The Impact of Green Bond Issuance on Corporate Environmental and Financial Performance: An Empirical Study of Japanese Listed Firms

Abstract

Based on firm-level data of Japanese listed companies for the period of 2013–2022, this study conducts an empirical analysis to investigate how the issuance of green bonds influences corporate environmental and financial performance. The results show that the green bond issuance demonstrates a reduction in corporate greenhouse gas emission intensity and energy consumption intensity in the long term. Moreover, the issuance of green bonds enhances the financial performance of firms in the long run. However, the positive effect of green bond issuance on corporate environmental and financial performance is significant only among firms that have set specific quantitative environmental targets. In addition, for manufacturing and transportation green bond issuers that have set specific quantitative environmental targets, the improvement in environmental performance is evident in both the long and short term.

1. Introduction

As an emerging financial instrument, green bonds are designed to promote environmental, social, and governance (ESG) investment and Sustainable Development Goals (SDGs) by raising funds for green projects, thereby reducing greenhouse gas (GHG) emissions and energy consumption of the green bond issuers. These issuers can be governments, financial institutions, corporations, or other organizations. According to the Green Bond Principles published by the International Capital Market Association, a green bond is “any bond instrument where the proceeds will be exclusively applied to finance or refinance, in part or in full, new and/or existing eligible green projects” (ICMA, 2022). From the investors’ perspective, green bonds provide a direct way to contribute to the achievement of the SDGs through the financial markets. For green bonds, the use of funds must be in line with predefined environmental targets and sustainable development criteria to mitigate climate change. Specifically, green bond issuers need to determine their sustainability performance targets (SPTs) and key performance indicators (KPIs) before issuance and disclose them in the prospectuses or frameworks. Additionally, green bond issuers are also required to disclose comprehensive information about their anticipated green projects, including their types, objectives, budgets, and expected environmental benefits after completion. Some issuers also publish annual environmental reports before the repayment of the green bonds, disclosing the completion progress and environmental benefits of these projects, as well as the use of funds.

Green bond markets have been rapidly expanding worldwide, supported by a series of policy incentives and regulatory frameworks aimed at advancing sustainable finance. According to the Climate Bonds Initiative, the cumulative issuance of green bonds reached 3.5 trillion dollars by the end of 2024. In particular, the issuance of green bonds increased to 672 billion dollars in 2024, 9% higher than the 2023 level of 614 billion dollars (CBI, 2024). To regulate this growing market, the European Commission (EC) has established the European Green Bond Standard, which was formally adopted in November 2023. This standard relies on the European Union taxonomy to define green activities, ensuring high transparency and environmental credibility. It also places external reviewers under the regulation of the European Securities and Markets Authority, thereby ensuring regulatory oversight throughout the issuance process (EC, 2023). In China, the People’s Bank of China released the Green Bond Endorsed Projects Catalogue to consolidate definitions of green projects and regulatory criteria. The catalog excludes carbon-intensive projects such as coal-based energy and adopts the internationally recognized “Do No Significant Harm” principle, which has not only improved pricing efficiency but also reduced issuance and management costs, providing a stable yet flexible framework for the development of green bond markets (PBC, 2021). Since Nomura Research Institute issued the first Japanese green bond in 2016, the number and amount of green bond issuance have grown rapidly in recent years. According to data published by the Japan Exchange Group ESG bond information platform, from 2016 to 2022, the number of green bond issuances increased from 1 to 93, with an average annual growth rate of 112.85%. The amount of issuance has increased from 10 billion yen to 1085.6 billion yen, demonstrating an average annual growth rate of 118.41%. This rapidly expanding green bond market includes several types of corporate green bonds, such as green bonds, sustainability-linked bonds, and transition bonds. All of these bonds follow the Green Bond and Sustainability-Linked Bond Guidelines. The growth of the Japanese green bond market can be attributed to the government’s increased interest in the potential role that green bonds are expected to play in addressing environmental issues. The Japanese government is dedicated to formulating economic policies that balance environmental protection and economic development. In 2018, the Japanese Ministry of the Environment (MOE) released the Fifth Environmental Basic Plan, which outlines the Japanese government’s environmental protection goal of “comprehensively improving the environment, economy, and society,” and proposes green bonds as an essential tool to achieve this goal (MOE, 2018). In addition, the Japanese government provides tax exemptions to companies that demonstrate good environmental performance. From May 2018 to March 2021, the Japanese Ministry of Economy, Trade and Industry (METI) implemented a new tax system to promote energy efficiency. According to the Energy Efficiency Act, companies that are certified as energy efficient in two consecutive years are entitled to accelerated depreciation benefits, allowing them to deduct 20% of the purchase value of energy-saving investments (METI, 2021).

This study analyzes panel data at the firm level for 3916 listed Japanese firms from 2013 to 2022, comprising a total of 32,356 observations. Following Flammer (2021), this research employs the difference-in-differences (DID) approach to address endogeneity concerns. The treatment group comprises green bond issuers, while the control group consists of firms that did not issue green bonds but closely match the treatment firms in key firm-level characteristics. This matching method results in a subsample of 154 unique firms and 360 observations. The study contributes to the literature by offering empirical evidence on the effects of green bond issuance on corporate environmental and financial performance. First, this study indicates that green bond issuance reduces corporate GHG emission intensity and energy consumption intensity over the long term. Moreover, for manufacturing and transportation firms that set specific quantitative targets to reduce emissions or energy consumption, the improvement in their corporate environmental performance through green bond issuance is evident in both the short and long term. Second, this research also examines the impact of green bond issuance on corporate financial performance. The empirical results reveal that green bond issuance enhances corporate financial performance, as indicated by improvements in return on assets (ROA) and Tobin’s Q. Notably, only firms with specific quantitative environmental targets demonstrate significant improvements in both environmental and financial outcomes following green bond issuance. Finally, the evidence suggests that corporate environmental performance has a positive impact on financial performance.

The structure of the paper is as follows. Section 2 presents the literature review and the formulation of hypotheses. Section 3 describes the methodology and sample. Section 4 focuses on the analysis of corporate environmental performance. Section 5 presents the analysis of corporate financial performance. Section 6 describes additional analysis of manufacturing and transportation companies. Section 7 summarizes the conclusions and future research directions.

2. Literature Review and Hypothesis Development

2.1. Green Bond Issuance, Corporate Environmental and Financial Performance

How does green bond issuance affect corporate environmental and financial performance? Previous studies (Flammer, 2021; Zhou & Cui, 2019) suggest the following potential factors: (1) green commitment, (2) signaling, cost of capital, and disclosure.

2.1.1. Green Commitment

Green bond issuance represents the green commitment by a corporation to improve its environmental performance. First, as stated in the green bond prospectuses and frameworks, corporations are committing to invest significant amounts of funds in green projects by issuing green bonds. Second, to ensure that these funds are actually spent on these green projects, green bonds are usually certified by independent rating institutions, such as JCR, R&I, DNV, Sustainalytics, etc. Unlike traditional bond ratings, independent institution ratings of green bonds focus on the following aspects: use and management of proceeds, green project evaluation and selection process, environmental performance reporting, and green project completion. Third, according to the Green Bond and Sustainability-Linked Bond Guidelines, although there are no principal requirements for issuing green bonds, the guidelines recommend that corporations set specific, quantitative KPIs and SPTs and report annually on their achievements. Thus, green bond issuance can encourage corporations to promote their green commitments and enhance their environmental management capabilities, contributing to improving their environmental performance. Previous research confirms the direct or indirect impact of corporate green commitment, green management, and green training for employees on improving corporate environmental performance (Haldorai et al., 2022; Joshi & Dhar, 2020; Sharma et al., 2021; Zhu & Sarkis, 2004).

In addition to the widely publicized GHG emissions, the reduction in energy consumption is also an important issue for sustainable development. According to the Sustainable Development Goals Report 2023 published by the United Nations (UN), achieving global energy efficiency goals still requires substantial progress (UN, 2008). The average annual increase in global primary energy intensity from 2015 to 2020 is 1.4%, which is far below the 2.6% required to achieve the SDGs targets. Hence, in order to compensate for the lost time, energy intensity must increase by an average of 3.4% per year until 2030. According to the International Energy Agency (IEA), energy consumption intensity is often used as an indicator of energy efficiency because it essentially captures a proxy measure of energy demand (IEA, 2021). As a major manufacturing country, Japan has also implemented a series of policies in recent years to control the energy consumption of corporations. According to the Japan Natural Resources and Energy Agency, corporations with annual energy consumption of at least 1.5 million liters (oil equivalent) are obliged to implement an energy management system, report annual energy consumption, and submit energy efficiency plans (METI, 2022).

Although reducing energy consumption is crucial to achieving the SDGs, the existing literature has concentrated on the relationship between green bond issuance and other measures of corporate environmental performance, while ignoring energy consumption. Lian et al. (2024) use data from non-financial listed firms in China between 2010 and 2020 and find that green bond issuance can alleviate corporate financing constraints and enhance corporate green innovation capabilities, thus playing a positive role in achieving environmental goals. Kartal et al. (2024) assess the role of green bonds in advancing carbon neutrality in China from 2019 to 2023, indicating that green bond issuance significantly reduces carbon emissions in the transport and international aviation sectors. Luo and Lyu (2024) suggest that the issuance of green bonds significantly enhances corporate environmental performance, particularly in labor-intensive firms, supporting the signaling role of green bonds. Zhou and Cui (2019) conduct an empirical study based on 144 green bonds issued in China between 2016 and 2019, and they find that the issuance of green bonds contributes to the improvement in the Corporate Social Responsibility (CSR) score. Xu et al. (2023) point out that green credit reduces corporate carbon emission intensity by strengthening environmental supervision. They emphasize the importance of quantitative and standardized corporate environmental information disclosure for green finance in reducing carbon emissions. Diaz-Sarachaga (2021) also suggests that corporate sustainability reporting is poorly standardized and lacks comparability. The author highlights the importance of developing quantitative frameworks to standardize and measure business contributions to achieving the SDGs. Flammer (2021) states that independent third-party certification of green bonds is a solid corporate green commitment, emphasizing that only certified green bond issuance will improve the environmental performance of their issuers. In addition, only green projects financed by green bonds lead to improvements in corporate environmental performance, as the green bond issuance amount is too small relative to the size of corporate assets. Based on this, this study categorizes green projects into two types: GHG emission intensity reduction projects and energy consumption intensity reduction projects. Furthermore, while green bond issuers incorporate their specific quantitative environmental targets as a crucial part of their green commitments, little is known in this area.

Table A1. GHG emission and energy consumption reduction targets for green bond issuers in Japan.

Year of Issuance	Issuer Name	GHG Emission Reduction Targets	Energy Consumption Reduction Targets
2016	Nomura Research Institute, Ltd.	No	No
2018	ASICS Corporation	Yes	Yes
2018	Mitsubishi Estate Company, Limited	Yes	No
2018	MARUI GROUP CO., LTD.	Yes	No
2018	Mitsui O.S.K. Lines, Ltd.	Yes	No
2018	Mitsui O.S.K. Lines, Ltd.	Yes	No
2018	OBAYASHI CORPORATION	Yes	No
2018	Daio Paper Corporation	Yes	No
2018	Daio Paper Corporation	Yes	No
2018	TODA CORPORATION	No	No
2018	Nippon Yusen Kabushiki Kaisha	Yes	No
2019	KANEKA CORPORATION	No	No
2019	NIDEC CORPORATION	Yes	No
2019	NIDEC CORPORATION	Yes	No
2019	Mitsui Fudosan Co., Ltd.	No	No
2019	The Sumitomo Warehouse Co., Ltd.	Yes	Yes
2019	OBAYASHI CORPORATION	Yes	No
2019	MEIDENSHA CORPORATION	Yes	No
2019	Tokyo Tatemono Co., Ltd.	No	No
2019	SHIMIZU CORPORATION	Yes	Yes
2019	SEIBU HOLDINGS INC.	Yes	No
2019	Takasago Thermal Engineering Co., Ltd.	Yes	Yes
2020	Asahi Group Holdings, Ltd.	No	No
2020	AEON Mall Co., Ltd.	No	No
2020	AEON Mall Co., Ltd.	No	No
2020	Kirin Holdings Company, Limited	Yes	No
2020	KOMATSU LTD.	Yes	No
2020	SEIKO EPSON CORPORATION	No	No
2020	SEIKO EPSON CORPORATION	No	No
2020	SENKO Group Holdings Co., Ltd.	No	No
2020	Hulic Co., Ltd.	No	No
2020	RENOVA, Inc.	Yes	No
2020	RENOVA, Inc.	Yes	No
2020	MITSUI-SOKO HOLDINGS Co., Ltd.	No	No
2020	Mitsubishi Heavy Industries, Ltd.	Yes	No
2020	PENTA-OCEAN CONSTRUCTION CO., LTD.	No	No
2020	The Sumitomo Warehouse Co., Ltd.	Yes	Yes
2020	The Sumitomo Warehouse Co., Ltd.	Yes	Yes
2020	DAIWA HOUSE INDUSTRY CO., LTD.	Yes	Yes
2020	TODA CORPORATION	Yes	Yes
2020	ASAHI KASEI CORPORATION	No	No
2020	TOKYO GAS CO., LTD.	Yes	No
2020	Tohoku Electric Power Company, Inc.	No	No
2020	Tohoku Electric Power Company, Inc.	No	No
2020	TOKYU CORPORATION	Yes	Yes
2020	TOKYU CORPORATION	Yes	Yes
2020	Tokyu Fudosan Holdings Corporation	Yes	Yes
2020	Tokyu Fudosan Holdings Corporation	Yes	Yes
2020	East Japan Railway Company	Yes	Yes
2020	KAJIMA CORPORATION	Yes	Yes
2020	ENEOS Holdings, Inc.	Yes	No
2021	ASICS Corporation	Yes	Yes
2021	AEON Mall Co., Ltd.	Yes	No
2021	EXEO Group, Inc.	Yes	No
2021	SoftBank Corp.	No	No
2021	SoftBank Corp.	No	No
2021	TOYOTA MOTOR CORPORATION	Yes	No
2021	TOYOTA MOTOR CORPORATION	Yes	No
2021	YOKOREI CO., LTD.	Yes	No
2021	Mitsubishi Heavy Industries, Ltd.	Yes	No
2021	Chubu Electric Power Company, Inc.	No	No
2021	Kyushu Railway Company	No	No
2021	Kyushu Electric Power Company, Inc.	No	No
2021	Keihan Holdings Co., Ltd.	No	No
2021	Hokkaido Electric Power Company, Inc.	Yes	No
2021	Hokuriku Electric Power Company	Yes	No
2021	Nagoya Railroad Co., Ltd.	Yes	No
2021	Ajinomoto Co., Inc.	Yes	No
2021	TAISEI CORPORATION	Yes	Yes
2021	DAITO TRUST CONSTRUCTION CO., LTD.	Yes	No
2021	YASKAWA Electric Corporation	No	No
2021	IWATANI CORPORATION	No	No
2021	IWATANI CORPORATION	No	No
2021	Kawasaki Heavy Industries, Ltd.	No	No
2021	NGK INSULATORS, LTD.	Yes	No
2021	NH Foods Ltd.	No	No
2021	Nippon Yusen Kabushiki Kaisha	Yes	No
2021	Nippon Yusen Kabushiki Kaisha	Yes	No
2021	Hitachi Zosen Corporation	Yes	No
2021	Meiji Holdings Co., Ltd.	Yes	No
2021	TOKYU CORPORATION	Yes	Yes
2021	TOKYU CORPORATION	Yes	Yes
2021	Tokyu Fudosan Holdings Corporation	Yes	No
2021	East Japan Railway Company	Yes	Yes
2021	EIKEN CHEMICAL CO., LTD.	Yes	Yes
2021	SHIMIZU CORPORATION	Yes	Yes
2021	ARAKAWA CHEMICAL INDUSTRIES, LTD.	Yes	No
2021	SEIBU HOLDINGS INC.	Yes	No
2021	Nomura Real Estate Holdings, Inc.	Yes	Yes
2021	Nomura Research Institute, Ltd.	Yes	No
2021	Hankyu Hanshin Holdings, Inc.	Yes	Yes
2021	Electric Power Development Co., Ltd.	No	No
2021	IINO KAIUN KAISHA, LTD.	Yes	Yes
2021	Takashimaya Company, Limited	Yes	Yes
2021	TAKAMATSU CONSTRUCTION GROUP CO., LTD.	No	No
2021	INPEX CORPORATION	Yes	No
2021	J.FRONT RETAILING Co., Ltd.	Yes	Yes
2021	SCSK Corporation	Yes	Yes
2021	TDK Corporation	Yes	No
2021	LY Corporation	No	No
2021	KITZ CORPORATION	Yes	No
2021	ASAHI PRINTING CO., LTD.	Yes	Yes
2021	East Japan Railway Company	Yes	Yes
2021	Central Japan Railway Company	No	Yes
2021	Shiseido Company, Limited	Yes	No
2021	Electric Power Development Co., Ltd.	Yes	No

in Appendix A reports the specific quantitative corporate environmental targets for these green projects, including GHG emissions reduction targets and energy consumption reduction targets. According to the prospectuses and frameworks of all 106 green bonds in the sample, 74 bonds have set targets to reduce GHG emissions, while only 30 bonds have set targets to reduce energy consumption. Since not all green bond issuers have set specific quantitative corporate environmental targets to reduce GHG emissions and energy consumption, this study proposes the following hypothesis:

Hypothesis 1.

Issuing green bonds reduces GHG emissions intensity, especially for corporations that have set specific targets.

Hypothesis 2.

Issuing green bonds reduces energy consumption intensity, especially for corporations that have set specific targets.

2.1.2. Signaling, Cost of Capital and Disclosure

Signaling theory focuses on how firms convey information about their qualities or characteristics to investors, consumers, or other stakeholders. Spence (1973) first suggests that, in the situation of information asymmetry, the party with more information has an incentive to convey information to the other party. Signals are considered credible if they are conveyed through observable behaviors or attributes that are costly or difficult to imitate. Corporations take a variety of actions to signal their quality by conveying credible information to investors. Several previous studies point out that the influence of green bond issuance on corporate financial performance can be examined through the framework of signaling theory due to the fact that investors lack sufficient information to assess a corporation’s commitment to environmental sustainability (Flammer, 2021; Lyon & Maxwell, 2011; Lyon & Montgomery, 2015). From the investors’ perspective, green bond issuance sends a reliable signal, thus helping them to reliably distinguish between corporations that are genuinely committed to environmental goals and those that are not.

As discussed in Section 2.1.1, green bond issuance constitutes a form of corporate commitment to environmental sustainability, which is a credible signal for investors. Using daily Google search activity and multiple green bond indices, Pham and Huynh (2020) reveal a dynamic feedback relationship between investor attention and green bond market performance. Some previous studies indicate that issuing green bonds leads to an increase in corporate stock prices. Tang and Zhang (2020) offer an empirical investigation of stock market reactions to corporate ESG initiatives. Their analysis shows a significant increase in issuers’ stock prices following the disclosure of green bond issues. Notably, the degree of response is more pronounced for first-time issuers than for recurring issuers and corporate entities, as opposed to financial institution issuers. Zhou and Cui (2019) indicate that green bond issuance announcements positively affect firm stock returns. They argue that the issuance of green bonds is a sign of sustainable development and environmental protection, and can enhance investor confidence. Furthermore, their study provides empirical evidence supporting the notion that green bond issuance contributes to enhancing corporate financial performance. Flammer (2021) provides evidence of a significant positive stock market response to green bond issuance, with a cumulative abnormal return of 0.49%. In contrast, Lebelle et al. (2020) point out that green bond issuance causes negative shocks to the stock market and emphasize that corporate green bonds are considered a sign of market uncertainty. According to their empirical evidence, green bond issuance announcements lead to a 0.2% to 0.5% decline in stock prices.

In addition, green bonds might become a more cost-effective source of financing if investors are willing to accept lower yields in pursuit of the broader objective of combating climate change. Previous studies suggest that green bonds can reduce the cost of capital for corporations, thereby affecting their financial performance. Zhai et al. (2022) analyze 1577 listed Chinese manufacturing firms from 2016 to 2020 and find that participation in ESG activities allows firms to attract more investors and reduce risk, resulting in a positive market response and lower cost of capital. Li et al. (2022) conduct a study on Chinese listed companies in pollution-intensive industries from 2014 to 2019, and find that firms with higher CSR performance have less operational risk and information asymmetry, thereby reducing their cost of capital. They also highlight the role of green credit in reducing the cost of capital for firms. Baldi and Pandimiglio (2022) investigate the impact of corporate ESG score and greenwashing risk on green bond yields, indicating that investors accept lower yields for impactful projects while demanding higher returns when greenwashing risk is high. Notably, they find that such risk is particularly elevated in manufacturing firms. Zerbib (2019) notes that the green bond label on corporate bonds reflects the issuer’s intention and responsibility to pursue environmentally responsible investment activities. This environmental commitment is certified through the issuance of green bonds and monitored by investors. Agliardi and Agliardi (2019) suggest that the issuance of green bonds confers a green label upon firms, enhancing their credibility in capital markets and resulting in a decrease in their cost of capital. According to their model, they confirm a green premium ranging from 0.43 to 17.96 basis points. Gianfrate and Peri (2019) suggest that green bond issuers enjoy certain advantages over issuers of conventional brown bonds. These green bonds typically have yields that are 14 to 19 basis points lower than conventional bonds. This result suggests that it is acceptable for investors to support corporate green projects and investments by purchasing green bonds with lower yields. In addition, Wang et al. (2020) point out that since green bond contracts require issuers to fulfill environmental commitments, firms can mitigate environmental risks and obtain a better environmental reputation by issuing green bonds. Further, corporate green investment can improve environmental performance, reduce environmental risks, and enhance corporate credit status. As a result, green bonds are supposed to have pricing premiums over non-green bonds. They argue that the yields on green bonds are 34 basis points lower than the yields on conventional bonds. The green label has a positive impact, leading to a saving of USD 100.6 million in the cost of capital for companies issuing these bonds. In addition, several studies emphasize the role of green bonds as an effective tool for risk diversification and hedging, especially in times of market stress, which provides reliable market support for the lower cost of capital of green bonds (Belguith, 2025; Gazi et al., 2024; Haq et al., 2021; Rehan et al., 2024; Trancoso & Gomes, 2024). However, some other literature does not support this difference. Flammer (2021) finds no evidence that issuing green bonds can reduce the cost of capital. In addition, Tang and Zhang (2020) also argue that there is no difference between the yields of green bonds and conventional bonds.

As for corporate environmental disclosure, some previous studies categorize corporate environmental disclosure into symbolic and substantive disclosure (Adu et al., 2023; Haque & Ntim, 2020). To issue green bonds, firms are generally required to disclose sustainable finance frameworks and publish annual green bond reports detailing the implementation status of green projects, including metrics such as reductions in GHG emissions and energy consumption. Therefore, green bond issuance can be interpreted as a form of composite environmental disclosure, encompassing both symbolic commitments (i.e., self-reported environmental commitments) and substantive components, such as SPTs and KPIs related to GHG emissions and energy consumption. In both cases, both symbolic and substantive environmental disclosures are associated with the improvement in corporate financial performance. Many previous studies indicate that environmental disclosure is positively related to corporate financial performance (Broadstock et al., 2018; Clarkson et al., 2013; Iatridis, 2013; Plumlee et al., 2015). Such studies generally use the environmental disclosure index to measure corporate symbolic environmental disclosure. Haque and Ntim (2020) and Adu et al. (2023) measure corporate environmental disclosure by using an environmental disclosure index that is constructed by a series of dummy variables that measure corporate self-reporting of carbon reduction initiatives. They state that only symbolic carbon reduction initiatives enhance corporate financial performance, including ROA and Tobin’s Q, while substantive carbon reductions have no significant impact. Previous research provides some explanations for this positive relationship between environmental disclosure and corporate financial performance. Several studies suggest that environmental disclosure enhances the ability of investors to assess the future opportunities and potential risks of companies, thereby reducing investment uncertainty and ultimately lowering the cost of capital (Dhaliwal et al., 2011; Healy & Palepu, 2001). In addition, greater disclosure levels can enhance corporate legitimacy among institutional investors, thus making it easier for companies to obtain low-cost financing and tax incentives (Ntim, 2016; Suchman, 1995).

Previous studies have rarely investigated the impact of such composite environmental disclosure of green bonds on corporate financial performance. Therefore, in order to fill this existing research gap and provide new evidence for this research area, this study tests the following hypothesis:

Hypothesis 3.

Issuing green bonds improves corporate financial performance.

2.2. Relationship Between Corporate Environmental and Financial Performance

Previous studies are relatively consistent in their conclusions on the relationship between corporate environmental performance and financial performance, and generally agree that improved corporate environmental performance enhances financial performance. Many studies indicate that reducing the carbon emission intensity and energy consumption intensity can significantly increase corporate financial performance. Fan et al. (2017) indicate that energy efficiency is positively related to the corporate financial performance in energy-intensive firms of China, while this relationship is stronger in growing firms, suggesting that energy efficiency contributes to financial performance, especially when firms expand. Shahiduzzaman et al. (2025) suggest that stronger corporate environmental performance, especially in emissions reduction, enhances the long-term market value of acquiring companies in mergers and acquisitions deals, highlighting the positive relationship between corporate environmental and financial performance. Zournatzidou et al. (2025) find that corporate responsibility practices enhance the financial performance of financial institutions in Europe.

In addition, several studies point out that the role of corporate environmental performance in improving financial performance is affected by several factors. Brouwers et al. (2018) reveal that firms with lower carbon emissions can only improve their financial performance if they are unable to pass on the cost of carbon to consumers, emphasizing the complex role of carbon cost pass-through in associating corporate environmental and financial performance. Yu et al. (2022) emphasize that the emission trading system enhances corporate financial performance by promoting carbon emission reductions. Bendig et al. (2023) indicate the positive relationship between carbon emission performance and financial performance among firms adopting science-based emission reduction targets.

However, Adu et al. (2023) suggest that carbon emission performance negatively affects the market value and financial performance of companies. They indicate that corporate self-reported carbon performance does not have a significant effect on financial performance, but has a positive effect on market value. According to Misani and Pogutz (2015), companies attain optimal financial performance at moderate levels of carbon emission intensity, rather than at very low or high levels. Kim et al. (2021) suggest that improvements in corporate environmental performance are associated with improvements in their Tobin’s Q. However, in the case of chaebol firms, commendable environmental performance tends to reduce their financial performance.

Therefore, additional empirical analysis results are needed to verify the relationship between corporate environmental performance and financial performance. Based on this, this study proposes the following hypothesis:

Hypothesis 4.

Corporate environmental performance improves corporate financial performance.

3. Methodology and Data

3.1. Data and Variables

The data for this research are collected from the following two sources: GHG emissions and energy consumption data are collected from Nikkei ESG data, while corporate accounting data are sourced from the Nikkei Economic Electronic Databank System.

3.1.1. Dependent Variables

Table 1. Variables definitions.

Notation	Units	Definition
Panel A: Dependent variable
GHGI	tons/million yen	GHG emission intensity, measured as total GHG emissions scaled by sales
EI	GJ/million yen	Energy consumption intensity, measured as total energy consumption scaled by sales
ROA		Net income scaled by total assets
TQ		Tobin’s Q is defined as the sum of the book value of assets and the market value of equity, divided by the book value of assets
Panel B: Independent and control variables
GBP		Dummy variable that equals 1 in the year preceding green bond issuance, and 0 otherwise
GBS		Dummy variable that equals 1 in the year following green bond issuance, and 0 otherwise
GBL		Dummy variable that equals 1 in the second year and subsequent years after green bond issuance, and 0 otherwise
GHGT		Dummy variable equals 1 if the firm has set specific quantitative GHG emission reduction targets and 0 otherwise
NGHGT		Dummy variable equals 1 if the firm has not set specific quantitative GHG emission reduction targets and 0 otherwise
ET		Dummy variable equals 1 if the firm has set specific quantitative energy consumption reduction targets and 0 otherwise
NET		Dummy variable equals 1 if the firm has not set specific quantitative energy consumption reduction targets and 0 otherwise
Size		Logarithm of the total assets
CI		Capital expenditures divided by total assets
Lev		Total debt divided by total assets
Growth		The percentage change in sales

presents the variable definitions. At the firm level, this study selects two dependent variables to measure the corporate environmental performance. The first measure is the GHG emission intensity, which is measured as the ratio of GHG emissions (in tons) scaled by sales, subsequently abbreviated as GHGI. More specifically, GHG emissions are defined as the sum of Scope 1 (direct) and Scope 2 (indirect) emissions. The second measure is the energy consumption intensity, which is measured as the ratio of energy consumption (in GJ) scaled by sales, subsequently abbreviated as EI. To reduce heterogeneity problems due to differences in the size of firms, this study scaled both GHG emissions and energy consumption by sales, which are widely used by previous studies (Alam et al., 2019; Bolton & Kacperczyk, 2021). This research selects ROA and Tobin’s Q to measure the influence of green bond issuance on corporate financial performance. ROA is defined as net income divided by the book value of total assets. And Tobin’s Q is defined as the sum of the book value of assets and the market value of equity, divided by the book value of assets, subsequently abbreviated as TQ. The reliability of ROA and TQ as good measures of corporate financial performance has been affirmed by many previous studies (Brouwers et al., 2018; Elsayed & Paton, 2005; Fan et al., 2017; Lee & Min, 2015; Misani & Pogutz, 2015).

Table 2. Descriptive statistics.

Variables	Obs	Mean	Sd	Min	Max
GHGI	330	2.404	5.543	0.010	37.760
EI	238	17.011	30.943	0.120	159.802
ROA	360	0.035	0.036	−0.069	0.241
TQ	360	1.535	0.538	1.095	5.297
GBP	360	0.254	0.436	0	1
GBS	360	0.144	0.351	0	1
GBL	360	0.110	0.314	0	1
GHGT	360	0.293	0.456	0	1
NGHGT	360	0.707	0.456	0	1
ET	360	0.110	0.314	0	1
NET	360	0.890	0.314	0	1
Size	360	13.890	1.148	10.930	15.460
CI	360	0.047	0.026	0.004	0.167
Lev	360	0.592	0.135	0.147	0.902
Growth	360	0.037	0.133	−0.316	0.444

This table presents descriptive statistics for the matched subsamples.

presents the descriptive statistics for the matched sample. Among the dependent variables, GHGI and EI show substantial variation across firms, with standard deviations exceeding their means. This indicates considerable heterogeneity in environmental performance, possibly reflecting differences in industry sectors, firm size, or environmental strategies. It is also worth noting that the number of observations for environmental intensity measures is lower than for financial indicators, due to incomplete environmental disclosures. This data limitation highlights a persistent challenge in ESG research: the uneven availability of environmental performance data, even among firms active in the green bond market.

3.1.2. Independent and Control Variables

In order to measure the dynamic impact of green bonds before and after issuance, this study defines three dummy variables to distinguish between pre-issuance, short-term and long-term responses, following Flammer (2021). In detail, GBP is defined as a dummy variable that equals 1 in the year preceding the green bond issuance and 0 otherwise. GBS is defined as a dummy variable that equals 1 in the year following the green bond issuance and 0 otherwise. GBL is defined as a dummy variable that equals 1 in the second year and the subsequent years after the green bond issuance and 0 otherwise. Green bond issuance data are obtained from the Japan Exchange Group ESG Bond Information Platform. To examine the impact of setting specific quantitative targets by green bond issuers on corporate environmental and financial performance, this study defines GHGT as a dummy variable that equals 1 if the firm has set specific quantitative GHG emission reduction targets and 0 otherwise. In contrast, NGHGT is defined as a dummy variable that equals 1 if the firm has not set specific quantitative GHG emission reduction targets and 0 otherwise. The definitions of ET and NET are similar to GHGT and NGHGT, respectively, and reflect the setting of specific quantitative energy consumption reduction targets for green bond issuers. To minimize estimation bias resulting from omitted variables, this study includes control variables following prior studies (Alam et al., 2019; Bolton & Kacperczyk, 2021; Brouwers et al., 2018; Lee & Min, 2015). According to Table 2, only 29% of firms have set specific quantitative greenhouse gas reduction targets, and just 11% have established specific quantitative energy consumption reduction targets. These figures suggest that while green bond issuers often engage in environmental reporting, relatively few make measurable and binding commitments. Size indicates the logarithm of the total assets. CI denotes the capital intensity, defined as capital expenditures divided by total assets. Lev represents the leverage, defined as total debt divided by total assets. Growth is defined as the percentage change in sales. To mitigate the influence of outliers, all variables are winsorized at the 1st and 99th percentiles of their empirical distributions.

Table 3. Correlation matrix.

	GHGI	EI	ROA	TQ	GBP	GBS	GBL	GHGT	NGHGT	ET	NET	Size	CI	Lev	Growth
GHGI	1.000
EI	0.700 ***	1.000
ROA	−0.224 ***	−0.101 **	1.000
TQ	−0.228 ***	−0.148 ***	0.582 ***	1.000
GBP	0.001	−0.021	−0.074 ***	−0.042	1.000
GBS	−0.026	−0.001	−0.018	−0.040	0.023	1.000
GBL	−0.026	0.001	0.045	0.018	0.006	0.102 ***	1.000
GHGT	−0.024	0.048	−0.020	−0.064 **	0.043	0.457 ***	0.404 ***	1.000
NGHGT	0.024	−0.048	0.020	0.064 **	−0.043	−0.457 ***	−0.404 ***	−1.000	1.000
ET	−0.063 **	−0.037	−0.061 **	−0.043	0.075 ***	0.306 ***	0.228 ***	0.583 ***	−0.583 ***	1.000
NET	0.063 **	0.037	0.061 **	0.043	−0.075 ***	−0.306 ***	−0.228 ***	−0.583 ***	0.583 ***	−1.000	1.000
Size	0.214 ***	0.114 ***	−0.111 ***	−0.128 ***	0.093 ***	0.083 ***	0.068 **	0.102 ***	−0.102 ***	0.026	−0.026	1.000
CI	0.141 ***	0.221 ***	−0.126 ***	−0.036	0.001	−0.020	−0.055 **	−0.029	0.029	−0.032	0.032	0.103 ***	1.000
Lev	0.334 ***	0.046	−0.313 ***	−0.377 ***	0.069 ***	0.045 *	0.043	0.068 **	−0.068 **	0.020	−0.020	0.314 ***	0.050 *	1.000
Growth	−0.056 *	−0.026	0.331 ***	0.123 ***	−0.128 ***	0.050*	0.066 **	0.015	−0.015	−0.018	0.018	−0.054 *	0.023	−0.070 **	1.000

Refer to Table 1 for variable definitions. Significance levels: * $p<0.1$; ** $p<0.05$; *** $p<0.01$.

presents the correlation matrix. The correlations between pairs of variables were calculated based on the covariance between pairs of variables scaled by the standard deviation of the two variables. A significant positive correlation is shown between GHGI and EI, and between ROA and TQ, suggesting consistency between the variables measuring the environmental and financial performance of the firms. The correlation among control variables does not exceed 0.40. Therefore, multicollinearity problems do not seriously affect the regression results.

3.2. Matching Methodology

In order to examine the impact of green bond issuance on corporate environmental and financial performance, this study employs a method that combines the DID method with the two-stage least squares. This approach takes into account potential individual firm differences and the endogeneity of environmental performance, which helps to assess the effect of green bond issuance more robustly and accurately. Both methods have been widely used in previous studies on corporate environmental performance and green bond issuance (Alam et al., 2019; Brouwers et al., 2018; Flammer, 2021). The initial sample consists of 32,356 firm-years of data from 3916 Japanese listed companies, covering 12 years from 2013 to 2022. A matching methodology is employed to examine changes in firm-level environmental and financial performance following the issuance of green bonds. Following Flammer (2021), this research constructs a matched sample by matching each of the 163 green bond issuers with a non-issuer control firm that is as similar as possible, based on the following matching criteria. First, this study requires that the control firms operate in the same two-digit JSIC industry as the treated firms. Second, the nearest neighbor is selected from the remaining candidates based on four commonly used firm-level characteristics to construct a set of similar companies: Size, TQ, ROA, and Lev. For each characteristic, this study also considers the change: $\Delta$Size, $\Delta$ROA, $\Delta$TQ, and $\Delta$Lev are the changes in the variables within the two years before the bond issuance, reflecting changes from year $t-2$ to year $t-1$. Based on this, this research identifies the nearest neighbor as the control firm with the minimum Mahalanobis distance to the treated firm across these eight matching characteristics.

To illustrate the similarity between treated and control firms,

Table 4. Matching results.

Variable	Group	Obs	Mean	Median	Sd	Min	Max	p-Value (Mean)	p-Value (Median)
Size	Treated	163	14.018	14.045	1.227	10.928	17.947	0.207	0.211
	Control	163	13.736	13.711	1.303	10.872	16.950
ROA	Treated	163	0.030	0.028	0.303	−0.061	0.162	0.743	0.912
	Control	163	0.031	0.029	0.264	−0.065	0.161
TQ	Treated	163	1.516	1.378	0.486	1.046	3.644	0.731	0.223
	Control	163	1.534	1.403	0.494	1.049	4.640
Lev	Treated	163	0.605	0.639	0.127	0.252	0.902	0.247	0.223
	Control	163	0.586	0.602	0.141	0.171	0.838
$\Delta$Size	Treated	163	0.004	0.003	0.007	−0.011	0.066	0.635	0.197
	Control	163	0.003	0.003	0.004	−0.015	0.027
$\Delta$ROA	Treated	163	−0.957	−0.063	6.241	−75.204	3.754	0.145	0.740
	Control	163	−0.230	−0.035	1.160	−11.292	2.713
$\Delta$TQ	Treated	163	−0.017	−0.022	0.059	−0.140	0.296	0.824	0.740
	Control	163	−0.016	−0.024	0.051	−0.134	0.168
$\Delta$Lev	Treated	163	0.006	−0.005	0.083	−0.246	0.739	0.762	0.150
	Control	163	0.008	0.001	0.052	−0.111	0.281

The last two columns report the p-values for difference-in-means and difference-in-medians tests, respectively. Obs denotes the number of observations, and Sd represents the standard deviation.

reports descriptive statistics for eight matched characteristics. For each characteristic, the table reports the mean, median, and standard deviation for the 163 treated and control firms1. p-values for the mean and median difference tests are reported in the final two columns of the table. Specifically, the null hypothesis of equal means and equal medians could not be rejected with p-values ranging from 0.145 to 0.912. Overall, these statistics confirm the comparability between treated and control firms, supporting the validity of the control group as a reliable counterfactual for estimating outcomes in the absence of green bond issuance.

3.3. Model and Estimation Method

$${\mathrm{EP}}_{i,t}=\alpha +{\beta }_1{\mathrm{GB}}_{i,t}+{\beta }_2{\mathrm{X}}_{i,t-1}+{\mu }_i+{\lambda }_{j,t}+{\epsilon }_{i,t}$$

(1)

Equation (1) presents the regression model for the first-stage instrumental variable estimation. $E{P}_{i,t}$ represents environmental performance, including GHGI and EI, respectively. $G{B}_{i,t}$ is a vector of treatment dummy variables including GBP, GBS and GBL, used to estimate the dynamics of the treatment. To mitigate potential estimation bias due to the omitted variables, this study includes control variables following previous research (Alam et al., 2019; Brouwers et al., 2018). $X_{i,t-1}$ is a vector of control variables including Size, CI, Lev, and Growth. ${\mu }_i$ is the firm fixed effect, and ${\lambda }_{j,t}$ is the industry year fixed effect, ${\epsilon }_{i,t}$ is the error term. In addition, to examine the effect of specific quantitative targets on the environmental performance of green bond issuers, this research introduces a series of interaction terms consisting of the treatment dummy variables multiplied by the target dummy variables. Further analysis of this part is discussed in Section 4, Section 5 and Section 6.

$${\mathrm{FP}}_{i,t}=\alpha +{\beta }_1{\mathrm{GB}}_{i,t}+{\beta }_2{\mathrm{pEP}}_{i,t-1}+{\beta }_3{\mathrm{X}}_{i,t-1}+{\mu }_i+{\lambda }_{j,t}+{\epsilon }_{i,t}$$

(2)

Equation (2) presents the regression model for the second-stage estimation. $F{P}_{i,t}$ represents financial performance, including ROA and TQ. Based on previous studies, the most commonly used independent variables are selected to measure corporate financial performance (Brouwers et al., 2018; Fan et al., 2017; Lee & Min, 2015; Misani & Pogutz, 2015). $pE{P}_{i,t-1}$ is the predicted value of environmental performance from Equation (1), including pGHGI and pEI, respectively. Similar to Equation (1), the same interaction terms are introduced to investigate the effect of specific quantitative targets on the corporate financial performance.

In addition, recent studies emphasize that the influence of green bond issuance on the environmental performance varies by industry and country, and green bonds are particularly effective in reducing emissions in the manufacturing and transport sectors, while the impact on other sectors remains mixed or limited (Kartal et al., 2024; Luo & Lyu, 2024; Pata et al., 2025). However, Baldi and Pandimiglio (2022) note that the risk of greenwash is particularly pronounced among manufacturing firms, which increases their cost of capital. Based on the potential differences in the impact of green bonds on the environmental and financial performance of different industries, this study conducts additional analyses of manufacturing and transportation firms. Specifically, the models are re-estimated using subsamples of firms from the manufacturing and transportation sectors.

4. Does Green Bond Issuance Improve Corporate Environmental Performance?

Table 5. Corporate environmental performance after green bond issuance.

Variables	GHGI			EI
	(1)	(2)		(3)	(4)
GBP	0.065		GBP	0.055
	(0.621)			(0.127)
GBS	−0.047		GBS	−0.595
	(−0.322)			(−0.925)
GBL	−0.412 **		GBL	−0.906 *
	(−1.986)			(−1.776)
GBP × GHGT		−0.033	GBP × ET		0.010
		(−0.303)			(0.024)
GBS × GHGT		−0.014	GBS × ET		−0.672
		(−0.115)			(−0.850)
GBL × GHGT		−0.467 **	GBL × ET		−1.243 *
		(−2.019)			(−1.809)
GBP × NGHGT		0.074	GBP × NET		1.234
		(0.669)			(1.494)
GBS × NGHGT		−0.074	GBS × NET		−0.311
		(−0.286)			(−0.658)
GBL × NGHGT		−0.391	GBL × NET		0.924
		(−1.450)			(1.149)
Size	0.430	0.434		0.908	0.525
	(0.601)	(0.607)		(0.404)	(0.232)
CI	−1.398	−1.480		−2.456	−4.684
	(−0.519)	(−0.539)		(−0.276)	(−0.529)
Lev	−0.505	−0.565		−7.223	−9.807
	(−0.304)	(−0.334)		(−1.080)	(−1.344)
Growth	1.777 *	1.814 *		2.988 *	3.203 *
	(1.905)	(1.796)		(1.724)	(1.782)
Firm FE	YES	YES		YES	YES
Industry-year FE	YES	YES		YES	YES
Cluster	YES	YES		YES	YES
Constant	8.745	8.849		7.259	13.430
	(0.849)	(0.857)		(0.228)	(0.419)
Observations	330	330		238	238
Adjusted $R^2$	0.601	0.601		0.917	0.918

Columns 1 and 2 report the empirical results for GHGI, while columns 3 and 4 report the empirical results for EI. The t-statistic of each coefficient is shown in the parentheses. Significance levels: * $p<0.1$, ** $p<0.05$.

reports the empirical results on the impact of green bond issuance on corporate environmental performance. Column 1 shows that green bond issuance reduces GHG emission intensity only in the long term. Specifically, the green bond issuers reduced their GHG emissions intensity by 0.412 two years after issuance compared to the control group of firms. Column 2 reports the results for the impact of specific quantitative GHG reduction targets on the GHG emission intensity of green bond issuers. This study finds that the issuance of green bonds reduces the GHG emission intensity of firms only when the firms have set specific quantitative targets. In addition, the GHG emission intensity of these firms is significantly reduced by 0.467 compared to control firms in the long term, which is higher than the overall long-term level of the treatment group.

The estimation results in column 3 report a significant negative coefficient of GBL on EI. The results show that the energy consumption intensity of firms that issued green bonds is reduced by 0.906 in the long run compared to the control firms that did not issue green bonds. Column 4 examines the impact of specific quantitative energy reduction targets on the energy consumption intensity of green bond issuers by including interaction terms into the model. The result indicates that green bond issuance appears to reduce energy consumption intensity only for companies that have set specific quantitative targets. Specifically, the energy consumption intensity of green bond issuers that have set specific quantitative targets is reduced by 1.243 after two years of green bond issuance compared to control firms. The estimation results in columns 1 and 3 report the results of the first-stage instrumental variables estimation. Based on these results, pGHGI and pEI are derived and subsequently used as key inputs in the second-stage instrumental variables estimation.

Generally, the empirical results provide partial support for Hypothesis 1 and Hypothesis 2, suggesting that green bond issuance reduces GHG emissions intensity and energy consumption intensity only in the long term, particularly among firms that have set specific quantitative environmental targets. These findings are consistent with previous research emphasizing the positive environmental impact of green bond issuance. However, this study extends existing evidence by distinguishing between firms with and without specific quantitative environmental targets. This contributes to a more detailed understanding of how green bond issuance can drive environmental improvements when combined with measurable and credible commitments.

5. Does Green Bond Issuance Improve Corporate Financial Performance?

5.1. Green Bond Issuance and ROA

Table 6. ROA after green bond issuance.

Variables	(1)	(2)		(3)	(4)
GBP	0.002		GBP	0.001
	(0.659)			(1.010)
GBS	−0.001		GBS	−0.002
	(−0.607)			(−1.063)
GBL	0.009 *		GBL	0.020 *
	(1.883)			(1.751)
GBP × GHGT		0.008	GBP × ET		−0.016
		(0.761)			(−0.968)
GBS × GHGT		0.000	GBS × ET		0.009
		(0.612)			(1.228)
GBL × GHGT		0.013 *	GBL × ET		0.017 *
		(1.909)			(1.807)
GBP × NGHGT		0.002	GBP × NET		0.001
		(0.545)			(0.155)
GBS × NGHGT		−0.004	GBS × NET		−0.002
		(−0.623)			(−0.213)
GBL × NGHGT		−0.009	GBL × NET		0.024
		(−1.524)			(1.100)
pGHGI	−0.035 ***	−0.032 ***
	(−6.585)	(−7.382)
pEI				−0.003 ***	−0.003 ***
				(−6.585)	(−6.242)
Size	−0.012	−0.011		0.005	0.007
	(−0.592)	(−0.528)		(0.254)	(0.341)
CI	−0.199	−0.186		−0.159	−0.146
	(−0.943)	(−0.880)		(−0.767)	(−0.708)
Lev	0.033	0.035		0.026	0.032
	(0.411)	(0.419)		(0.328)	(0.397)
Growth	0.046 *	0.045		−0.025	−0.022
	(1.686)	(1.436)		(−0.747)	(−0.656)
Firm FE	YES	YES		YES	YES
Industry-year FE	YES	YES		YES	YES
Cluster	YES	YES		YES	YES
Constant	0.338	0.289		0.059	0.033
	(1.199)	(1.010)		(0.210)	(0.117)
Observations	360	360		360	360
Adjusted $R^2$	0.686	0.690		0.686	0.693

The t-statistic of each coefficient is shown in the parentheses. Significance levels: * $p<0.1$, *** $p<0.01$.

reports the empirical results of the impact of green bond issuance and environmental performance on ROA. Estimation results in column 1 report a significant positive coefficient between GBL and ROA. This result indicates that the ROA of green bond issuers is 0.009 higher than that of the control firms in the long term. Column 2 categorizes green bond issuers into firms that have set specific quantitative GHG emission reduction targets and firms that have not. The results show that for the green bond issuers that set targets, ROA is 0.013 higher than the control firms in the long term. However, for green bond issuers that have not set targets, issuing green bonds does not significantly increase their ROA. Columns 3 and 4 replace pGHGI with pEI and GHG emission intensity reduction targets with energy consumption intensity reduction targets and come to similar conclusions, which suggests that the results are robust. Additionally, the results reveal that pGHGI and pEI have statistically significant negative coefficients on ROA. Specifically, a 1 unit decrease in GHGI is associated with a 0.035 unit increase in ROA. In column 3, the results show that a decrease of 1 unit in energy consumption intensity is associated with an increase of 0.003 units in ROA.

5.2. Green Bond Issuance and Tobin’s Q

Table 7. Tobin’s Q after green bond issuance.

Variables	(1)	(2)		(3)	(4)
GBP	0.003		GBP	0.001
	(0.243)			(0.078)
GBS	0.071 *		GBS	0.070 *
	(1.882)			(1.874)
GBL	0.165 **		GBL	0.178 **
	(2.187)			(2.317)
GBP × GHGT		0.006	GBP × ET		0.002
		(0.411)			(0.143)
GBS × GHGT		0.137 **	GBS × ET		0.083 *
		(2.070)			(1.788)
GBL × GHGT		0.179 **	GBL × ET		0.201 **
		(2.099)			(2.264)
GBP × NGHGT		0.010	GBP × NET		−0.002
		(0.440)			(−0.083)
GBS × NGHGT		0.030	GBS × NET		0.034
		(0.992)			(1.022)
GBL × NGHGT		0.161	GBL × NET		0.039
		(1.517)			(0.901)
pGHGI	−0.039 **	−0.051 **
	(−2.413)	(−2.320)
pEI				−0.004 **	−0.004 **
				(−2.413)	(−2.323)
Size	0.278	0.256		0.298	0.316 *
	(1.508)	(1.371)		(1.574)	(1.750)
CI	−0.218	−0.082		−0.173	−0.109
	(−0.544)	(−0.221)		(−0.430)	(−0.273)
Lev	0.862	0.916		0.855	0.902
	(1.055)	(1.116)		(1.045)	(1.108)
Growth	−0.142	−0.203		−0.222 *	−0.216 *
	(−1.242)	(−1.633)		(−1.728)	(−1.686)
Firm FE	YES	YES		YES	YES
Industry-year FE	YES	YES		YES	YES
Cluster	YES	YES		YES	YES
Constant	−3.269	−2.925		−3.583	−3.848 *
	(−1.403)	(−1.246)		(−1.492)	(−1.677)
Observations	360	360		360	360
Adjusted $R^2$	0.732	0.740		0.732	0.737

The t-statistic of each coefficient is shown in the parentheses. Significance levels: * $p<0.1$, ** $p<0.05$.

reports the empirical results of the impact of green bond issuance and environmental performance on Tobin’s Q. Estimation results in column 1 indicate that the TQ of green bond issuers is 0.071 higher than that of the control firms in the short term, and is 0.165 higher than that of the control firms in the long term. Column 2 categorizes green bond issuers into firms that have set specific quantitative energy consumption reduction targets and firms that have not. The empirical result shows that for green bond issuers that have set targets, TQ is 0.137 higher than that of control firms in the short term and 0.179 higher in the long term. However, for green bond issuers that have not set targets, issuing green bonds does not significantly increase their TQ. Columns 3 and 4 replace pGHGI with pEI, and replace GHG emission intensity reduction targets with energy consumption intensity reduction targets, and come to similar results, suggesting that the results are robust. Furthermore, the results show that pGHGI and pEI have statistically significant negative coefficients on TQ. In particular, a decrease of 1 unit in the GHGI is associated with an increase of 0.039 unit in the TQ. In column 3, the results show that a 1 unit decrease in energy consumption intensity is associated with a 0.004 unit increase in TQ.

In general, the issuance of green bonds significantly enhances corporate financial performance, as reflected in improvements in both ROA and Tobin’s Q, particularly in the long term. These results provide partial support for Hypothesis 3, confirming that issuing green bonds improves corporate financial performance, which is consistent with previous studies that highlight the signaling role of green bonds and their ability to reduce the cost of capital. However, the results of this study suggest that these effects are observed only among firms that have set specific quantitative environmental targets. Furthermore, the empirical results provide support for Hypothesis 4, suggesting that corporate environmental performance improves financial performance, which is consistent with existing research and emphasizes that improvements in corporate environmental performance can generate long-term financial benefits.

6. Additional Analyses

The findings in Section 5 are based on the full sample of green bond issuers. However, recent research indicates that the impact of green bond issuance on environmental performance varies by industry, with particularly significant effects in the manufacturing and transportation sectors. Given these industry differences, this study re-estimates the model using a subsample of firms from these two industries and conducts additional analysis.

The empirical results in

Table 8. Corporate environmental performance after green bond issuance for manufacturing and transportation firms.

Variables	GHGI			EI
	(1)	(2)		(3)	(4)
GBP	−0.024		GBP	0.426
	(−0.725)			(0.254)
GBS	−0.084		GBS	−0.916 *
	(−0.873)			(−1.829)
GBL	−0.223 *		GBL	−1.664 *
	(−1.984)			(−1.738)
GBP × GHGT		−0.178 *	GBP × ET		0.203
		(−1.770)			(0.788)
GBS × GHGT		−0.213 **	GBS × ET		−0.809 *
		(−2.517)			(−1.871)
GBL × GHGT		−0.282 ***	GBL × ET		−1.962 *
		(−3.758)			(−1.837)
GBP × NGHGT		−0.015	GBP × NET		1.603
		(−0.451)			(1.149)
GBS × NGHGT		0.090	GBS × NET		−1.250
		(0.987)			(−0.854)
GBL × NGHGT		−0.167	GBL × NET		−1.501
		(0.954)			(−0.437)
Size	−0.658 **	−0.544		−4.097 *	−2.340
	(−2.013)	(−1.591)		(−1.866)	(−0.732)
CI	1.153	2.906		18.154	23.570
	(0.736)	(1.627)		(0.968)	(1.109)
Lev	1.655 *	1.774		−10.522 *	−12.710
	(1.787)	(1.639)		(−1.871)	(−0.705)
Growth	−0.032	−0.299		−3.896	−6.937
	(−0.113)	(−1.102)		(−1.151)	(−1.549)
Firm FE	YES	YES		YES	YES
Industry-year FE	YES	YES		YES	YES
Cluster	YES	YES		YES	YES
Constant	9.437 **	7.687 *		7.495	5.273
	(2.197)	(1.672)		(1.433)	(1.181)
Observations	137	137		105	105
Adjusted $R^2$	0.843	0.860		0.837	0.858

Columns 1 and 2 report the empirical results for GHGI, columns 3 and 4 report the empirical results for EI. The t-statistic of each coefficient is shown in the parentheses. Significance levels: * $p<0.1$, ** $p<0.05$, *** $p<0.01$.

,

Table 9. ROA after green bond issuance for manufacturing and transportation firms.

Variables	(1)	(2)		(3)	(4)
GBP	−0.004		GBP	−0.014 **
	(−1.254)			(−2.193)
GBS	−0.014		GBS	−0.026 **
	(−1.661)			(−2.347)
GBL	0.026 *		GBL	0.016 **
	(1.719)			(2.543)
GBP × GHGT		−0.014	GBP × ET		−0.044 ***
		(−1.040)			(−3.656)
GBS × GHGT		−0.007	GBS × ET		−0.011
		(−0.702)			(−0.822)
GBL × GHGT		0.034 ***	GBL × ET		0.085 ***
		(2.825)			(4.133)
GBP × NGHGT		−0.002	GBP × NET		−0.018 ***
		(−1.121)			(−3.101)
GBS × NGHGT		−0.009	GBS × NET		−0.034 ***
		(−1.550)			(−2.899)
GBL × NGHGT		0.030 **	GBL × NET		0.015 **
		(2.003)			(2.250)
pGHGI	−0.158 **	−0.207 ***
	(−2.274)	(−2.733)
pEI				−0.015 **	−0.018 ***
				(−2.274)	(−2.879)
Size	−0.156 ***	−0.174 ***		−0.115 ***	−0.137 ***
	(−3.372)	(−3.406)		(−3.436)	(−5.165)
CI	−0.096	−0.147		−0.002	−0.044
	(−0.540)	(−0.694)		(−0.009)	(−0.205)
Lev	0.294 **	0.330 **		−0.127	−0.204
	(2.264)	(2.178)		(−0.888)	(−1.273)
Growth	0.031	0.049		−0.024	−0.020
	(0.760)	(1.248)		(−0.554)	(−0.487)
Firm FE	YES	YES		YES	YES
Industry-year FE	YES	YES		YES	YES
Cluster	YES	YES		YES	YES
Constant	2.271 ***	2.554 ***		1.946 ***	2.334 ***
	(3.448)	(3.607)		(3.584)	(5.235)
Observations	156	156		156	156
Adjusted $R^2$	0.840	0.851		0.840	0.856

The t-statistic of each coefficient is shown in the parentheses. Significance levels: * $p<0.1$, ** $p<0.05$, *** $p<0.01$.

and

Table 10. Tobin’s Q after green bond issuance for manufacturing and transportation firms.

Variables	(1)	(2)		(3)	(4)
GBP	−0.045		GBP	−0.069
	(−0.967)			(−0.710)
GBS	0.094		GBS	0.065
	(1.136)			(0.511)
GBL	0.059 **		GBL	0.160 ***
	(2.291)			(2.804)
GBP × GHGT		−0.096	GBP × ET		−0.104
		(−1.258)			(−1.074)
GBS × GHGT		0.046	GBS × ET		0.005
		(0.651)			(0.034)
GBL × GHGT		0.097 **	GBL × ET		0.274 **
		(2.127)			(2.555)
GBP × NGHGT		−0.031	GBP × NET		−0.072
		(−0.778)			(−0.741)
GBS × NGHGT		0.249	GBS × NET		0.076
		(1.063)			(0.581)
GBL × NGHGT		0.067 *	GBL × NET		0.170 *
		(1.712)			(1.855)
pGHGI	−0.382 **	−0.174 **
	(−2.342)	(−2.228)
pEI				−0.037 *	−0.040 *
				(−1.842)	(−1.880)
Size	−0.264	−0.408		−0.164	−0.239
	(−0.490)	(−0.811)		(−0.469)	(−0.644)
CI	−1.209	−1.054		−0.981	−0.873
	(−0.702)	(−0.704)		(−0.494)	(−0.427)
Lev	0.849	0.845		−0.170	−0.027
	(0.461)	(0.506)		(−0.194)	(−0.029)
Growth	0.709 *	0.641 **		0.577	0.601
	(1.929)	(2.090)		(1.450)	(1.455)
Firm FE	YES	YES		YES	YES
Industry-year FE	YES	YES		YES	YES
Cluster	YES	YES		YES	YES
Constant	5.570	7.239		4.783	5.751
	(0.742)	(1.041)		(0.806)	(0.933)
Observations	156	156		156	156
Adjusted $R^2$	0.698	0.740		0.698	0.702

The t-statistic of each coefficient is shown in the parentheses. Significance levels: * $p<0.1$, ** $p<0.05$, *** $p<0.01$.

are consistent with the main findings. The results in Table 5 and Table 8 indicate that green bond issuance can significantly enhance corporate environmental performance in the short term for manufacturing and transportation companies that have established specific targets for reducing GHG emissions or energy consumption. The GHGI of these manufacturing and transportation firms starts to decrease one year before green bond issuance, while the EI starts to decrease one year after green bond issuance. In contrast, based on the full sample regression results, this effect does not appear until two years after the issuance of green bonds.

The results in Table 6 and Table 9 indicate that, in the long run, green bond issuance leads to a significant increase in ROA for manufacturing and transportation companies, regardless of whether specific GHG emission reduction or energy consumption reduction targets are set or not. Despite this, the coefficients in Table 9 suggest that firms with specific goals exhibit greater increases in ROA. Table 10 replaces ROA with Tobin’s Q, and the results remain robust.

7. Conclusions and Discussions

Emerging as a notable financial instrument, green bonds have garnered escalating attention in recent times. These bonds play a key role in redirecting financial flows towards energy efficiency and emission reduction projects, thereby forcing companies to increase their environmental investments, promote green innovations, and improve corporate environmental performance. In addition, issuing green bonds can also reduce the cost of capital for companies, increase corporate transparency, and ultimately improve corporate financial performance. Consequently, green bond issuance contributes to simultaneous improvements in both corporate environmental and financial performance.

The findings of this study are fourfold. First, green bond issuance contributes to the reduction in GHG emission intensity and energy consumption intensity, although this takes several years. However, this improvement in environmental performance is only effective for green bond issuers that have set specific quantitative environmental targets. Second, green bond issuance significantly increases corporate ROA and Tobin’s Q. Moreover, this effect is only significant for green bond issuers that have set targets, and is weaker or absent for issuers that have not set targets. Third, for manufacturing and transportation companies that have set specific quantitative environmental targets, green bond issuance can significantly enhance their environmental performance in the short term. Finally, improvements in corporate environmental performance significantly enhance corporate financial performance.

The findings of this study have important theoretical and practical implications. In particular, this study clarifies that setting corporate environmental targets significantly affects the impact of green bond issuance on corporate environmental and financial performance. While previous studies have generally concluded the positive relation between green bonds and corporate environmental performance, this research reveals that such effects are not universal. Specifically, setting specific quantitative environmental targets is the necessary condition to ensure that green bond issuance can improve corporate environmental and financial performance. This highlights environmental target setting as a crucial mechanism that strengthens the credibility and impact of green bonds. Theoretically, this finding refines the application of signaling theory in the context of sustainable finance. It suggests that only green bonds based on specific environmental commitments can send a credible signal to investors and stakeholders about the long-term sustainability strategy of a company. Not all green bonds transmit the same level of green commitment or reliability. From a practical perspective, this research suggests that companies considering green bond issuance should include specific quantitative environmental targets in their sustainability frameworks to improve corporate environmental and financial performance. Overall, this study is important for corporate managers, policy makers, and regulators as it provides empirical evidence for promoting environmental target setting and environmental disclosure to enhance the credibility of green bonds and their effects on improving corporate environmental and financial performance.

This study has some limitations that suggest future research directions. First, the environmental performance measures of this research include GHG emissions intensity and energy consumption intensity, while financial performance is measured through ROA and Tobin’s Q. Although these commonly used indicators represent important areas of corporate environmental and financial performance, future research could extend this analysis to other specific dimensions. Second, this study only considered data from Japanese listed firms. Future research could explore this issue in the context of developing countries, where financial market conditions and levels of environmental investment are significantly different from those in developed countries. Finally, to investigate the impact of specific quantitative environmental targets on the environmental and financial performance of green bond issuers, this study only considers the target settings of green bond issuers. However, for other companies that have not issued green bonds, the impact of environmental targets on their environmental and financial performance is still worth investigating.