Sachini Wewalwala Hewage The Impact of Sustainability Reporting on Firm Performance Empirical Evidence from Finland Vaasa 2025 School of Accounting and Finance Master’s thesis in Finance 2 UNIVERSITY OF VAASA School of Accounting and Finance Author: Sachini Wewalwala Hewage Title of the thesis: The impact of Sustainability Reporting on Firm Performance: Empirical Evidence from Finland Degree: Master of Science in Economics and Business Administration Discipline: Master's Degree Programme in Finance Supervisor: Timo Rothovius Year: 2025 Pages: 101 ABSTRACT: Corporate Social Responsibility (CSR) disclosure has emerged as a crucial component of modern corporate communication, due to the growing expectations that firms exhibit accountability in environmental, social, and governance (ESG) domains. However, the academic literature regarding the relationship between CSR disclosure and firm performance presents mixed and often contradictory evidence, showing positive, negative, or statistically insignificant effects. This study investigates this ongoing discussion by examining publicly listed companies in Finland, a renowned sustainability leader in northern region and across the globe, that ranked first out of 193 nations according to Sustainable Development Goals Index in 20024 with a score of 86.35. The study explores whether ESG transparency results in better operational, financial and market outcome, drawing on theories of stakeholder, legitimacy, and agency. The study employs fixed- effects regression models to investigate the effects of ESG disclosure scores, obtained from Bloomberg, on Return on Equity (ROE), Return on Assets (ROA), Tobin's Q, and annual stock returns using a panel dataset of 88 firms over a six- years period (2018–2023). The results show that the relationship between ESG disclosure and firm performance is largely statistically insignificant, with some dimensions even revealing a negative relationship. These findings suggest that while ESG reporting may serve legitimacy purposes in high-standard CSR environments like Finland, it does not necessarily yield measurable financial advantages. That said, the robustness test suggests that the link between ESG disclosures and firm performance can differ depending on the industry and timing of the ESG impacts. These results highlight the importance of considering the broader context when analysing the impact of ESG disclosures. This opens opportunities for future research to explore how ESG disclosures influences companies differently across sectors and over extended time periods. By focusing on Finnish firms, this study provides country specific insights to the growing body of CSR literature and highlight the need for further investigation into the qualitative and strategic dimensions of ESG practices within organizations. KEYWORDS: Corporate Social Responsibility (CSR), ESG disclosure, sustainability reporting, firm performance, operational performance, financial performance, market performance, Finland 3 Contents 1 Introduction 7 1.1 Background of the study 8 1.2 Purpose of the study 10 1.3 Structure of the study 11 2 Theoretical Framework 12 2.1 Corporate Social Responsibility (CSR) 12 2.1.1 Definition and Evolution of CSR 12 2.1.2 CSR Reporting 15 2.1.3 CSR in the Nordic Region and Finland 17 2.2 Key theories in CSR 19 2.2.1 Legitimacy Theory 19 2.2.2 Stakeholder Theory 21 2.2.3 Agency Theory 23 3 Literature review 25 3.1 CSR disclosure and firm performance 25 3.2 Positive relationship between CSR reporting and firm performance 26 3.3 Negative relationship between CSR reporting and firm performance 28 3.4 Neutral relationship between CSR reporting and firm performance 29 3.5 Research Hypothesis 30 4 Data and Methodology 31 4.1 Data collection 31 4.1.1 Sample selection 32 4.1.2 Sample distribution 33 4.2 Variables of the study 35 4.2.1 Dependent variables 36 4.2.2 Independent variables 39 4.2.3 Control variables 41 4 4.3 Data cleaning and preprocessing 45 4.3.1 Managing outliers 45 4.3.2 Handling missing values 47 4.4 Correlation Analysis 48 4.5 Descriptive statistics 52 4.6 Regression analysis and model 54 4.7 Regression model selection and diagnostic tests 57 5 Empirical Results 58 5.1 Regression results for hypothesis 1 58 5.1.1 ROA and ESG disclosure 60 5.1.2 ROE and ESG disclosure 61 5.1.3 Tobin’s Q and ESG disclosure 62 5.1.4 Stock returns and ESG disclosure 63 5.2 Regression results for hypothesis 2 63 5.2.1 Impact on ROA 64 5.2.2 Impact on ROE 64 5.2.3 Impact on Tobin’s Q 65 5.2.4 Impact on stock returns 65 5.3 Robustness Test 67 5.3.1 Lagged ESG variables 67 5.3.2 Industry Interaction Effects 71 5.4 Discussion and findings 73 6 Conclusion 78 6.1 Summary of the study 78 6.2 Practical implications 79 6.3 Contribution, limitations and suggestions 80 References 82 Appendices 96 Appendix 1. Sample of the study 96 5 Appendix 2. Variance Inflation Factors (VIF) 99 Appendix 3. Breusch- Pagan Test 100 Appendix 4. Hausman Test 101 6 Tables Table 1. Industry distribution 34 Table 2. ESG disclosure categorization criteria 34 Table 3. Summary of variable measurements 35 Table 4. Comparison of original and winsorized dependent variables 46 Table 5. Summary of missing data 48 Table 6. Correlation Analysis 51 Table 7. Descriptive Statistics 53 Table 8. Regression results for H1 59 Table 9. Regression results for H2 66 Table 10. Lagged ESG and firm performance 69 Table 11. Interaction effects of ESG on performance across industries 72 Abbreviations CSR Corporate Social Responsibility ESG Environmental, Social and Governance SDG Sustainability Development Goals UN United Nations ROA Return on Assets ROE Return in Equity TQ Tobin’s Q RTS Return to Stocks / stock returns TCFD Task Force on Climate-Related Financial Disclsoures SASB Sustainability Accounting Standards Board GRI Global Reporting Initiative TBL Triple Bottom Line ICB Industry Classification Benchmark VIF Variance Inflation Factor 7 1 Introduction Corporate Social Responsibility (CSR) disclosure has become a central aspect of modern corporate communication, as firms increasingly recognize the importance of sustainability and social responsibility. Academic literature combines sustainability and corporate social responsibility (CSR) into a single concept (Dhaliwal et al., 2011). A company's duty to measure and handle social and environmental results from business operations is commonly known as CSR. The company's responsibility to implement CSR must become part of its core business strategy and operations by maintaining ethical conduct together with environmental stewardship and human rights protection and consumer considerations following legal requirements (European Commission, n.d.). Companies increasingly publish sustainability reports in order to showcase their CSR efforts to the stakeholders (Christensen et al., 2021). This form of reporting has become an almost widely accepted global norm with the majority corporations now disclosing information on their sustainability efforts (KPMG, 2022). Depending on the sustainability reporting measure, financial performance measure, sample composition, time-period, and control variables, the results of the relationship between CSR disclosure and firm performance, provide mixed, inconsistent and often contradictory evidence ranging from positive, to negative, to statistically insignificant according to the academic literature. However, majority of evidence from literature suggest positive relationships while explaining that comprehensive CSR reporting enhances a firm’s financial performance by enhancing market competitiveness, attracting ethical investors, improving consumer loyalty (Cahan et al., 2016; De Klerk et al., 2015; Tsang et al., 2022). Meanwhile, other researchers like Chen et al. (2018), Grewal et al. (2017), Lioui and Sharma (2012), have found a negative relationship and studies done by Atan et al. (2018), Guidry and Patten (2010) and Phan et al. (2020), have not discovered any meaningful connection about this relationship. Furthermore, several studies conducted by Jones (2005), Mǎnescu (2011), Omar and Zallom (2016) produced contradictory results about this relationship. This conflicting finding encourages more 8 research into the potential effects of CSR disclosures on firm performance across various market environments, especially in Finland. 1.1 Background of the study Nowadays environmental, social and governance (ESG) issues have become central to how business operate. The companies’ expectations have changed towards demonstrating their commitment to sustainability in tangible ways. Due to this change companies are increasingly adopting CSR frameworks, guidelines, and reporting standards to show they understand the importance of CSR and sustainability in today’s business world (Carroll, 1999; Elkington, 1997). Companies use CSR as an essential strategic element to define their corporate reputation and business identity in the market. In today’s modern world, companies need to show transparency in their environmental and social operations in order to follow laws and build a better reputation in the market (Freeman, 1984). CSR reporting throughout history has been considered as an optional voluntary procedure which helps businesses showcase their values. Organizations would disclose data about their ESG efforts to prove their social engagement alongside ethical conduct. However, society’s general understanding about CSR has changed overtime. The implementation of legal CSR disclosure requirements has become standard practice in many nations across the world. Research examining the effects of this mandatory CSR disclosures on firm performance becomes more important after this transition (Haji et al., 2023). Business operations face significant changes because companies must develop sustainable approaches to balance profitability together with social benefits (Aggarwal, 2013). When CSR gained prominence, it gained status as a vital business management perspective which required organizations to exceed regulatory requirements by actively 9 discussing and sharing their CSR initiatives to build competitive advantages (Du et al., 2017). This is demonstrated by the Non-Financial Reporting Directive (2014/95/EU), that the European Commission enacted. This directive applies to publicly listed companies that have an average of at least 500 employees during the financial year. These companies must report their CSR activities as per the mandatory requirement established since 2017 (European Commission, 2014). Lueg and Peshva (2021) together with Midttun et al (2015) and Strand et al (2015), confirm through research that Nordic counties have established themselves as leaders in sustainability. These countries performing well in various sustainability rankings, and corporate social responsibility stands as a vital strategic component that shapes their business approaches (Lueg & Pesheva, 2021). Although scholarly research in Nordic context remains scarce which proves that additional research is required according to Khatri & Kjærland (2023). Finland is often seen as one of the Nordic countries that puts a strong focus on sustainability. More and more Finnish companies are now making corporate social responsibility (CSR) a key part of how they do business. This shift is largely driven by the fact that society, both customers and the general public, expect companies to take responsibility for their actions, especially when it comes to the ESG issues. Although interest in CSR is growing across the country, there still isn’t enough literature that shows how CSR activities affect business performance in Finland. This study hopes to fill that gap by looking into the link between what companies report about their CSR efforts and how well they perform. In Finland, doing business responsibly isn’t just encouraged, it’s become something that people expect as part of good business practice. 10 1.2 Purpose of the study The purpose of this study to find out whether sustainability disclosures have a significant impact on the performance of Finnish companies. Finland is widely recognised for its strong commitment to sustainability and currently Finland ranked first according to the Sustainable Development Report 2024 followed by Sweden and Denmark (Sachs et al., 2024). In light of this background, Finnish firms offer a unique setting to explore whether this improved ESG disclosure translate into better outcomes for businesses. This study looks at publicly listed companies in Finland over the years 2018 to 2023 and uses Bloomberg ESG disclosure scores to measure sustainability disclosure. It then examines how these ESG disclosure scores impact on firm performance assessed across three dimensions; • Operational Performance: determined by Return on Assets (ROA). • Financial Performance: determined by Return on Equity (ROE). • Market Performance: determined by Tobin’s Q and stock returns. The main research question of the study is; - Do CSR disclosures have a positive effect on firm performance in the Finnish context? By answering this question, this study aims to provide a clearer picture of whether ESG disclosure truly benefits firms in a country where sustainability is already a high priority. The goal of this study to offer valuable and up-to-date insights not only for researchers but also for businesses, investors, and policymakers who are navigating the growing expectations around responsible business conduct. 11 1.3 Structure of the study The structure of the remaining chapters in this thesis uses the following order. Chapter two provide the theoretical framework that support the empirical analysis. It begins by introducing the concept of CSR, tracing its definitions and development over time, and then explores CSR reporting with a particular focus on context of Nordic region and Finland. This chapter also present three theoretical perspectives that help explain CSR behaviour of firms, namely legitimacy, stakeholder, and agency theory. These theories give an in depth understanding into why companies engage in CSR reporting. Chapter three continue with an in-depth review of existing literature and introduces research hypotheses. The literature is categorized into three groups, studies showing a positive link between CSR and firm performance, studies indicating a negative relationship and those that report no significant connection between the variables under investigation. The methodology and data are presented in the fourth chapter. It first describes data collection and sample selection procedure, variables of the study and data cleaning procedure. This chapter also includes a presentation of descriptive statistics and correlation analysis to provide an overview of the dataset. Subsequently, the regression models employed in the study are explained along with a discussion of model selection and the diagnostic tests conducted to ensure reliability. Chapter five presents the empirical findings including the results of robustness checks. The findings of the study are then analysed and discussed in depth. The sixth chapter is the conclusion which include summarization of the whole thesis, practical implications, contribution and limitations, and ideas for future research. 12 2 Theoretical Framework The study’s theoretical framework is divided into two sections, corporate social responsibility (CSR) and its foundational theories. The first section explores the concept of CSR, its definitions and evolution, CSR reporting, and the application of CSR within the Nordic region and Finland. The second section outlines key CSR related theories commonly utilized in related studies. Altogether, this chapter provides a theoretical foundation and introduces key concepts important for the empirical analysis of the study. 2.1 Corporate Social Responsibility (CSR) There is no universally accepted definition for Corporate social responsibility (CSR) as its meaning can vary depending on the context. This concept has emerged in the 1950s. Over time, many definitions have been proposed, and debates have arisen about which one is the most accurate. Companies began recognizing the public's increasing concern with social issues, which were previously not seen as significant. External stakeholders now expect more from businesses, particularly regarding transparency. Corporations are increasingly held accountable for being responsible citizens and acting ethically (Porter & Kramer, 2006). 2.1.1 Definition and Evolution of CSR Over the decades, various definitions of Corporate Social Responsibility (CSR) have been proposed. Scholars using different terms such as ‘social responsibility’ to describe CSR. (Carroll, 1999). One of the earliest definitions was provided by Howard Bowen In 1953, who describe it as the obligation of business to align their policies, decisions and actions with societal expectations and values (Bowen, 1953). Another early definition by Keith Davis viewed social responsibility as actions and decisions made by corporations for reasons beyond direct economic and technical considerations (Davis, 1960). Over time, 13 researchers began acknowledging the broader range of responsibilities that businesses hold beyond their economic duties (Carroll, 1999). The understanding of CSR has progressively developed over the time, shaped by various academic and practical contributions (Carroll, 1999; Latapí Agudelo et al., 2019; Lee, 2008). According to Carrol (1999), the concept of CSR began to take shape in the late 1950s, though earlier scholarly works had presented it from a different perspective. Bowen’s 1953 study is frequently considered as the first theoretical framework for understanding the interaction between businesses and society (Carroll, 1999; Lee, 2008). He argued that corporate actions and decisions have significant societal impact, and while CSR doesn’t solve all societal issues, it is a key step toward improving welfare (Bowen, 1953). Around the same time, new laws regarding business conduct, employee, and consumer protection emerged in the U.S., and the consumer rights movement gained momentum, which introduced new complexities for businesses to navigate (Lee, 2008). During the 1960s, the body of CSR empirical evidence grew considerably during the 1960s, with many scholars seeking to define the concept (Carroll, 1999). In 1960, the scholars introduced two distinct categories of responsibilities within CSR, namely, socio- economic, which focuses on public welfare like employment and competition, and socio- human, which concerns values such as cooperation and morale (Davis, 1960). Carroll (1999) highlights that Davis’s view on the link between CSR and business power became widely accepted in the late 1970s and 1980s. The committee for Economic Development further expanded the CSR concept, addressing its controversies and arguing that socially responsible behaviour benefits shareholders in 1970 (Lee, 2008; Wallich & McGowan, 1970). throughout the 1970s, definitions of CSR became more specific, with scholars like Harold Johnson emphasizing the importance of considering groups beyond stockholders, such as employees and the local community (Johnson, 1971). 14 While new definitions of Corporate Social Responsibility (CSR) slowed in the 1980s, other ideas, such as Corporate Social Performance (CSP), gained popularity (Carroll, 1999). Carrol (1999) explains that CSR reflects a broader set of responsibilities that business holds, whereas CSP focuses more on the actual outcomes of those responsibilities. Stakeholder theory, corporate social responsiveness and business ethics were also discussed during this time (Carroll, 1999). While new interpretations of CSR continued to emerge, significance global events played a major role in increasing CSR awareness in the 1990s. A good example for that is the Summit of United Nations on the Environment and Development. Throughout this decade, corporate social responsibility (CSR) became an essential part of organizational reputation and stakeholder management, making the foundation for other related ideas like stakeholder theory and business ethics (Carroll, 1999; Lee, 2008). The introduction of new organizational frameworks and strategies made corporate social responsibility (CSR) recognized internationally throughout the 2000s. According to Latapí Agudelo et al. (2011) the European Commission released a corporate social responsibility report in 2001 alongside international corporate certification development including ISO 26000. According to Latapí Agudelo et al. (2011) CSR transformed into a vital strategic management instrument. The new concepts of shared value creation emerged, further developing the CSR concept during the subsequent decade. The social and economic value generation framework Named Creating Shared Value (CSV) was initially proposed by Porter and Kramer (2011). Unlike CSR, CSV directly enhance company profitability and competitiveness, while improving corporate reputation (Porter & Kramer, 2011). Werther and Chandler (2014) also emphasized strategic CSR, encouraging companies to find market-based solutions to societal issues in order to create social and economic value. According to Cupertino et al. (2022) CSR has become an essential component for businesses while various organizations present different definitions of this concept. For example, The United Nations Industrial Development Organization (UNIDO) defines CSR 15 as the practice of incorporating social and environmental factors into business operations (UNIDO, n.ds.). But according to the European Commission, it is the duty of corporations to consider their effects on society, including ethical and environmental issues (European Commission, n.d.). These fundamental ideas have changed over time while the core principle of CSR remained consistent over time. 2.1.2 CSR Reporting In today’s business environment, being transparent about sustainability and social responsibility become an important aspect of corporate reporting. The practice of disclosing CSR initiatives as a report has become popular throughout the globe, as many businesses share information about their sustainability efforts and results (KPMG, 2020, 2022). The core purpose of CSR reporting involves collecting, analysing and presenting data that describes both measurable and subjective evidence about firm’s social, ethical, and environmental responsibilities (Christensen et al., 2021). The purpose of CSR disclosures includes both strategic and communicative elements. According to Crane and Glozer (2016), the strategic purpose of CSR disclosures classified into influence both behaviours and attitudes, legitimacy establishment, reputational development, stakeholder interaction as well as corporate identity creation. These disclosures included in the annual report or in a separate report that contain information about CSR. The content of these reports often changes, but it commonly includes sections about community relations, supply chain management, human rights combined with workplace safety, CSR tactics, risk mitigation, ethical conduct and environmental initiatives (Vartiak, 2016). Companies during recent years have prioritized sharing information about climate risks, while working to reduce their carbon emissions and contributing to the Sustainable Development Goals (KPMG, 2022). CSR reporting for companies involves the environmental, social, and governance aspects known as ESG. Fair labour practices, workplace diversity, and consumers safety belong 16 to ‘social’ component of ESG assessment. The ‘environmental’ component looks at how companies evaluate their natural impact, such as resource utilization and waste management (Clément et al., 2023). The ‘governance’ pillar of ESG investigates company leadership structure, fight against corruption, and its ethical standards (Singh & Chakraborty, 2021). As noted by Clément et al. (2023) and Han et al. (2016), ESG scores are often used to evaluate how well a company is performing in these ESG-related domains. Although CSR and ESG terms are sometimes treated identically, many experts agree that ESG provides a wider perspective by prioritizing governance elements (Abdul Rahman & Alsayegh, 2021). Many corporations adopted uniform standards to manage their reporting because stakeholders expected higher levels of corporate social responsibility. Three main frameworks exist for reporting which include the Task Force on Climate-related Financial Disclosures (TCFD) as well as Sustainability Accounting Standards Board (SASB) and Global Reporting Initiative (GRI). However, GRI is recognized worldwide as the most common standard (KPMG, 2022). Through its standardized approach organizations can assess and report their environmental social and economic impacts through GRI platform. GRI standards consist of ‘Topic Standards’ which focus on workplace safety, waste management and taxes, ‘Sector Standards’ particular to industry-specific reporting standards along with ‘Universal Standards’ that apply to all organizations (GRI, n.d.). CSR reporting is optional in many jurisdictions, just as the adoption of such frameworks is. To uphold their ethical reputation, satisfy stakeholders, or show social responsibility, businesses frequently decide to voluntarily reveal CSR information (Dobbs & van Staden, 2016; Mukherjee & Nuñez, 2019; Thorne et al., 2014). However, an increasing number of governments have started enacting mandatory CSR disclosure laws; this is seen in Asia and the European Union (European Commission, 2021; Mukherjee et al., 2018). As KPMG (2022) notes that there has been a noticeable shift from voluntary to mandatory 17 CSR reporting, therefore more business will soon be required to report their sustainability performances by the law. Mandatory and voluntary CSR reporting have impacted differently to the Companies. For example, mandatory disclosures usually lead to higher short-term operating expenses since they force businesses to improve their CSR initiatives to meet regulatory requirements (Cupertino et al., 2022). Frequently, the driving forces behind the push are social pressures, stakeholder influence, and industry comparisons (Christensen et al., 2021). However, mandatory reporting also enhances the transparency and dependability of CSR disclosures by standardizing what needs to be reported. Organizations that adopt voluntary reporting had freedom to publish only positive information that builds reputational value, but it creates concerns regarding their trustworthiness (Gatti et al., 2019). Mixing both mandatory requirements and their own voluntary efforts in reporting CSR, it encourages more honest and responsible behaviour (Christensen et al., 2021; Gatti et al., 2019). 2.1.3 CSR in the Nordic Region and Finland Nordic countries are often seen as leaders when it comes to corporate social responsibility (CSR) and sustainability according to Midttun et al. (2015). Throughout different periods in history these countries have worked together to improve social wellbeing, protect the environment, and strengthen their economies through various partnerships (Bird, 2017). Scientific evidence show that people in the Nordic region are generally willing to spending more on environmental protection and are supportive of eco-friendly efforts (Reyes, 2021). Many companies in this region have made sustainability a part of their everyday business activities (Lueg & Pesheva, 2021). Nordic governments also play an important role in global CSR efforts. They support international programs, such as those led by the United Nations, and promote sustainable practices through national policies (Lueg & Pesheva, 2021; Midttun et al., 2015). 18 International sustainability rankings show Finland as a leader with its first-place position from 193 countries scoring 86.35 on the 2024 Sustainable Development Goals (SDG) Index that assesses countries on their progress toward SDGs (Sachs et al., 2024). This top ranking highlights the country’s serious commitment to protecting the environment and promoting sustainable development. Sustainability reporting is also widely practiced in Finland. KPMG’s 2024 report shows that 94% of Finnish companies included in the survey published sustainability reports, number that has stayed steady since 2022 (KPMG, 2022, 2024). This thesis investigates the impact of ESG disclosures on corporate performance, with a particular focus on Finland. The Nordic welfare model is distinguished by the government's proactive engagement in social and environmental issues. Several CSR-related policies have been developed as a result of this over time (Midttun et al., 2015). Finland's public discourse and CSR- related policy implementation, however, have developed more slowly than those of its Nordic neighbours. However, the legislative measures like the Finnish Accounting Act, which mandate that certain companies disclose non-financial information as part of their annual reporting obligations were introduced by the Finland government (Khatri & Kjærland, 2023; Midttun et al., 2015). Even though sustainability performance is generally high, a 2023 report by Finnish Business and Society notes that although many Finnish companies have a strong foundation for CSR, their efforts can still be improved (Finnish Business & Society, 2023). However, in areas such as CSR integration and sustainability reporting, Finland continues to perform well globally (KPMG, 2022; Sachs et al., 2024). Currently, CSR reporting practices in Finland are greatly influenced by EU regulations. The Nordic nations work together to position themselves as the most sustainable region in the world by 2030, in addition to fulfilling regulatory requirements. Achieving carbon neutrality, advancing the ideas of the circular economy, encouraging innovation, and 19 improving social inclusion and equality are just a few of the objectives included in this vision (Nordic Council of Minister, 2020). Sustainability is often seen as a way to support both economic growth and environmental responsibility. Aagaard et al. (2022) point out that the Nordic countries have several strengths in this area. Thanks to their strong start-up environment and advanced technical expertise, these nations are well-positioned to build and support sustainable systems. Moreover, this region maintains a strong position to develop economic value through job creation because it effectively utilizes its resources to align with established policy goals that support worldwide sustainability transitions. 2.2 Key theories in CSR In academic literature, corporate social responsibility (CSR) has been studied from a number of theoretical perspectives (Frynas & Yamahaki, 2016). Research studies have utilized various theoretical frameworks to explain CSR practice mechanisms including political economy theory, resource dependency theory, agency theory, legitimacy theory, stakeholder theory and resource-based view. These theories are frequently cited by scholars like Frynas and Yamahaki (2016) and Mehedi and Jalaluding (2020). This research will focus primarily on stakeholder theory and legitimacy theory together with agency theory since these theories offer significant insights about CSR reporting effects on firm performance. This paper analyzes each of these theories in separate sections. 2.2.1 Legitimacy Theory Voluntary disclosure practices of social and environmental matters are commonly studied using Legitimacy theory within the CSR field. ‘Social contract theory’ stands as the foundational idea behind this framework according to Guthrie and Parker (1990). Lindblom (1993) defines business legitimacy through operational conformity to wider 20 social norms and organizational principles and procedural standards. In other words, companies seek to maintain continuous social acceptance in order to ensure their survival and success. Legitimacy evolves over time as public values, expectations and awareness change. Failure to adopt to these changes exposes firms to legitimate threats that may originate from ethical and moral problems, environmental incidents, or business transparency issues (Fernando & Lawrence, 2014). To overcome these many businesses, take steps to act more responsibly and share information about their actions through CSR efforts (Deegan et al., 2002). The idea of this theory based on the view that companies are part of a broader society and their decisions and actions shaped by political, social and economic influences. (Deegan et al., 2002; Gray et al., 1995). In other words, organizations must earn their ‘license to operate’ through continuous stakeholder satisfaction since the automatic right to operate no longer exists. Because of this, legitimacy theory considers CSR reporting essential for organizations to implement their communication strategy. The disclosure of CSR initiatives helps businesses convince the public they are committed to follow ethical standards while reducing adverse feedback and constructs favourable reputation. The reporting process helps businesses showcase their environmental projects and community involvement alongside the management of their reputation risks by handling adverse incidents (Villiers & Maroun, 2017). As an example, businesses increased their environmental reporting effort after environmental tragedies to rebuild their lost legitimacy. Interestingly, the legitimacy theory explains the existence of correlation between CSR reporting and company performance. Through CSR reporting companies can build their reputation, prevent external stakeholder conflicts and can obtain competitive benefits by improving stakeholder perceptions and establishing trust. Companies that build 21 positive public reputation through enhanced disclosure may obtain multiple financial benefits from expanded capital access, greater customer loyalty and operational efficiency. In today’s world, CSR reporting transparency plays a vital role due to stricter regulations and improved stakeholder awareness (KPMG, 2020). CSR reporting is no longer something companies do just to look good, it’s now a crucial part of building trust and improving how a business performs. 2.2.2 Stakeholder Theory Corporate responsibility can be better understood through stakeholder theory which Freeman introduced in 1984 because this approach demonstrates that companies maintain relationships with diverse stakeholder groups. Besides shareholders the framework of stakeholder theory includes communities together with government agencies, suppliers, workers as well as consumers and even the environment. The model of stakeholder theory instructs businesses to interact with various stakeholder groups' multiple interests and expectations since they affect company operations (Freeman, 1984). According to Friedman (1970) classical thinking maintains the company's only social obligation consists of shareholder profit maximization but this approach contradicts his view. According to Rather stakeholder theory accepts a wider corporate goal spectrum because it understands that financial outcomes have social and environmental responsibilities (Fernando & Lawrence, 2014). According to Freeman and McVea (2001), the successful engagement of businesses with stakeholders leads to trust development and enhances reputation alongside maintaining long-term success of the business. CSR reporting demands transparent disclosure about environmental and social projects since such openness helps build better stakeholder 22 relationships. CSR reporting that actively manages stakeholder expectations by being proactive brings dual benefits of risk reduction while promoting positive goodwill. Donaldson and Preston (1995) have defined Stakeholder theory into the three parts of normative, instrumental and descriptive. Through this theory we understand business stakeholder relationships and we can forecast financial effects of stakeholder management while companies face moral obligations to address stakeholder needs. The understanding of how CSR reporting affects business performance requires these core elements to establish moral responsibility while achieving strategic targets. CSR driven by stakeholder theory demonstrate positive value creation despite being misperceived as mere regulatory obligations. Research finds that when companies seriously address stakeholder matters, through initiatives like better worker treatment and environmental reduction efforts and transparency commitments will lead businesses to experience higher morale among employees and better loyalty from customers as well as increased confidence from investors (Barnett & Salomon, 2002; Waddock & Graves, 1997). There remains an increasing doubt about the authenticity of disclosures that organizations publish regarding their Corporate Social Responsibility activities. Some critics argue that businesses utilize CSR reporting as a symbolic gesture for accountability more than real improvement, a tactic often called "window-dressing" (Connors et al., 2017; Ting, 2021). Which is why stakeholder theory emphasizes that true responsibility means understanding all stakeholder conflicts rather than catering to interests of specific group. Therefore, to deliver meaningful CSR reporting organizations need true and equal measures that prove their dedication toward social environmental and economic duties (Lin et al., 2022). To sum up, the stakeholder theory creates a useful system to investigate how CSR reporting impacts business performance. The theory demonstrates that correctly 23 presented ethical business practices through CSR disclosures create legitimacy while delivering long-term profitability thanks to their focus on open stakeholder engagement. 2.2.3 Agency Theory Jensen and Meckling (1976) established that agency theory defines the relationships between principals (business owners like shareholders) and agent who manage the company. This relationship is prone to conflicts of interest wherein managers choose personal benefits over shareholder value growth because they seek promotions and bonuses and personal recognition (Quinn & Jones, 1995). Agency problems are a common term used to describe these conflicts. Information asymmetry is one of the main causes of agency issues. Compared to external stakeholders, managers typically possess a more comprehensive understanding of the company's operations, risks, and performance because they are internal actors. Moral hazard, in which managers make choices that benefit themselves more than the company or its investors, can result from this imbalance. Due to inadequate disclosure, it may also result in adverse selection, where shareholders or potential investors find it difficult to determine the firm's actual state (de Klerk & de Villiers, 2012). As a result, Investors may consequently perceive increased risk, which could result in higher capital expenditures or an undervaluation of the company's stock. According to this viewpoint, CSR reporting significantly contributes to fewer agency- related issues. Businesses can increase transparency and accountability by voluntarily releasing comprehensive information on their governance, social, and environmental practices. This improved transparency enables managers to direct their decisions toward business objectives which strengthens trust between stakeholders and management staff. CSR disclosures serve as a monitoring system which reveals how managers handle stakeholder matters and implement company responsibilities. 24 Trustworthy CSR reporting allows organizations to decrease agency costs through reduced expenses for monitoring workplace behaviour. Through trustworthy CSR reporting an organization establishes better marketplace legitimacy and reputation by showing stakeholders commitment to moral sustainable practices (Dhaliwal et al., 2011). According to a comprehensive stakeholder-oriented perspective Brealey et al. (2011) state that the principal-agent framework recognizes various stakeholder groups beyond shareholder interests. During these circumstances managers need to consider consumer interests alongside worker interests, supplier interests, community interests and shareholder value creation. Thoroughly prepared CSR reports serve as tools for showing how companies fulfil their various requirements in addition to presenting a complete picture of decision-making and performance metrics (de Klerk & de Villiers, 2012). To sum up, CSR reporting serves to enhance corporate governance by resolving agency problems and enhancing transparency and through alignment of managerial actions with stakeholder expectations. These efforts can build trust, reduce perceived risks, attract socially responsible investors, and support long-term relationships with key stakeholders, all of which can contribute to better overall company performance (Dhaliwal et al., 2011). 25 3 Literature review By reviewing previous research that provides pertinent empirical insights, this review of the literature seeks to investigate the complex relationship between corporate social responsibility (CSR) disclosure and firm performance. It lays the groundwork for the empirical investigation of this thesis, which aims to assess how ESG disclosures affect different aspects of business performance, such as operational efficiency, financial success, and market valuation. The review starts with discussing previous research on CSR disclosure and firm performance. The review then organized into three subsections, research that shows no significant relationship, research that shows a negative relationship, and research that shows a positive relationship. The thesis then goes on to discuss how the research hypotheses were developed. 3.1 CSR disclosure and firm performance The connection between firm performance and CSR disclosures has drawn growing interest among researchers as highlighted in Chapter 1. The results of a great deal of research on this subject are still inconsistent and occasionally contradictory. Although some research indicates a positive relationship between CSR and financial results, other studies find no connection at all, and some even point to a negative relationship. CSR has changed from being a completely voluntary practice to a reaction influenced by societal scrutiny and rising stakeholder expectations. Businesses are now expected to show accountability for social and environmental issues in addition to making a profit. High-profile incidents such as Nike's labour disputes, which led to widespread consumer outrage and boycotts after the New York Times exposed unethical labour practices, made this change especially clear (Porter & Kramer, 2006). These public responses show how changes in society and the economy can affect investor behaviour and heighten calls for corporate accountability and transparency. 26 Scholars have investigated how corporate social responsibility (CSR) contributes to long- term financial resilience as well as reputation enhancement. Kim et al. suggests that businesses that incorporate sustainable practices may improve their financial and public standing, primarily because of their perceived dedication to the well-being of future generations. Likewise, Burke and Logsdon (1996) point out that companies that reveal environmental, social, and governance (ESG) information typically encounter lower market risks, especially in developed markets, since these disclosures mitigate the possibility of legal action and harm to their reputation. Porter and Kramer (2006) contend that corporations actively integrating Environmental, Social, and Governance (ESG) practices may secure a competitive advantage. By harmonizing their product and service offerings with objectives such as environmental preservation and societal well-being, corporations can bolster their market positions. In order to shed light on the various findings in this field, Huang et al. (2020) carried out a thorough meta-analysis that included 437 empirical studies. According to their findings, about 40% of the studies discovered a positive relationship between CSR and CFP, 50% found no significant effect, and 10% found a negative one. Methodological variances among studies, such as differences in the definition and measurement of CSR disclosures and firm performance, can be attributed for these conflicting findings (Huang et al., 2020; Q. Wang et al., 2016), including variations in sample sizes, analytical methodologies, and industry and geographic contexts (Quazi & Richardson, 2012). CSR results are also influenced by macroeconomic circumstances, as well as the type of CSR disclosure, whether it is required by law or voluntary (Chen et al., 2018; Dhaliwal et al., 2011; Huang et al., 2020). 3.2 Positive relationship between CSR reporting and firm performance The idea that sustainability initiatives can be strategically advantageous for businesses is supported by an increasing amount of research that indicates a positive relationship 27 between CSR disclosures and firm performance. According to empirical data, companies that implement CSR practices frequently see better operational and financial results. Studies by Kapoor and Sandhu (2010), Ameer and Othman (2012), and Akisik and Gal (2014), for example, offer strong proof that sustainability initiatives have a positive correlation with conventional performance metrics like ROA and ROE. According to these studies, businesses that have active sustainability disclosures or reporting systems typically see increases in operational effectiveness and profitability. This connection is further supported by the stakeholder-oriented view, which highlights that socially conscious businesses are more likely to win over important stakeholders like clients, staff, and investors, improving long-term competitiveness and financial results in the process (Marom, 2006; Orlitzky et al., 2003). Market-based performance metrics can also be impacted by CSR. Transparency in ESG disclosures has been demonstrated in numerous studies to decrease information asymmetry and boost investor confidence, both of which have a positive impact on stock valuations and Tobin's Q (De Klerk et al., 2015; Reverte, 2016; Saleh et al., 2011). More specifically, De Klerk et al. (2015) and Reverte (2016) contend that CSR disclosures enhance traditional financial reporting and allow investors to value companies more accurately. Furthermore, studies carried out in the Nordic region, including Lueg and Pesheva (2021), confirm that ESG elements, particularly governance, can make a substantial contribution to overall shareholder returns. These results support the idea that CSR investments can act as intangible assets that increase customer loyalty, and reduce financial analysts' perceptions of risk, thus enhancing the firm's value (Hichri & Ltifi, 2021; Tsang et al., 2022). Together, these studies support the study's theoretical framework by indicating that market (Tobin's Q, stock return), financial (ROE), and operational (ROA) performance metrics may be significantly influenced by ESG disclosure, a measurable representation of CSR. 28 3.3 Negative relationship between CSR reporting and firm performance While numerous studies highlight the potential benefits of corporate social responsibility (CSR), some empirical evidence suggests a negative association between CSR initiatives and firm performance, especially in the short term. An article by Friedman (1970) is frequently cited in the early literature exploring the negative link between corporate sustainability and financial outcomes. He argues that the concept of "social responsibility" is misapplied to corporations, as only individuals, not businesses can hold responsibilities. In his view, it is the duty of company executives to prioritize profit maximization on behalf of shareholders. Any diversion from this goal, such as adopting socially responsible initiatives, imposes additional costs that may reduce shareholder wealth. As a result, Friedman contends that corporate sustainability efforts are misaligned with the core objective of maximizing shareholder value, given that social and administrative expenses do not contribute directly to financial returns (Friedman, 1970). From an agency theory perspective, managers may engage in CSR for personal motives, potentially misaligning with shareholder interests (Jensen, 2002; Preston & O’Bannon, 1997). Ho and Taylor (2007) found that firms with extensive sustainability disclosures under the triple bottom line (TBL) framework reported lower profitability (ROA), while López et al. (2007) observed that companies listed in the Dow Jones Sustainability Index experienced reduced profit before tax (PBT) growth. Market inefficiencies also contribute to this relationship. Brammer et al. (2006) argued that socially responsible firms often underperform due to investor reluctance to divest, leading to asset mispricing. Similarly, Makni et al. (2009) found that the costs of CSR often outweigh short-term benefits, reducing net income. Kao et al. (2018) and Hichri & Ltifi (2021) emphasized that CSR may divert resources from core business areas or lead to inefficient capital allocation. 29 CSR’s reputational risks can also harm firm value. Wang and Zhang (2022) showed that negative CSR media coverage has stronger market impacts than positive news. Chen et al. (2018) observed that mandatory CSR disclosures in China, although non-costly, correlated with declines in profitability (ROA, ROE), likely due to increased social pressure. Industry-specific evidence by Omar and Zallom (2016) demonstrated negative CSR effects in sectors like food and pharmaceuticals, particularly in themes such as environment and community. Moreover, Lioui and Sharma (2012) found environmental CSR negatively impacts performance unless paired with innovation, such as R&D. Overall, the negative CSR and firm performance relationship highlights the potential short-term costs, misalignments, and market skepticism that firms may face when implementing CSR without strategic integration or stakeholder alignment. 3.4 Neutral relationship between CSR reporting and firm performance While a significant body of research supports either a positive or negative link between CSR and firm performance, some studies have reported neutral or inconclusive results. Inoue and Lee (2011), for example, conducted a sector-specific analysis by breaking down sustainability into five dimensions, diversity, community involvement, product quality, employee relations, and environmental responsibility. Their findings revealed that the impact of these sustainability elements varied significantly across industries, with some dimensions showing no substantial effect on firms’ short-term or long-term profitability as measured by ROA and Tobin’s Q. Similarly, Dilling (2010) observed that while firms with higher profit margins tended to issue high-quality sustainability reports, there was no evident association between sustainability reporting and governance-related variables such as board committees or audit oversight. This suggests that profitability might encourage disclosure, but governance practices may not necessarily align with sustainability performance. 30 Guidry and Patten (2010) also found that the initial release of standalone CSR reports in the U.S. did not elicit a significant market reaction. However, report quality, measured through adherence to GRI standards, played a notable role, with high-quality reports attracting better market responses than low-quality ones. Lastly, Atan et al. (2018), focusing on Malaysian firms, reported no statistically significant relationship between individual or composite ESG scores and profitability or firm value. The only exception was a positive association among the overall ESG score and the cost of capital, suggesting that while ESG efforts may not immediately influence firm performance metrics, they might have long-term implications for financing costs. The authors also noted that stakeholders’ skepticism and the limited three-year study period could explain the absence of stronger associations. 3.5 Research Hypothesis This part presents the research hypothesis developed considering the theoretical framework and insights from the literature review. The first hypothesis addresses the main research question of this study. Drawing upon prior literature, which largely supports a positive association between Sustainability disclosures and firm performance and considering Finland’s strong track record in ESG disclosure practices, this study hypothesizes a positive relationship between sustainability disclosures and firm performance. The second hypothesis builds on findings in the literature that suggest firms with more comprehensive ESG disclosures may experience greater benefits in terms of performance outcomes. Accordingly, the following hypotheses are proposed. H1: CSR disclosures have a positive impact on overall firm performance (as measured by ROA, ROE, Tobin’s Q, and stock returns). H2: The relationship between ESG disclosure and firm performance is stronger in firms with high ESG disclosure scores than in those with lower scores. 31 4 Data and Methodology This study explores the association between Corporate Social Responsibility (CSR) disclosures and firm performance, including operational, financial, and market performance over a period of 6 years. The selected time period is 2018–2023 as complete ESG disclosure score data was not available for most firms prior to 2018. Firm performance is measured by two accounting-based indicators, namely ROA and ROE, and two market-based indicators, namely Tobin’s Q and stock returns. These variables are selected based on prior literature. Sustainability disclosures are measured using Bloomberg’s ESG disclosure scores. To examine the relationship, this study employs a quantitative methodology and performs a regression analysis. The study partially adopts the approaches of Al Hawaj and Buallay (2022), Buallay (2019) and Botchwey et al. (2022). Given the numerical and measurable nature of the variables of interest, quantitative research methods are preferred over qualitative ones in this study setting. Since quantitative methods enable accurate measurement of the direction and strength of the relationship between variables, they are appropriate for this thesis. In this chapter the data and the used methodology are presented. The procedure for gathering data and choosing a sample is described in the first subchapter. The sample distribution by industry and ESG disclosure score levels is covered in the second subchapter. In the third subchapter, the variables of this study are presented. Data cleaning and processing steps were explained in the fourth subchapter and the fifth subchapter focus on the correlation between the variables. The descriptive statistics are presented in the sixth subchapter, while the regression analysis and model are covered in the last subchapter. 4.1 Data collection The data for this study was collected by combining information from various data sources. The data for this study was sourced from active Finnish companies listed on the Helsinki 32 Stock Exchange as of the 06th of February 2025, downloaded from Refinitiv which is a well-known financial data platform. The original dataset consisted of 193 active companies listed on the Helsinki Stock Exchange as of this date. It was then I identified that ESG disclosure data was not available for all these companies. Therefore, two sample selection criteria were applied to verify the quality and consistency of the dataset. First, companies must have been listed and established before January 1, 2019. This criterion was applied to improve the robustness of the analysis by ensuring a consistent time series of data. As a result, 56 companies listed after this date were excluded. Second, companies without ESG data available throughout the study period were also excluded, leading to the removal of an additional 49 companies. After applying these criteria, the final sample used in the analysis consisted of 88 companies. The financial data gathered from the Refinitiv database includes various metrics such as ROA, ROE, market capitalization, minority interests, preferred stocks, total assets, total liabilities, total debt to total assets, market to book, stock price beginning values, as well as starting and ending values of stock prices. These raw data were used to calculate the dependent and control variables using Excel. The data for the independent variable, ESG disclosure scores, was obtained from the Bloomberg Terminal. 4.1.1 Sample selection To ensure representativeness, the sample will include companies across all major industries, proportional reflecting their presence in the Finnish economy. The industry classification in this study follows the Industry Classification Benchmark System (ICB). This study measures the level of sustainability reporting using ESG disclosure scores. The sampling method employs a stratified approach to ensure the sample represents diverse industries and varying levels of ESG disclosures. 33 The process involves two main steps; • Industry Categorization: The study adopts the Industry Classification Benchmark system (ICB) system to categorize sample firms. This internationally accepted taxonomy developed by FTSE Russell and Dow Jones in 2005 and it organizes companies into 11 distinct industrial sectors (FTSE Russell, 2024). To account for sector-specific effects, industry dummy variables are incorporated as controls in the analysis, with each sector identified by its unique two-digit ICB code. Table 1 provides comprehensive details regarding the distribution of companies across these industrial classifications. • ESG disclosure categorization: Within each industry, companies are further classified based on their corporate social responsibility (CSR) performance, measured by their average ESG disclosure scores over the study period (2018–2023). To ensure a meaningful comparison, I calculated each company's mean ESG score across these six years and then categorized them as High ESG disclosure firms and Low ESG disclosure firms. The threshold for these categories was determined using ESG score benchmarks, as illustrated in Table 2. This stratification ensures that both industry diversity and different levels of ESG transparency are represented, enabling a comprehensive analysis of CSR disclosures and their impact on operational, financial, and market performance. 4.1.2 Sample distribution Table 1 presents distribution of the sample of companies according to their industry. The industrials sector has the largest representation making up for around one-third of the total sample. Following that, the consumer discretionary sector is the second and technology sector is the third most represented industries. On the other hand, the telecommunications, utilities, and energy sectors are the least represented, each comprising just over 1% of the total sample. Additionally, Table 1 includes the corresponding ICB codes with two-digits used to create dummy for industries as a control variable for the analysis. 34 Table 1. Industry distribution ICB Code Industry (ICB) No. of firms Observations Share of Total 55 Basic Materials 7 42 7.95 % 40 Consumer Discretionary 18 108 20.45 % 45 Consumer Staples 7 42 7.95 % 30 Financials 7 42 7.95 % 20 Health Care 4 24 4.55 % 50 Industrials 29 174 32.95 % 35 Real Estate 2 12 2.27 % 10 Technology 11 66 12.50 % 15 Telecommunications 1 6 1.14 % 65 Utilities 1 6 1.14 % 60 Energy 1 6 1.14 % Total 88 528 100.00 % Table 2 represents the threshold for categorizing companies into low and medium to High ESG levels. The disclosure groups were determined based on the mean ESG scores calculated for each firm over the study period from 2018–2023. The overall mean ESG score for the full sample was 39.88, which shows the general level of ESG transparency among the firms studied. For simplicity and practical application, the threshold was rounded to 40. Companies with an average ESG score equal to or greater than 40 were categorized as medium to high ESG disclosure firms, while those with scores below 40 were classified as low ESG disclosure firms. This ESG-based categorization was specifically designed to test Hypothesis 2 (H2) of the study. Table 2. ESG disclosure categorization criteria Categorization Threshold: average score (2018-2023) Low ESG Disclosure Firms Average ESG disclosure score < 40 Medium to High ESG Disclosure Firms Average ESG disclosure score ≥ 40 The full list of companies included in the analysis, along with their industry classification, is provided in Appendix 1. Each firm was categorized into either the low or medium-to- high ESG disclosure group according to its average score across the study period. This 35 classification enables a comparative analysis of how varying levels of ESG disclosure relate to firms’ operational, financial, and market performance across different sectors. 4.2 Variables of the study This research explores how Finnish companies' operational, financial, and market outcomes are associated with their CSR disclosures or reporting practices. Regression analysis will be the main analytical tool used in this quantitative study to look at how CSR disclosures affect performance indicators like ROA, ROE, Tobin's Q and stock returns. Table 3 presents all the variables used in the regression models of this study. Table 3. Summary of variable measurements Variables Labels Measurements Dependent Variables Operational Performance ROA Net income divided by total assets Financial Performance ROE Net income divided by shareholder equity Market Performance TQ (Market value of equity + total liabilities) ÷ total assets Market Performance RTS Percentage change in stock price over a defined period Independent Variables ESG Disclosure ESG Bloomberg index which combines E, S and G dimensions Environmental Disclosure E Bloomberg index which measures the disclosure of the firm’s energy consumption, waste, pollution levels, conservation of natural resource and treatment of animals. Corporate Social Responsibility Disclosure S Bloomberg index which measures the disclosure of the firm’s business relationships, charitable activities, volunteer work, employees’ health, welfare and safety Corporate Governance Disclosure G Bloomberg index which measures the disclosure of corporate adherence to governance standards and practices. Control Variables Firm Size TA Logarithm of total assets Leverage LEV Ratio of total debt to total assets Growth MTB Market-to book ratio Industry Type IND Dummy variable based on the industry of the firm Additional Dummy ESG Disclosure Group (H2) ESGHigh_Lo w Dummy = 1 if ESG score ≥ 40 (Medium-to-High), 0 if ESG score < 40 (Low) 36 4.2.1 Dependent variables This study aims to explore the impact of sustainability reporting on firm performance. Accordingly, firm performance serves as the dependent variable, as it is the outcome expected to respond to changes in the independent variable. To evaluate firm performance, two accounting-based and two market-based indicators have been selected. Each indicator is introduced in the following sections, along with its definition, formula, and the rationale for its inclusion in the analysis. 4.2.1.1 Return on Assets (ROA) ROA is employed in this study as an indicator of operational efficiency. It reflects how effectively a firm utilizes its assets to generate profit. This ratio is calculated by dividing the net income of a company by its total assets. Formula for ROA: 𝑁𝑒𝑡 𝐼𝑛𝑐𝑜𝑚𝑒 𝑇𝑜𝑡𝑎𝑙 𝐴𝑠𝑠𝑒𝑡𝑠 × 100 (1) A higher ROA suggests that the company is more efficient in turning its asset base into net earnings. Since some industries rely more heavily on assets than others, ROA values may naturally differ across sectors. Despite this, ROA remains a widely accepted measure of profitability and is often used in empirical research examining the relationship between corporate social responsibility (CSR) disclosure and firm performance (Al Hawaj & Buallay, 2022; Botchwey et al., 2022; P. Wang & Zhang, 2020). Some researchers prefer using EBIT or profit before tax in ROA calculations, particularly when dealing with companies across different tax jurisdictions, to minimize distortions from varying tax policies. However, since our sample is drawn entirely from firms operating within the same Finnish tax framework, such adjustments were not necessary for this study. 37 4.2.1.2 Return on Equity (ROE) To assess the financial performance aspect of firms, this study uses Return on Equity (ROE). ROE evaluates how effectively a company generates profit relative to the equity invested by its shareholders, thereby providing insight into shareholder profitability. It is calculated by dividing the net income for the fiscal year by the total shareholder equity. Formula for ROE: 𝑁𝑒𝑡 𝐼𝑛𝑐𝑜𝑚𝑒 𝑆ℎ𝑎𝑟𝑒ℎ𝑜𝑙𝑑𝑒𝑟 𝐸𝑞𝑢𝑖𝑡𝑦 × 100 (2) While ROE is a widely recognized and commonly used profitability ratio, it can be influenced by managerial decisions such as equity buybacks, which may reduce the denominator and artificially inflate the ratio. However, within the scope of our sample, this risk is considered minimal and does not significantly affect the reliability of the metric. Similar to ROA, ROE is viewed as a neutral and established measure of profitability and is frequently cited in CSR disclosure and firm performance research (Al Hawaj & Buallay, 2022; Botchwey et al., 2022; P. Wang & Zhang, 2020). Although ROA and ROE are likely to be strongly correlated, they provide complementary perspectives. ROA focuses on asset efficiency, whereas ROE captures return specifically from the viewpoint of shareholders. Including both metrics allows us to obtain a more nuanced understanding of how sustainability reporting relates to different dimensions of firm performance. Return on equity (ROE) is the measure used to measure the financial performance of our study. ROE measures the profits generated by the company against the money shareholders have invested in the company. More specifically, the ratio measures profitability for the shareholders. The ratio is calculated by dividing the net profit of the fiscal year with the total equity of the company. ROE is however considered to be easily manipulated by managers, as they can reduce equity through equity buybacks. We do however not see this risk to be material in our investigation. We choose to use ROE as 38 our second dependent variable as it is, similar to ROA, a neutral and popular profitability measure. ROE is also according to Wang and Zhang (2020) a frequently used measure in previous CSR – firm performance research (Wang & Zhang, 2020). As previously stated, ROE calculates shareholder profitability, which the researcher believes to be an intriguing comparison to ROA. Although a comparatively high correlation between ROA and ROE is anticipated, the researcher think that each will still contribute unique insights to this research. 4.2.1.3 Tobin’s Q (TQ) To evaluate market performance, Tobin’s Q (TQ) is employed as one of the dependent variables. Originally introduced by Brainard and Tobin (1968) and further developed by Tobin (1969), Tobin’s Q reflects the ratio between a firm’s market value and the book value of its assets. Higher Q values indicate that the market values a company above the accounting value of its asset base, which may reflect positive future expectations (Atan et al., 2018). Formula for Tobin’s Q: (𝑀𝑎𝑟𝑘𝑒𝑡 𝑐𝑎𝑝𝑖𝑡𝑎𝑙𝑖𝑧𝑎𝑡𝑖𝑜𝑛 + 𝑇𝑜𝑡𝑎𝑙 𝐿𝑖𝑎𝑏𝑖𝑙𝑖𝑡𝑖𝑒𝑠 + 𝑃𝑟𝑒𝑓𝑒𝑟𝑟𝑒𝑑 𝑆𝑡𝑜𝑐𝑘𝑠 + 𝑀𝑖𝑛𝑜𝑟𝑖𝑡𝑦 𝑖𝑛𝑡𝑒𝑟𝑒𝑠𝑡) 𝑇𝑜𝑡𝑎𝑙 𝑎𝑠𝑠𝑒𝑡𝑠 × 100 (3) This measure is selected for several reasons. First, it captures market perceptions of firm value and can reflect the added value that sustainability reporting may bring (Tsang et al., 2022). Second, as noted by Agostini et al. (2022), it is one of the most frequently applied market-based indicators in CSR disclosure and firm performance relationship. For these reasons, Tobin’s Q is an appropriate measure for evaluating how the market perceives the influence of sustainability reporting. 39 4.2.1.4 Return to stocks (RTS) The second market-based indicator used in this study is stock returns, which provide a direct measure of how a company’s share price has evolved over a specific period. Stock returns were selected because they encompass a broad range of market expectations and investor perceptions regarding a firm’s future performance (de Villiers & Marques, 2016). Stock prices are forward-looking and integrate market sentiment about potential earnings, associated risks, and other relevant information. As mentioned by De Klerk et al. (2015) and Reverte (2016), CSR disclosures can contribute additional signals to the market, helping investors better assess a company’s long-term value and risk profile. Therefore, stock returns serve as a comprehensive reflection of market performance. Following Dorfleitner et al. (2018), this study measures stock returns using the share price at the end of the financial year, including dividends when applicable. Formula for stock returns: (𝑆ℎ𝑎𝑟𝑒 𝑝𝑟𝑖𝑐𝑒𝐸𝑛𝑑 + 𝑑𝑖𝑣𝑖𝑑𝑒𝑛𝑑𝑠) − 𝑠ℎ𝑎𝑟𝑒 𝑝𝑟𝑖𝑐𝑒𝐵𝑒𝑔 𝑆ℎ𝑎𝑡𝑒 𝑝𝑟𝑖𝑐𝑒𝐵𝑒𝑔 × 100 (4) 4.2.2 Independent variables In this study, the independent variables, also known as explanatory variables, are central to understanding the relationships examined. The primary focus is on the overall ESG (Environmental, Social, and Governance) disclosure score, along with its three subcomponents: Environmental, Social, and Governance disclosure scores. This research emphasizes the disclosure of sustainable practices. As highlighted by Brammer et al. (2006), CSR reports that are ambiguous or inconsistent may not 40 contribute positively to corporate value. From a stakeholder’s point of view, a company's "greenness" is closely tied to how effectively it communicates its sustainability efforts. The most common method for this is through annual sustainability or CSR reports. In Finland, CSR reporting becomes mandatory when companies surpass certain thresholds, though there are no standardized guidelines governing how these reports should be structured or what content must be included. Typically, companies with robust CSR integration tend to have substantial content to include in their sustainability reports. Conversely, firms with limited CSR engagement might produce only minimal reports. As such, a company’s disclosure often reflects the extent of their actual CSR activities. This study applies the "fair view" principle, assuming that sustainability disclosures accurately represent a company’s CSR efforts. Accordingly, firms with more comprehensive reports are considered to have higher ESG disclosure scores. We use this principle to maintain analytical focus rather than delve into ethical concerns. 4.2.2.1 ESG disclosure score The core independent variable in this research is the ESG disclosure score. As a quantitative study, this study utilizes secondary data, specifically ESG disclosure scores sourced from Bloomberg Terminal. This score serves as a proxy for transparency and quality in ESG reporting. Bloomberg’s ESG Disclosure Score is widely recognized by investors, analysts, and corporations for assessing how thoroughly a company communicates its ESG-related risks, strategies, and initiatives to stakeholders (Oberndorfer, 2021). Rather than measuring ESG performance outcomes directly, the score reflects the level of detail and frequency of ESG disclosures across more than 120 data points spanning environmental, social, and governance topics (ESG Advising LLC, 2024). These scores, ranging from 0 to 100, are not a measure of actual ESG performance but rather of the 41 quality, consistency, and depth of reporting. Higher scores indicate greater transparency and completeness in disclosure. Environmental Disclosures (E) cover aspects such as greenhouse gas emissions, energy and water consumption, waste management practices, environmental risk management, and climate-related reporting. This reflects how openly a firm discusses its environmental footprint and sustainability measures (ESG Advising LLC, 2024). Corporate Social Responsibility Disclosures (S) relate to areas including labor practices, employee health and safety, diversity and inclusion, community engagement, and product responsibility. This sub-score assesses the extent to which a company reports on its social impact and interactions with various stakeholders (ESG Advising LLC, 2024). Governance governance disclosures (G) address corporate structure and policies, including board diversity, executive remuneration, shareholder rights, anti-corruption policies, and overall business ethics. This helps capture the degree of transparency around decision-making structures and ethical standards within the firm (ESG Advising LLC, 2024). Bloomberg gathers these disclosures from publicly available sources such as annual and sustainability reports, regulatory filings, company websites, and direct disclosures (Oberndorfer, 2021). The frequency and detail of updates also play a role in determining the score. 4.2.3 Control variables Control variables play a critical role in regression analysis by helping to isolate the relationship between the key independent and dependent variables. While they are not the primary focus of this research, their inclusion is essential as they may provide alternative explanations for observed effects. These variables are typically selected 42 because they could influence both the explanatory and outcome variables in the model. In this study, control variables have been carefully chosen based on their potential relevance to the sample and their frequent use in previous literature. There are several company- and industry-specific characteristics that may influence the link between sustainability disclosure and firm performance. To reduce the risk of omitted variable bias and improve the robustness of the model, this study incorporates four control variables widely adopted in related empirical studies firm size, financial leverage, growth opportunities, and industry classification (Al Hawaj & Buallay, 2022; Cupertino et al., 2022). 4.2.3.1 Firm size Firm size is chosen as a control variable because bigger companies typically perform better than smaller ones as well as because prior studies have suggested that the size of a company may influence stakeholders' interest in its sustainability initiatives (Feng et al., 2017; Velte, 2017). They also contend that smaller businesses might be less inclined to engage in socially conscious practices than larger ones since, as they mature and expand, they will draw greater interest from outside parties and be more inclined to accommodate stakeholder demands (Waddock & Graves, 1997). In order to determine whether firm size has a significant impact on the regression results, firm size must be added as a control variable because the sizes of the companies in our sample are not balanced. Building on earlier research by Agostini et al. (2022), De Klerk et al. (2015), Feng et al. (2017) and Velte (2017), this study incorporate natural logarithm of total assets as a control variable to represent firm size. Formula for log of total assets (represent firm size): 𝐿𝑜𝑔 (𝑇𝑜𝑡𝑎𝑙 𝐴𝑠𝑠𝑒𝑡𝑠) (5) 43 4.2.3.2 Financial Leverage Financial leverage is also considered as it can impact both a firm's sustainability initiatives and its overall performance. High debt levels may increase financial risk, which in turn can negatively affect performance (Atan et al., 2018; Omar & Zallom, 2016). Additionally, leverage may influence a firm's capacity or willingness to invest in sustainability (Cupertino et al., 2022). In line with the methodology used by Agostini et al. (2022) and Tsang et al. (2022), financial leverage is calculated as the ratio of total debt to total assets. Formula for financial leverage: 𝑇𝑜𝑡𝑎𝑙 𝐷𝑒𝑏𝑡 𝑇𝑜𝑡𝑎𝑙 𝐴𝑠𝑠𝑒𝑡𝑠 × 100 (6) 4.2.3.3 Growth This study also controls for firm growth using the market-to-book ratio, which reflects investors’ expectations of future profitability and growth (Fama & French, 1992). A higher ratio often signals that the market anticipates strong future performance, potentially affecting firm outcomes independent of ESG factor (Christensen et al., 2021). Conversely, a low ratio could imply undervaluation or weak growth prospects. By accounting for growth potential, the analysis ensures that the observed relationships between ESG disclosure and firm performance are not merely capturing the effects of growth expectations. Growth is measured as the ratio of the market value of equity to its book value. The market-to-book ratio is included as a control variable in this study because it reflects growth opportunities and market valuation relative to the firm's book value (Fama & French, 1992). A higher market-to-book ratio often indicates that investors expect strong future growth and profitability, which can positively influence firm performance (Galant 44 & Cadez, 2017). Conversely, a lower ratio may suggest that the firm is undervalued or faces limited growth prospects. This variable is particularly relevant for the study because it helps control for variations in firm performance that stem from growth potential rather than the effects of ESG factors. By accounting for growth opportunities, the analysis can isolate the true impact of ESG practices on firm performance and ensure that the observed relationships are not driven by growth expectations. In this study, the market-to-book ratio is calculated by dividing the market value of equity by the book value of equity. Formula for growth (market-to-book ratio): 𝑀𝑎𝑟𝑘𝑒𝑡 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝐸𝑞𝑢𝑖𝑡𝑦 𝐵𝑜𝑜𝑘 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝐸𝑞𝑢𝑖𝑡𝑦 × 100 (7) 4.2.3.1 Industry type Firms operating in different sectors often face distinct operational demands, regulatory pressures, and stakeholder expectations, all of which may affect both performance and sustainability reporting (Singh & Chakraborty, 2021). Thus, industry type is included as a categorical control variable to capture sector-specific influences. Industry classification in this study follows the ICB (Industry Classification Benchmark) (FTSE Russell, 2024). In line with the studies done by Cupertino et al (2022) and Singh and Chakraborty (2021), the sample firms are grouped into 11 broad industry sectors, and industry dummy variables are created to control for these effects. Research by Ruf et al. (2001) emphasizes that companies within the same industry are subject to similar stakeholder pressures and must often address comparable expectations. As a result, financial and sustainability performance can differ significantly across industries. Controlling for industry type helps ensure that the influence of ESG 45 disclosure is not confounded by sector-based performance differences (Waddock & Graves, 1997). 4.3 Data cleaning and preprocessing Prior to conducting the empirical analysis, it was necessary to verify the accuracy and reliability of the dataset. Data cleaning and preprocessing steps were implemented to address inconsistencies that could compromise the validity of the findings. Particular attention was given to managing outliers and handling missing values, both of which are prevalent challenges in financial datasets. These measures were taken to improve data quality and ensure the robustness of the statistical analysis 4.3.1 Managing outliers Outliers are data points that significantly deviate from the overall pattern of a dataset. These extreme values can distort statistical analyses and potentially lead to misleading conclusions if not addressed properly before running regression models. One of the initial steps in identifying such anomalies is through examining the descriptive statistics, particularly the minimum, maximum, mean, and median values of each variable. Additionally, visual tools such as scatterplots can help in detecting values that fall far outside the normal distribution. In this study, noticeable outliers were identified in variables like Return on Equity (ROE), Tobin’s Q (TQ), and stock returns. These variables showed extreme minimum and maximum values that differed substantially from their central tendency measures. To address this, a uniform outlier treatment was applied to all dependent variables through a technique known as Winsorization. The same method was followed by a similar study done by Söderholm & Metsä-Tokila (2021). 46 Table 4. Comparison of original and winsorized dependent variables Mean Median Maximum Minimum Std. Dev. ROA 5.068 4.440 42.930 -41.870 7.486 ROA (wins.) 5.126 4.440 24.711 -9.522 6.609 ROE 7.174 10.360 117.180 -1117.920 54.551 ROE (wins.) 9.788 10.360 49.373 -40.307 16.697 TQ 1.719 1.278 29.131 0.497 1.892 TQ (wins.) 1.618 1.278 5.655 0.689 1.020 RTS 0.086 0.012 3.255 -0.900 0.468 RTS (wins.) 0.075 0.012 1.327 -0.607 0.409 As shown in Table 4, the winsorization method was employed to reduce the influence of extreme values by capping them at more moderate percentiles. Specifically, a 2% winsorization threshold was used, which involved adjusting the lowest 1% of values to the 1st percentile and the highest 1% to the 99th percentile. This adjustment is a common practice to enhance data quality while preserving the majority of the dataset’s structure. The most significant transformation occurred in the ROE variable, where the extreme maximum value of 1,117% was reduced to 49.37%, and the overall variability (standard deviation) dropped by approximately 69%. While such outliers may reflect real-world situations, like companies facing bankruptcy or undergoing restructuring, they can unduly skew analytical outcomes if left untreated. The winsorization process helped in stabilizing the data and improving its suitability for regression analysis. Overall, the application of a standard 2% winsorization threshold across the dataset proved effective. It brought the extreme values closer to the central range, resulting in a more balanced distribution without compromising the integrity of the data. 47 4.3.2 Handling missing values The dataset used in this study, which was compiled from secondary sources, included some missing entries, particularly within the ESG disclosure score variables. This is a common occurrence in sustainability-related datasets, as firms often vary in the consistency and frequency of reporting ESG-related information across different years. As depicted in Table 5, the bulk of missing data was observed in ESG-related variables. The Environmental Disclosure Score, in particular, had the highest proportion of missing values, with 92 observations absent, accounting for roughly 17.4% of the total for that variable. Meanwhile, the ESG, Social, and Governance disclosure scores each had 53 missing entries, representing approximately 10% of their respective datasets. On the other hand, most financial variables, such as ROA and Tobin’s Q, were nearly complete, with only minor gaps (less than 0.4% missing). In total, 253 out of 5,808 values were missing, equating to a missing data rate of 4.36%. This falls below the widely accepted threshold of 5%, above which more sophisticated statistical treatments are typically recommended (Allison, 2001). Furthermore, no patterns indicating systematic bias were identified, suggesting that the missingness was random and not driven by any specific variable or firm characteristic. Given the minimal scope of missing data and the lack of evidence for non-randomness, imputation was deemed unnecessary. Instead, the dataset was used in its existing form to preserve the authenticity of the data and avoid introducing estimation-based artifacts. This approach maintains the robustness and interpretability of the regression results while keeping the analysis straightforward and reliable. 48 Table 5. Summary of missing data Variable Observations Zero values (No data) Total Observations ESG 475 53 528 E 436 92 528 S 475 53 528 G 475 53 528 ROA 528 0 528 ROE 526 2 528 TQ 528 0 528 RTS 528 0 528 TA 528 0 528 LEV 528 0 528 MTB 528 0 528 Total 5555 253 5808 4.4 Correlation Analysis To investigate the potential presence of multicollinearity between the variables, the A correlation analysis was conducted in this study. Correlation coefficients range from -1 to 1, where values close to -1 or 1 indicate strong linear relationships, while those near 0 imply weak or no linear association. Typically, a correlation coefficient above 0.7 suggests a possible risk of multicollinearity warranting further investigation (Velte, 2017). Table 6 presents the Pearson correlation matrix of the key variables. Significance levels are denoted at 1% (p < 0.01) and 5% (p < 0.05). The matrix allows for a preliminary observation of inter-variable relationships and potential concerns regarding collinearity. As shown in the table 6, no significant correlation exists between ESG scores and performance measures, indicating that ESG disclosure levels do not directly associate with short-term financial outcomes. A particularly high correlation is observed between Return on Assets (ROA) and Return on Equity (ROE) (r = 0.896), which is expected since both indicators are derived from net income. Despite the high correlation, both ROA and ROE are retained in the analysis due to their distinct interpretation of operational and financial performance as explained in chapter 4.2. To avoid multicollinearity, they are analyzed in separate regression models. 49 Tobin’s Q demonstrates moderate positive correlations with both ROA (r = 0.625) and ROE (r = 0.513), but these values remain below the 0.7 threshold, suggesting no immediate concern. The overall ESG score is strongly correlated with the individual ESG pillar scores: Environmental (E = 0.898), Social (S = 0.908), and Governance (G = 0.876). These high correlations are expected since the aggregate ESG score is derived from these components. However, to avoid multicollinearity, the pillar scores are not included in the same regression models with the overall ESG score. Financial leverage has a notable significant negative correlation with both Tobin’s Q and stock returns, suggesting that companies with reduced debt tend to perform well in the market. This is consistent with the significant negative correlations observed between leverage and both ROA and ROE. These findings are in line with prior research done by Feng et al. (2017), Phan et al (2020) and Singh and Chakraborty (2021) which similarly find that firms with greater efficiency and profitability typically experience superior market performance. ESG scores are positively and significantly correlated with firm size, measured by the log of total assets. This suggests that larger firms may possess more resources to invest in ESG initiatives. Conversely, financial leverage is negatively correlated with all ESG variables, implying that highly leveraged firms may allocate fewer resources to ESG practices. Among the control variables, firm size positively correlates with ESG disclosure (r = 0.427) but negatively correlates with Tobin’s Q (r = -0.125), potentially reflecting differing stages of firm growth. Financial leverage is negatively associated with both ROA (r = -0.315) and ROE (r = -0.289), consistent with the financial burden associated with higher debt levels. Additionally, Tobin’s Q shows an extremely high correlation with the market-to-book ratio (r = 0.939), indicating that both variables serve as closely related proxies for market 50 valuation. Although this presents a potential multicollinearity concern, MTB is retained in the analysis due to its theoretical distinctiveness from Tobin’s Q. To further assess multicollinearity, variance inflation factors (VIFs) were computed. This test is a standard that measures the effect of independent variables (Buallay, 2019). As can be seen in Appendix 2, All VIF values were found to be below 5, suggesting that multicollinearity is not a serious issue. VIF values above 10 generally considered to indicate problematic multicollinearity as explained in prior literature (de Villiers & Marques, 2016; Gujarati, 2003; Omar & Zallom, 2016; Reverte, 2016; Velte, 2017). Therefore, the MTB variable was retained in the models even when Tobin’s Q was used as the dependent variable. Table 6. Correlation Analysis * Correlation is considered statistically significant at the 10% threshold (p < 0.10) ** Correlation is considered statistically significant at the 5% threshold (p < 0.05) ROA ROE TQ RTS ESG E S G TA LEV ROA 1.0000 ROE 0.8963** 1.0000 TQ 0.6247** 0.5128** 1.0000 RTS 0.4116** 0.4248** 0.3958** 1.0000 ESG 0.0583 0.0357 0.0701 0.0022 1.0000 E 0.0836 0.0557 0.0466 -0.0333 0.8977** 1.0000 S 0.0692 0.0554 0.0770 0.0135 0.9083** 0.8174** 1.0000 G 0.0124 -0.0046 0.0680 0.0269 0.8759** 0.6075** 0.6844** 1.0000 TA -0.0941* 0.0177 -0.1248** -0.0678 0.4270** 0.5024** 0.4575** 0.2216** 1.0000 LEV -0.3147** -0.2891** -0.2249** -0.1005* -0.0994* - 0.1223** -0.0990* -0.0513 0.0443 1.0000 MTB 0.5918** 0.5211** 0.9390** 0.4207** 0.0543 0.0224 0.0548 0.0682 -0.1127** -0.1528** 4.5 Descriptive statistics Table 7 summarizes the descriptive statistics for the variables used in the analysis, including values of mean, median, minimum, maximum, standard deviation, and the number of observations. The dataset comprises 528 firm-year observations. Focusing first on the dependent variables, the average Return on Assets (ROA) is 5.13%, indicating that the firms in the sample generally utilize their assets efficiently to generate profits. The ROA ranges from a low of -9.52%, suggesting operational inefficiencies in some firms, to a high of 24.71%, reflecting significant profitability in others. The standard deviation of 6.61% signals moderate dispersion in profitability levels. For Return on Equity (ROE), the mean value is 9.79%, with a median of 10.36%. This suggests that, on average, firms have managed to deliver solid returns to shareholders. However, the wide range from -40.31% to 49.37%, along with a relatively high standard deviation of 16.70%, illustrates substantial differences in shareholder return performance across firms in the sample. Tobin’s Q, with a mean of 1.62, a minimum of 0.69, and a maximum of 5.66, indicates that the majority of firms are valued above the replacement cost of their assets, an interpretation consistent with positive investor sentiment. Despite this, the standard deviation of 1.02 reflects significant variability in market valuations. These findings align with previous literature (Feng et al., 2017; Tsang et al., 2022). Turning to stock performance, the mean stock return is 7.5%, suggesting a moderate average gain. Nevertheless, the high standard deviation of 0.41 and the range from -0.61 to 1.33 point to pronounced volatility, with some firms facing notable losses and others enjoying substantial returns. 53 The average ESG disclosure score across the sample is 39.88, indicating that firms are, on the whole, only partially compliant with ESG standards. The scores range from 0 to 82.72, with a standard deviation of 19.80, highlighting significant variation in ESG-related performance. Breaking down ESG components, the Environmental (E) dimension has a mean score of 27.79, suggesting room for improvement in environmental practices. Similarly, the Social (S) score averages at 23.84, reinforcing this view. In contrast, the Governance (G) component stands out with a higher mean of 67.91, suggesting stronger performance in corporate governance among the sampled firms. Regarding the control variables, the average of the natural logarithm of total assets (log assets) is 13.48, indicating a wide range in firm size, further supported by a standard deviation of 2.07. The mean leverage ratio is 27.43%, with extreme values ranging from as low as 0.04% to as high as 79.64%, reflecting diverse capital structures. Finally, the average Market-to-Book (MTB) ratio is 2.45, suggesting that, on average, firms are valued above their book value. The maximum MTB of 11.09 points to high growth expectations for some firms. Table 7. Descriptive Statistics Variables Mean Median Min. Max. Std. Dev. Observations ROA 5.126 4.440 -9.522 24.711 6.609 528 ROE 9.788 10.360 -40.307 49.373 16.697 528 TQ 1.618 1.278 0.689 5.655 1.020 528 RTS 0.075 0.012 -0.607 1.327 0.409 528 ESG 39.883 40.273 0.000 82.718 19.797 528 E 27.795 26.246 0.000 81.758 23.993 528 S 23.841 21.191 0.000 71.735 16.417 528 G 67.907 75.090 0.000 98.615 26.175 528 TA 13.482 13.061 10.138 22.616 2.073 528 LEV 27.434 26.850 0.040 79.640 15.154 528 MTB 2.452 1.760 0.392 11.086 2.212 528 54 In summary, the descriptive statistics reveal considerable heterogeneity in the sample, particularly in financial performance, ESG scores, and firm characteristics, supporting the need for further analysis. 4.6 Regression analysis and model Regression models are empirical tools designed to detect and quantify the influence of one or more explanatory variables on a target variable (Singh & Chakraborty, 2021). These models assess how a change in one or more predictors impacts an outcome variable (Feng et al., 2017). The key components involved in regression analysis include independent variables, dependent variables, and control variables, each of which will be elaborated upon in subsequent sections. In this study, the dependent variable is firm performance, which is analysed through financial, operational, and market-based metrics. The main explanatory variable is the ESG disclosure score, complemented by several control variables designed to account for other significant elements that could affect the sustainability- firm performance relationship. A fixed effects panel data regression model was selected for this study. This approach controls for unobserved, firm-specific characteristics that do not vary over time but may differ across firms, such as management practices, organizational culture, or industry classification (Velte, 2017). The fixed effects method is particularly appropriate when unobservable firm-specific features are potentially correlated with the explanatory variables, making it ideal for panel data analysis where certain characteristics stay constant within firms but differ across them. By accounting for these time-invariant characteristics, the model helps to minimize bias in the estimated relationships. The primary objective of the regression model in this research is to investigate the effect of sustainability reporting, captured through ESG scores, on firm performance. Firm performance is assessed using thre