon transformation and carbon risk. We estimate the model shown below to test this mechanism:(4) $E E_{i, t} = λ_{0} + λ_{1} I S S U A N C {E_{i,}}_{t} + β X_{i, t} + θ_{i} + η_{t} + ε_{i, t}$ where $E E_{i, t}$ indicates the energy efficiency of firm $i$ in year $t$ , which is represented by energy production efficiency (Su et al., 2022). Following Su et al. (2022), energy production efficiency is calculated by the ratio of the energy consumed by firms to their operating income. The regression results of Eq. (4) are presented in Table 10. The coefficients for $I S S U A N C E$ indicate that GBI improves energy efficiency. The results above prove that GBI reduces the carbon risk of firms by improving energy efficiency, thus verifying Hypothesis 2.

Table 10. The impact of GBI on energy efficiency.

Variable	$E E$
Variable	(1)	(2)
$I S S U A N C E$	0.045*** (3.23)	0.047*** (3.48)
$S I Z E$		0.087** (2.27)
$A G E$		0.236 (1.02)
$L E V$		−0.033 (−0.55)
$B O A R D$		0.048** (2.25)
$G R O W T H$		0.072** (2.16)
$R O A$		−0.012 (−0.78)
$S O E$		0.042** (2.25)
$E P U$		−0.324 (−0.37)
$T B A$		0.003 (1.58)
$E R$		0.061*** (2.83)
Firm FE	Yes	Yes
Year FE	Yes	Yes
Obs	7538	7538
Adj R2	0.415	0.431

5.2. Energy consumption structure

Theoretical analysis demonstrates that GBI reduces the consumption of coal and oil, increases their consumption of natural gas and new energy, and improves their energy consumption structure. Improving the energy consumption structure is an important measure for China to promote low-carbon economic transformation (Xu et al., 2020). The Chinese government has always insisted on reducing carbon emissions by replacing coal and oil with cleaner energy sources such as natural gas and new energy, thereby promoting low-carbon economic transformation (Xu et al., 2020). The improvement of firms' energy consumption structures can reduce their carbon emissions and help win the support of more stakeholders (Yu et al., 2018), thereby obtaining more resources to promote their sustainable development and improving their ability to cope with risks (Choi and Wang, 2009; Blake and Moschieri, 2017), thereby reducing their financial vulnerability in low-carbon transformation and carbon risk. To support Hypothesis 3, we develop the model shown below:(5) $E C S_{i, t} = λ_{0} + λ_{1} I S S U A N C {E_{i,}}_{t} + β X_{i, t} + θ_{i} + η_{t} + ε_{i, t}$ where $E C S_{i, t}$ indicates the energy consumption structure of firm $i$ in year $t$ . The structure of energy consumption is measured by the ratio of coal and oil consumption to the total consumption of firms. We convert different types of energy consumption into standard coal in a unified unit. The regression results of Eq. (5) are presented in Table 11. The coefficients for $I S S U A N C E$ suggest that GBI improves energy consumption structure. The aforementioned findings validate Hypothesis 3 by demonstrating that GBI promotes firms to transform from coal and oil consumption as the basis for cleaner energy, thereby improving their energy consumption structure and reducing their carbon risk.

Table 11. The impact of GBI on energy consumption structure.

Variable	$E C S$
Variable	(1)	(2)
$I S S U A N C E$	−0.157*** (−2.81)	−0.149*** (−2.73)
$S I Z E$		−0.344 (−1.05)
$A G E$		0.061 (1.25)
$L E V$		0.245* (1.95)
$B O A R D$		−0.022 (−1.05)
$G R O W T H$		0.368*** (2.77)
$R O A$		−0.015** (−2.27)
$S O E$		−0.136* (−1.82)
$E P U$		0.007 (1.16)
$T B A$		0.033 (1.25)
$E R$		−0.052*** (−3.28)
Firm FE	Yes	Yes
Year FE	Yes	Yes
Obs	4256	4256
Adj R2	0.361	0.382

6. Heterogeneity analysis

6.1. Green bond issuance costs

Compared with conventional bonds, a notable feature of green bonds is their lower issuance costs and longer issuance period (Wang et al., 2020). The lower issuance costs and longer issuance period also encourage firms to be willing to use the funds raised through green bonds for green projects (Tang and Zhang, 2020). Due to factors such as issuers and bond ratings, there are significant differences in the issuance costs of different green bonds. The higher issuance costs of green bonds will increase the financial pressure on firms and weaken their motivation to use the funds raised from green bonds for green projects, as green projects typically require larger investments and have a longer payback period, which may encourage them to use funds for non-green projects with high carbon emissions through green packaging and lead to the emergence of greenwashing behavior (Flammer, 2021). The greenwashing behavior of firms will significantly hinder their green transformation, which is not conducive to the reduction of their carbon risk and may even increase their carbon risk (Delmas and Burbano, 2011). Therefore, we believe that higher GBI costs will weaken the negative effect of GBI on firms' carbon risk. We estimate the model below to verify the above conjecture:(6) $C R_{i, t} = α + β_{1} I S S U A N C E_{i, t} + β_{2} I S S U A N C E_{i, t} \times G B I C_{i, t} + β_{3} G B I C_{i, t} + δ X_{i, t} + θ_{i} + η_{t} + ε_{i, t}$ where $G B I C_{i, t}$ is the GBI cost of firm $i$ in year $t$ , which is represented by the coupon rate of green bonds. The higher the coupon rate of green bonds, the higher the interest required to be paid at maturity, and the corresponding issuing costs of green bonds are also higher. When it is higher than the sample median, the indicator variable for the GBI cost equals 1, otherwise it equals 0. In Table 12, the coefficients for $I S S U A N C E \times G B I C$ are significant positive, demonstrating that the negative effect of GBI on firms' carbon risk is weakened by higher GBI costs.

Table 12. Heterogeneity effect of GBI costs.

Variable	$C R 1$		$C R 2$
Variable	(1)	(2)	(3)	(4)
$I S S U A N C E$	−0.253*** (−2.84)	−0.292*** (−2.31)	−0.167*** (−2.76)	−0.142** (−2.33)
$I S S U A N C E \times G B I C$	0.048*** (4.25)	0.046*** (3.68)	0.035** (2.46)	0.034*** (2.71)
$G B I C$	0.062 (1.02)	0.060** (2.37)	0.053* (1.76)	0.482** (2.32)
$S I Z E$		−0.216*** (−2.77)		−0.123*** (−2.83)
$A G E$		−0.094 (−0.73)		−0.062 (−0.91)
$L E V$		0.042 (1.24)		0.013 (1.36)
$B O A R D$		−0.023*** (−3.14)		−0.012** (−2.37)
$G R O W T H$		0.263*** (2.83)		0.152*** (3.40)
$R O A$		−0.027* (−1.72)		−0.012** (−2.18)
$S O E$		−0.026 (−1.51)		−0.012 (−0.37)
$E P U$		0.235 (0.84)		0.108* (1.67)
$T B A$		0.004 (0.83)		0.002 (0.43)
$E R$		−0.263** (−2.67)		−0.184*** (−3.07)
Firm FE	Yes	Yes	Yes	Yes
Year FE	Yes	Yes	Yes	Yes
Obs	7538	7538	7538	7538
Adj R2	0.273	0.288	0.331	0.349

6.2. Digital transformation

Digital transformation refers to the firm's use of big data and block chain technologies, such as artificial intelligence, to realize changes in its business model, organizational structure, and business processes (Zhou et al., 2021). Digital transformation can apply digital technology to the production and operation processes of firms to conduct real-time and dynamic monitoring of energy utilization and emissions, so that the energy supply and consumption of firms in different production stages can be more accurate and predictable, thus reducing energy waste and improving energy efficiency (Maroufkhani et al., 2022). Digital technology can improve the efficiency of information transmission between various links in the production process of firms, achieve information sharing, improve resource allocation efficiency and green innovation efficiency, and reduce energy consumption and carbon emissions (Sheng et al., 2022). Digital transformation helps improve the competitiveness of firms and their ability to cope with risks so that they can carry out low-carbon transformation at a lower cost when the external environment changes, thus reducing the impact of low-carbon transformation on their financial vulnerability (Sheng et al., 2022). In addition, digital technology can also help reduce information asymmetry between firms, financial institutions, and investors, making it easier for firms to obtain loans to alleviate their financing constraints. The alleviation of financing constraints promotes firms' investment in green and low-carbon technology R&D, thereby improving their energy efficiency and reducing carbon risk. Therefore, we believe that a higher level of digital transformation will strengthen the negative effect of GBI on firms' carbon risk. To investigate how the digitization transformation of firms affects the relationship between GBI and carbon risk, we develop the following model:(7) $C R_{i, t} = α + β_{1} I S S U A N C {E_{i,}}_{t} + β_{2} I S S U A N C {E_{i,}}_{t} \times D T_{i, t} + β_{3} D T_{i, t} + δ X_{i, t} + θ_{i} + η_{t} + ε_{i, t}$

We adopt the method used in Zhou et al. (2021, 2022) to measure digital transformation. We define the indicator variable of digital transformation ( $D T$ ) as 1 when a firm adopts digital technologies and 0 otherwise. We can see that the coefficients for $I S S U A N C E \times D T$ in Table 13 indicate that a higher level of digital transformation strengthens the negative effect of GBI on the carbon risk of firms.

Table 13. Heterogeneity effect of digital transformation.

Variable	$C R 1$		$C R 2$
Variable	(1)	(2)	(3)	(4)
$I S S U A N C E$	−0.452** (−2.13)	−0.461** (−2.34)	−0.341*** (−2.46)	−0.339*** (−3.12)
$I S S U A N C E \times D T$	−0.316*** (3.02)	−0.304*** (−2.73)	−0.218*** (−2.84)	−0.203** (−2.52)
$D T$	0.175 (1.35)	0.157 (1.34)	−0.084** (−2.54)	−0.076** (−2.35)
$S I Z E$		−0.362*** (−2.84)		−0.231** (−2.24)
$A G E$		−0.143 (−1.24)		−0.084 (−0.71)
$L E V$		0.073 (0.66)		0.031 (1.36)
$B O A R D$		−0.064** (−2.13)		−0.026*** (−2.88)
$G R O W T H$		0.377** (2.41)		0.206** (2.59)
$R O A$		−0.045** (−2.17)		−0.028*** (−2.75)
$S O E$		−0.041 (−1.02)		−0.024 (−0.75)
$E P U$		0.328* (1.72)		0.154 (1.24)
$T B A$		0.006 (1.33)		0.003 (0.57)
$E R$		−0.334*** (−3.75)		−0.215*** (−2.93)
Firm FE	Yes	Yes	Yes	Yes
Year FE	Yes	Yes	Yes	Yes
Obs	7538	7538	7538	7538
Adj R2	0.283	0.291	0.341	0.350

7. Conclusion and policy implications

The smooth transition to a green and low-carbon economy depends greatly on reducing firms' carbon risk. However, the reduction of the carbon risk of firms is facing serious challenges due to the low return and long investment period characteristics of green projects. Green bonds are the most important green direct financing tool. To ease financing limitations and promote green development, it is crucial for Chinese firms to use green direct financing tools effectively. Therefore, promoting the economy's green development depends critically on whether or not green bonds can effectively reduce the carbon risk of firms.

Based on the Guidelines on GBI, we study the relationship between GBI and the carbon risk of firms. We further analyze the internal mechanisms and heterogeneity effects of GBI affecting the carbon risk of firms. We find a negative relationship between GBI and the carbon risk of firms. We also find that GBI can improve the energy efficiency and energy consumption structure of firms. Furthermore, lower GBI costs and a higher level of digital transformation strengthen the negative effect of GBI on the carbon risk of firms.

We can get some meaningful implications from the empirical results of this paper. GBI can reduce the carbon risk of firms. Therefore, regulators should vigorously develop the green bond market and encourage and guide firms to issue green bonds. Green bonds are an important green direct financing tool that not only improves China's financing system but also plays an important role in promoting low-carbon transformation. Green bonds reduce the carbon risk of firms by improving their energy efficiency and consumption structure. Therefore, firms should actively issue green bonds and participate in green projects to improve energy efficiency and energy consumption structure, thereby reducing their carbon risk. The improvement of energy efficiency and energy consumption structure in firms can not only reduce carbon risk but also help establish good relationships with stakeholders and enhance their sustainable development capabilities. The negative effect of GBI on the carbon risk of firms is stronger for firms with lower GBI costs and a higher level of digital transformation. Therefore, regulators should improve the green bond standard system and regulatory system and increase subsidies for firms issuing green bonds to reduce the issuance cost of green bonds. Firms should actively carry out digital transformation to better play the role of green bonds in reducing carbon risk.

Author contributions

Hu Wang: Conceptualization; Data curation; Formal analysis; Validation; Funding acquisition; Methodology; Software; Writing; Hong Shen: Formal analysis; Methodology; Software; Shouwei Li: Supervision

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

Data will be made available on request.

References

Bae et al., 2021
K.H. Bae, S. El Ghoul, O. Guedhami, X. Zheng
Board reforms and dividend policy: international evidence
J. Financ. Quant. Anal., 56 (4) (2021), pp. 1296-1320
CrossRef View in Scopus Google Scholar
Baker et al., 2016
S.R. Baker, N. Bloom, S.J. Davis
Measuring economic policy uncertainty
Q. J. Econ., 131 (4) (2016), pp. 1593-1636
CrossRef View in Scopus Google Scholar
Balachandran and Nguyen, 2018
B. Balachandran, J.H. Nguyen
Does carbon risk matter in firm dividend policy? Evidence from a quasi-natural experiment in an imputation environment
J. Bank. Finance, 96 (2018), pp. 249-267
View PDF View article View in Scopus Google Scholar
Baulkaran, 2019
V. Baulkaran
Stock market reaction to green bond issuance
J. Asset Manag., 20 (5) (2019), pp. 331-340
CrossRef View in Scopus Google Scholar
Beck et al., 2010
T. Beck, R. Levine, A. Levkov
Big bad banks? The winners and losers from bank deregulation in the United States
J. Finance, 65 (5) (2010), pp. 1637-1667
CrossRef View in Scopus Google Scholar
Bhutta et al., 2022
U.S. Bhutta, A. Tariq, M. Farrukh, A. Raza, M.K. Iqbal
Green bonds for sustainable development: review of literature on development and impact of green bonds
Technol. Forecast. Soc. Change, 175 (2022), Article 121378
View PDF View article View in Scopus Google Scholar
Blake and Moschieri, 2017
D.J. Blake, C. Moschieri
Policy risk, strategic decisions and contagion effects: firm‐specific considerations
Strat. Manag. J., 38 (3) (2017), pp. 732-750
CrossRef View in Scopus Google Scholar
Bolton and Kacperczyk, 2021
P. Bolton, M. Kacperczyk
Do investors care about carbon risk?
J. Financ. Econ., 142 (2) (2021), pp. 517-549
View PDF View article View in Scopus Google Scholar
Bose et al., 2021
S. Bose, K. Minnick, S. Shams
Does carbon risk matter for corporate acquisition decisions?
J. Corp. Finance, 70 (2021), Article 102058
View PDF View article View in Scopus Google Scholar
Cai and Zhang, 2011
J. Cai, Z. Zhang
Leverage change, debt overhang, and stock prices
J. Corp. Finance, 17 (3) (2011), pp. 391-402
View PDF View article View in Scopus Google Scholar
Callaway and Sant'Anna, 2021
B. Callaway, P.H. Sant'Anna
Difference-in-differences with multiple time periods
J. Econom., 225 (2) (2021), pp. 200-230
View PDF View article Google Scholar
Chapple et al., 2013
L. Chapple, P.M. Clarkson, D.L. Gold
The cost of carbon: capital market effects of the proposed emission trading scheme (ETS)
Abacus, 49 (1) (2013), pp. 1-33
CrossRef View in Scopus Google Scholar
Chen et al., 2020
D. Chen, S. Chen, H. Jin, Y. Lu
The impact of energy regulation on energy intensity and energy structure: firm-level evidence from China
China Econ. Rev., 59 (2020), Article 101351
View PDF View article View in Scopus Google Scholar
Choi and Wang, 2009
J. Choi, H. Wang
Stakeholder relations and the persistence of corporate financial performance
Strat. Manag. J., 30 (8) (2009), pp. 895-907
CrossRef View in Scopus Google Scholar
Clarkson et al., 2011
P.M. Clarkson, Y. Li, G.D. Richardson, F.P. Vasvari
Does it really pay to be green? Determinants and consequences of proactive environmental strategies
J. Account. Publ. Pol., 30 (2) (2011), pp. 122-144
View PDF View article View in Scopus Google Scholar
Custódio et al., 2013
C. Custódio, M.A. Ferreira, L. Laureano
Why are US firms using more short-term debt?
J. Financ. Econ., 108 (1) (2013), pp. 182-212
View PDF View article View in Scopus Google Scholar
Dang et al., 2022
V.A. Dang, E. Lee, Y. Liu, C. Zeng
Bank deregulation and stock price crash risk
J. Corp. Finance, 72 (2022), Article 102148
View PDF View article View in Scopus Google Scholar
Del Gaudio et al., 2022
B.L. Del Gaudio, D. Previtali, G. Sampagnaro, V. Verdoliva, S. Vigne
Syndicated green lending and lead bank performance
J. Int. Financ. Manag. Account., 33 (3) (2022), pp. 412-427
CrossRef View in Scopus Google Scholar
Dell'Atti et al., 2022
S. Dell'Atti, C. Di Tommaso, V. Pacelli
Sovereign green bond and country value and risk: evidence from European Union countries
J. Int. Financ. Manag. Account., 33 (3) (2022), pp. 505-521
CrossRef View in Scopus Google Scholar
Delmas and Burbano, 2011
M.A. Delmas, V.C. Burbano
The drivers of greenwashing
Calif. Manag. Rev., 54 (1) (2011), pp. 64-87
CrossRef View in Scopus Google Scholar
Dong et al., 2021
Y. Dong, Q. Hou, C. Ni
Implicit government guarantees and credit ratings
J. Corp. Finance, 69 (2021), Article 102046
View PDF View article View in Scopus Google Scholar
El Ghoul et al., 2018
S. El Ghoul, O. Guedhami, H. Kim, K. Park
Corporate environmental responsibility and the cost of capital: international evidence
J. Bus. Ethics, 149 (2018), pp. 335-361
CrossRef View in Scopus Google Scholar
Flammer, 2021
C. Flammer
Corporate green bonds
J. Financ. Econ., 142 (2) (2021), pp. 499-516
View PDF View article View in Scopus Google Scholar
Huang et al., 2019
Z. Huang, G. Liao, Z. Li
Loaning scale and government subsidy for promoting green innovation
Technol. Forecast. Soc. Change, 144 (2019), pp. 148-156
View PDF View article View in Scopus Google Scholar
Jung et al., 2018
J. Jung, K. Herbohn, P. Clarkson
Carbon risk, carbon risk awareness and the cost of debt financing
J. Bus. Ethics, 150 (2018), pp. 1151-1171
CrossRef View in Scopus Google Scholar
Kabir et al., 2021
M.N. Kabir, S. Rahman, M.A. Rahman, M. Anwar
Carbon emissions and default risk: international evidence from firm-level data
Econ. Modell., 103 (2021), Article 105617
View PDF View article View in Scopus Google Scholar
Kim et al., 2015
Y.B. Kim, H.T. An, J.D. Kim
The effect of carbon risk on the cost of equity capital
J. Clean. Prod., 93 (2015), pp. 279-287
View PDF View article CrossRef View in Scopus Google Scholar
Li et al., 2022
R. Li, L. Li, Q. Wang
The impact of energy efficiency on carbon emissions: evidence from the transportation sector in Chinese 30 provinces
Sustain. Cities Soc., 82 (2022), Article 103880
View PDF View article View in Scopus Google Scholar
Liu et al., 2022
F. Liu, J. Yu, Y. Shen, L. He
Does the resource-dependent motivation to disclose environmental information impact company financing? Evidence from renewable energy companies of China
Renew. Energy, 181 (2022), pp. 156-166
View PDF View article View in Scopus Google Scholar
Maroufkhani et al., 2022
P. Maroufkhani, K.C. Desouza, R.K. Perrons, M. Iranmanesh
Digital transformation in the resource and energy sectors: a systematic review
Resour. Pol., 76 (2022), Article 102622
View PDF View article View in Scopus Google Scholar
Nguyen and Phan, 2020
J.H. Nguyen, H.V. Phan
Carbon risk and corporate capital structure
J. Corp. Finance, 64 (2020), Article 101713
View PDF View article View in Scopus Google Scholar
Oestreich and Tsiakas, 2015
A.M. Oestreich, I. Tsiakas
Carbon emissions and stock returns: evidence from the EU emissions trading scheme
J. Bank. Finance, 58 (2015), pp. 294-308
View PDF View article View in Scopus Google Scholar
Reboredo et al., 2022
J.C. Reboredo, A. Ugolini, J. Ojea-Ferreiro
Do green bonds de-risk investment in low-carbon stocks?
Econ. Modell., 108 (2022), Article 105765
View PDF View article View in Scopus Google Scholar
Sartzetakis, 2021
E.S. Sartzetakis
Green bonds as an instrument to finance low carbon transition
Econ. Change Restruct., 54 (3) (2021), pp. 755-779
CrossRef Google Scholar
Sheng et al., 2022
H. Sheng, T. Feng, L. Liu
The influence of digital transformation on low-carbon operations management practices and performance: does CEO ambivalence matter?
Int. J. Prod. Res. (2022), pp. 1-15
Google Scholar
Shu and Tan, 2023
H. Shu, W. Tan
Does carbon control policy risk affect corporate ESG performance?
Econ. Modell., 120 (2023), Article 106148
View PDF View article View in Scopus Google Scholar
Su et al., 2022
Z. Su, Q. Guo, H.T. Lee
Green finance policy and enterprise energy consumption intensity: evidence from a quasi-natural experiment in China
Energy Econ., 115 (2022), Article 106374
View PDF View article View in Scopus Google Scholar
Sun et al., 2019
H. Sun, B.K. Edziah, C. Sun, A.K. Kporsu
Institutional quality, green innovation and energy efficiency
Energy Pol., 135 (2019), Article 111002
View PDF View article View in Scopus Google Scholar
Tang and Zhang, 2020
D.Y. Tang, Y. Zhang
Do shareholders benefit from green bonds?
J. Corp. Finance, 61 (2020), Article 101427
View PDF View article View in Scopus Google Scholar
Trianni et al., 2016
A. Trianni, E. Cagno, S. Farné
Barriers, drivers and decision-making process for industrial energy efficiency: a broad study among manufacturing small and medium-sized enterprises
Appl. Energy, 162 (2016), pp. 1537-1551
View PDF View article View in Scopus Google Scholar
Wang and Jiang, 2023
H. Wang, S. Jiang
Green bond issuance and stock price informativeness
Econ. Anal. Pol., 79 (2023), pp. 120-133
View PDF View article CrossRef Google Scholar
Wang et al., 2023
H. Wang, S. Li, Y.Y. Ma
Climate policy and financial system stability: evidence from Chinese fund markets
Clim. Pol., 23 (4) (2023), pp. 395-408
CrossRef View in Scopus Google Scholar
Wang et al., 2020
J. Wang, X. Chen, X. Li, J. Yu, R. Zhong
The market reaction to green bond issuance: evidence from China
Pac. Basin Finance J., 60 (2020), Article 101294
View PDF View article View in Scopus Google Scholar
Wang et al., 2022
T. Wang, X. Liu, H. Wang
Green bonds, financing constraints, and green innovation
J. Clean. Prod., 381 (2022), Article 135134
View PDF View article View in Scopus Google Scholar
Xie et al., 2017
R.H. Xie, Y.J. Yuan, J.J. Huang
Different types of environmental regulations and heterogeneous influence on “green” productivity: evidence from China
Ecol. Econ., 132 (2017), pp. 104-112
View PDF View article View in Scopus Google Scholar
Xu et al., 2020
G. Xu, P. Schwarz, H. Yang
Adjusting energy consumption structure to achieve China's CO2 emissions peak
Renew. Sustain. Energy Rev., 122 (2020), Article 109737
View PDF View article View in Scopus Google Scholar
Yu et al., 2018
S. Yu, S. Zheng, X. Li
The achievement of the carbon emissions peak in China: the role of energy consumption structure optimization
Energy Econ., 74 (2018), pp. 693-707
View PDF View article View in Scopus Google Scholar
Zhang et al., 2021
R. Zhang, Y. Li, Y. Liu
Green bond issuance and corporate cost of capital
Pac. Basin Finance J., 69 (2021), Article 101626
View PDF View article View in Scopus Google Scholar
Zhang et al., 2018
Y.J. Zhang, Y.F. Sun, J. Huang
Energy efficiency, carbon emission performance, and technology gaps: evidence from CDM project investment
Energy Pol., 115 (2018), pp. 119-130
View PDF View article CrossRef Google Scholar
Zhou et al., 2021
D. Zhou, M. Kautonen, W. Dai, H. Zhang
Exploring how digitalization influences incumbents in financial services: the role of entrepreneurial orientation, firm assets, and organizational legitimacy
Technol. Forecast. Soc. Change, 173 (2021), Article 121120
View PDF View article View in Scopus Google Scholar
Zhou et al., 2022
S. Zhou, P. Zhou, H. Ji
Can digital transformation alleviate corporate tax stickiness: the mediation effect of tax avoidance
Technol. Forecast. Soc. Change, 184 (2022), Article 122028
View PDF View article View in Scopus Google Scholar
Zhu and Hou, 2022
B. Zhu, R. Hou
Carbon risk and dividend policy: evidence from China
Int. Rev. Financ. Anal., 84 (2022), Article 102360
View PDF View article View in Scopus Google Scholar

Cited by (7)

Self-restraint or external supervision: Green bond issuance and greenwashing
2024, Research in International Business and Finance
How green bond issuance affects firms' greenwashing behavior is an important and unexplored topic. We use the data of Chinese green bond issuance and listed firms to investigate the impact of green bond issuance on firms' greenwashing behavior. We find that green bond issuance significantly inhibits firms' greenwashing behavior. The mechanism test results show that green bond issuance inhibits firms' greenwashing behavior by improving information disclosure quality, and the negative impact of green bond issuance on firms' greenwashing behavior by improving information disclosure quality is greater than the positive impact of alleviating financing constraints on firms' greenwashing behavior. The results of heterogeneity analysis show that the negative impact of green bond issuance on greenwashing behavior is more significant for firms located in regions with a higher level of fintech, political connections, and a higher level of digital transformation. Overall, our empirical evidence suggests that inhibiting firms' greenwashing behavior requires external supervision rather than firms' self-restraint.
Decarbonization policy and high-carbon enterprise default risk: Evidence from China
2024, Economic Modelling
Studying the impact of decarbonization policy on high-carbon enterprise default risk is of great importance to improve climate risk governance. However, the relationship between decarbonization policy and default risk has not been examined to date. Adopting data from Chinese A-share listed enterprises from 2000 to 2020, this paper examines the issue by using a difference-in-differences model. The results indicate that decarbonization policy significantly increases high-carbon enterprise default risk, which is more pronounced among small-sized enterprises and those in the growth period. The results of mechanism analysis show that the decarbonization policy increases the default risk of high-carbon enterprises by increasing debt cost and decreasing profitability. Furthermore, these effects are negatively moderated by low-carbon technology innovation and digital transformation. These findings can help governments, high-carbon enterprises, and investors to address default risk triggered by decarbonization policy.
Innovation through Green Finance: a thematic review
2024, Current Opinion in Environmental Sustainability
The race for economic development worldwide is being brought at the expense of the rapid depletion of natural resources and adverse environmental changes. This has brought our attention to the global urgency of climate change. To overcome this problem, drastic changes are required in our way of living, including consumption, transportation, and production. This requires green innovation flourished by finance (also known as green finance [GF]). Considering this, we conducted a thematic analysis of the previous strand of literature that presents evidence that finance supports green innovation in different phases of life. Our thematic analysis was threefold: 1) we explored the theoretical lens adopted by recent studies in exploring the causal effect of finance in promoting green innovation, 2) we identified the scope of these studies in terms of environmental outcomes covered by them, we called them research themes in this study, and 3) based on our thematic analysis, we highlighted areas needing greater attention by academicians and practitioners. We found that GF positively impacts the firm's green innovation. The main reason discussed in the literature is that the dedicated resource availability for green innovative projects helps to minimize the constraints on funds for innovative projects. One stream of literature also highlights the rise of greenwashing as managers only use these funds for publicity, however, they are not fully committed to sustainability goals. This review provides valuable suggestions to academicians and practitioners regarding effective financial resource usage to progress toward achieving sustainability goals.
Does the financial investment preference of renewable energy firms promote their advance towards sustainable development goals?
2023, Renewable Energy
Renewable energy firms are mainly private. The long payback period and low return characteristics of renewable energy projects and the natural financing disadvantages of private firms seriously hinder the sustainable development of renewable energy firms. We study the impact of the financial investment preference of renewable energy firms on their sustainable development using a sample of Chinese renewable energy firms from 2008 to 2022. The results show that the increasing financial investment preference of renewable energy firms significantly inhibits their sustainable development. Further, we also find that the increase in digital inclusive finance and the increase in transparency weaken the negative impact of the financial investment preference of renewable energy firms on their sustainable development.
Green digital finance and technology diffusion
2024, Humanities and Social Sciences Communications
Firm-level carbon risk perception and ESG performance
2024, Environmental Science and Pollution Research

View all citing articles on Scopus

^☆: We are very grateful for the constructive comments from the co-editor, Professor Sushanta Mallick, the associate editor, and three anonymous reviewers, which greatly improved the quality of our paper.

View Abstract

[1] Bae et al., 2021
K.H. Bae, S. El Ghoul, O. Guedhami, X. Zheng
Board reforms and dividend policy: international evidence
J. Financ. Quant. Anal., 56 (4) (2021), pp. 1296-1320
CrossRef View in Scopus Google Scholar

[2] Baker et al., 2016
S.R. Baker, N. Bloom, S.J. Davis
Measuring economic policy uncertainty
Q. J. Econ., 131 (4) (2016), pp. 1593-1636
CrossRef View in Scopus Google Scholar

[3] Balachandran and Nguyen, 2018
B. Balachandran, J.H. Nguyen
Does carbon risk matter in firm dividend policy? Evidence from a quasi-natural experiment in an imputation environment
J. Bank. Finance, 96 (2018), pp. 249-267
View PDF View article View in Scopus Google Scholar

[4] Baulkaran, 2019
V. Baulkaran
Stock market reaction to green bond issuance
J. Asset Manag., 20 (5) (2019), pp. 331-340
CrossRef View in Scopus Google Scholar

[5] Beck et al., 2010
T. Beck, R. Levine, A. Levkov
Big bad banks? The winners and losers from bank deregulation in the United States
J. Finance, 65 (5) (2010), pp. 1637-1667
CrossRef View in Scopus Google Scholar

[6] Bhutta et al., 2022
U.S. Bhutta, A. Tariq, M. Farrukh, A. Raza, M.K. Iqbal
Green bonds for sustainable development: review of literature on development and impact of green bonds
Technol. Forecast. Soc. Change, 175 (2022), Article 121378
View PDF View article View in Scopus Google Scholar

[7] Blake and Moschieri, 2017
D.J. Blake, C. Moschieri
Policy risk, strategic decisions and contagion effects: firm‐specific considerations
Strat. Manag. J., 38 (3) (2017), pp. 732-750
CrossRef View in Scopus Google Scholar

[8] Bolton and Kacperczyk, 2021
P. Bolton, M. Kacperczyk
Do investors care about carbon risk?
J. Financ. Econ., 142 (2) (2021), pp. 517-549
View PDF View article View in Scopus Google Scholar

[9] Bose et al., 2021
S. Bose, K. Minnick, S. Shams
Does carbon risk matter for corporate acquisition decisions?
J. Corp. Finance, 70 (2021), Article 102058
View PDF View article View in Scopus Google Scholar

[10] Cai and Zhang, 2011
J. Cai, Z. Zhang
Leverage change, debt overhang, and stock prices
J. Corp. Finance, 17 (3) (2011), pp. 391-402
View PDF View article View in Scopus Google Scholar

[11] Callaway and Sant'Anna, 2021
B. Callaway, P.H. Sant'Anna
Difference-in-differences with multiple time periods
J. Econom., 225 (2) (2021), pp. 200-230
View PDF View article Google Scholar

[12] Chapple et al., 2013
L. Chapple, P.M. Clarkson, D.L. Gold
The cost of carbon: capital market effects of the proposed emission trading scheme (ETS)
Abacus, 49 (1) (2013), pp. 1-33
CrossRef View in Scopus Google Scholar

[13] Chen et al., 2020
D. Chen, S. Chen, H. Jin, Y. Lu
The impact of energy regulation on energy intensity and energy structure: firm-level evidence from China
China Econ. Rev., 59 (2020), Article 101351
View PDF View article View in Scopus Google Scholar

[14] Choi and Wang, 2009
J. Choi, H. Wang
Stakeholder relations and the persistence of corporate financial performance
Strat. Manag. J., 30 (8) (2009), pp. 895-907
CrossRef View in Scopus Google Scholar

[15] Clarkson et al., 2011
P.M. Clarkson, Y. Li, G.D. Richardson, F.P. Vasvari
Does it really pay to be green? Determinants and consequences of proactive environmental strategies
J. Account. Publ. Pol., 30 (2) (2011), pp. 122-144
View PDF View article View in Scopus Google Scholar

[16] Custódio et al., 2013
C. Custódio, M.A. Ferreira, L. Laureano
Why are US firms using more short-term debt?
J. Financ. Econ., 108 (1) (2013), pp. 182-212
View PDF View article View in Scopus Google Scholar

[17] Dang et al., 2022
V.A. Dang, E. Lee, Y. Liu, C. Zeng
Bank deregulation and stock price crash risk
J. Corp. Finance, 72 (2022), Article 102148
View PDF View article View in Scopus Google Scholar

[18] Del Gaudio et al., 2022
B.L. Del Gaudio, D. Previtali, G. Sampagnaro, V. Verdoliva, S. Vigne
Syndicated green lending and lead bank performance
J. Int. Financ. Manag. Account., 33 (3) (2022), pp. 412-427
CrossRef View in Scopus Google Scholar

[19] Dell'Atti et al., 2022
S. Dell'Atti, C. Di Tommaso, V. Pacelli
Sovereign green bond and country value and risk: evidence from European Union countries
J. Int. Financ. Manag. Account., 33 (3) (2022), pp. 505-521
CrossRef View in Scopus Google Scholar

[20] Delmas and Burbano, 2011
M.A. Delmas, V.C. Burbano
The drivers of greenwashing
Calif. Manag. Rev., 54 (1) (2011), pp. 64-87
CrossRef View in Scopus Google Scholar

[21] Dong et al., 2021
Y. Dong, Q. Hou, C. Ni
Implicit government guarantees and credit ratings
J. Corp. Finance, 69 (2021), Article 102046
View PDF View article View in Scopus Google Scholar

[22] El Ghoul et al., 2018
S. El Ghoul, O. Guedhami, H. Kim, K. Park
Corporate environmental responsibility and the cost of capital: international evidence
J. Bus. Ethics, 149 (2018), pp. 335-361
CrossRef View in Scopus Google Scholar

[23] Flammer, 2021
C. Flammer
Corporate green bonds
J. Financ. Econ., 142 (2) (2021), pp. 499-516
View PDF View article View in Scopus Google Scholar

[24] Huang et al., 2019
Z. Huang, G. Liao, Z. Li
Loaning scale and government subsidy for promoting green innovation
Technol. Forecast. Soc. Change, 144 (2019), pp. 148-156
View PDF View article View in Scopus Google Scholar

[25] Jung et al., 2018
J. Jung, K. Herbohn, P. Clarkson
Carbon risk, carbon risk awareness and the cost of debt financing
J. Bus. Ethics, 150 (2018), pp. 1151-1171
CrossRef View in Scopus Google Scholar

[26] Kabir et al., 2021
M.N. Kabir, S. Rahman, M.A. Rahman, M. Anwar
Carbon emissions and default risk: international evidence from firm-level data
Econ. Modell., 103 (2021), Article 105617
View PDF View article View in Scopus Google Scholar

[27] Kim et al., 2015
Y.B. Kim, H.T. An, J.D. Kim
The effect of carbon risk on the cost of equity capital
J. Clean. Prod., 93 (2015), pp. 279-287
View PDF View article CrossRef View in Scopus Google Scholar

[28] Li et al., 2022
R. Li, L. Li, Q. Wang
The impact of energy efficiency on carbon emissions: evidence from the transportation sector in Chinese 30 provinces
Sustain. Cities Soc., 82 (2022), Article 103880
View PDF View article View in Scopus Google Scholar

[29] Liu et al., 2022
F. Liu, J. Yu, Y. Shen, L. He
Does the resource-dependent motivation to disclose environmental information impact company financing? Evidence from renewable energy companies of China
Renew. Energy, 181 (2022), pp. 156-166
View PDF View article View in Scopus Google Scholar

[30] Maroufkhani et al., 2022
P. Maroufkhani, K.C. Desouza, R.K. Perrons, M. Iranmanesh
Digital transformation in the resource and energy sectors: a systematic review
Resour. Pol., 76 (2022), Article 102622
View PDF View article View in Scopus Google Scholar

[31] Nguyen and Phan, 2020
J.H. Nguyen, H.V. Phan
Carbon risk and corporate capital structure
J. Corp. Finance, 64 (2020), Article 101713
View PDF View article View in Scopus Google Scholar

[32] Oestreich and Tsiakas, 2015
A.M. Oestreich, I. Tsiakas
Carbon emissions and stock returns: evidence from the EU emissions trading scheme
J. Bank. Finance, 58 (2015), pp. 294-308
View PDF View article View in Scopus Google Scholar

[33] Reboredo et al., 2022
J.C. Reboredo, A. Ugolini, J. Ojea-Ferreiro
Do green bonds de-risk investment in low-carbon stocks?
Econ. Modell., 108 (2022), Article 105765
View PDF View article View in Scopus Google Scholar

[34] Sartzetakis, 2021
E.S. Sartzetakis
Green bonds as an instrument to finance low carbon transition
Econ. Change Restruct., 54 (3) (2021), pp. 755-779
CrossRef Google Scholar

[35] Sheng et al., 2022
H. Sheng, T. Feng, L. Liu
The influence of digital transformation on low-carbon operations management practices and performance: does CEO ambivalence matter?
Int. J. Prod. Res. (2022), pp. 1-15
Google Scholar

[36] Shu and Tan, 2023
H. Shu, W. Tan
Does carbon control policy risk affect corporate ESG performance?
Econ. Modell., 120 (2023), Article 106148
View PDF View article View in Scopus Google Scholar

[37] Su et al., 2022
Z. Su, Q. Guo, H.T. Lee
Green finance policy and enterprise energy consumption intensity: evidence from a quasi-natural experiment in China
Energy Econ., 115 (2022), Article 106374
View PDF View article View in Scopus Google Scholar

[38] Sun et al., 2019
H. Sun, B.K. Edziah, C. Sun, A.K. Kporsu
Institutional quality, green innovation and energy efficiency
Energy Pol., 135 (2019), Article 111002
View PDF View article View in Scopus Google Scholar

[39] Tang and Zhang, 2020
D.Y. Tang, Y. Zhang
Do shareholders benefit from green bonds?
J. Corp. Finance, 61 (2020), Article 101427
View PDF View article View in Scopus Google Scholar

[40] Trianni et al., 2016
A. Trianni, E. Cagno, S. Farné
Barriers, drivers and decision-making process for industrial energy efficiency: a broad study among manufacturing small and medium-sized enterprises
Appl. Energy, 162 (2016), pp. 1537-1551
View PDF View article View in Scopus Google Scholar

[41] Wang and Jiang, 2023
H. Wang, S. Jiang
Green bond issuance and stock price informativeness
Econ. Anal. Pol., 79 (2023), pp. 120-133
View PDF View article CrossRef Google Scholar

[42] Wang et al., 2023
H. Wang, S. Li, Y.Y. Ma
Climate policy and financial system stability: evidence from Chinese fund markets
Clim. Pol., 23 (4) (2023), pp. 395-408
CrossRef View in Scopus Google Scholar

[43] Wang et al., 2020
J. Wang, X. Chen, X. Li, J. Yu, R. Zhong
The market reaction to green bond issuance: evidence from China
Pac. Basin Finance J., 60 (2020), Article 101294
View PDF View article View in Scopus Google Scholar

[44] Wang et al., 2022
T. Wang, X. Liu, H. Wang
Green bonds, financing constraints, and green innovation
J. Clean. Prod., 381 (2022), Article 135134
View PDF View article View in Scopus Google Scholar

[45] Xie et al., 2017
R.H. Xie, Y.J. Yuan, J.J. Huang
Different types of environmental regulations and heterogeneous influence on “green” productivity: evidence from China
Ecol. Econ., 132 (2017), pp. 104-112
View PDF View article View in Scopus Google Scholar

[46] Xu et al., 2020
G. Xu, P. Schwarz, H. Yang
Adjusting energy consumption structure to achieve China's CO2 emissions peak
Renew. Sustain. Energy Rev., 122 (2020), Article 109737
View PDF View article View in Scopus Google Scholar

[47] Yu et al., 2018
S. Yu, S. Zheng, X. Li
The achievement of the carbon emissions peak in China: the role of energy consumption structure optimization
Energy Econ., 74 (2018), pp. 693-707
View PDF View article View in Scopus Google Scholar

[48] Zhang et al., 2021
R. Zhang, Y. Li, Y. Liu
Green bond issuance and corporate cost of capital
Pac. Basin Finance J., 69 (2021), Article 101626
View PDF View article View in Scopus Google Scholar

[49] Zhang et al., 2018
Y.J. Zhang, Y.F. Sun, J. Huang
Energy efficiency, carbon emission performance, and technology gaps: evidence from CDM project investment
Energy Pol., 115 (2018), pp. 119-130
View PDF View article CrossRef Google Scholar

[50] Zhou et al., 2021
D. Zhou, M. Kautonen, W. Dai, H. Zhang
Exploring how digitalization influences incumbents in financial services: the role of entrepreneurial orientation, firm assets, and organizational legitimacy
Technol. Forecast. Soc. Change, 173 (2021), Article 121120
View PDF View article View in Scopus Google Scholar

[51] Zhou et al., 2022
S. Zhou, P. Zhou, H. Ji
Can digital transformation alleviate corporate tax stickiness: the mediation effect of tax avoidance
Technol. Forecast. Soc. Change, 184 (2022), Article 122028
View PDF View article View in Scopus Google Scholar

[52] Zhu and Hou, 2022
B. Zhu, R. Hou
Carbon risk and dividend policy: evidence from China
Int. Rev. Financ. Anal., 84 (2022), Article 102360
View PDF View article View in Scopus Google Scholar

Does green direct financing work in reducing carbon risk?☆绿色直接融资对降低碳风险有效吗？