Carbon Footprint, Financial Structure, and Firm Valuation: An Empirical Investigation

Abstract

This study aims to investigate the complex link between carbon emissions, firm value, and financial choice in regard to the GCC, a dynamic emerging economy. It also seeks to answer the question on whether the financial structure of a firm moderates the correlation between carbon emissions and firm value. We focus on analyzing data from non-financial firms registered on the GCC stock markets between 2010 and 2020. By applying the GLS technique, we assess the impact of carbon emissions on firm value and examine the manner in which a firm’s financial structure either enhances or hinders this relationship. The results demonstrate that there is a strong and adverse connection between carbon emissions and corporate value, as increased emissions translate into lower corporate value. The study then moves on to emphasize the critical role that capital financing plays in mitigating the detrimental effects of carbon emissions. This is accomplished by balancing both debt and equity in terms of their proper proportions (optimal capital structure). However, excessive borrowing could have adverse consequences in terms of carbon emissions on company value. Moreover, the GMM estimator is also applied to carry out a robustness check and the results are consistent with the main findings. This study highlights the significance of financial strategy in advancing sustainability and protecting business value. These findings are supported by both stakeholder and signaling theory, proving that companies can use their capital financing to signal their dedication to sustainability. These results could be used by GCC policymakers to create rules and regulations that encourage environmentally friendly corporate activities and efforts to lower emissions. The research expands the existing literature by examining the difficulties and opportunities faced by GCC firms when combining financial strategy with environmental objectives. It may be necessary to perform additional research in regard to various circumstances and for an extended period, because this study is restricted to non-financial sectors.

1. Introduction

Since the Industrial Revolution, the emission of greenhouse gases (GHGs) has increased rapidly. This explosion in the release of carbon dioxide (CO₂) is linked to catastrophic and frequent changes in the weather around the world (Le and Nguyen-Phung 2024). One of the most important environmental problems faced worldwide in the twenty-first century is global warming or the effect of climate change (Adamolekun and Kyiu 2024; Galati et al. 2023; Le and Nguyen-Phung 2024; Matsumura et al. 2014; Haque et al. 2016). Corporate social responsibility, which is commonly known as “CSR”, has moved from being an incidental issue to a major strategic goal (Karwowski and Raulinajtys-Grzybek 2021). Companies must demonstrate their responsibility in regard to the environment in order to meet the increasing demands of investors, customers, and other stakeholders. The stakeholder group of a corporation expects to gain from its reputation for implementing environmentally responsible practices. These advantages encompass potential improvements in the value of a company, higher income, and favorable perceptions by a range of stakeholders, such as employees, investors, customers, creditors, suppliers, and local communities (Simnett et al. 2009). The control systems in companies are also designed to gather information and data on carbon emissions, as a result of demands from external entities and stockholders. As a result, in response to these demands, companies have started to release their carbon footprint data, even when it is recognized that this may hurt the value of the firm (Lee and Cho 2021). As part of the academic debate on the matter, Azhgaliyeva and Le (2023) argue that companies that apply efficient measures to detect environmental issues could develop objectives for long-term economic growth. They also point out that technological advancements and environmental legislation have beneficial effects in regard to improving firm performance. On the other hand, Esty and Porter (1998) claim that the preliminary costs related to minimizing carbon levels are high, and that a long period of time is required for a business to absorb them. Hence, there is still much to learn about the effects of environmental protection on a firm’s value.

Furthermore, another important factor that determines the value of a business is financial leverage (financial structure), which is the ratio of borrowing (debt) to total equity (Ahmed et al. 2023a; Ali and Ahmed 2021). A capital framework that achieves an ideal balance between the tax advantages of debt and the expense of a financial crisis is recommended by conventional theory, such as trade-off theory (by Modigliani and Miller 1963) and agency theory (by Jensen and Meckling 1976). The selection of an appropriate financial structure may also be significantly influenced by environmental efficiency in terms of business risk and debt costs. Sensitivity to risks related to a business’s credibility, compliance with regulations, litigation, and uncertain responsibility for the environment raise business risks and make it harder for the company to pay off its debts (Chen et al. 2021). Therefore, companies that depend on high levels of debt financing prioritize short-term achievements over a long-term plan (Abdullah and Tursoy 2021; Ahmed et al. 2023b, 2024b; Sdiq and Abdullah 2022), which might lead to the generation of more emissions. Alternatively, companies that adopt a balanced capital structure follow a strategy designed to protect the environment, as they protect themselves from risky projects. Thus, they can promote their reputation in the marketplace (Al Amosh et al. 2024).

There are some theories that can be used to guide our research in order to examine the correlation between financial structure, carbon emissions, and firm value. Agency theory is one of these, and it suggests that uncontrolled managers may give priority to short-term benefits rather than sustainable development. This perspective causes an increase in CO₂ and, finally, creates a conflict of interest among the major group of stockholders. This is in line with the hypothesis that this issue would worsen due to high levels of leverage, which is often associated with higher expectations in terms of short-term achievements (Jensen and Meckling 1976). In contrast, stakeholder theory suggests that companies should have a responsibility to consider all stakeholder groups when they make final choices. Firms that demonstrate a strong dedication to long-term sustainability, as seen by their minimized emission levels, may be demonstrating to stakeholders and investors that they operate responsibly (Griffin and Mahon 1997). The signaling theory (developed by Ross 1977) states that companies can communicate information regarding their long-term opportunities and risk profile through their financial strategies, which include borrowing. As a result, these related philosophical views are likely to have an impact on the complexity of the link between carbon emissions, financial structure, and corporate valuation.

This study has several goals. Initially, it aims to examine the effects of carbon emissions on the value of non-financial firms listed on the GCC capital markets. After that, it explores the direct association between capital financing and firm value, and answers the question on whether the financial structure has any influence on the link between carbon emissions and firm value within emerging economies. From the empirical perspective, several studies have investigated the link of carbon footprint with firm value and performance; for instance, Adamolekun and Kyiu (2024), Le and Nguyen-Phung (2024), Matsumura et al. (2014) and (Benkraiem et al. 2024) found an adverse relationship between CO₂ emission and profitability. However, Busch et al. (2022) evidenced that a reduction in firm value is correlated with decreased carbon emissions, and Lee and Cho (2021) and Wang et al. (2014) argued that the value of the firm is positively and significantly affected by carbon emissions. However, to the best of our knowledge, few studies have been conducted to examine the correlation between financial structure, carbon emissions, and firm value, specifically in the context of Gulf Cooperation Council (GCC) stock markets. Hence, by filling the previously described gap, our study enriches the existing literature, and explores the challenges and opportunities faced by firms that operate in emerging economies when combining financial strategy with environmental objectives. The Gulf Cooperation Council (GCC) stands itself apart from other areas in two ways: first, most of its governments have developed Vision 2030; second, global investors are increasingly choosing to invest in GCC nations (Ahmed et al. 2024a). Thus, this region offers a valuable sample for research.

The findings of this study offer significant contributions. Firstly, the results enrich the current literature on carbon emissions and firm valuation by concentrating on emerging markets with a dynamically emerging economy, such as the GGC. Even though the effects of CO₂ emissions on company performance have been extensively studied in industrialized nations, our study fills a gap through an examination of how the emission of greenhouse gases influences emerging markets. Therefore, understanding the link between carbon emissions and the value of firms in this region is essential, as it presents both possibilities and challenges for companies seeking to adopt a sustainable environmental policy. Secondly, our findings have real-world implications for companies operating in underdeveloped nations. We emphasize the importance of making financial decisions that can successfully negotiate the complexities of environmental performance.

The paper continues in the following sections: A literature review is presented in Section 2, while the materials and method are provided in Section 3. The results are presented in Section 4, and the discussion is given in Section 5. The Section 6 includes a conclusion and recommendations.

2. Literature Review

2.1. Theoretical Background

The stakeholder hypothesis holds that an organization’s performance is based on its ability to maintain its relationships with a variety of stakeholders, including employees, investors, customers, creditors, suppliers, local communities (Dhaliwal et al. 2012), and those that might impact a business’s operations (Nekhili et al. 2017). According to this theory, environmental activities undertaken by companies to lower greenhouse gas emissions may lead to financial possibilities and lend them an advantage over their competitors, both of which will raise the value of a company (Freeman 2015; Makni et al. 2009). Firms that perform environmental practices may boost their value and reputation, and increase financial return (Ferrat 2021). However, companies that are perceived as being environmentally unfriendly may experience severe damage to their reputation and a decline in market share (Bose et al. 2024). This may result in further increases in the unpredictability of future cash flows, raising the cost of borrowing and undermining the value of the company (Zhou et al. 2020; Böcskei and Hágen 2017). Moreover, this theory recognizes Corporate Social Responsibility (CSR) as a crucial component that companies may utilize to satisfy the demands of their stakeholders (Benkraiem et al. 2024). In addition, when the requirements of its stakeholders are met entirely, highly creative companies typically perceive a favorable increase in stock value (Godfrey et al. 2009).

Agency theory developed by Jensen and Meckling (1976) emphasizes the conflicts that exist between managers (agents) and owners (principals) (Ahmed et al. 2024a; Ahmed and Hágen 2023). The principal gives authorization to the agent to manage the firm in the principal’s best interests. Nonetheless, agency costs are typically the result of the owner’s and manager’s information asymmetry (Shen et al. 2020). Similarly, shareholders believe that managers should take proactive steps to reduce exposure to environmental risks associated with climate change, even if such steps might harm shareholder value in the short term (Matsumura et al. 2014). The hypotheses behind agency theory suggest that managers could prioritize short-term gain over long-term achievement, which may increase carbon emissions. This behavior might be interpreted by shareholders as proof of poor governance and risk control; therefore, it could lower the value of the firm. However, when managers comply with environmental practices through modern technology, this can improve their reputation, which results in enhancing market value and investor’s confidence (Nishitani and Kokubu 2012). In this regard, agency theory claims that debt can restrict managers from spending excessive money on unprofitable projects, as they are committed to returning the loan and related expenses (Ahmed et al. 2023a; Sdiq and Abdullah 2022).

Moreover, the voluntary publication of carbon emissions in annual or sustainability reports is a common use of the concept of signaling. The signaling theory developed by Ross (1977) claims that information released by corporations will make their signals visible.

Managers of a company, also known as “transmitters”, implement sustainability reports that allow interested parties to know that they intend to follow ethical practices. Other parties, including debtholders, may utilize the non-financial information they find in sustainability reporting to guide their decision-making about the company and to supplement their current understanding of its operations (Connelly et al. 2011). Investors, for example, review environmental reports to find information about the risk of regulation and environmental risk before making decisions about investments (Krueger et al. 2020). Creditors also scan the sustainability reports that are related to a firm’s credibility. Non-compliance with regulations related to the environment raises corporate risk, and makes it harder for them to return their obligations. From this perspective, a company’s commitment to sustainability and overall management standards may be inferred from its carbon emissions. Shareholders may place a higher value on a business with low carbon emissions because it communicates to the public that it is well-organized, progressive, and complies with social norms (Friske et al. 2023).

These theories prove the hypothesis that better firm performance and value can be achieved through reductions in carbon emissions. Companies may generate long-term benefits and reach sustainable development by implementing signaling processes, mitigating conflicts of interest between managers and owners, and controlling interactions with stakeholders efficiently.

2.2. Empirical Evidence and the Formation of Hypotheses

Empirical research constantly links the outside environment and internal business attributes to elements that analyze the firm’s value. This section provides an overview of the empirical research on the interplay of financial structure and carbon emissions in relation to firm value.

2.2.1. Effect of Carbon Emissions on Firm Value

The complex relationship between environmental performance and firm value has been the subject of continuous discussion since the 1990s. The main premise in this discourse (Porter 1991) posits a positive connection, suggesting that environmental performance positively influences firm performance. A revised perspective supports credibility to this approach by highlighting the fact that expenses and costs associated with reducing pollution can be viewed as investments in green technologies (Le and Nguyen-Phung 2024). In the long run, these investments may help to lower clearance expenses (Benkraiem et al. 2024). Recently, companies around the globe have started to minimize the amount of carbon they produce voluntarily (Houqe et al. 2022), but there are still possible consequences associated with releasing these dangerous gases. From a stakeholder theory perspective, companies that implement sustainable practices could improve their financial return, reputation, and worth (Ferrat 2021). Nonetheless, firms considered to be environmentally destructive face considerable damage to their identity and a loss of share in the marketplace. For instance, companies must spend additional funds on technological advancement for producing products and services with low emissions of greenhouse gases, acquire or adopt innovative technologies to reduce emissions, and make other efforts to lessen the ecological impacts of different stakeholders, which result in more financial outflow (Lee and Cho 2021). However, modern corporations still face a significant problem in attempting to reduce the elements contributing to climate change, which raises the average temperature (Benkraiem et al. 2024). In the literature, the opposite correlation between emissions of carbon and company value is supported, and this highlights how voluntary environmental disclosure by businesses increases the value of the company. For example, Garzón-Jiménez and Zorio-Grima (2021) suggested that firms that have higher capital expenses release more CO₂ emissions. Their study also confirms that firms that follow a CSR strategy have lower costs of debt. Moreover, by collecting data from 82 firms registered on the Indonesia Stock Market from 2014 to 2018, Hardiyansah et al. (2021) concluded that carbon disclosure has a favorable influence on firm value and reputation. They also argued that carbon disclosure can be formed as the basis for shareholders’ assessments of sustainability. Likewise, Adamolekun and Kyiu (2024), Le and Nguyen-Phung (2024), Matsumura et al. (2014), Benkraiem et al. (2024), Ahmadi and Bouri (2017), Friske et al. (2023) and Houqe et al. (2022) found an adverse relationship between CO₂ emissions and firm value and performance. Also, they concluded that there is an encouraging association between the environmental outcomes of companies and sustainability reporting. However, Busch et al. (2022) argued that a reduction in firm value is related to low CO₂ emissions, and Lee and Cho (2021) and Wang et al. (2014) evidenced that the value of the firm is positively and significantly impacted by carbon emissions. They also found that firms that behave effectively in terms of the environment typically offer reports of their CO₂ emissions voluntarily. Therefore, we propose the first hypothesis as follows:

Hypothesis 1. 

Carbon emissions are a significant determinant of firm value.

2.2.2. Effect of Financial Leverage on Firm Value

Since the current study aims to investigate the effects of carbon emissions on firm value and answer the question of whether financial leverage significantly moderates the above relationship, it is crucial to explain the link between financial structure and firm value. The focus of our examination of the literature will continue in this area. Studies by Ahmed et al. (2023b), Ardalan (2017) and Fosu (2013) declare that the precise link between financial decisions and corporate performance might change based on the circumstances. They observed that several factors, including corporate size, economic condition, and other internal and external indicators tend to have an impact on the characteristics of the relationship between capital choices and firm value or performance. According to agency theory, managers can put short-term revenue ahead of long-term success (Podukhovich 2023), which might lead to higher emissions of carbon dioxide. Therefore, the theory points out that debt at optimum levels may be utilized to regulate and discipline management action, which will eventually enhance corporate performance and value (Ronoowah and Seetanah 2024). From this point of view, numerous studies demonstrate that financial leverage positively affects corporate value and performance (Abdullah 2020; Abdullah and Tursoy 2021; Adair and Adaskou 2015; Berger and Bonaccorsi di Patti 2006; Fosu 2013; Jouida 2018; Margaritis and Psillaki 2007; Ngatno et al. 2021). Additional research, however, has revealed a negative correlation between the structure of firms’ capital and their value (Ahmed et al. 2023a, 2023b, 2024a; Barburski and Hołda 2023; Dawar 2014; Khan et al. 2023; Muhammed et al. 2024; Sdiq and Abdullah 2022; Sutomo et al. 2020). Others demonstrated a non-linear connection, that is, both positive and adverse correlations have been identified (Ngatno et al. 2021). The above argument claims that the findings of previous investigations are still unclear and contradictory. Hence, we propose the second hypothesis as follows:

Hypothesis 2. 

Financial leverage is a significant determinant of firm value.

2.2.3. Financial Leverage, Carbon Emissions, and Firm Value

The existing literature on the relationship between firm capital financing and risk from external sources indicates that when challenged with crises from the outside, businesses should proactively select their financial structure (Klasa et al. 2018). Even with international efforts to minimize carbon emissions, firms face a significant external threat from the increasingly stringent climate legislation. According to Shu et al. (2023), the choice of financing is impacted by elevated regulatory risk, which eventually affects business valuation. Agency theory suggests that managers could adopt sustainable techniques, but only if they serve their personal benefits at the cost of stakeholder interests (Vural-Yavaş 2021). In contrast to the conventional understanding, the mutual benefit approach asserts that companies may gain competitive benefits and improve their financial results by increasing productivity and efficiency in operations through greater environmental sustainability (Hang et al. 2018; Hart and Dowell 2011). In recent years, the company’s financial choices have been impacted by the risks associated with carbon legislation (Shu et al. 2023). For instance, Australia’s adoption of the Kyoto Agreement was viewed as an external surprise, and empirical studies have demonstrated that enhanced policy risk related to carbon can result in a reduction in company borrowing capacity and a decreased probability of dividend payments (Balachandran and Nguyen 2018; Nguyen and Phan 2020). Theoretically, strengthening laws and regulations to reduce greenhouse gas emissions raises the risk of carbon management and environmental control expenses. It also enhances the possibility that major polluters would experience financial default, and makes it more challenging for them to get financing from lenders (Huang et al. 2021). Hence, to maintain financial stability as well as take advantage of development prospects, owners, creditors, and investors request that companies with significant environmental policy risk change their funding choices when they experience adverse effects from the shift to lower emissions (Shu et al. 2023).

Moreover, a study by Coulson and Monks (1999) stated that a firm with a significant environmental issue cannot receive funding if the issue has not been solved first. Companies that use green management techniques can profit via financial institutions concerned with sustainability performance when making funding choices, as well as better terms of loans. Kalash (2021) also showed that environmental performance favors financial leverage and firm performance. However, Kumar and Firoz (2018) found that companies that generate pollution seek more funding from external financiers, and their debt-financing expenses climb. Hence, firm value is negatively affected. Likewise, (Nguyen and Phan 2020; Shu et al. 2023) firms are incentivized to reduce debt financing due to the correlation between the risk of carbon emissions and financial instability. Li et al. (2014) also found that the cost of debt impacts emissions intensity positively, thereby affecting firm value and performance unfavorably. From the above arguments, we propose to generate the third hypothesis as follows:

Hypothesis 3. 

Financial leverage significantly moderates the correlation between carbon emissions and firm value.

3. Data and Methodology

3.1. Data

The data of company-specific, carbon emissions, and macroeconomic indicators, such as Gross Domestic Product (GDP), employed in our investigation are gathered from academic databases, such as Thomson Reuters, and the official website of World Bank data. The final data consist of 195 non-financial listed companies over eleven years from 2010 to 2020 in the Gulf Cooperation Council (GCC). Although the GCC nations originally contained six countries, data limitations restricted us from eliminating the United Arab Emirates. Financial sectors are also excluded from this investigation because, according to Jena et al. (2020) and Zeitun (2014), they have a different system and commitments compared to non-financial firms. Therefore, our final sample was limited to considering data from 195 non-financial companies; however, our initial sample included a larger number of companies. The drop was brought about by differences in accessibility to the non-financial company’s data and variable classifications.

Moreover, we collect data on CO₂ emissions at the country level in this analysis as a measure of carbon emissions, as the data at the company level are not accessible. Although this technique captures the environmental impacts of major geographical corporations, it could overgeneralize firm-level variation and restrict the result’s direct applicability to specific companies. However, the research offers insightful information about overall patterns, considering the high number of major non-financial companies in the GCC nations and their significant CO₂ output. The GCC countries mainly depend on the energy sector, meaning that bigger companies release higher emissions due to the nature of their activities. Hence, according to this macro-level paradigm, the calming impact of financial structure is investigated, providing insights into how businesses modify their capital financing in the face of environmental difficulties. Future studies might improve the validity of the current findings by including data about emissions at the firm level if it is accessible.

3.2. Explanation of Variables

3.2.1. Firm Value

The current research views firm value as an explained (dependent) variable, and, in theory, it is taken as directly influenced by carbon emissions and financial leverage. In the literature, different indicators have been used to measure firm value and performance, such as Tobin’s Q (TOBQ), market to book value (MBV), return on assets (ROA), return on equity (ROE), and earnings per share (EPS) (Adamolekun and Kyiu 2024; Bose et al. 2024; Busch et al. 2022; Friske et al. 2023; Hardiyansah et al. 2021; Houqe et al. 2022; Le and Nguyen-Phung 2024; Zeitun 2014). In this study, we used both TOBQ and MBV as proxies of firm value.

3.2.2. Carbon Emissions

The independent variable in this study is carbon emissions, and according to stakeholder theory, companies that engage in environmental efforts to reduce their carbon emissions may find new opportunities for growth, and establish an advantage over competitors, both of which will increase their value (Freeman 2015; Makni et al. 2009); further, Adamolekun and Kyiu (2024) and Le and Nguyen-Phung (2024) found that carbon emissions have a detrimental effect on firm value. Previous studies used total greenhouse gas emissions, log of CO₂ emissions, year-on-year change in carbon emissions (which is computed by the differences in emissions of carbon between the current and prior year) and the intensity of greenhouse gas emissions (Adamolekun and Kyiu 2024; Bose et al. 2024; Busch et al. 2022; Houqe et al. 2022; Le and Nguyen-Phung 2024; Shu et al. 2023). Our study used year-on-year changes in carbon emissions. According to Houqe et al. (2022), this metric is resistant to outliers and cannot appreciably distort the findings.

3.2.3. Financial Structure

Financial leverage refers to the systematic utilization of borrowed funds in order to optimize the shareholder’s return, and theoretically, it has a significant impact on firm value (Abdullah and Tursoy 2021; Ahmed et al. 2023a). Financial leverage is used in this study as the explanatory (independent) and moderating variable. Debt to assets (DTA), debt to equity (DTE), debt to market capitalization (DTM), long-term debt ratio, (LTD), short-term debt ratio (STD), and equity multiplier (EM) are commonly used in the literature to measure financial structure (Ahmed et al. 2023a, 2024a; Jouida 2018; Kalash 2021; Ngatno et al. 2021; Nguyen and Phan 2020; Sdiq and Abdullah 2022; Shu et al. 2023). Our study used DTE, STD, LTD, and DTM as metrics of financial structure. The definitions of the study’s variables are presented in

Table 1. Variable definitions.

Type	Variables	Label	Descriptions	Source
Dependent	TOQ	Tobin’s Q	(Market capitalization + book value of debt)/book value of assets	(Adamolekun and Kyiu 2024)
	MBV	Market to book value	Market value of equity/book value of equity	(Bose et al. 2024)
Independent	CEN	Carbon emissions	Percentage change in annual CO2 emissions.	(Houqe et al. 2022)
Independent and moderating	DTE	Debt to equity	Total debt/Total shareholder’s equity	(Ahmed et al. 2024a)
	STD	Short-term debt	Total short-total debt/total assets	(Shu et al. 2023)
	LTD	Long-term debt	Total long-total debt/total assets	(Ngatno et al. 2021)
	DTM	Debt to market capitalization	Total debt/(Total debt + Market capitalization)	(Shu et al. 2023)
Control	FMS	Firm size	log of total assets	(Adamolekun and Kyiu 2024)
	GPM	Gross profit margin	Total gross profit/total sales	(Hardiyansah et al. 2021)
	TAN	Tangibility	Total fixed assets total assets	(Houqe et al. 2022)
	CEX	Capital expenditure	Total capital expenditure/Total sales	(Bose et al. 2024)
	GDP	Gross domestic product	Annual GDP growth rate	(Ahmed et al. 2024a)

.

3.2.4. Control Variables

Our study includes several control variables to accurately analyze the association between carbon emissions, financial structure, and firm value. To preserve and manage the market features and lessen selection bias, we incorporated these factors into the regression models. Consistent with prior investigations, this study uses firm size, profitability, tangibility, capital expenditure, and GDP growth as control variables (Adamolekun and Kyiu 2024; Ahmadi and Bouri 2017; Benkraiem et al. 2024; Ahmed et al. 2024b; Bose et al. 2024; Hardiyansah et al. 2021; Houqe et al. 2022; Shu et al. 2023). The control variables used in this study are justified in the context of the GCC nations. Firstly, both financial choice and CO₂ emissions are influenced by the region’s leading asset-heavy and energy-intensive sectors, which are reflected in company size and physical assets. Secondly, the company’s financial stability, investment opportunities, and firm valuation can be determined by profitability and capital expenditure. Lastly, the macroeconomic climate formed by trade and investment fueled by fossil fuels is considered via GDP growth.

3.3. Method

The secondary data provided by listed companies on the GCC stock market are analyzed quantitatively in this investigation. The suggested relationships are examined using multiple regression methods. The Generalized Least Square (GLS) estimator is used for data analysis. According to Ahmed et al. (2023a), Ahmed et al. (2024a), Bai et al. (2021), Wooldridge (2015) and Saif-Alyousfi et al. (2020), a cross-sectional relationship, heteroskedasticity between panels, and first-order autocorrelation do not affect GLS. This is because GLS considers the difficulties of homoscedasticity and normality. Thus, it is a more appropriate technique. Moreover, the GLS technique is an improved OLS model that predicts models with serial correlation problems, and operates more strongly with data that are not normal (Abubakar et al. 2018; Saif-Alyousfi 2020).

Following the direction of Zaid et al. (2020), Ahmed et al. (2023a), Houqe et al. (2022) and Ngatno et al. (2021), the following basic econometric models were proposed to test the study hypotheses and achieve the study objectives:

$$\mathrm{Single}\text{-}\mathrm{effect},{FV}_{it}=a_0+{\beta }_1{CEN}_{it}{+\beta }_2{FS}_{it}+{\beta }_3{c}_{it}+e_{it}$$

$$\mathrm{Joint}\text{-}\mathrm{effect},{FV}_{it}=a_0+{\beta }_1{CEN}_{it}+{{\beta }_2{FS}_{it}+\beta }_3\left({CEN}_{it}\times {FS}_{it}\right)+{\beta }_4{c}_{it}+e_{it}$$

Here, $i$ is an index of the firms that have been observed; $t$ is the time; $a_0$ is a constant; ${\beta }_1$ to ${\beta }_4$ are the coefficients of the explanatory variables; $CEN$ displays independent (explanatory) variables, such as carbon emissions; $FS$ displays moderating factors, such as financial structure; ${CEN}_{it}\times {FS}_{it}$ is the outcome variable determined by independent and moderating variables; $c_{it}$ is the control variable; and $e_{it}$ is an error term. Furthermore, this study used multiple regression methods to examine the effects of the interplay of financial leverage and carbon emissions on firm performance. Using this method, the relationship between one dependent (explained) variable and several independent (explanatory) factors is investigated (Ahmed et al. 2023b; Abdullah and Tursoy 2021). Thus, a multiple regression strategy may be used to assess two associations, which is useful (Jaccard et al. 1990). Based on the above arguments, the following econometric models could be suggested as the expanded regression model to evaluate the link mentioned above:

Single-effect model,

$$\begin{array}{c}{FV}_{it}={\beta }_0+{\beta }_1{CEN}_{it}+{\beta }_2{DTE}_{it}+{\beta }_3{STD}_{it}+{\beta }_4{LTD}_{it}+{\beta }_5{DTMC}_{it}+{\beta }_6{FMS}_{it}\\ +{{\beta }_7{GPM}_{it}+{\beta }_8{TAN}_{it}+{\beta }_9{CEX}_{it}+{\beta }_{10}{GDP}_{it}+e}_{it}\end{array}$$

Joint-effect model,

$$\begin{array}{ll}{FV}_{it}={\beta }_0+{\beta }_1{CEN}_{it}& +{\beta }_2{DTE}_{it}+{\beta }_3{STD}_{it}+{\beta }_4{LTD}_{it}+{\beta }_5{DTMC}_{it}+{\beta }_6({CEN}_{it}\times {DTE}_{it})+{\beta }_7({CEN}_{it}\times {STD}_{it})\\ & +{\beta }_8({CEN}_{it}\times {LTD}_{it})+{\beta }_9({CEN}_{it}\times {DTMC}_{it})+{\beta }_{10}{FMS}_{it}\\ & +{{\beta }_{11}{GPM}_{it}+{\beta }_{12}{TAN}_{it}+{\beta }_{13}{CEX}_{it}+{\beta }_{14}{GDP}_{it}+e}_{it}\end{array}$$

Moreover, the Generalized Method of Moments (GMM) or dynamic regression is employed as the robustness test in this investigation. The lag of the dependent (explained) variable and the residuals are assumed to be correlated consistently through the implementation of GMM.

4. Results

4.1. Descriptive Statistics

The descriptive results for the variables used in the analysis during the sample period are shown in

Table 2. Descriptive results.

	TOQ	MBV	CEN	DTE	STD	LTD	DTM	FSM	GPM	TAN	CEX	GDP
Mean	0.933	1.862	−0.235	0.827	0.112	0.165	0.350	13.401	0.266	0.664	0.282	2.471
Std. Dev.	1.112	2.303	0.770	1.210	0.109	0.161	0.273	2.426	0.321	0.208	0.708	4.493
Minimum	0.003	0.008	−1.979	0.000	0.000	0.000	0.000	7.782	−4.138	0.041	0.000	−8.855
Maximum	9.286	24.077	2.431	18.227	0.711	0.770	0.991	20.010	0.998	0.995	7.935	19.592
Skewness	2.665	3.249	0.494	6.518	1.527	1.014	0.494	0.226	−3.063	−0.737	5.840	0.350
Kurtosis	13.117	19.441	3.950	74.208	5.398	3.350	1.969	2.457	33.626	2.787	46.454	4.864
VIF			1.16	1.91	2.22	3.15	3.90	2.21	1.12	1.47	1.09	1.16
Observations	2145	2145	2145	2145	2145	2145	2145	2145	2145	2145	2145	2145

. Among the proxies of firm value, TOQ has a lower mean value and lower standard deviation (M = 0.933; SD = 1.112), where the MBV arithmetic mean is M = 1.862; SD = 2.303. The minimum and maximum values of TOQ are 0.003 and 9.286, while the highest and lowest values of MBV are 0.008 and 24.077, respectively. The arithmetic mean of CEN is −0.235 with a deviation of 0.770. The minimum and maximum values of CEN are −1.979 and 2.431, respectively. The mean values of financial structure measures are as follows: DTE (M = 0.827; SD = 1.210; MIN = 0.000; MAX = 18.227); STD (M = 0.112; SD = 0.109; MIN = 0.000; MAX = 0.711); LTD (M = 0.165; SD = 0.161; MIN = 0.000; MAX = 0.770); and DTM (M = 0.350; SD = 0.273; MIN = 0.000; MAX = 0.991). The arithmetic means for control variables are as follows: FMS (M = 13.401; SD = 2.426; MIN = 7.782; MAX = 20.010); GPM (M = 0.266; SD = 0.321; MIN = −4.138; MAX = 0.998); TAN (M = 0.664; SD = 0.208; MIN = 0.041; MAX = 0.995); CEX (M = 0.282; SD = 0.708; MIN = 0.000; MAX = 7.935); and GDP (M = 2.471; SD = 4.493; MIN = −8.855; MAX = 19.592).

Studies by Brooks (2014) and Sdiq and Abdullah (2022) argue that data are considered normal if the standard skewness and kurtosis are between (0) and (3). The results of Table 2 show that the dataset has a non-normal distribution because the data do not fit into the standard range. However, it has been predicted that the non-normal distribution of data will not present an issue in studies that have large sample sizes (Ahmad et al. 2022; Ahmed et al. 2023a; Ghasemi and Zahediasl 2012; Panda and Nanda 2020). In this analysis, 2145 firm-year observations from five different countries were considered. Hence, non-normal distributions would not be a problem in our dataset. Variables such as GPM and Tan are skewed negatively, while the rest are positively skewed. In addition, research conducted by Liang et al. (2008) revealed that a distribution is classified as “platykurtic” if the kurtosis is less than 3, as “mesokurtic” if it is equal to 3, and as “leptokurtic” if it is greater than 3. Therefore, the findings of the kurtosis in Table 2 demonstrate that, except for DTM, FMS, and TAN, which have “platykurtic” distributions, all variables were distributed in a “leptokurtic” manner.

4.2. Correlation Matrix

The Pearson correlation coefficients for TOQ as a dependent variable and independent and control variables are displayed in

Table 3. Correlation analysis between TOQ and independent variables.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)	(11)
(1) TOQ	1
(2) CEN	−0.09 ***	1
(3) DTE	−0.20 ***	0.01	1
(4) STD	−0.08 ***	−0.08 ***	0.31 ***	1
(5) LTD	−0.22 ***	−0.02	0.55 ***	−0.17 ***	1
(6) DTM	−0.60 ***	0.04 **	0.51 ***	0.28 ***	0.60 ***	1
(7) FMS	−0.53 ***	0.11 ***	0.17 ***	−0.15 ***	0.34 ***	0.61 ***	1
(8) GPM	0.05 **	−0.06 ***	−0.02	−0.14 ***	0.13 ***	−0.02	0.11 ***	1
(9) TAN	−0.16 ***	−0.04 **	0.11 ***	−0.32 ***	0.45 ***	0.19 ***	0.24 ***	0.25 ***	1
(10) CEX	−0.03	0.03	0.01	−0.03	0.12 ***	0.01	0.01	0.16 ***	0.24 ***	1
(11) GDP	−0.02	0.34 ***	−0.06 ***	−0.04 *	−0.06 ***	0.01	0.14 ***	0.05 **	−0.02	0.06 ***	1

Note(s): *** p < 1%; ** p < 5%; * p < 10%.

. It is proven that there was little to no substantial correlation between the independent variables and between control and independent variables. Moreover, TOQ was positively associated with GPM. However, for others, it was seen that TOQ is inversely related.

Table 4. Correlation matrix between MBV and independent variables.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)	(11)
(1) MBV	1
(2) CEN	−0.11 ***	1
(3) DTE	0.12 ***	0.01	1
(4) STD	0.09 ***	−0.08 ***	0.31 ***	1
(5) LTD	−0.08 ***	−0.02	0.55 ***	−0.17 ***	1
(6) DTM	−0.44 ***	0.04 **	0.51 ***	0.28 ***	0.60 ***	1
(7) FMS	−0.45 ***	0.11 ***	0.17 ***	−0.15 ***	0.34 ***	0.61 ***	1
(8) GPM	0.02	−0.06 ***	−0.02	−0.14 ***	0.13 ***	−0.02	0.11 ***	1
(9) TAN	−0.18 ***	−0.04 **	0.11 ***	−0.32 ***	0.45 ***	0.19 ***	0.24 ***	0.25 ***	1
(10) CEX	−0.02	0.03	0.01	−0.03	0.12 ***	0.01	0.01	0.16 ***	0.24 ***	1
(11) GDP	−0.06 ***	0.34 ***	−0.06 ***	−0.04 *	−0.06 ***	0.01	0.14 ***	0.05 **	−0.02	0.06 ***	1

Note(s): *** p < 1%; ** p < 5%; * p < 10%.

shows the relationship between MBV and the other suggested variables. Overall, it proves that there was little to no substantial correlation between the independent variables and between control and independent variables. MBV was positively associated with DTE, STD, and GPM. However, for others, it was seen that MBV is inversely related. Additionally, according to Lee and Cho (2021), Palaniappan (2017), and Yoshikawa and Phan (2003), if the degree of association between independent factors is more than 0.7, then the issue of multicollinearity should be considered. However, as presented in both Table 3 and Table 4, the independent variables do not significantly relate to each other. This conclusion is supported by the Variance Inflation Factor (VIF) values as illustrated in Table 2, which were determined for each independent variable. In studies by Gujarati and Porter (2009) and Panda and Nanda (2020), it was confirmed that when the VIF value is greater than ten, multicollinearity problems are present and considered a significant issue. Table 2 shows that the VIF has the highest value of 3.90. Thus, this study is free from the issues of multicollinearity.

4.3. Panel Unit Root Test Statistics

To determine if data series display level stationarity, four independent unit root tests were performed in the study; they are Augmented Dickey–Fuller (ADF) Fisher (Dickey and Fuller 1979), Phillips–Perron (PP) (Phillips and Perron 1988), Levin–Lin–Chu (LLC) (Levin et al. 2002), and Harris–Tzavalis (Hadri-Z) (Harris and Tzavalis 1999). As shown in

Table 5. Panel data stationarity tests.

Variables	Status	ADF-Fisher	PP	LLC	Hadri-Z	Results
TOQ	I(0)	472.43 ***	699.00 ***	−11.34 ***	22.12 ***	Reject null hypothesis H0
MBV	I(0)	485.65 ***	649.98 ***	−10.38 ***	21.21 ***	Reject null hypothesis H0
CEN	I(0)	454.23 ***	527.72 ***	−16.86 ***	26.57 ***	Reject null hypothesis H0
DTE	I(0)	585.22 ***	565.97 ***	−19.09 ***	18.30 ***	Reject null hypothesis H0
STD	I(0)	519.58 ***	688.52 ***	−12.54 ***	17.55 ***	Reject null hypothesis H0
LTD	I(0)	501.56 ***	504.75 ***	−12.06 ***	20.13 ***	Reject null hypothesis H0
DTM	I(0)	514.34 ***	521.89 ***	−16.66 ***	21.54 ***	Reject null hypothesis H0
FMS	I(0)	504.20 ***	645.89 ***	−15.58 ***	24.52 ***	Reject null hypothesis H0
GPM	I(0)	511.44 ***	631.88 ***	−13.48 ***	13.59 ***	Reject null hypothesis H0
TAN	I(0)	507.05 ***	703.87 ***	−17.22 ***	17.94 ***	Reject null hypothesis H0
CEX	I(0)	595.99 ***	1006.30 ***	−30.56 ***	15.02 ***	Reject null hypothesis H0
GDP	I(0)	1475.91 ***	2405.28 ***	−58.66 ***	73.34 ***	Reject null hypothesis H0

Note(s): *** p < 1%.

, it was discovered that all twelve variables were stationary at their level I(0). Therefore, at the 1% significance level, all variables are determined to be stationary, proving the absence of a unit root.

4.4. Model Specification Tests

Several common estimate approaches were employed, as taken from the previous literature, including Generalized Least Squares (GLS), Ordinary Least Square (OLS), Fixed Effect, and Random Effect (Adamolekun and Kyiu 2024; Ahmed et al. 2023a, 2024a, 2024b; Benkraiem et al. 2024; Friske et al. 2023; Houqe et al. 2022; Le and Nguyen-Phung 2024). It is necessary to confirm the fundamental assumptions that ensure linear regression’s correctness before we can examine the outcomes of our analyses. The panel econometric approach, namely, the GLS method, was used in this work to investigate the relationship between carbon emissions, financial structure, and firm value.

Table 6. Diagnostic test results.

Test Summary	Dependent Variable (TOQ)	Dependent Variable (MBV)
Lagrange Multiplier Test	46.43 ***	46.09 ***
Chow Test	2335.61 ***	2276.78 ***
Hausman Test	102.30 ***	50.96 ***
Breusch–Pagan–Godfrey (${Chi}_2$)	108.29 ***	149.19 ***

Note(s): *** p < 1%; ** p < 5%; * p < 10%.

provides a series of diagnostic tests used to assess whether the fixed effects (FE) model is more appropriate than the random effect (RE) model, and if heteroscedasticity is apparent. To make a decision across pooled OLS and RE models, the Lagrange Multiplier (LM) test could be used, and then the pooled OLS over FE model could be compared with the Chow test. The Hausman test evaluates the accuracy and integrity of the RE estimator to distinguish between the FE and RE models. Moreover, to address potential estimation difficulties such as heteroscedasticity, additional diagnostic tests were implemented.

The findings displayed in Table 6 demonstrate that the LM test has a probability < 1%, claiming that RE models are more suitable than polled OLS. In other words, individual effects are random. The probability of the Chow test is also <1%, suggesting that pooled OLS is not an appropriate model. In other words, FE models are more effective. Further, to choose between RE and FE models, the Hausman test was conducted, and the results show the probability is < 1%, indicating that FE models are more effective and accurate. However, the results of Breusch–Pagan–Godfrey display a significance level probability < 1%, proving that heteroscedasticity is present. Hence, the null hypothesis of homoscedasticity cannot be accepted. Based on the above arguments, it is evident that GLS with cross-section weight is the most robust and efficient estimation method for this study. According to Bai et al. (2021) and Saif-Alyousfi (2020), the GLS approach allows for cross-sectional correlation and/or heteroscedasticity among the panels. Abubakar et al. (2018) also argued that the GLS model prevails, as it perceives core data problems such as homoscedasticity and normalcy.

4.5. Cointegration Analysis of Panel Data

A longer-term equilibrium connection between the panel variables was examined using the Kao cointegration test. This approach was developed by Engle and Granger (1987).

Table 7. Cointegration test.

Kao Test	Dependent Variable (TOQ)		Dependent Variable (MBV)
	t-Statistic	Probability	t-Statistic	Probability
ADF	−4.480032	0.000 ***	−3.069540	0.001 ***
Residual variance	0.133532		0.658201
HAC variance	0.142310		0.558755

Note(s): *** p < 1%.

presents the outcomes of the test, which reveals clear evidence of cointegration. The alternative hypothesis of no cointegration was accepted because the computed t-statistics were significant at crucial values at the 1% level. This suggests a stable long-term link between the variables and shows that all indicators are integrated.

4.6. Regression Analysis (GLS)

A GLS regression with cross-sectional weights was performed to evaluate the impacts of independent and moderating factors on firm value under the standard presumptions of the linear regression technique.

Table 8. GLS regression results.

Variables	Model 1 (TOQ)		Model 2 (TOQ)		Model 3 (MBV)		Model 4 (MBV)
	Coef.	t-Stat.	Coef.	t-Stat.	Coef.	t-Stat.	Coef.	t-Stat.
CEN	−0.025 *** (0.005)	−5.157	−0.061 *** (0.010)	−6.151	−0.074 *** (0.010)	0.022	−0.126 *** (0.022)	−5.792
DTE	−0.020 *** (0.005)	−4.225	−0.021 *** (0.004)	−4.844	0.116 *** (0.023)	0.025	0.107 *** (0.025)	4.207
STD	0.640 *** (0.066)	9.731	0.554 *** (0.065)	8.492	3.725 *** (0.172)	0.181	3.451 *** (0.181)	19.100
LTD	0.690 *** (0.061)	11.232	0.622 *** (0.061)	10.213	3.599 *** (0.164	0.173	3.404 *** (0.173)	19.696
DTM	−1.164 *** (0.044)	−26.254	−1.046 *** (0.044)	−23.894	−3.433 *** (0.092)	0.097	−3.252 *** (0.097)	−33.525
CEN $\times$ DTE			0.009 *** (0.003	3.464		0.016	0.008 (0.016)	0.513
CEN $\times$ STD			−0.290 *** (0.053)	−5.429		0.131	−0.733 *** (0.131)	−5.583
CEN $\times$ LTD			−0.246 *** (0.046)	−5.340		0.119	−0.543 *** (0.119)	−4.545
CEN $\times$ DTM			0.217 *** (0.027)	8.065		0.057	0.486 *** (0.057)	8.577
FSM	−0.238 *** (0.019)	−12.759	−0.242 *** (0.018)	−13.088	−0.344 *** (0.035)	0.036	−0.336 *** (0.036)	−9.421
GPM	0.098 *** (0.020)	4.899	0.083 *** (0.019)	4.278	0.092 ** (0.036)	0.036	0.080 ** (0.036)	2.235
TAN	0.005 (0.058)	0.088	−0.047 (0.058)	−0.810	0.044 (0.107)	0.108	−0.005 (0.108)	−0.047
CEX	0.005 (0.008)	0.705	0.005 (0.007)	0.667	0.031 ** (0.014)	0.014	0.016 (0.014)	1.169
GDP	0.003 *** (0.001)	3.530	0.002 *** (0.001)	2.826	0.005 *** (0.002)	0.002	0.004 ** (0.002)	2.267
C	4.318 *** (0.250)	17.277	4.376 *** (0.247)	17.695	6.471 *** (0.463)	0.474	6.395 *** (0.474)	13.503
$R_2$	0.901		0.905		0.906		0.907
Adj. $R_2$	0.890		0.894		0.896		0.897
F-statistic	86.81		88.82		92.41		91.70
Prob.	0.000		0.000		0.000		0.000

Note(s): *** p < 1%; ** p < 5%.

displays the findings of this investigation. The findings of GLS for models 1 and 3 illustrate that without a moderating factor, CMN has an adverse effect on both TOQ and MBV, with coefficients of β = −0.025 and β = −0.074, respectively. These results are significant at the 1% level, and indicated by the small standard errors of 0.005 and 0.010. If other factors remain constant, this suggests that a one-unit increase in the release of carbon emissions would decrease TOQ and MBV by 0.025 and 0.074 units, respectively. Thus, carbon emissions have a negative influence on firm value among non-financial firms listed in GCC stock markets. As shown in models 1 and 3, DTE as a proxy of financial structure affects TOQ negatively and significantly with a coefficient of −0.020, but positively affects MBV, with a beta coefficient of 0.116. This means that a one-unit increase in DTE would decrease TOQ by around 0.020 units and increase MBV by 0.116. Moreover, STD and LTD have a positive and significant influence on firm value measures (TOQ), with coefficients of 0.640 and 0.690, respectively. Both variables also positively and significantly affect MBV with beta coefficients of 3.725 and 3.599, respectively. These results are statistically significant at the 1% level. If other factors do not change, this suggests that a one-unit increase in STD and LTD would increase TOQ by around 0.640 and 0.690 units, respectively, and increase MBV by 3.725 and 3.599, respectively. DTM also has a significant but negative impact on TOQ and MBV, with coefficients of −1.164 and −3.433, respectively, suggesting that a one-unit increase in DTM would decrease both TOQ and MBV by around 1.164 and 3.433 units each.

Further, FMS as a control variable registered values of −0.238 and −0.344 for TOQ and MBV, respectively, indicating that elevated FMS relates to diminished firm value. The GPM coefficients are 0.098 and 0.092 on TOQ and MBV, implying a favorable connection between GPM, TOQ and MBV. These links are statistically significant at the 1% and 5% levels, respectively. TAN has a coefficient of 0.005 and 0.044 on TOQ and MBV, respectively, but the results are statistically insignificant, elucidating that TAN does not influence either TOQ or MBV. CEX with a 0.005 coefficient affects TOQ insignificantly, while it affects MBV significantly and positively with a coefficient of 0.031. Finally, macroeconomic variables, such as GDP, have effects of 0.003 and 0.005 on both TOQ and MBV, respectively, signifying a beneficial impact on firm value metrics.

4.7. Moderating Effect of Financial Structure

The findings of regression analyses that examined the relationship between carbon emissions and firm value while considering the moderating effect of financial structure are shown in Table 8. As exhibited in Table 8, for models 2 and 4, the coefficients of CEN on TOQ (β = −0.061 and β = −0.126, respectively) illustrate that the carbon emissions have a detrimental impact on firm value, and these results are statistically significant at a 1% level. These results prove that the negative association between carbon emissions and firm value persists even after controlling for the moderating effect of financial structure. The consequences of financial structural measures have the same effects on firm value even after the moderating influence, as exhibited in models 2 and 4 in Table 8. The value of a company is heavily influenced by its financial structure. This means that, while market leverage has an adverse effect on firm value, both short- and long-term debt have a beneficial influence on TOQ and MBV.

Moreover, the inclusion of the interaction term CEN*DTE has a coefficient of 0.009 on TOQ and 0.008 on MBV, with a low standard error, suggesting the potential moderating effect of DTE on the relationship between carbon emissions and firm value. In addition, the interaction term CEN*STD indicates that the effect of carbon emissions on firm value may differ depending on the level of financial structure. The presence of STD has a negative effect of −0.290 on TOQ and −0.733 on MBV. Likewise, the moderating effect of CEN*LTD has a coefficient of −0.246 on TOQ and −0.543 on MBV. This demonstrates that LTD significantly moderates the association between carbon emissions and firm value. The interaction term between CEN and DTM has a statistically significant and positive influence on both indicators of firm value (TOQ and MBV), with coefficients of 0.217 and 0.486, respectively. This indicates that DTM has a crucial impact on the nexus between carbon emissions and firm value. Together, these findings suggest that CMN has an adverse correlation with TOQ and MBV among listed firms in the GCC stock markets. This detrimental correlation is significantly moderated by the level of financial structure. According to this, an optimum financial structure (mixture of debt and equity) can diminish the negative impacts of carbon emissions on firm value. However, a higher reliance on excessive debt financing may worsen the above association.

4.8. Robustness Check

Generalized Method of Moments (GMM) is applied as a robustness check in this investigation. GMM guarantees that the residuals and the lag of explained (dependent) factors are consistently assumed to be correlated. This approach reduces bias from missing variables or measurement errors, thereby improving the results’ validity. The findings of GMM are presented in

Table 9. GMM regression results.

Variables	Model 1 (TOQ)		Model 2 (TOQ)		Model 3 (MBV)		Model 4 (MBV)
	Coef.	t-Stat.	Coef.	t-Stat.	Coef.	t-Stat.	Coef.	t-Stat.
CEN	−0.013 *** (0.003)	−4.436	−0.029 *** (0.007)	−3.906	−0.037 *** (0.009)	−4.103	−0.064 *** (0.019)	−3.363
DTE	−0.005 (0.004)	−1.239	−0.010 * (0.006)	−1.734	0.073 ** (0.030)	2.396	0.079 ** (0.031)	2.545
STD	0.393 *** (0.055)	7.152	0.409 *** (0.066)	6.157	1.977 *** (0.226)	8.752	1.886 *** (0.232)	8.143
LTD	0.468 *** (0.056)	8.429	0.502 *** (0.067)	7.543	2.065 *** (0.243)	8.495	2.024 *** (0.244)	8.304
DTM	−0.599 *** (0.045)	−13.175	−0.584 *** (0.045)	−12.909	−1.829 *** (0.139)	−13.153	−1.807 *** (0.143)	−12.646
CEN $\times$ DTE			0.005 ** (0.002)	2.219			0.006 (0.013)	0.460
CEN $\times$ STD			−0.184 *** (0.037)	−5.028			−0.539 *** (0.143)	−3.775
CEN $\times$ LTD			−0.140 *** (0.029)	−4.855			−0.314 *** (0.106)	−2.965
CEN $\times$ DTM			0.111 *** (0.019)	5.867			0.286 *** (0.060)	4.724
FSM	−0.095 *** (0.015)	−6.213	−0.097 *** (0.015)	−6.394	−0.218 *** (0.043)	−5.111	−0.206 *** (0.045)	−4.539
GPM	0.018 * (0.011)	1.700	0.018 (0.012)	1.526	−0.002 (0.029)	−0.078	−0.029 (0.026)	−1.132
TAN	0.053 (0.043)	1.236	0.022 (0.045)	0.486	−0.009 (0.105)	−0.084	−0.008 (0.106)	−0.079
CEX	−0.011 * (0.006)	−1.814	−0.013 ** (0.006)	−2.350	0.012 (0.008)	1.410	0.006 (0.009)	0.692
GDP	0.001 (0.001)	−0.766	−0.001 * (0.000)	−1.697	−0.001 (0.002)	−0.355	0.001 (0.001)	−0.333
C	1.689 *** (0.204)	8.298	1.726 *** (0.202)	8.527	3.796 *** (0.569)	6.672	3.631 *** (0.599)	6.052
TQ_lag1	0.564 *** (0.032)	17.602
TQ_lag1			0.559 *** (0.031)	18.159
MBV_lag1					0.503 *** (0.032)	15.820
MBV_lag1							0.502 *** (0.031)	15.967
$R_2$	0.950		0.951		0.944		0.944
Adj. $R_2$	0.945		0.945		0.937		0.937
Prob.	0.000		0.000		0.000		0.000

Note(s): *** p < 1%; ** p < 5%; * p < 10%.

. The outcomes of the robustness tests reinforce the previous argument that CO₂ emissions negatively affect firm value, and financial structure significantly moderates the aforementioned correlation.

5. Discussion

Through the GLS regression analysis, a detrimental relationship has been found between carbon emissions and corporate value among non-financial enterprises registered in the GCC stock exchanges. In particular, there is a negative correlation with statistical significance at a 1% level between the CMN and the TOQ and MBV. This means that carbon emissions could place companies at risk for more costly regulations, such as taxes on carbon, which might eventually affect their bottom line. Further, reputational damage could also be observed as a result of releasing higher CO₂ emissions. Hence, investors’ confidence might be eroded and firm value diminished. Based on the above explanations, we can confirm the first hypothesis, that carbon emissions are a significant determinant of firm value, which is consistent with stakeholder theory claims that firms that reduce emissions have more ability to generate more profit. The theory of signaling also proves the above results—that higher emissions of carbon are associated with less concern by shareholders, which brings about regulatory issues and damages corporate value. These findings are also consistent with the arguments of Adamolekun and Kyiu (2024), Ahmadi and Bouri (2017), Benkraiem et al. (2024), Friske et al. (2023), Houqe et al. (2022), Le and Nguyen-Phung (2024) and Matsumura et al. (2014).

Moreover, the research also investigates the relationship between financial choice and the value of the firm. The results illustrate the complex association between the structure of firms’ capital and its value. DTE, as a measure of financial structure, significantly and negatively affects TOQ, suggesting that elevated debt-to-equity ratios could lower TOQ. DTE, on the other hand, has a favorable connection with MBV, suggesting that, under some circumstances, greater leverage may indicate optimism about the company’s profitability and future expansion. The positive and significant link of STD and LTD with TOQ and MBV implies that even though the financial choice may have a different output, balanced financing can be seen as a crucial factor in assessing the value of the company. This means that capital financing might be an effective strategy for improving corporate value. Together, these findings illustrate that the value of a company can be increased by virtue of an optimum capital structure, which is defined by an appropriate balance of debt to equity, yet an over-reliance on funding from debt or substantial market leverage might have negative consequences. These arguments prove the next hypothesis that financial leverage is a significant determinant of firm value, and contribute to the extension of agency theory, indicating that debt may be used as a control mechanism by restricting managers from spending funds on projects that could not be profitable. However, a large amount of borrowing can also result in risks that lower the value of a firm, as argued by agency theory. Hence, a mixture of debt and equity can be seen as a satisfactory solution to increase firm value. The results are consistent with the findings of Ahmed et al. (2024a), Barburski and Hołda (2023), Dawar (2014), Khan et al. (2023), Muhammed et al. (2024), Sdiq and Abdullah (2022), and Sutomo et al. (2020), and oppose the arguments of Abdullah and Tursoy (2021), Adair and Adaskou (2015), Jouida (2018) and Ngatno et al. (2021).

Furthermore, the present investigation provides a significant addition by analyzing the moderating role of financial choice on the association between emissions of carbon and the value of the firm. As indicated by the presence of interaction variables in the regression analyses, the influence of CO₂ emissions on business value is strongly moderated by financial selection. The results indicate that the appropriate level of debt financing might potentially lessen the detrimental impact of carbon emissions on the value of a business. This could be achieved by providing firms with the necessary investment in implementing sustainable development goals. Thus, the adverse consequences of emissions may be mitigated by adopting an optimum level of financial structure. However, excessive dependence on debt funding, especially short-term loans, could generate severe consequences. From this perspective, it is recommended that companies adopt a balanced capital structure to finance their sustainable projects. Long-term borrowing should be prioritized beyond short-term financing to reduce financial risk and promote ecological sustainability. Furthermore, using green finance solutions increases investor confidence and business value by demonstrating a commitment to lowering carbon emissions. Long-term credibility and robustness are ensured when financial policies are in line with sustainability objectives.

These findings help in broadening the concepts of signaling and stakeholders in developing economies such as the GCC. Stakeholder theory argues that having a correct capital structure (debt and equity) may offer the sufficient funds required to cover sustainable projects. This can strengthen the business’s credibility among stakeholders and enhance its value. The concept of signaling also states that companies can reduce the detrimental effect of carbon by sending signals to investors and creditors through making investments to reduce emissions and managing their funding restrictions. These findings are also corroborated by the studies of Coulson and Monks (1999), Kalash (2021), and Shu et al. (2023), who concluded that firms with prominent levels of carbon emissions cannot obtain enough funds from external sources, and increasing debt levels beyond the optimum level may increase emissions of carbon.

6. Conclusions

The main objective of this investigation is to offer empirical evidence about the effect of financial structure on the nexus between carbon emissions and firm value among listed non-financial firms in GCC markets. The research examined a large dataset containing 195 firms over eleven years (2010–2020). The official World Bank website and Thomson Reuters were the sources of the data. Although the GLS method with cross-section weight has been considered a valid and suitable method in this study, the GMM estimator is also employed for robustness check. Firm value is a dependent (explained) variable and is proxied by TOQ and MBV. Carbon emission, as indicated by CMN, is an independent (explanatory) variable. The financial structure measured through DTE, STD, LTD, and DTM plays a role as a moderating and independent factor.

The results show that there is a strong and adverse correlation between corporate value and emission levels of carbon, claiming that increased emissions translate into a lower value for the firm. The analysis also emphasizes how intricately capital financing plays an essential role in this connection. Despite the adverse consequences of emissions that may be mitigated by adopting an optimum level of financial structure, excessive dependence on debt funding, especially short-term loans, could generate severe consequences. Hence, following environmental practices while increasing firm value necessitates balanced capital financing (mixture of debt and equity). The aforementioned claims are supported by both stakeholder and signaling theories. Stakeholder theory argues that having an optimum capital structure may offer the sufficient funds required to cover sustainable projects. This can strengthen the business’s credibility among stakeholders and enhance its value. Signaling theory also points out that firms can reduce the negative impact of carbon by sending signals to investors and creditors.

The study significantly contributes to the current literature about the consequences of carbon emissions on business value and the moderating effects of financial strategy on the above link. It begins by filling a knowledge gap on the effects of carbon emissions in transition economies, such as GCC markets. The findings of this study provide an idea about the difficulties and possibilities that face companies in GCC markets who are willing to adopt sustainable development. Firms that aim at long-term growth should consciously select funding options. The results also highlight how crucial it is for GCC firms to analyze their financial and environmental policies to increase their value in the marketplace. Policymakers could also use the knowledge gained from our investigation to create legislation that can promote environmental practices in developing nations. Our study emphasizes the importance of applying the optimal capital structure, and we implore legislators to consider this as a crucial component when financing sustainable projects and monitoring environmental policies. The results show how an optimal capital structure and sustainability enhance firm value, which is consistent with Vision 2030. Companies can comply with legislative objectives to draw investment and promote sustainable development, while regulators might establish opportunities for such activities.

Regarding the research limitations, first, since the data at the company level are not accessible, this study collected data on CO₂ emissions at the country level as a proxy of carbon emissions. Future studies might improve validity by including data about emissions at the firm level. Secondly, it must be highlighted that the outcomes produced in this study are limited to sectors other than finance, and as financial organizations are subject to laws, one could reach different views regarding the association explored in this study. Thirdly, a further understanding of the long-term impacts of carbon emissions might be achieved by investigating the connection between carbon footprint, firm value, and financial structure over longer periods. Future study directions could also involve expanding this investigation to other emerging economies, including the UAE, to achieve cross-country analysis.