Board Characteristics, Ownership Structure, and Shareholder Activism as Determinants of Sustainability Transparency: Panel Data Analysis for Türkiye

Abstract

Using data from the 41 companies listed on the Borsa Istanbul Sustainability Index between 2020 and 2024, this study examines the relationship between board characteristics, concentrated ownership structure, shareholder activism and sustainability transparency. Reports published on a voluntary basis were subjected to content analysis based on criteria selected from the Global Reporting Initiative (GRI) standards, and a sustainability disclosure score was calculated. The relationship between board characteristics, concentrated ownership structure, shareholder activism, financial metrics identified as control variables, and sustainability scores was evaluated via robust random effects (static) and System Generalized Method of Moments (System GMM) (dynamic) panel data estimators. According to the static estimation results, board meeting frequency and the ratio of female members serve as positive drivers for sustainability transparency. In the dynamic model estimates, these governance mechanisms lose their explanatory power and show no statistically observable effect. However, across both methodological approaches, firm size, which was integrated as a control factor, consistently demonstrates a robust positive correlation with levels of disclosure. These findings reveal that governance mechanisms such as the percentage of female members and meeting frequency have a short-term and marginal effect, but structural factors such as company size are the main determinants for a long-term and sustainable level of transparency in Türkiye. Consequently, market regulators should deploy policy frameworks that incentivize disclosure trajectories aligned with international frameworks while fostering voluntary reporting. Concurrently, corporate managers should look beyond mere statutory compliance and continuously embrace extensive global reporting standards.

1. Introduction

At present, simply protecting shareholders’ assets and maximizing their wealth is not enough to ensure the continuity of a company, as companies are now viewed as platforms that take into account the rights and interests of various stakeholder groups, including partners, creditors, suppliers, employees, customers, the community, and the environment [1]. Managers are expected to manage the assets and resources entrusted to them so as to balance the benefits of shareholders with the benefits of all other stakeholders. This expectation has led to an increase in managerial responsibilities, thereby broadening the conceptual boundaries of corporate governance. In this context, corporate governance is defined as the establishment of a management structure that creates sustainable value for all stakeholders [2,3]. This is achieved by reconciling economic and social objectives, as well as individual and societal goals [4]. Achieving this alignment requires integrating the expectations of the entire stakeholder spectrum into core corporate strategies and operations. Furthermore, corporate governance “involves a set of relationships between the company’s management, board, shareholders, and stakeholders” [5]. From this perspective, “sustainable corporate governance” [6] can be defined as a system in which governing bodies oversee the value chain based on the principles of double materiality and the duty of care to strike a balance between short-term profits and long-term sustainability goals, while supporting performance through incentive systems and ensuring transparent reporting [7,8,9,10]. This system ensures that scarce resources are utilized efficiently, profitability targets are secured for shareholders, decent working conditions are provided for employees, natural resources are conserved, and broader societal benefits are generated.

The welfare of all stakeholders is inherently linked to the presence of a functional and effective corporate governance structure. In this context, the ownership structure, the qualitative and quantitative characteristics of the board of directors and senior management, executive compensation policies, the presence of independent members, investor relations, stakeholder participation, shareholder activism, and the board’s disclosure responsibility are key determining factors [1,11,12]. Corporate governance mechanisms, along with ownership structure and shareholder activism, are the most important factors influencing transparent reporting [5,9,11,12]. The board of directors is entrusted with the duty of maintaining transparency for both shareholders and the broader stakeholder community [5,9]. Companies inform their stakeholders by publishing financial reports, integrated reports, and sustainability reports. While financial reports are prepared in accordance with strict statutory mandates, specific frameworks for voluntary sustainability reporting are determined by corporate governing bodies [13]. When governing bodies align operational outcomes with corporate governance principles and foster an integrated thinking mindset, transparent reporting enhances business performance, drives sustainable value creation, facilitates efficient control, and preserves organizational legitimacy [14]. This is because integrated and sustainability reports serve as evidence of effective governance. The publication of such reports underscores a management strategy dedicated to optimizing investor returns while simultaneously upholding stakeholder interests. Communicating this orientation, which aims to improve performance for all stakeholders, through sustainability reporting serves to establish corporate legitimacy. However, establishing legitimacy extends beyond the simple act of publishing reports; the granular level of disclosure contained within them is of paramount importance. In other words, the desired outcomes can be achieved if all positive and negative results related to business activities are disclosed transparently in reports, whether required by law or published voluntarily. Given that the board of directors is entrusted with maintaining transparency for both shareholders and the broader stakeholder community, mechanisms such as board size, board independence, board gender diversity, and board meeting frequency, alongside corporate elements like ownership concentration and shareholder activism, become pivotal monitoring channels for ensuring transparency and reducing information asymmetry.

There is a significant research gap in the literature regarding how the voluntary disclosure strategies of businesses have changed following the regulations enacted in Türkiye in 2021 and 2024. Existing studies examining the Turkish sample have predominantly focused on the linear relationships between sustainability performance and financial performance, or they have evaluated the impact of corporate governance mechanisms within static environments. However, since the transition period from the implementation of the first regulation in 2021 to the present, the dynamic impact of board characteristics, ownership, and governance structures on voluntary disclosures has not been analyzed by controlling for path dependency or endogeneity issues stemming from past reporting habits. Therefore, the current literature is insufficient for explaining the dynamic effectiveness of board characteristics, ownership, and governance structures over time in voluntary reporting processes conducted under the shadow of government obligations, such as the Capital Markets Board of Türkiye (CMBT) [15] framework and the subsequent Türkiye Sustainability Reporting Standards (TSRS) [16,17,18]. In this context, the fundamental empirical question of this study is as follows: In an emerging market like Türkiye, during a mandatory regulatory transition period, do board qualifications, concentrated ownership structure, and shareholder activism truly drive long-term sustainability transparency, or are disclosure levels entirely determined by structural and financial capacities, such as firm size, profitability, and leverage ratio?

To systematically address this problem, this research utilizes data spanning 2020 to 2024 from 41 companies consistently traded on the Borsa Istanbul (BIST) Sustainability Index (XUSRD). While several studies have investigated the Türkiye sample, this one captures the exact five-year window of the market’s intense regulatory transition. This study moves beyond merely addressing a geographical gap or updating chronological data. As noted, Türkiye is a market dominated by family-owned firms, where shareholder activism is relatively new and in its nascent stages, and the institutional environment is driven by recent regulatory requirements aimed at standardizing corporate sustainability disclosures. Considering this environment, examining how corporate governance mechanisms interact with voluntary disclosure provides vital and generalizable insights into the boundary conditions of governance effectiveness in transition economies, significantly expanding the existing literature. This study offers three distinct contributions. First, from a methodological perspective, this study utilizes both static (random effects) and dynamic (System GMM) estimators; in doing so, it proves that governance mechanisms often lose their explanatory power in long-term dynamic settings, thereby effectively mitigating endogeneity and path-dependency issues that frequently bias static empirical findings. Second, it introduces a unique institutional insight by identifying a regulatory compliance paradox in transitioning regimes, illustrating how mandatory minimum compliance requirements can counterintuitively encourage firms to regress from comprehensive global frameworks, such as Global Reporting Initiative (GRI) standards, Sustainability Accounting Standards Board (SASB) standards, or the International Integrated Reporting Framework, toward narrower, standardized local disclosures. Finally, this study expands the boundary conditions of traditional agency and stakeholder theories within the context of emerging and transitioning economies. It does so by revealing that, while diverse board configurations may function primarily as tools for symbolic legitimacy to appease regulatory pressures, substantive sustainability disclosure operations remain structurally dependent on organizational scale, thus clearly delineating the practical limits of internal governance oversight.

Theoretically examining how the existing ecosystem shapes corporate behavior requires focusing on specific governance mechanisms that directly influence the disclosure strategy. To provide a structured and comprehensive analysis of these dynamics, the remainder of this study is organized into six interconnected sections. Section 2 provides the “Institutional Background of Türkiye,” establishing the regulatory and contextual baseline for corporate operations. Following this context, Section 3, “Governance and Sustainability Disclosure: Theory and Hypotheses,” delineates the theoretical frameworks and develops the core research hypotheses. Subsequently, Section 4 outlines the empirical methodology, sample selection, and variable definitions. Section 5 presents the empirical findings and regression results, which are then critically evaluated and discussed in Section 6. Finally, Section 7 concludes the study with policy implications and future research directions.

2. Institutional Background of Türkiye

While sustainability reporting historically emerged as a voluntary practice, it has increasingly transitioned into a mandatory statutory requirement globally. For example, the Institute of Directors in South Africa [19] mandated publicly traded companies to publish integrated reports from 2010 onwards as part of the King III Report; and the European Commission [20] made it mandatory for companies meeting certain criteria to report on their sustainability performance from 2024 onwards. In Türkiye, while companies previously disclosed their corporate performance on a voluntary basis in accordance with globally accepted reporting standards, frameworks, or guidelines, sustainability reporting only became mandatory in 2021 [16,21,22].

According to the provisions of the Communiqué on Amendments to the Communiqué on Corporate Governance, prepared by the CMBT and published in the Official Gazette dated 2 October 2020, companies subject to corporate governance principles are also subject to sustainability principles, and these companies are required to publish a sustainability principles compliance report as of 2021 [15,22]. The Capital Markets Boards of Türkiye [15] has ruled that companies are required to make statements in four sections in their reports—namely, general, environmental, social, and corporate governance principles—based on the “Comply or Explain” principle. In this context, companies must present the principles included in the general principles, environmental principles, social principles, and corporate governance principles to stakeholders by indicating their compliance status as yes, no, partially, or irrelevant and by providing the necessary explanations [23].

Another milestone regulation regarding sustainability disclosures in Türkiye is the TSRS 1 [17] and the TSRS 2 [18], referencing international standards. These standards were published in the Official Gazette dated 29 December 2023, and have entered into force [16]. The Public Oversight Accounting and Auditing Standards Authority (POA) [16] has sustainability reports in accordance with the TSRS mandatory for publicly traded companies that exceed at least two of the following criteria in two consecutive reporting periods—(a) total assets of 500 million Turkish Lira, (b) annual net sales revenue of 1 billion Turkish Lira, and (c) 250 employees.

The obligations to publish both a sustainability principles compliance report and a TSRS-compliant sustainability report apply to publicly traded companies. These enterprises are stringently monitored and audited to establish and maintain a governance structure that strictly aligns with corporate governance principles. Therefore, Türkiye is a suitable environment for corporate governance research. Türkiye serves as a critical institutional laboratory for this study due to three fundamental pillars. Firstly, the market is characterized by highly concentrated, family-dominated ownership structures [24], where family firms constitute the backbone of the economy of Türkiye [25]. Secondly, owing to this concentrated ownership and characteristically low free-float ratios, minority shareholder activism remains a relatively nascent and developing phenomenon [26]. Lastly, the institutional environment is shaped by strict state-led regulatory mandates, such as the Sustainability Principles Compliance Framework [15] and the recent enforcement of the TSRS [17,18], which standardize and institutionalize corporate sustainability disclosures. Collectively, these three pillars provide Türkiye with a unique corporate ecosystem.

3. Governance and Sustainability Disclosure: Theory and Hypotheses

In the literature, the relationship between corporate governance mechanisms and the level of sustainability disclosure is explained through agency theory, legitimacy theory, and stakeholder theory [27,28,29,30,31,32,33,34,35,36,37]. These theories are complementary approaches that support corporate transparency from different perspectives.

Agency theory explains the relationship between shareholders—referred to as principals—and managers—referred to as agents—who are granted the right to manage the shareholders’ assets [38,39,40]. In this association, an agency relationship exists between the principal and the agent, and agency costs are incurred to harmonize and maximize the interests of both parties [31,35,39,41]. The structuring of the board of directors, the financial benefits offered to directors, and the costs of reporting and meeting processes with stakeholders, and also reporting practice, can be considered within the scope of agency costs, which ensure the functionality and oversight of the corporate governance system [5,11,12,42,43]. Within this framework, sustainability-related investments may initially be perceived as an additional corporate cost [35]. However, the presence of robust corporate governance mechanisms that enhance the transparency of sustainability disclosures effectively reduces information asymmetry and contributes to informed decision-making among stakeholders [33,43].

The principal–agent conflict, which is the focus of agency theory, has been complemented in the modern corporate governance framework by a broader paradigm: stakeholder theory. Stakeholder theory argues that in addition to shareholders, organizations interact with various constituent groups, and that managers should harmonize the interests of all stakeholders, including shareholders [35,44,45]. Managers are responsible for meeting the expectations of internal and external stakeholder groups, regulating relationships among stakeholders, and aligning the interests of the company and its stakeholders using strategic management tools [5]. This responsibility is essential for ensuring corporate sustainability. According to stakeholder theory, a company’s performance and survival require continuous stakeholder support, which is achieved through active dialog with these groups. According to stakeholder theory, a company’s performance and survival require continuous stakeholder support, which is achieved through active dialog with these groups [35,36,46,47]. Therefore, stakeholder theory supports the adoption of risk management policies, the implementation of sustainability practices, and the provision of sustainability disclosures [33]. In this context, sustainability reports serve as strategic management tools used to demonstrate that the company’s risk management policies and sustainability practices align with the interests of all stakeholders, thereby establishing a meaningful dialog with them.

By harmonizing and meeting stakeholders’ expectations, the company’s legitimacy will be established, maintained, or enhanced. Essentially, legitimacy theory is built on a social contract that is assumed to exist between a company and society [48]. Within this social contract, legitimacy is established through the generally accepted perception that the company acts in accordance with societal norms and values [35,49,50]. It is important for a company to plan activities that align with social norms and for society to accept this; however, acceptance is not limited to adherence to norms alone [49]. Suchman [49] divides legitimacy into institutional and strategic legitimacy, emphasizing that, within the scope of strategic legitimacy, legitimacy is related not only to conforming to norms and values but also to how businesses explain their activities and how they inform society about their activities. Accordingly, companies seeking to gain or increase legitimacy must undertake activities that are accepted by society and provide disclosure regarding the results of their activities. In this context, sustainability reports are a strategic tool used to prove that companies are fulfilling their social contract obligations [36,48,51,52].

To operationalize these theoretical perspectives and internalize transparency, the board of directors serves as the core apex governing body within the corporate architecture. Positioned at the intersection of agency, stakeholder, and legitimacy requirements, the board of directors is the highest governing body that “manages and represents the company by making strategic decisions, maintaining the balance of risk, growth and return at the most optimal level, and considering the long-term interests of the company with a rational and prudent risk management approach” [21]. The fundamental mission of the board is to generate enduring value for shareholders in the long run via a proactive management philosophy while simultaneously fulfilling its responsibilities toward all stakeholders to safeguard and enhance corporate reputation [53]. To execute this overarching mission, the board possesses distinct statutory duties and oversight powers mandated by legal regulations and the corporate articles of association. In this context, the structural configurations of the board are pivotal in determining its efficiency in performing these mandated functions. Furthermore, integrating the underlying ownership structure alongside the specific expectations of shareholders and broader stakeholders into goal-setting, implementation, and auditing processes emerges as a critical determinant. Collectively, this mission, alongside these institutional duties, powers, and governance elements, directly shapes the transparency of corporate disclosures. Consequently, to systematically analyze the factors driving the extent of a firm’s voluntary disclosures, it is essential to examine structural board dimensions (such as size, independence, and diversity) in combination with systemic ownership concentration and shareholder activism. In line with the literature, the hypotheses developed are presented below under subheadings.

3.1. Board Size

As a fundamental structural dimension of corporate governance, board size reflects the total capacity of the board of directors to execute its strategic oversight and resource provision functions, and it is empirically operationalized by the total number of members serving on the board. An increase in board size could weaken the board’s oversight over senior executives, thus potentially increasing agency costs [41]. Contrary to the view that large boards of directors increase agency costs, size can also reduce agency costs by facilitating the alignment of stakeholder interests [42,54]. Larger boards of directors ensure better representation of stakeholder expectations on the board [31]. Furthermore, an increase in board size enables better oversight of decisions made on sustainability issues [55]. Consequently, it enhances the level of disclosure in sustainability reports [43,52,55,56,57]. Thereby improving corporate reputation [35,57]. In addition, board size can reduces the likelihood of a company being involved in sustainability scandals [41].

Studies examining whether board size has an impact on sustainability disclosures show that there is a significant and positive relationship between these two variables [31,35,43,54,55,56,57,58,59,60]. These studies show that the level of sustainability disclosure increases as the number of board members increases. Other studies have found a significant and negative relationship between board size and sustainability disclosure level [61] but some have not found a significant relationship [30,36,37,62].

When evaluated within a theoretical framework, a larger board size allows for a broader range of member competencies to address stakeholder needs and expectations, thereby enhancing corporate reputation and image. Conversely, expanding the board primarily to project an image of stakeholder inclusivity may result in excessive bureaucratic burden and impede decision-making processes, ultimately reducing operational efficiency. In conclusion, the impact of board size on sustainability disclosure is shaped by the balance between the expanded stakeholder representation provided by the board and the potential efficiency losses encountered in decision-making processes. In this context, the first hypothesis is established as follows:

Hypothesis 1.

There is a statistically significant relationship between board size and the level of sustainability disclosure.

3.2. Board Independence

Another governance mechanism examined for its impact on sustainability disclosure is the independence of the board. Within the scope of board diversity, an independent director is a member who is free from conflicts of interest and makes decisions independently, without being influenced by external pressures [9]. Independent board members are assumed to protect stakeholder groups and will enhance company legitimacy by reducing information asymmetry and increasing transparency and accountability [28,30,36,42,43,51,54,56,57,63,64,65].

The literature presents mixed evidence regarding the influence of board independence on disclosure level. A significant body of research suggests that having more independent directors enhances sustainability transparency [28,29,30,34,43,51,66,67]. Conversely, some studies have identified a negative correlation, implying that increased board independence may lead to a reduction in the scope of disclosures [47,61,68]. Furthermore, a third group of researchers found no statistically significant link, suggesting that board independence does not inherently drive sustainability reporting behavior [31,36,54,57,69].

In light of these diverse findings, the following hypothesis is proposed:

Hypothesis 2.

There is a statistically significant relationship between board independence and the level of sustainability disclosure.

3.3. Board Gender Diversity

It has been stated that the proportion of female members, considered within the context of board diversity, has an impact on sustainability disclosures. According to Galbreath [70], the appointment of women to the board promotes social responsibility by enabling more effective oversight of managers, strengthening adherence to ethical standards and developing a management approach that is more responsive to stakeholder needs. Post et al. [67] argue that women are more concerned about environmental issues than men and improve corporate performance through sustainability-focused strategic alliances. Female directors encourage sustainability disclosures, thereby reducing information asymmetry and agency costs [64,71]. Boards of directors with a high proportion of female members exhibit a higher level of sustainability disclosure, thus better meeting the disclosure requirements of stakeholders [63,65].

The assignment of female members is expected to improve sustainability performance as well as the level of disclosure related to that performance. This expectation is supported by findings showing that companies with a high proportion of female board members also have higher levels of social, environmental, economic, and governance disclosures [28,34,35,47,55]. Some research suggests that board gender diversity exerts a neutral impact, showing that the presence of female directors neither promotes nor inhibits the extent of sustainability reporting.

Despite the prevailing view that female board representation fosters broader sustainability transparency, empirical inconsistencies persist in the literature. To address this ambiguity, the relationship is examined under the following hypothesis:

Hypothesis 3.

There is a statistically significant relationship between the proportion of female board members and the level of sustainability disclosure.

3.4. Board Meeting

Board meetings are where the company’s strategic plans are prepared. Taking into account stakeholder expectations, operational efficiency is monitored periodically, thus reducing agency costs and responding to stakeholder expectations. While the decisions made and implemented at these meetings can affect the financial result [72,73], some findings show that the frequency of meetings has an effect on sustainability [31,74].

The influence of board meeting frequency on sustainability performance is inherently linked to disclosure levels, as these internal governance deliberations are eventually communicated to stakeholders through formal reporting. Board of directors meetings serve as primary channels for providing information flows to company management [31]. Regularly conducted board meetings not only ensure efficiency in decision-making and monitoring subsequent outcomes but also facilitate the transparent disclosure of these decisions and results [65]. In this context, there is an opinion that the frequency of meetings improves the quality of disclosure on social, environmental and governance issues [34]. Caputo et al. [69] observed that, while environmental transparency is generally shaped by corporate governance and reporting characteristics, the frequency of board meetings remains statistically insignificant regarding disclosure levels.

Within a theoretical perspective, the frequency of board meetings is considered both a strategic tool that strengthens audit effectiveness and stakeholder engagement, and a symbolic process that, if lacking in quality, can hinder operational efficiency. These differing assessments indicate that the balance between the potential contributions of meeting frequency to sustainability transparency and the accompanying administrative costs has not yet been clearly established. In this context, another hypothesis is formulated as follows:

Hypothesis 4.

There is a statistically significant relationship between the frequency of board meetings and the level of sustainability disclosure.

3.5. Concentrated Ownership Structure

In the literature, one of the governance mechanisms examined within the context of sustainability disclosure is concentrated ownership structure. Majority control, where a large majority of shares are concentrated in one hand, can reduce agency costs [39,75] and provide efficiency in management control [42]. However, due to the concentrated ownership structure, the major shareholder may also wish to maximize its own benefits rather than, or even before, the benefits of others [76]. Companies seeking to maintain or enhance their legitimacy are expected to disclose performance results through corporate reports demonstrating that they consider the interests of major shareholders and that the company is managed accordingly. Transparency in sustainability disclosures increases if company owners adopt an orientation that prioritizes sustainability and the fulfillment of stakeholder expectations [62,77]. Conversely, a concentrated ownership structure may reduce companies’ sustainability disclosure levels by diminishing public pressure [78,79]. Ho & Taylor [80] found that the concentrated ownership structure increases the level of voluntary social responsibility disclosure, while Said et al. [62] found that sustainability disclosure decreases in companies where the majority of shares are held by one person or institution. However, Eng & Mak [68] and Erben Yavuz et al. [35] found that ownership structure has no effect on voluntary disclosures.

Thus, different studies examining the relationship between ownership structure and sustainability disclosures have produced differing findings. Therefore, the hypothesis proposed in this research is presented below:

Hypothesis 5.

There is a statistically significant relationship between concentrated ownership structure and the level of sustainability disclosures.

3.6. Shareholder Activism

Shareholder activism refers to the strategies and actions of shareholders used to bring about changes in company management to achieve a specific goal [81]. Shareholders can express their expectations to the company management by using their rights arising from share ownership in general assembly meetings, other meetings held by the company with its shareholders, personal meetings, or public relations activities [81]; in this manner, they can actively shape corporate operations and transparency practices. Shareholder activism is a critical factor that directs corporate activities [82]. One of the key areas that shareholders influence through activism is the sustainability impacts and disclosures of companies [82,83].

Flammer et al. [84] concluded that, when environmental shareholder activism is initiated by institutional investors, it increases the level of voluntary disclosure. Alessa et al. [83] found that shareholder activism increases the level of sustainability disclosure by supporting green technological innovation. Furthermore, Nur DP et al. [85] emphasized the role of stakeholder engagement in achieving transparency in sustainability disclosures. Yang et al. [86] stated that shareholder proposals only increase corporate social responsibility disclosures at the end of the year in which the proposal is made, but this increase is not observed in subsequent years. Similarly, a study by Alfian and Ismi [87] on the Indonesian banking sector revealed that shareholder participation exerted no significant influence on sustainability disclosure levels.

In a theoretical framework, meetings with shareholders play a role in encouraging sustainability disclosures when environmental and social performance is questioned by shareholders. The effectiveness of shareholder meetings in promoting sustainability statements depends on shareholders being aware of or concerned about sustainability issues. In the absence of such awareness or concern, it is possible to say that shareholder meetings, or shareholder activism in general, will have no impact on sustainability statements. Based on the literature findings, the proposed hypothesis regarding shareholder activism is as follows:

Hypothesis 6.

There is a statistically significant relationship between shareholder activism and the level of sustainability disclosure.

4. Materials and Methods

4.1. Sampling Selection

This study aims to analyze the impact of governance mechanisms on the level of sustainability disclosure and includes companies listed on the BIST Sustainability Index (XUSRD) as of January 2026. Borsa Istanbul [88] announced that companies with high sustainability performance are listed on XUSRD in order to serve as an example for practices within the scope of sustainability. Companies with high sustainability performance report information on their performance both legally and voluntarily, in accordance with global standards. Therefore, it was decided to use the sustainability disclosures of companies listed on the XUSRD as the basis for this analysis.

To ensure time-series continuity, comparability, and a balanced panel structure, the final sample was constructed through a strict, chronological multi-step filtering logic. According to Public Disclosure Platform [89] data, the initial research universe consisted of 98 companies traded on the XUSRD as of January 2026. In the first stage, financial institutions and insurance companies were excluded from the universe because their business models, strict banking regulations, asset structures, and financial reporting standards fundamentally differ from real-sector firms, reducing the potential sample to 68 companies. In the subsequent stage, an additional 27 companies were excluded due to a lack of consistent, continuous data across the analyzed timeline or because they were listed late during the target timeframe. This chronological reduction resulted in a final balanced panel dataset of 41 firms (N = 41) that consistently published voluntary corporate sustainability disclosures in accordance with globally recognized frameworks. The study examines these 41 companies over a 5-year period covering 2020–2024 (T = 5), yielding a total of 205 firm-year observations. Within this empirical framework, 2025 serves as the strict temporal cutoff point of the study to capture the most recent and fully completed disclosure cycles.

4.2. Variables and Data Collection

The sustainability disclosure score (DSCORE) is the dependent variable in this study. To calculate the DSCORE, scoring was performed using the content analysis method, a frequently used method for measuring sustainability scores through secondary data [90]. The sectoral diversity of the companies in the sample was considered in selecting the criteria to be used in the content analysis. To ensure measurement uniformity and comparability, content analysis was carried out by selecting 18 key criteria that are universal and commonly applicable to all sectors from the current standards published by the Global Reporting Initiative [91]. The criteria used in conducting the content analysis are presented in

Table 1. Metrics used in calculating the sustainability disclosure score.

Sustainability Dimension	GRI Standards Used as Metrics	Key Indicators Confirming the Existence of the Criterion
Economic	GRI 201: Economic Performance	Net sales, operating expenses, business expenses, employee wages, tax payments, dividends paid, distributed economic value
	GRI 202 Market Presence	Local employment, local minimum wage rate by gender, local administrator
	GRI 203 Indirect Economic Impacts	Infrastructure investment, public benefit, support service.
	GRI 204: Procurement Practices	Local supplier rate, locally sourced spending, local purchasing rate
	GRI 205: Anti-corruption	Risk of corruption, anti-corruption policy, anti-corruption training
	GRI 206: Anti-competitive Behavior	Monopolization, unfair competition policies
Environmental	GRI 301: Materials	Raw materials, recycled inputs, packaging, input materials
	GRI 302: Energy	Energy consumption (all renewable and non-renewable energy types consumed internally and externally), energy intensity, energy efficiency
	GRI 303: Water and Effluents	Water extraction, water consumption, water discharge, water recovery, groundwater
	GRI 305: Emissions	Scope 1, Scope 2, Scope 3, emission intensity, emission reduction
	GRI 306: Waste	Hazardous waste, non-hazardous waste, disposal, recycling, water consumption, wastewater
	GRI 308: Supplier Environmental Assessment	Supplier environmental audit
Social	GRI 401: Employment	Number of employees, new hires, employee turnover rate, benefits, parental leave, employee demographics
	GRI 403: Occupational Health and Safety	Workplace accidents, lost days, occupational health and safety measures, occupational health and safety training, occupational disease, accident frequency rate, number of accidents
	GRI 404: Training and Education	Talent management, performance system, total training hours, training hours per employee, performance appraisal
	GRI 405: Diversity and Equal Opportunity	Female employees, female managers, gender-based pay rates, equality and inclusion
	GRI 413: Local Communities	Supporting local communities, community contribution, social responsibility, and social impact assessment of the business
	GRI 414: Supplier Social Assessment	Supplier social audit
Source: Global Reporting Initiative [91]

.

The reports were published by a sample of 41 companies between 2020 and 2024; these were integrated, sustainability, integrated sustainability, and integrated annual reports. They were rigorously analyzed and scored according to the comprehensive criteria outlined in Table 1. The scoring system used by Bewley & Li [27] and Kidaye & Saoussany [90] was used as the basis for the scoring. Accordingly, 0 points were awarded if no explanation was given on the fundamental issues that prove the existence of the criterion, 1 point if the explanation was given in general terms, 2 points if only qualitative or only quantitative explanation was given regarding the criterion, and 3 points if both qualitative and quantitative or monetary explanation was given regarding the criterion. After scoring, the company’s explanation score for that year was calculated as follows [27,90]:

$$DSCORE=company\mathrm{’}s total score / maximum score that can be obtained$$

(1)

In this equation, the total score represents the sum of the points awarded for the disclosures made by the company as a result of the content analysis, while the maximum score that can be obtained represents the highest score that can be received if both qualitative and quantitative or monetary disclosures are made for all criteria (18 criteria) (18 criteria × 3 points = 54 points). Furthermore, to estimate dynamic panel models and control for potential historical reporting behavior, the lagged dependent variable (LAG_DSCORE), representing the sustainability disclosure score from the previous fiscal year (t − 1), is integrated into the empirical specification for use in subsequent dynamic frameworks.

To ensure the robustness of the DSCORE variable and shield the manual content analysis against criticisms of subjectivity, a strict double-scoring validation protocol was implemented. To this end, a second independent coding round of corporate sustainability reports was conducted across all 205 firm-year observations for the 18 GRI criteria following a predetermined time interval. To rigorously verify intra-coder reliability (representing the longitudinal consistency between the two coding rounds), Krippendorff’s alpha (α) coefficients were estimated for each criterion by employing the empirical bootstrap sampling distribution framework introduced by Hayes and Krippendorff [92]. As presented in

Table 2. Krippendorff’s alpha results.

GRI Criterion Code	Krippendorff’s Alpha (α)	GRI Criterion Code	Krippendorff’s Alpha (α)
GRI 201	0.7603	GRI 305	1.0000
GRI 202	0.7096	GRI 306	1.0000
GRI 203	0.8937	GRI 308	0.9726
GRI 204	0.8950	GRI 401	1.0000
GRI 205	0.9647	GRI 403	1.0000
GRI 206	0.9659	GRI 404	0.9610
GRI 301	0.9796	GRI 405	0.9534
GRI 302	1.0000	GRI 413	0.9375
GRI 303	1.0000	GRI 414	1.0000

, the calculated Krippendorff’s α coefficients range from 0.7096 to 1.0000. According to the authoritative reliability benchmarks established by Hayes and Krippendorff [92], alpha values above 0.70 and 0.80 represent good and excellent data reliability, respectively. Notably, for criteria such as GRI 302, GRI 303, GRI 305, GRI 306, GRI 401, GRI 403, and GRI 414, perfect longitudinal consistency (α = 1.0000) was achieved due to identical cross-scoring and zero baseline variation between the two iterations. Collectively, these reliability coefficients decisively demonstrate that the content analysis protocol is highly stable, reproducible, and free from subjective coding bias.

The independent variables in the study are board size (BOSIZE); the proportion of independent board members (BOIN); the proportion of female board members (FEMEM); the number of board meetings (BOMEET); the number of meetings held with shareholders (SHMEET), which represents shareholder activism; and the ownership ratio of the largest shareholder (LARSH), which represents concentrated ownership. Data on independent variables were obtained from reports published annually by companies under the title “Corporate Governance Information Form”, which were accessed from Public Disclosure Platform [89] company notifications. BOSIZE is operationalized as the aggregate count of directors serving on the firm’s board. The ratio of BOIN is the ratio of independent board members to the total number of board members, and the ratio of FEMEM is the ratio of female board members to the total number of board members. Data on the variables BOMEET, SHMEET, and the LARSH were obtained from the Corporate Governance Information Forms published by the companies in the relevant year. Significant differences were observed between the raw data of the variables BOMEET and SHMEET from the companies in the sample. To control for the effects of outliers and improve model consistency, a logarithmic transformation was applied to these values. As the variable for SHMEET contained zero values in ten observations, the ln(x + 1) transformation, which is accepted in the literature [93], was applied to both the SHMEET and BOMEET. This procedure was manually executed to accommodate zero values and prevent data loss, thereby ensuring the robustness of the subsequent analysis.

The control variables of the study were return on assets (ROA), leverage ratio (LEV), and firm size (SIZE), in accordance with the literature [29,31,32,33,34,35,59]. Due to their heightened visibility and exposure to public scrutiny, larger firms (SIZE) exhibit a greater propensity to provide extensive disclosures in order to maintain or enhance their organizational legitimacy [33,50,59,94]. ROA captures corporate profitability; highly profitable firms possess greater financial capacity to allocate resources toward sustainability disclosures to protect or amplify their legitimacy, despite the additional costs [32,50]. Lastly, LEV is incorporated as a control variable because highly leveraged firms may increase their sustainability disclosures by intensifying their responsiveness to transparency requirements demanded by creditors, aiming to demonstrate their debt-servicing capabilities and secure or maintain access to external funding under favorable conditions [32,79]. Regarding their operationalization, the ROA variable was calculated as the ratio of net profit to total assets for each respective year. The LEV was calculated by scaling total liabilities against total assets. The SIZE variable was calculated using the natural logarithm of total assets.

The symbols, operational definitions, and measurement criteria of the variables utilized in this study are summarized in

Table 3. Definitions, symbols, and operationalization of the study variables.

Variable Type	Variable Name	Symbol	Measurement/Operationalization
Dependent Variable	Sustainability Disclosure Level	DSCORE	The overall score or index based on the content analysis of sustainability/integrated reports.
Lagged Dependent Variable	Lagged Sustainability Disclosure Level	LAG_DSCORE	The sustainability disclosure score (DSCORE) from the previous fiscal year (t − 1)
Independent Variables	Board Size	BOSIZE	Total number of directors serving on the board of directors.
	Board Independence	BOIN	The proportion of independent directors to the total number of board members (Independent Directors/Board Size).
	Board Gender Diversity	FEMEM	The percentage of female directors on the board (Female Directors/Board Size).
	Board Meeting Frequency	BOMEET	The total number of formal board meetings held during the fiscal year.
	Concentrated Ownership Structure	LARSH	The percentage of shares held by the largest shareholder.
	Shareholder Activism	SHMEET	Number of meetings held with shareholders during the year.
Control Variables	Firm Size	SIZE	The natural logarithm of total assets.
	Firm Profitability	ROA	Return on Assets (Net Income/Total Assets).
	Leverage	LEV	Total Debt divided by Total Assets (Financial Leverage ratio).

.

4.3. Model Specification and Estimation Method

In this study, panel data analysis was chosen to test the relationship between corporate governance mechanisms and sustainability disclosure level. Analyses were performed in the RStudio (v2024.12.1+563) interface using the open-source R statistical reporting language (v4.4.3). The plm library was used for estimating panel data models, lmtest for applying significance and assumption tests of the model, car for multicollinearity analysis, and stargazer for reporting the findings.

The static panel data model established within the scope of the research is presented in Equation (2).

$$\begin{array}{cc}\hfill {DSCORE}_{it}=& {\beta }_0+{\beta }_1{BOSIZE}_{it}+{\beta }_2{BOIN}_{it}+{\beta }_3{FEMEM}_{it}+{\beta }_4{BOMEET}_{it}+{\beta }_5{SHMEET}_{it}+{\beta }_6{LARSH}_{it}+{\beta }_7{ROA}_{it}\hfill \\ & +{\beta }_8{LEV}_{it}+{\beta }_9{SIZE}_{it}+{\mu }_i+{ϵ}_{it}\hfill \end{array}$$

(2)

To test the suitability of the data for panel data analysis, the Breusch–Pagan Lagrange Multiplier (LM) test was performed. Following the test, the Hausman test was applied to test the correlation of unit effects with explanatory variables and to choose between fixed effects (FE) and random effects (RE) models. Based on the Hausman test results (p > 0.05), the random effects model was found to be appropriate. As part of the assumption checks, the homoscedasticity assumption was checked using the Studentized Breush–Pagan test, and the autocorrelation assumption was checked using the Breusch–Godfrey/Wooldridge test. To increase the reliability of predictions from diagnostic tests, robust HC3-type standard errors were used in the analyses.

To account for the dynamic nature of sustainability disclosure, which is influenced by past performance, and to control for potential endogeneity issues within corporate governance variables, a Two-step System-Generalized Method of Moments (GMM) analysis was employed by incorporating the lagged value of the dependent variable (LAG_DSCORE). To mitigate bias resulting from instrument proliferation and to maintain the power of the Sargan/Hansen tests, the instrument set was restricted by collapsing the lags of the dependent variable (collapse = TRUE). The validity of the models was confirmed using the Sargan test to verify the exogeneity of the instruments and the AR(2) test to ensure the absence of second-order autocorrelation. In this context, the final dynamic panel data model is presented in Equation (3):

$$\begin{array}{cc}\hfill {DSCORE}_{it}=& {\delta DSCORE}_{it-1}+{\beta }_1{BOSIZE}_{it}+{\beta }_2{BOIN}_{it}+{\beta }_3{FEMEM}_{it}+{\beta }_4{BOMEET}_{it}+{\beta }_5{SHMEET}_{it}+{\beta }_6{LARSH}_{it}\hfill \\ & +{\beta }_7{ROA}_{it}+{\beta }_8{LEV}_{it}+{\beta }_9{SIZE}_{it}+{\mu }_i+{ϵ}_{it}\hfill \end{array}$$

(3)

5. Results

5.1. Descriptive Statistics

The descriptive properties of the variables employed in this study are detailed in

Table 4. Summary statistics for the study variables.

Variables	N	Mean	St. Dev.	Min	Max
DSCORE	205	0.6678	0.0855	0.3333	0.8333
BOSIZE	205	8.8244	2.4149	5.0000	16.0000
BOIN	205	33.2303	7.5928	14.2857	60.000
FEMEM	205	16.7979	11.7146	0.0000	50.000
BOMEET	205	2.1183	0.8622	0.0000	4.5109
SHMEET	205	4.1918	1.6440	0.0000	6.7742
LARSH	205	48.2240	16.9893	9.0600	99.9000
ROA	205	0.0791	0.0833	−0.2086	0.3071
LEV	205	0.6037	0.1791	0.1548	1.0191
SIZE	205	24.3488	1.7160	20.7867	30.7283

, covering the central tendencies and distributions of the dataset.

The descriptive statistics show that the mean DSCORE is 0.6678. This value indicates that the companies in the sample have an above-mean performance according to the measurement criteria. The minimum DSCORE, calculated as a minimum of 0.3333 and a maximum of 0.8333, shows that some of the companies in the sample are conservative in reporting their sustainability performance and disclose less information, while others disclose more information regarding their sustainability performance. Notably, the observation that no corporate entity within the sample achieved the maximum possible transparency score of 1.0000 carries significant institutional implications. Given that the dataset exclusively comprises firms listed on the Borsa Istanbul (BIST) Sustainability Index, which represents the most progressive enterprises in Türkiye regarding sustainability compliance, this ceiling effect indicates that substantial room for improvement remains. Even among these leading market actors, further effort is required to enhance the qualitative and quantitative depth of their environmental and social disclosures. Descriptive results for BOSIZE reveal a mean of 8 directors. The observation that no firm falls below 5 members underscores compliance with the Capital Markets Boards of Türkiye [21]. The representation of FEMEM averages 16.79%. The female representation rate, ranging from a minimum of 0 to a maximum of 50%, reveals that progress in female representation varies among companies. Specifically, a minimum value of 0 indicates a lack of gender equality in certain boards of directors, while a maximum value of 50% reflects a promising trend towards greater board diversity. At the same time, companies in the sample had a concentrated ownership structure with an average of 48.22%, operating with an average return on assets of 0.0791 and a leverage ratio of 0.6037. Finally, the structural scale of the sampled enterprises exhibits significant variation, with Firm Size (SIZE) demonstrating a widespread distribution characterized by a mean value of 24.3488 and ranging from a minimum of 20.7867 to a maximum of 30.7283.

5.2. Correlation Matrix and VIF Values

Table 5. VIF values and correlation matrix.

Variables	VIF	DSCORE	BOSIZE	BOIN	FEMEM	BOMEET	SHMEET	LARSH	ROA	LEV	SIZE
DSCORE	-	1
BOSIZE	1.519	0.039	1
BOIN	1.392	−0.021	−0.431 ***	1
FEMEM	1.106	0.112	−0.171 **	0.129 *	1
BOMEET	1.181	0.249 ***	−0.138 **	0.261 ***	−0.020	1
SHMEET	1.228	0.054	0.042	0.055	0.229 ***	−0.161 **	1
LARSH	1.241	−0.025	0.051	0.077	−0.134 *	0.203 ***	−0.251 ***	1
ROA	1.341	0.037	0.183 ***	−0.239 ***	−0.030	0.053	0.040	−0.195 ***	1
LEV	1.354	−0.270 ***	0.041	−0.068	−0.009	−0.188 ***	0.180 ***	−0.211 ***	−0.249 ***	1
SIZE	1.354	0.196 ***	0.358 ***	−0.081	−0.011	−0.080	0.145 **	0.101	−0.074	−0.212 ***	1

*** p < 0.01, ** p < 0.05, * p < 0.1.

presents the correlation matrix and Variance Inflation Factor (VIF) statistics employed to assess potential multicollinearity among the independent variables. The VIF values range from 1.106 to 1.519, well within the conservative threshold of 2.5 recommended by the literature [95]. These results confirm that multicollinearity does not compromise the reliability of the regression estimates. Regarding the pairwise correlations, the sustainability disclosure score exhibits significant associations with the frequency of board meetings (BOMEET, r = 0.249, p < 0.01), firm size (SIZE, r = 0.196, p < 0.01), and the leverage ratio (LEV, r = −0.270, p < 0.01). Consistent with Tabachnick & Fidell [96], all pairwise coefficients remain below the acceptable 0.70 threshold, mitigating risks regarding multicollinearity and supporting the statistical validity of the following empirical tests.

5.3. Model Selection and Diagnostic Tests

Table 6. Results of model selection and diagnostic tests.

Test	Statistics	p-Value
Breusch–Pagan(LM) test χ2	41.442	0.000
Hausman Test χ2	14.481	0.106
Studentized Breusch–Pagan χ2 (49)	62.702	0.090
Breusch–Godfrey/Wooldridge χ2 (5)	9.521	0.090

presents the results of model selection tests conducted to determine the suitability of the data for panel data analysis and the type of model, as well as diagnostic tests to identify heteroskedasticity and autocorrelation issues. The Breusch–Pagan LM test results (χ² = 41.442, p < 0.01) confirm that the model is suitable for panel data analysis. The Hausman test was conducted to determine which model was more appropriate for panel data analysis: fixed effects or random effects. The results of the Hausman test (χ² = 14.481, p = 0.106) show that the random effects model provides more consistent estimates. According to this result, a random effects model was used in the model estimation. The Studentized Breusch–Pagan test result (χ² = 62.702, p = 0.090) indicates the absence of variance issues, and the Breusch–Godfrey/Wooldridge test (χ² = 9.521, p = 0.090) suggests that the model is free from autocorrelation.

5.4. Corporate Governance Mechanisms and Sustainability Transparency

Empirical findings on the relationship between sustainability disclosure level and corporate governance mechanisms are presented in

Table 7. Panel regression results for the research models.

Independent Variables	Model 1: RE (Robust)	Model 2: GMM-1	Model 3: GMM-2	Model 4: GMM-3
LAG_DSCORE	-	0.2082 (0.1416)	0.2125 (0.1680)	0.2032 (0.1712)
BOSIZE	−0.0013 (0.0031)	0.0012 (0.0043)	0.0013 (0.0042)	−0.0008 (0.0047)
BOIN	−0.0008 (0.0009)	0.0014 (0.0014)	0.0013 (0.0015)	0.0009 (0.0014)
FEMEM	0.0009 (0.0004) *	0.0008 (0.0007)	0.0008 (0.0007)	0.0007 (0.0006)
BOMEET	0.0285 (0.0084) ***	0.0118 (0.0151)	0.0122 (0.0192)	0.0161 (0.0191)
SHMEET	0.0042 (0.0041)	0.0028 (0.0059)	0.0029 (0.0068)	0.0030 (0.0072)
LARSH	−0.0007 (0.0004)	0.0001 (0.0004)	0.0001 (0.0004)	0.0000 (0.0004)
ROA	−0.0116 (0.0761)	0.0180 (0.0638)	0.0163 (0.0692)	0.0303 (0.0624)
LEV	−0.1105 (0.0441) *	0.0265 (0.0460)	0.0252 (0.0503)	0.0382 (0.0476)
SIZE	0.0130 (0.0043) ***	0.0168 (0.0041) ***	0.0168 (0.0049) ***	0.0177 (0.0048) ***
Intercept	0.3977 (0.1195) ***	-	-	-
N	205	205	205	205
Instrumental variable lag	-	(t − 2, t − 4)	(t − 2, t − 3)	(t − 2, t − 2)
R2	0.2303
Wald χ2 (Prob)	88.446 (0.0000) ***	10,299.46 (0.0000) ***	10,639.26 (0.0000) ***	11,912.98 (0.0000) ***
Sargan Test (p-value)	-	0.2453	0.1867	0.3398
AR(2) Test (p-value)	-	0.2539	0.2581	0.2837

Values in parentheses are robust standard errors. *** p < 0.01, * p < 0.10. “-“ indicates that the statistic is not reported or dropped due to the econometric specification of the model.

. Model 1 is a Random Effects model and was estimated using robust HC3 standard errors. Although diagnostic test results showed no heteroskedasticity or autocorrelation problem at the 5% significance level, robust HC3 standard errors were used in model estimation to increase the reliability of the estimates, considering the proximity of the p-values to the marginal limit (0.10). The Wald χ² test (χ² = 88.446, p < 0.001) represents the overall significance of the model. The model’s R² value of 0.2303 indicates that it accounts for 23.03% of the total variation in sustainability disclosure levels.

In the static model (Model 1), FEMEM had a positive effect (β = 0.0009, p < 0.10) on the DSCORE. Similarly, BOMEET also positively affects the DSCORE (β = 0.0285, p < 0.01). Although the economic magnitude of these coefficients is small, their statistical significance, particularly the highly significant nature of BOMEET (p < 0.01) alongside the significant effect of FEMEM (p < 0.10), decisively confirms the validity and existence of these relationships. These empirical findings suggest that both gender diversity on the board of directors and active board meeting dynamics promote sensitivity to and transparency regarding sustainability issues. Furthermore, more frequent board meetings result in more effective monitoring and reporting of sustainability strategies, thus increasing the level of sustainability disclosure. The static model (Model 1) reveals a weakly significant negative association between the LEV and DSCORE (β = −0.1105, p < 0.10). This suggests that highly leveraged firms, which rely heavily on external debt, may prioritize financial obligations over resource allocation for sustainability reporting. Consequently, as debt ratios rise, the extent of disclosure tends to diminish, potentially indicating a preference for highlighting financial solvency rather than operational efficiency and sustainability outcomes. Conversely, SIZE exhibits a robust positive relationship with DSCORE (β = 0.0130, p < 0.01), confirming that larger enterprises leverage their broader asset base to maintain higher transparency. By contrast, variables such as BOSIZE, BOIN, SHMEET, LARSH, and ROA yielded no statistically significant impact, suggesting that they are not primary determinants of reporting behavior in this specific context.

In dynamic models, Sargan test p-values greater than 0.10 (Model 2 p = 0.2453; Model 3 p = 0.1867, Model 4 p = 0.3398) confirm the validity of the instrument variables. The AR(2) test results (Model 2 p = 0.2539, Model 3 p = 0.2581, Model 4 p = 0.2837) reveal no second-order autocorrelation in the models, supporting the reliability of the empirical findings. Although the coefficients of the lagged dependent variable added to the dynamic models were calculated as positive (Model 2: β = 0.208, Model 3: β = 0.212, Model 4: β = 0.203), the p-values remained above the significance level, leading to the conclusion that it had no significant effect on the sustainability disclosure level. This situation could be due to the rapid regulatory transition and structural changes in reporting mandates within the Türkiye business context during the analyzed period, particularly around 2021 and 2024. It is highly plausible that as companies were compelled to abruptly adapt to newly introduced, stringent sustainability reporting standards and institutional expectations, the historical continuity of their voluntary disclosure habits may have been systematically altered, thereby potentially weakening the reporting inertia observed in previous years. According to the regression analysis results in the dynamic models, BOSIZE, BOIN, SHMEET, LARSH, ROA, and LEV had no significant effect on the DSCORE. While FEMEM and BOMEET exhibited significant positive effects in the static framework (Model 1), they lost their statistical significance upon transitioning to the GMM specifications. This divergence suggests that once endogeneity and dynamic dependencies are controlled for, board gender diversity and meeting frequency may not exert a robust influence on sustainability disclosure levels.

FEMEM and BOMEET were found to have a significant and positive effect in the static model but remained below the significance level in the GMM models. Accordingly, it was concluded that the number of female members and the number of board meetings did not have an effect on the sustainability disclosure level in dynamic models. The SIZE variable maintained its relationship from the static model in the GMM models as well, retaining its significance at the 0.01 level and coefficient magnitude in all the GMM-2 (Model 2: β = 0.0168), GMM-3 (Model 3: β = 0.0168), and GMM-4 (Model 4: β = 0.0177) models. These findings reinforce the conclusion that corporate scale remains the primary structural driver of sustainability transparency, regardless of the econometric model employed.

6. Discussion

Table 8. Hypothesis acceptance/rejection decision table.

Hypothesis	Variable	Robust RE (Static)	GMM (Dynamic)	Decision
H1	BOSIZE	Insignificant	Insignificant	Reject
H2	BOIN	Insignificant	Insignificant	Reject
H3	FEMEM	Significant (+)	Insignificant	Reject
H4	BOMEET	Significant (+)	Insignificant	Reject
H5	LARSH	Insignificant	Insignificant	Reject
H6	SHMEET	Insignificant	Insignificant	Reject

provides a summary of the hypotheses tested in this study, indicating whether each was supported or rejected based on the empirical findings.

As summarized in Table 8, hypotheses H1, H2, H5, and H6 were rejected according to the static and dynamic model results, while H3 and H4 were also rejected according to the dynamic model results. These results support previous studies reporting no significant correlation between sustainability disclosures and board size, independence, concentrated ownership, or shareholder activism [30,31,35,36,37,54,57,62,66,68,69,86,87]. Additionally, while the leverage ratio significantly influenced sustainability disclosure levels in the static model, this effect became non-significant in the dynamic model. Regarding financial performance, ROA shows no discernible effect on sustainability disclosure levels in either the static or the dynamic models. These findings align with studies that identify no significant relationship between sustainability disclosure levels and either return on assets [32,35,59] or leverage [33,59].

According to the static model results, the significant and positive effect of FEMEM on the sustainability disclosure level supports theories that gender diversity increases social responsibility awareness and plays a motivating role in transparency regarding sustainability performance disclosure. Similarly, the significant and positive effect of BOMEET on the sustainability disclosure level in the static model supports the view that stakeholder expectations are considered in meetings, thus increasing the company’s sustainability disclosure level and consequently its legitimacy. However, the positive impact of variables such as the percentage of female board members and the frequency of board meetings lost its statistical significance in GMM models. In contrast, both the static and the dynamic model results obtained for the ultimate objective revealed that the SIZE variable—an indicator of financial performance and included in the study as a control variable—has a significant and positive effect on the sustainability disclosure level. This finding supports previous studies that company size has an effect on sustainability disclosures [29,34,35]. This indicates that the quality of meetings is more decisive than the number of board meetings and that diversity and other corporate governance mechanisms only increase transparency when supported by financial and structural strengths such as company size. Rather than a direct causal relationship between board mechanisms and the level of sustainability disclosure, it may indicate that companies adopting appropriate governance structures adopt a strategy of increasing their legitimacy by allocating sufficient resources due to their size and meeting stakeholder expectations. In other words, it can be said that corporate governance mechanisms are viewed as complementary elements of large-scale companies rather than as independent determinants of sustainability disclosures.

When connected back to the foundational frameworks of agency, legitimacy, and stakeholder theories, this empirical divergence yields profound conceptual insights. This indicates that the quality of meetings is more decisive than the number of board meetings and that diversity and other corporate governance mechanisms only increase transparency when supported by financial and structural strengths such as company size. Rather than a direct causal relationship between board mechanisms and the level of sustainability disclosure, it may indicate that, in the Türkiye corporate context, adopting diverse board configurations or frequent meetings functions primarily as a mechanism of Symbolic Legitimacy to temporarily appease external stakeholder pressures and comply with soft institutional codes. In other words, companies that adopt appropriate governance structures may be pursuing a strategy of increasing their legitimacy by allocating sufficient resources due to their size and taking stakeholder expectations into account. While these symbolic governance setups satisfy immediate institutional demands, the actual capacity to execute high-quality, substantive disclosure operations -and to bear the substantial agency costs associated with comprehensive transparency- resides purely within the domain of organizational scale and available financial slack resources (SIZE). Accordingly, it can be said that corporate governance mechanisms are viewed as complementary elements of large-scale companies rather than as independent determinants of sustainability disclosures.

In the dynamic panel, while it was expected that past disclosure levels (LAG_DSCORE) would influence future disclosure levels, this historical effect was not observed, indicating no statistically significant relationship between the lagged disclosure score and the current year’s disclosure score. This result suggests that institutional compliance with emerging reporting regulations directly drives changes in corporate reporting behavior. Alternatively, because a small sample size can constrain the statistical power of dynamic GMM estimators, the insignificance of the lagged dependent variable may simply reflect these mathematical limitations, potentially obscuring long-term historical dependencies.

7. Conclusions

This study comprehensively evaluates the empirical nexus between internal corporate governance configurations and sustainability disclosure trajectories within the unique institutional setting of an emerging market undergoing a mandatory regulatory transition. By deploying both static and dynamic panel estimators on data from the BIST XUSRD spanning 2020 to 2024, the overarching conclusion of this research demonstrates that traditional board characteristics often lose their robust explanatory power in long-term dynamic settings. Empirical results reveal that, while leverage ratios and board characteristics (such as female board representation and meeting frequency) exert only short-term, marginal effects on sustainability disclosure levels, firm size stands out as a structural determinant with a profound long-term impact in Türkiye. Consequently, contemporary sustainability transparency is primarily shaped by organizational scale and financial resource capacity rather than internal board dynamics.

The marginal effect of board characteristics on sustainability disclosures may stem primarily from the legal regulations that define board characteristics and guide company operations, or it may be due to companies complying with basic legal requirements and choosing not to adopt strategies and policies beyond those requirements. Publicly traded companies in Türkiye have been mandated to report under the ‘Sustainability Principles Compliance Framework’ since 2021, and subsequently under the mandatory TSRS 1 [17] and TSRS 2 [18] standards since 2024. Following the implementation of these legal requirements, a relative decrease has been observed in the number of sustainability disclosures based on comprehensive global frameworks such as GRI, SASB, or the International Integrated Reporting Framework. It has also been observed that some companies that published sustainability reports compliant with global standards before the mandatory implementation have regressed to basic compliance levels or have chosen not to publish voluntary sustainability reports at all.

7.1. Managerial and Policy Implications

The empirical findings of this study offer critical pathways for both regulators and corporate executives, particularly in emerging markets undergoing challenging regulatory transformations.

From a policymaker’s perspective, the dominant impact of firm size over board characteristics on sustainability disclosures suggests that publicly traded small and medium enterprises (SMEs) may face a structural disadvantage in their sustainability reporting. Regulatory bodies, such as the CMBT and the POA, could develop tailored incentive mechanisms or provide compliance subsidies for these smaller-scale businesses. Such support mechanisms are crucial for ensuring standardization in reporting. Currently, businesses in Türkiye are already providing sustainability disclosures as required by the CMBT and publishing TSRS-compliant sustainability reports mandated by the POA. As previously mentioned, the mandate for TSRS-compliant sustainability reporting by the Bureau of CMBT and the POA represents a milestone for Türkiye in aligning national reporting with global standardization benchmarks, demonstrating that Türkiye closely follows global trends. To facilitate a single, comprehensive sustainability report that blends this national standard with other global frameworks, regulatory bodies like the CMBT and the POA must act in tight coordination.

As established earlier, corporate reports serve as a direct output of the overarching corporate governance system. The quantity and quality of these reporting outputs inherently depend on the scope and quality of the operational activities executed by the company, and consequently, on managerial performance. In this context, policy-makers should redefine the statutory corporate governance architecture. In Türkiye, the management structures of corporations are legally anchored within regulations such as the Turkish Commercial Code and the Communiqué on Corporate Governance Principles, reflecting the strategic importance Türkiye places on institutional governance. However, it is vital to recognize that to truly achieve substantive sustainability, management structures must be built upon a foundation of integrated thinking. Accordingly, existing regulations could be revised to incorporate provisions requiring directors to possess explicit theoretical and technical competencies in sustainability issues.

From a managerial perspective, executives should view sustainability disclosures not as an operational cost burden but as a strategic investment that mitigates information asymmetry, demonstrates the extent to which stakeholder expectations are met, and ultimately solidifies corporate legitimacy. To yield a return on this investment, corporate operations must first be driven by an integrated thinking mindset. Therefore, it is essential to ensure that managers at all organizational echelons, particularly board members, possess robust sustainability competencies, or alternatively, to actively appoint individuals with such expertise to the board and its sub-committees. Ultimately, to ensure long-term sustainability and maintain a competitive advantage, managers must transcend mandatory compliance and continue to strategically adopt comprehensive, voluntary sustainability reporting standards.

7.2. Limitations and Future Research Directions

Despite its contributions, this study also has some limitations. In this study, the evaluation of sustainability disclosures relies on a scoring matrix calculated via content analysis. Although a rigorous validation protocol was executed to minimize evaluation errors (as detailed in Section 3.2), the inherently manual and retrospective nature of content analysis remains a baseline limitation. The final sample size (N = 41) can be considered another structural limitation of this research. Although the BIST Sustainability Index encompasses a broader universe of 98 companies, the exclusion of financial institutions due to their distinct regulatory environments naturally reduces the potential sample to 68 enterprises. This figure was further refined down to 41 companies following the exclusion of firms that failed to publish sustainability reports in any of the fiscal years during the analyzed period. While this rigorous balanced panel requirement ensures data continuity and consistency, it inherently restricts the overall sample scale.

Considering the study’s sample, limitations, and empirical outcomes together, several avenues for future research emerge. Future studies could analyze the impact of sustainability disclosure scores, calculated using a broader set of criteria, on alternative corporate governance mechanisms, focusing solely on post-TSRS compliant reports in the Türkiye context. Additionally, empirically analyzing the distinct, isolated impacts of environmental, economic, and social disclosure dimensions on corporate governance configurations would provide vital insights into the transparency of specific sustainability dimensions. At a broader level, by conducting comparative analyses across a sample of countries with similar socioeconomic conditions, these results can be evaluated at the macro level.

Crucially, future researchers are strongly encouraged to apply dynamic econometric modeling (such as GMM) within these cross-country designs to effectively mitigate potential endogeneity and path dependency issues when investigating governance mechanisms. Utilizing such dynamic specifications will help verify whether the long-term statistical irrelevance of corporate governance attributes observed in this study is an isolated phenomenon unique to Türkiye or a broader structural trend characteristic of emerging markets globally.